Handbook of Molecular Gastronomy

Scientific Foundations, Educational Practices,

and Culinary Applications
 Handbook of Molecular Gastronomy
Scientific Foundations, Educational Practices, and
Culinary Applications

Edited by
Róisín M. Burke, Alan L. Kelly, Christophe Lavelle
and Hervé This vo Kientza
First edition published 2021
by CRC Press
6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-​2742
and by CRC Press
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
© 2021 Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, LLC
The right of Róisín M. Burke, Alan L. Kelly, Christophe Lavelle and Hervé This vo
Kientza to be identified as the authors of the editorial material, and of the authors for their
individual chapters, has been asserted in accordance with sections 77 and 78 of the
Copyright, Designs and Patents Act 1988.
Reasonable efforts have been made to publish reliable data and information, but the
author and publisher cannot assume responsibility for the validity of all materials or the
consequences of their use. The authors and publishers have attempted to trace the copyright
holders of all material reproduced in this publication and apologize to copyright holders if
permission to publish in this form has not been obtained. If any copyright material has not
been acknowledged please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted,
reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other
means, now known or hereafter invented, including photocopying, microfilming, and
recording, or in any information storage or retrieval system, without written permission
from the publishers.
For permission to photocopy or use material electronically from this work, access www.
copyright.com or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood
Drive, Danvers, MA 01923, 978-​750-​8400. For works that are not available on CCC please
contact mpkbookspermissions@tandf.co.uk
Trademark notice: Product or corporate names may be trademarks or registered trademarks
and are used only for identification and explanation without intent to infringe.
Library of Congress Cataloging‑in‑Publication Data
Names: Burke, Róisín, editor.
Title: Handbook of molecular gastronomy: scientific foundations and culinary
applications / edited by Róisín Burke, Alan Kelly, Christophe Lavelle,
and Hervé This vo Kientza.
Description: First edition. | Boca Raton: CRC Press, 2021. |
Includes bibliographical references and index.
Identifiers: LCCN 2020053893 (print) | LCCN 2020053894 (ebook) |
ISBN 9781466594784 (hardback) | ISBN 9780429168703 (ebook)
Subjects: LCSH: Molecular gastronomy–Handbooks, manuals, etc.
Classification: LCC TX651 .H335 2021 (print) |
LCC TX651 (ebook) | DDC 641.01/3--dc23
LC record available at https://lccn.loc.gov/2020053893
LC ebook record available at https://lccn.loc.gov/2020053894
ISBN: 978-​1-​4665-​9478-​4 (hbk)
ISBN: 978-0-367-74161-7 (pbk)
ISBN: 978-​0-​429-​16870-​3 (ebk)
Typeset in Times
by Newgen Publishing UK
Contents

About the Editors.................................................................................................................................................................................. xiii

Contributors............................................................................................................................................................................................xv
Foreword...............................................................................................................................................................................................xxv

Introduction to Molecular Gastronomy and Its Applications.............................................................................................................1

Part I Scientific Foundations...................................................................................................................5

Acids in Foods and Perception of Sourness..........................................................................................................................................7
Christian Salles

Anthocyanins in Food...........................................................................................................................................................................13
Véronique Cheynier

Alcoholic Beverages: Production, Trends, Innovations.....................................................................................................................19

Konstantin Bellut, Kieran M. Lynch and Elke K. Arendt

Ash in the Kitchen.................................................................................................................................................................................25

Marta Ghebremedhin, Christine Schreiber, Bhagyashri L. Joshi, Andreas Rieger and Thomas A. Vilgis

Baking: Laminated Bakery Products..................................................................................................................................................35

Roxane Detry, Christophe Blecker and Sabine Danthine

Baking: Chemical Leaveners...............................................................................................................................................................41

Linda A. Luck

Baking: Injera –​the Multi-​Eyed Flat Bread......................................................................................................................................43

Mahelet Girma, Sumaya M. Abdullahi and Benjamin L. Stottrup

Baking: Viennoiserie –​Laminated Pastry Production......................................................................................................................47

James A. Griffin

Baking: How Does Starch Gelatinization Influence Texture?...........................................................................................................53

Anaïs Lavoisier

Baking: Sourdough Bread....................................................................................................................................................................57

Mark Traynor and Imran Ahmad

Barbecue: The Chemistry behind Cooking on a Barbecue...............................................................................................................63

Florent Allais

Bioactivity and Its Measurement.........................................................................................................................................................71

Hervé This vo Kientza

Browning: The Glycation and Maillard Reactions –​Major Non-​Enzymatic Browning Reactions in Food................................81
Frédéric J. Tessier

Canning: Appert and Food Canning...................................................................................................................................................87

Jean-​Christophe Augustin

Capillarity in Action.............................................................................................................................................................................91
Hervé This vo Kientza

v
vi Contents

Champagne Tasting from a Scientific Perspective.............................................................................................................................97

Gérard Liger-​Belair, Clara Cilindre, Daniel Cordier, Guillaume Polidori, Fabien Beaumont and Thomas Séon

Chantillys: The Cousins of Whipped Cream....................................................................................................................................105

Hervé This vo Kientza

Cheese: Hot Culinary Uses of Cheese...............................................................................................................................................107

Sébastien Roustel and John A. Hannon

Chocolate: Chocolates from around the World, Simple Physics, Complex Flavour....................................................................121
Bhagyashri L. Joshi, Sarah Gindra and Thomas A. Vilgis

Chocolate: Oral Processing of Chocolate – Successive Interplay of Sensory and Physicochemical Parameters.......................131
Thomas A. Vilgis

Coffee Preparation –​from Roasted Beans to Beverage..................................................................................................................139

Laura Febvay and Hervé This vo Kientza

Colour: Natural Pigments in Foods and Their Technical Uses.......................................................................................................151

Juan Valverde

Cooking................................................................................................................................................................................................157
Hervé This vo Kientza

Cooking: Culinary Precisions and Robustness of Recipes..............................................................................................................163

Hervé This vo Kientza

Cryogenics in the Kitchen..................................................................................................................................................................171

Peter Barham

Dairy: Milk Gels –​a Gastrophysics View.........................................................................................................................................181

Judith Hege, Marta Ghebremedhin, Bhagyashri L. Joshi, Christine Schreiber, H.-​C. Gill and Thomas A. Vilgis

Dairy: Culinary Uses of Milk, Butter and Ice Cream.....................................................................................................................191

Alan L. Kelly and David S. Waldron

Dairy: Ginger Milk Curd...................................................................................................................................................................199

Martin Lersch

Dehydration.........................................................................................................................................................................................203
José M. Aguilera

Dispersed System Formalism.............................................................................................................................................................207

Hervé This vo Kientza

Distillation: The Behaviour of Volatile Compounds during Distillation of Hydro-​Alcoholic Solutions and
during Hydro-​Distillation...................................................................................................................................................................213
Martine Esteban-​Decloux

Eggs: Let Us Have an Egg..................................................................................................................................................................221

Hervé This vo Kientza

Emulsions: Emulsified Systems in Food...........................................................................................................................................227

Markus Ketomäki, Trivikram Nallamilli, Christine Schreiber and Thomas A. Vilgis

Emulsions and Foams: Ostwald Ripening and Disproportionation in Practice...........................................................................241

Hervé This vo Kientza

Emulsions: Lecithin............................................................................................................................................................................249
Elzbieta Kozakiewicz and Daniel Cossuta
Contents vii

Emulsions: Emulsions and Surfactants in the Kitchen...................................................................................................................257

Hervé This vo Kientza

Essential Oils.......................................................................................................................................................................................265
Eric Angelini and Laure Dziuba

Essential Oils: How to Safely Use Essential Oils..............................................................................................................................275

Eric Angelini and Laure Dziuba

Evaporation.........................................................................................................................................................................................281
Hervé This vo Kientza

Expansion............................................................................................................................................................................................291
Hervé This vo Kientza

Fats and Oils: Physicochemical Properties of Edible Oils and Fats...............................................................................................295

Sabine Danthine

Fats and Oils: From Fat Droplets in Plant Seeds to Novel Foods...................................................................................................299
Juan C. Zambrano, Behic Mert and Thomas A. Vilgis

Fats and Oils: Oxidation of Dietary Lipids......................................................................................................................................305

Luc Eveleigh

Fats and Oils: Extra Virgin Olive Oil in Cooking –​Molecular Keys for Traditional and
Modern Mediterranean Gastronomy................................................................................................................................................311
Raffaele Sacchi

Fermentation: Kimchi........................................................................................................................................................................321
Weon-​Sun Shin

Fermentation: Fermenting Flavours with Yeast...............................................................................................................................327

Angela M. Coral Medina and John P. Morrissey

Fermentation: A Short Scientific and Culinary Overview of Kefir................................................................................................331

Christophe Lavelle and Jean-​Baptiste Boulé

Filtration Membranes for Food Processing and Fractionation......................................................................................................335

Marie-​Laure Lameloise

Food Matrices and the Matrix Effect in the Kitchen.......................................................................................................................343

José M. Aguilera and Hervé This vo Kientza

Food Pairing: Is It Really about Science?.........................................................................................................................................347

Hervé This vo Kientza and Christophe Lavelle

Freeze-​Drying......................................................................................................................................................................................349
Yrjö H. Roos

Foams: Pickering Edible Oil Foam –​Toward New Food Products................................................................................................357

Anne-​Laure Fameau

Frying...................................................................................................................................................................................................365
Franco Pedreschi

Gastrophysics: A New Scientific Approach to Eating......................................................................................................................371

Charles Spence

Gels.......................................................................................................................................................................................................375
Hervé This vo Kientza
viii Contents

Heat Transfer in Culinary Sciences...................................................................................................................................................381

Denis Flick

Hydrocolloid Usages as Gelling and Emulsifying Agents for Culinary and Industrial Applications..........................................385
Rachel Edwards-​Stuart and Reine Barbar

Imaging Foodstuffs and Products of Culinary Transformations....................................................................................................409

Mathias Porsmose Clausen, Morten Christensen and Ole G. Mouritsen

Meat: Meat Tenderness and the Impact of Cooking........................................................................................................................415

Jean-​François Hocquette and Alain Kondjoyan

Meat: Heat Transfer in Meat.............................................................................................................................................................419

Douglas Baldwin

Meat: Reduction of Nitrate and Nitrite Salts in Meat Products –​What Are the Consequences
and Possible Solutions?.......................................................................................................................................................................423
Régine Talon and Sabine Leroy

Microwave Heating and Modern Cuisine.........................................................................................................................................429

Alan L. Kelly and Hervé This vo Kientza

Mineral Ions and Cooking.................................................................................................................................................................433

Christian Salles

Osmosis in the Kitchen.......................................................................................................................................................................441

Hervé This vo Kientza

Pasta: Durum Wheat Proteins –​a Key Macronutrient for Pasta Qualities..................................................................................447
Coline Martin, Marie-​Hélène Morel and Bernard Cuq

Pasteurization in the Kitchen.............................................................................................................................................................451

Gabriela Precup and Dan-​Cristian Vodnar

Plating: The Science of Plating..........................................................................................................................................................459

Charles Spence

Proteins and Proteases........................................................................................................................................................................463

Linda A. Luck and Alan L. Kelly

Puddings: The Secret of the Rice Pudding.......................................................................................................................................471

Martin Lersch

Roasting...............................................................................................................................................................................................473
Laura Febvay and Hervé This vo Kientza

Salt: When Should Salt Be Added to Meat Being Grilled?.............................................................................................................491

Hervé This vo Kientza, Marie-​Paule Pardo and Rolande Ollitrault

Sauces...................................................................................................................................................................................................495
Hervé This vo Kientza

Sauces: Hollandaise Sauce.................................................................................................................................................................499

Guro Helgesdotter Rognså

Sauces and Purées: The Underside of Applesauce...........................................................................................................................505

Cassandre Leverrier

Seaweeds: Phycogastronomy –​the Culinary Science of Seaweeds.................................................................................................517

Ole G. Mouritsen
Contents ix

Size Reduction.....................................................................................................................................................................................523
José M. Aguilera

Smoked Foods......................................................................................................................................................................................527
Jane K. Parker and Alice Pontin

Sous Vide Cooking..............................................................................................................................................................................531

Douglas Baldwin

Spherification......................................................................................................................................................................................537
Linda A. Luck

Squid: Gastrophysics of Squid –​from Gastronomy to Science and Back Again..........................................................................541

Ole G. Mouritsen, Charlotte Vinther Schmidt, Peter Lionet Faxholm and Mathias Porsmose Clausen

Sugars: Soft Caramel and Sucre à la Crème –​an Undergraduate Experiment about Sugar Crystallization...........................545
Irem Altan, Patrick Charbonneau and Justine de Valicourt

Sugars: Sugar (and Its Substitutes) in Pastries................................................................................................................................549

Anne Cazor and Ramon Morató

Sugars: Erythritol–​Sucrose Mixtures out of Equilibrium –​Exciting Thermodynamics in the Mouth......................................557

Hannah M. Hartge, Birgitta I. Zielbauer and Thomas A. Vilgis

Sugars: Intramolecular Dehydration of Hexoses.............................................................................................................................563

Marie-​Charlotte Belhomme, Stéphanie Castex and Arnaud Haudrechy

Taste and Sound..................................................................................................................................................................................569

Bruno A. Mesz

Temporal Domination of Sensation: When Building Dishes, Let’s Take Temporality into Account...........................................575
Pascal Schlich

Texture: The Physics of Mouthfeel –​Spreadable Food and Inulin Particle Gels.........................................................................581
Thomas A. Vilgis

Texture: How Texture Makes Flavour..............................................................................................................................................585

Ole G. Mouritsen

Texture: Tsukemono –​the Art and Science of Preparing Crunchy Vegetables............................................................................593

Ole G. Mouritsen

Thickeners: Cellulose and Its Derivatives........................................................................................................................................599

Rachel Edwards-​Stuart

3D Printing of Food............................................................................................................................................................................605
Megan M. Ross, Róisín M. Burke and Alan L. Kelly

Umami: The Molecular Science of Umami Synergy........................................................................................................................619

Ole G. Mouritsen

Part II Educational Practices.............................................................................................................625

The Right Words for Improving Communication in Food Science, Food Technology, and between Food
Science and Technology and a Broader Audience............................................................................................................................627
Hervé This vo Kientza

Experimental Flavour Workshops.....................................................................................................................................................635

Hervé This vo Kientza
x Contents

Teaching Argumentation and Inquiry through Culinary Claims...................................................................................................643

Erik Fooladi

Cooking and Science Workshops: The “Soft of the World”, Gelling Agents.................................................................................651
Pere Castells

Culinary Sciences for the Enhancement of the Public Understanding of Science........................................................................655

Ole G. Mouritsen

“Science and Cooking Activities” for Secondary School Students.................................................................................................659

Marie-​Claude Feore, Laure Fort, Marie-​Blanche Mauhourat and Hervé This vo Kientza

How to Reduce Oil in French Fries: A Student Experiment...........................................................................................................663

Hervé This vo Kientza

An Educational Satellite Project around the Scientific Elucidation of Culinary Precisions in Lebanon and
in the Middle East...............................................................................................................................................................................665
Reine Barbar, Jean-​Marie Malbec, Christophe Lavelle and Hervé This vo Kientza

Bon Appétit, Marie Curie! A Stanford University Introductory Science of Cooking Course.....................................................673
Markus W. Covert and Imanol Arrieta-​Ibarra

Molecular Gastronomy in Science Education and Science Communication at the National University of Singapore.............679
Linda Sellou and Lau Shi Yun

Molecular Gastronomy: A Universal Portal to the Molecular Sciences........................................................................................683

Patricia B. O’Hara

Heat Transfer in the Kitchen –​Exercises.........................................................................................................................................687

Manuel Combes

Ionic Diffusion in Spherified Calcium Alginate Gels: A Laboratory Experiment........................................................................689

Lorenzo Soprani, Lara Querciagrossa, Silvia Cristofaro, Luca Muccioli, Silvia Orlandi, Elena Strocchi,
Alberto Arcioni and Roberto Berardi

Simple Calculations Based on Cooking............................................................................................................................................703

Hervé This vo Kientza

Teaching and Cooking with Culinary Teachers...............................................................................................................................717

Christophe Lavelle

The Monthly INRAE-​AgroParisTech Seminars on Molecular Gastronomy................................................................................721

Hervé This vo Kientza

Part III Culinary Applications...........................................................................................................725

New Greek Cuisine.............................................................................................................................................................................727
Georgianna Hiliadaki and Nikos Roussos

3D Printed Note by Note Recipe: Soya Lobster Prototype.............................................................................................................735

Róisín M. Burke

Cooking (with) Olive Oil....................................................................................................................................................................737

Christophe Lavelle

Cooking for the Elderly......................................................................................................................................................................741

Christophe Lavelle
Contents xi

Culinary Constructivism and Note by Note Cooking......................................................................................................................743

Pierre Gagnaire

Decantation..........................................................................................................................................................................................751
Hervé This vo Kientza

Note by Note Recipes for a Press Conference and Tasting Organized at ITHQ, 2012.................................................................755
Erik Ayala-​Bribiesca and Ismael Osorio

Using Liquid Nitrogen to Prepare Ice Creams in the Restaurant..................................................................................................759

Christophe Lavelle and Hervé This vo Kientza with chefs André Daguin, Noël Gutrin and Philippe Labbé

A Note by Note Traditional Chinese Dinner Created and Served in Singapore...........................................................................763

Kelly Lee, Aaron Goh, Tony Choo, Nicolas Vergnole, Gn Ying Wei and Tais Berenstein

Greek Diracs........................................................................................................................................................................................771
Makis Kalossakas and Nicolas Nikolakopoulos

An Eclipse Dish...................................................................................................................................................................................775
Hervé This vo Kientza

Modern Swiss Cooking.......................................................................................................................................................................777

Denis Martin

How Do Eggs Coagulate?...................................................................................................................................................................779

Hervé This vo Kientza

Vegetable Salad...................................................................................................................................................................................785
Jean Chauvel

Filtration..............................................................................................................................................................................................789
Hervé This vo Kientza

Waiter! There Is Garlic in My Meringue!........................................................................................................................................793

César Vega

Lobster and Juniper...........................................................................................................................................................................797

David Toutain

Molecular Cooking..............................................................................................................................................................................801
Róisín M. Burke and Pauline Danaher

Note by Note Cooking and Note by Note Cuisine............................................................................................................................809

Hervé This vo Kientza and Róisín M. Burke

Spherification......................................................................................................................................................................................819
Sasa Hasic

The Raspberry Pear Viennoiserie......................................................................................................................................................825

James A. Griffin

Molecular Mixology: Welcome Coffee, a Cocktail with Ten Layers..............................................................................................827

Hervé This vo Kientza and Pierre Gagnaire

Cube of “Chicken-​Carrot” with Chips of “Basil-​Lemon”..............................................................................................................829

Pasquale Altomonte and Dao Nguyen

Some of the Easiest Note by Note Recipes Served at Senses Restaurant.......................................................................................831

Andrea Camastra
xii Contents

The Forest Floor..................................................................................................................................................................................837

Sophie Dalton

A Note by Note Macaron....................................................................................................................................................................841

Julien Binz

Note by Note Cooking.........................................................................................................................................................................843

Michael Pontif

Note by Note Sushis............................................................................................................................................................................847

Guillaume Siegler

Slowly Cooked Lamb Neck with Fermented Flour Pancakes, Sunchoke Puree and Beer Glaze................................................849
Alex Tsionitis

Index.....................................................................................................................................................................................................851
About the Editors
Róisín M. Burke is a senior lecturer in the School of Culinary
Arts and Food Technology, College of Arts and Tourism,
Technological University Dublin, City Campus, Ireland. She
obtained her PhD from University College Dublin and subse-
quently carried out postdoctoral research at the Agricultural
University in Wageningen, The Netherlands. Róisín lectures and
conducts research, specializing in Culinary Science and Food
Product Development. In the last 14 years, she has developed
Molecular Gastronomy as a subject discipline in The School
of Culinary Arts and Food Technology, TU Dublin. She has
published widely in international peer-​reviewed journals and has
joined a number of editorial teams. For many years, Róisín has
lectured to international students, and she is the TU Dublin co-​
ordinator of the Erasmus+ MSc programme in Food Innovation
and Product Design (FIPDes). She has given guest lectures in
Ireland and abroad.
Alan L. Kelly is a professor in the School of Food and Nutritional
Sciences at University College Cork in Ireland. His teaching
interests include food processing and preservation, dairy
product technology and new food product development, as
well as regularly giving courses on effective scientific commu-
nication. He leads an active research group on the chemistry
and processing of milk and dairy products, has published over
250 research papers, review articles and book chapters, and
has supervised over 40 MSc and PhD students to completion.
He has been an editor of the International Dairy Journal since
2005 and has acted as an external examiner in universities and
reviewed for journals and funding agencies around the world.
In July 2009, he received the Danisco International Dairy
Science award from the American Dairy Science Association
for his contributions to research in dairy science and tech-
nology. In recent years, he has become very interested in the
interface between the worlds of food and culinary sciences, and
has organized several workshops and seminars on this topic and
molecular gastronomy. In 2019, he published a book entitled
Molecules, Microbes and Meals: The Surprising Science of
Food (Oxford University Press), and in 2020, he published
How Scientists Communicate: Dispatches from the Frontiers of
Knowledge (Oxford University Press), both of which are aimed
at a general audience.
Christophe Lavelle is a research scientist at the French National
Centre for Scientific Research, working at the National Museum
of Natural History and Sorbonne University in Paris. He is an
expert in biophysics, epigenetics and food science and teaches in
many universities and professional schools (including Sorbonne
University, Le Cordon Bleu and Basque Culinary Center). He is
frequently invited to conferences for general public or profes-
sional audiences, and is also responsible for the scientific training
of cookery teachers at the national level. He is the author of more
than 60 research papers and 12 books on food, including Toute
la chimie qu’il faut savoir pour devenir un chef! (Flammarion,
2017) and Je mange donc je suis. Petit dictionnaire curieux de
l’alimentation (Museum National d’Histoire Naturelle, 2019).
He is a member of several scientific and food societies (including
the French Biophysical Society, the American Biophysical
Society, the Disciples d’Escoffier Society and the Association for
the Study of Food and Society).
Hervé This vo Kientza is a physical chemist, one of the two
creators of Molecular and Physical Gastronomy and a promoter
of “molecular cooking” (he devised the name in 1999). He is
currently working for the French National Research Institute
for Agriculture, Food and the Environment, is a professor at
AgroParisTech, and is the director of the AgroParisTech-​INRAE
International Centre for Molecular and Physical Gastronomy. As
such, he organizes the International Workshops on Molecular
and Physical Gastronomy. He is also the head of the Educational
Committee of the Institute for Advanced Studies in Gastronomy
and a member of several academies: the French Academy of
Agriculture, the Belgian Academy of Sciences, the Academy of
Alsace and the Academy Stanislas. He is doing scientific research
(molecular and physical gastronomy) in the Inrae-​AgroParisTech
Group of Molecular Gastronomy. At the same time, he is lec-
turing extensively worldwide for the promotion of molecular
gastronomy, stimulating the creation of groups for research and
education in universities and research centres, but also of “note
by note cooking”, an application of molecular gastronomy that
he proposed as early as 1994. He has published many books and
chapters of books, made TV programmes and radio shows, and
has several blogs, but he is also the author of more than 150 sci-
entific articles.
xiii
Contributors

Sumaya M. Abdullahi Alberto Arcioni

Augsburg University, Dipartimento di Chimica Industriale “Toso Montanari”,
Department of Physics, Università di Bologna,
2211 Riverside Ave, Viale del Risorgimento 4,
Minneapolis, MN 55454, Bologna I-​40136,
United States Italy

José M. Aguilera Elke K. Arendt

Department of Chemical Engineering and Bioprocesses, School of Food and Nutritional Sciences and APC Microbiome
Pontificia Universidad Católica de Chile, Ireland,
Avenida Libertador Bernardo O’Higgins 340, University College Cork,
Santiago, T12 YN60,
Chile Ireland

Imran Ahmad Imanol Arrieta-​Ibarra

Chaplin School of Hospitality and Tourism Management, Department of Management Science and Engineering,
Florida International University, Stanford University,
North Miami, Stanford, CA,
Florida, United States of America
United States of America
Jean-​Christophe Augustin
Florent Allais Ecole Nationale Vétérinaire d’Alfort,
URD Agro-​Biotechnologies Industrielles, 7 avenue du Général de Gaulle,
CEBB, F-​94704 Maisons-​Alfort,
AgroParisTech, France
51110, Pomacle,
France Erik Ayala-​Bribiesca
Food Science and Technology Consultant and
Irem Altan Professor at Cégep de St-​Hyacinthe,
Department of Chemistry, Quebec,
Duke University, Canada
Durham,
North Carolina, Douglas Baldwin
United States Breville Pty Ltd 19400,
South Western Ave,
Pasquale Altomonte Torrance,
The Kitchen Lab, California 90501,
23 chemin de la Tour, United States
Meyrin-​Village,
Geneva, Reine Barbar
Switzerland UMR IATE,
Univ. Montpellier, INRAE, Institut Agro,
Eric Angelini 2 Place Pierre Viala,
V. MANE FILS, F-​34060 Montpellier Cedex,
Route de Grasse, France
06620, and
Le Bar-​sur-​Loup, Department of Agriculture and Food Engineering,
France School of Engineering,
Holy Spirit University of Kaslik,
P.O. Box 446 Jounieh,
Mount Lebanon,
Lebanon
xv
xvi List of Contributors

Peter Barham Róisín M. Burke

H. H. Wills Physics Laboratory, School of Culinary Arts and Food Technology,
University of Bristol, College of Arts and Tourism,
Bristol, Technological University Dublin,
BS8 1TL, City Campus,
United Kingdom Dublin 1,
Ireland
Fabien Beaumont
Laboratoire de Thermomécanique (GRESPI), Andrea Camastra
Université de Reims Champagne-​Ardenne, Restaurant Senses,
Reims 51100, Bielanska 12,
France Warsaw 00-​085,
Poland
Marie-​Charlotte Belhomme
Institut de Chimie Moléculaire de Reims, Pere Castells
UMR CNRS 7312, Science and Cooking World Congress,
SFR Condorcet FR CNRS 3417, Av. de la Torre Blanca 57,
Université de Reims, Sant Cugat des Vallès,
BP 1039, Barcelona,
F-​51687 REIMS Cedex, Spain
France
Stéphanie Castex
Konstantin Bellut Institut de Chimie Moléculaire de Reims,
School of Food and Nutritional Sciences, UMR CNRS 7312,
University College, SFR Condorcet FR CNRS 3417,
Cork, Université de Reims,
T12 YN60, BP 1039,
Ireland F-​51687 REIMS Cedex,
France
Roberto Berardi (deceased, 2020)
Dipartimento di Chimica Industriale “Toso Montanari”, Anne Cazor
Università di Bologna 28 Tain Seng A,
Scinnoy,
Tais Berenstein 6 Rue d’Estienne,
Chef at the At-​Sunrice GlobalChef Academy D’Orves,
92110 Clichy,
Julien Binz France
Restaurant Julien Binz,
Ammerschwihr, Patrick Charbonneau
Alsace, Departments of Chemistry and Physics,
France Duke University,
Durham,
Christophe Blecker North Carolina,
Food Science and Formulation, United States
Gembloux Agro-​Bio Tech,
University of Liège, Jean Chauvel
Avenue de la faculté d’Agronomie 2B, Restaurant Jean Chauvel,
5030 Gembloux, 33 avenue du Général Leclerc,
Belgium Boulogne-​Billancourt,
France
Jean-​Baptiste Boulé
Genome Structure and Instability, Véronique Cheynier
CNRS UMR7196 /​INSERM U1154, SPO, INRAE,
National Museum of Natural History, Université de Montpellier,
Sorbonne University, Institut Agro-Montpellier SupAgro,
Paris 75005, 2 Place Viala,
France 34060 Montpellier,
France
List of Contributors xvii

Tony Choo Daniel Cossuta

Chef at the At-​Sunrice GlobalChef Academy, Bunge Zrt.,
28 Tai Seng Street, Katalin Kővári Innovation Centre,
Level 5, Illatos út 38. DÜP II. Building G/​3rd floor,
Singapore 534106, 1097 Budapest,
Singapore Hungary

Morten Christensen Markus W. Covert

Taste for Life, Department of Bioengineering,
University College Lillebælt, c/​o Kold College, Stanford University,
55 Landbrugsvej, Stanford CA,
DK-​5260 Odense S, United States
Denmark
Silvia Cristofaro
Clara Cilindre Dipartimento di Chimica Industriale “Toso Montanari”,
Equipe Effervescence, Università di Bologna,
Champagne et Applications (GSMA), Viale del Risorgimento 4,
UMR CNRS 7331, Bologna I-​40136,
Université de Reims Champagne-​Ardenne, Italy
Reims,
France Bernard Cuq
L’Institut Agro – Montpellier SupAgro,
Mathias Porsmose Clausen UMR IATE,
Department of Green Technology, 2 place Viala 34060 Montpellier Cedex,
SDU Biotechnology, France
University of Southern Denmark,
55 Campusvej, André Daguin (deceased, 2019)
DK-​5230 Odense M, Chef
Denmark
Sophie Dalton
Manuel Combes School of Culinary Arts and Food Technology,
Teacher of Physics in Prep Class, College of Arts and Tourism,
La Pérouse-Kerichen, Technological University Dublin,
Brest, City Campus,
France Ireland
and
Research Associate in Laboratoire Géosciences Océan Pauline Danaher
(UMR 6538), School of Culinary Arts and Food Technology,
Brest, City Campus,
France Technological University Dublin,
Ireland
Angela M. Coral Medina
School of Microbiology, Sabine Danthine
University College Cork, Food Science and Formulation,
T12 YN60, Gembloux Agro-​Bio Tech,
Ireland University of Liège,
Avenue de la faculté d’Agronomie 2B,
Daniel Cordier 5030 Gembloux,
Equipe Effervescence, Belgium
Champagne et Applications (GSMA),
UMR CNRS 7331, Roxane Detry
Université de Reims Champagne-​Ardenne, Food Science and Formulation,
Reims, Gembloux Agro-​Bio Tech,
France University of Liège,
Avenue de la faculté d’Agronomie 2B,
5030 Gembloux,
Belgium
xviii List of Contributors

Justine de Valicourt Laura Febvay

Vestjyllands højskole, Aerial,
Ringkøbing, 250 Rue Laurent Fries,
Denmark Parc d’innovation,
and 67412 Illkirch,
Les Colibris –​Entreprise permacole, France
St-​Jean-​de-​Matha,
Québec, Marie-​Claude Feore
Canada Group of Molecular Gastronomy,
INRAE-​AgroParisTech,
Laure Dziuba International Centre for Molecular Gastronomy,
Consultant chromatographic analysis of fragrances and F-​75005 Paris,
related compounds, France
1955 chemin des vergers, and
06620 Le Bar-​sur-​Loup, UMR 0782 SayFood,
France AgroParisTech (INRAE),
Université Paris-​Saclay,
Rachel Edwards-​Stuart F-​91300 Massy,
Culinary Science Department, France
Westminster Kingsway College,
76 Vincent Square, Denis Flick
London SW1P 2PD, AgroParisTech,
United Kingdom 16 rue Claude Bernard,
Paris,
Martine Esteban-​Decloux France
Unité Mixte de Recherche Ingénierie Procédés Aliments,
AgroParisTech, Erik Fooladi
INRAE, Department of Science and Mathematics,
Université Paris-​Saclay, Faculty of Humanities and Education,
F-​91300 Massy, Volda University College,
France P.O. Box 500,
Volda,
Luc Eveleigh Norway
Sayfood (UMR 0782),
INRAE, Laure Fort
AgroParisTech, Group of Molecular Gastronomy,
Université Paris-​Saclay, INRAE-​AgroParisTech,
91300, Massy, International Centre for Molecular Gastronomy,
France F-​75005 Paris
and
Anne-​Laure Fameau UMR 0782 SayFood,
Research & Innovation, International Physical-​Chemistry AgroParisTech (INRAE),
Department, Université Paris-​Saclay,
L’Oréal, F-​91300 Massy,
Saint-​Ouen 93400, France
France
Pierre Gagnaire
Peter Lionet Faxholm Restaurants Pierre Gagnaire: Bordeaux, Chatelaillon,
Department of Food Science, Courchevel, Danang, Dubai, Hong Kong, Las Vegas, London,
Taste for Life, Design and Consumer Behavior, Nîmes, Paris, Shanghai, Seoul, Tokyo
University of Copenhagen,
26 Rolighedsvej, Marta Ghebremedhin
DK-​1958 Frederiksberg C, Max Planck Institute for Polymer Research,
Denmark Ackermannweg 10,
55128 Mainz,
Germany
List of Contributors xix

H.-​C. Gill Judith Hege

Max Planck Institute for Polymer Research, Max Planck Institute for Polymer Research,
Ackermannweg 10, Ackermannweg 10,
55128 Mainz, 55128 Mainz,
Germany Germany

Sarah Gindra Georgianna Hiliadaki

Max Planck Institute for Polymer Research, Funky Gourmet Restaurant
Ackermannweg 10, Athens
55128 Mainz, Greece
Germany
Jean-​François Hocquette
Mahelet Girma Clermont University,
Augsburg University, INRAE,
Department of Physics, VetAgro Sup,
2211 Riverside Ave, UMR 1213 Herbivores,
Minneapolis, MN 55454, Theix,
United States 63122 Saint-​Genès Champanelle,
France
Aaron Goh
Chef at the At-​Sunrice GlobalChef Academy, Bhagyashri L. Joshi
28 Tai Seng Street, Max Planck Institute for Polymer Research,
Level 5, Ackermannweg 10,
Singapore 534106, 55128 Mainz,
Singapore Germany

James A. Griffin Makis Kalossakas

School of Culinary Arts and Food Technology, Chef teacher at the Institut Le Monde,
College of Arts and Tourism, 45 Thessalonikis Str,
Technological University Dublin, Moschato, Athens 18346,
City Campus, Greece
Dublin 1,
Ireland Alan L. Kelly
School of Food and Nutritional Sciences,
Noël Gutrin
University College,
Retired; former Head Food & Beverage at Le Futuroscope,
Cork,
Chasseneuil-​du-​Poitou,
T12 YN60,
France
Ireland
John A. Hannon (deceased, 2021)
Markus Ketomäki
Glanbia Ireland DAC
Max Planck Institute for Polymer Research,
Ackermannweg 10,
Hannah M. Hartge
55128 Mainz,
Max Planck Institute for Polymer Research,
Germany
Ackermannweg 10,
55128 Mainz,
Alain Kondjoyan
Germany
INRAE,
Sasa Hasic Unité Qualité des Produits Animaux,
R&D Chef, Theix,
Croatia 63122 Saint-​Genès Champanelle,
France
Arnaud Haudrechy
Institut de Chimie Moléculaire de Reims, Elzbieta Kozakiewicz
UMR CNRS 7312, Market Innovation and Technology Engineer, Koninklijke
SFR Condorcet FR CNRS 3417, Bunge B.V.,
Université de Reims, 221 rue de la loi,
BP 1039, 1040 Bruxelles,
F-​51687 REIMS Cedex, Belgium
France
xx List of Contributors

Philippe Labbé Gérard Liger-​Belair

Former chef of La Tour d’Argent, Equipe Effervescence,
Paris, Champagne et Applications (GSMA),
France UMR CNRS 7331,
Université de Reims Champagne-​Ardenne,
Marie-​Laure Lameloise Reims,
UMR SayFood, France
AgroParisTech,
INRAE, Linda A. Luck
Université Paris-​Saclay, Professor of Chemistry-​Emeritus,
Massy 91300, State University of New York at Plattsburgh,
France Plattsburgh,
12901 New York,
Christophe Lavelle United States
National Museum of Natural History /​Sorbonne University,
CNRS UMR7196 /​INSERM U1154, Kieran M. Lynch
43 rue Cuvier, Paris 75005, School of Food and Nutritional Sciences,
France University College,
and Cork,
Institut National Supérieur du Professorat et de l’Education T12 YN60,
(INSPE), Ireland
Toulouse University and Cergy-​Pontoise University,
France Jean-​Marie Malbec
Conseiller pédagogique sur la zone MOPI,
Anaïs Lavoisier Lycée Bonaparte,
UMR SayFood (Paris-Saclay Food and Bioproduct Engineering Doha,
Research Unit), Qatar
INRAE,
AgroParisTech, Coline Martin
Université Paris-Saclay, INRAE,
F-91300 Massy, UMR IATE,
France 2 place Viala,
34060 Montpellier cedex,
Kelly Lee France
Chef at the At-​Sunrice GlobalChef Academy,
28 Tai Seng Street, Denis Martin
Level 5, Restaurant Denis Martin,
Singapore 534106, Rue du Château 2,
Singapore 1800 Vevey,
Switzerland
Sabine Leroy
Université Clermont Auvergne, Ramon Morató
INRAE, Creative Director for Cacao Barry brand
MEDIS,
Clermont-​Ferrand, Marie-​Blanche Mauhourat
France Education nationale,
Inspection générale,
Martin Lersch 110 rue de Grenelle,
https://khymos.org Paris 75007,
France
Cassandre Leverrier
UMR SayFood, Behic Mert
AgroParisTech, Max Planck Institute for Polymer Research,
INRAE, Ackermannweg 10,
Université Paris-​Saclay, 55128 Mainz,
Massy 91300, Germany
France
List of Contributors xxi

Bruno Mesz Patricia B. O’Hara

Universidad Nacional de Tres de Febrero (UNTREF), Amherst College,
Instituto de Investigación en Arte y Cultura (IIAC), Amherst, MA 01002,
Sáenz Peña, United States
Argentina
Rolande Ollitrault
Ramon Morató Crêperie Ti Joos,
Creative Director for Cacao Barry brand Rue Delambre,
75014 Paris,
Marie-​Hélène Morel France
INRAE, and
UMR IATE, Group of Molecular Gastronomy,
2 place Viala, INRAE-​AgroParisTech,
34060 Montpellier cedex, International Centre for Molecular Gastronomy,
France. F-​75005 Paris,
France
John P. Morrissey
School of Microbiology, Silvia Orlandi
University College Cork, Dipartimento di Chimica Industriale “Toso Montanari”,
T12 YN60, Università di Bologna,
Ireland Viale del Risorgimento 4,
Bologna I-​40136,
Ole G. Mouritsen Italy
Department of Food Science,
Taste for Life, Design and Consumer Behavior, Ismael Osorio
University of Copenhagen, Freelance chef and cook at the ITHQ,
26 Rolighedsvej, Montreal,
DK-​1958 Frederiksberg C, Quebec,
Denmark Canada

Luca Muccioli Marie-​Paule Pardo

Dipartimento di Chimica Industriale “Toso Montanari”, Crêperie Ti Joos,
Università di Bologna, Rue Delambre,
Viale del Risorgimento 4, 75014 Paris,
Bologna I-​40136, France,
Italy and
Group of Molecular Gastronomy,
Trivikram Nallamilli INRAE-​AgroParisTech,
Max Planck Institute for Polymer Research, International Centre for Molecular Gastronomy,
Ackermannweg 10, F-​75005 Paris,
55128 Mainz, France
Germany
Jane K. Parker
Dao Nguyen Department of Food and Nutritional Sciences,
The Kitchen Lab, University of Reading,
23 chemin de la Tour, United Kingdom
Meyrin-​Village,
Geneva, Franco Pedreschi
Switzerland Departamento de Ingeniería Química y Bioprocesos,
Pontificia Universidad Católica de Chile,
Nicolas Nikolakopoulos P.O. Box 306,
Pastry Chef at the Institut Le Monde, Santiago 6904411,
45 Thessalonikis Str, Chile
Moschato, Athens 18346,
Greece Guillaume Polidori
Laboratoire de Thermomécanique (GRESPI),
Université de Reims Champagne-​Ardenne,
Reims 51100,
France
xxii List of Contributors

Michael Pontif Sébastien Roustel

Igemusu Inc., Chr-​Hansen,
Paris, Cheese application,
France Boge Allé 10-​12,
2970 Horsholm,
Alice Pontin Denmark
Department of Food and Nutritional Sciences,
University of Reading, Raffaele Sacchi
United Kingdom Department of Agricultural Sciences,
Unit of Food Science and Technology,
Gabriela Precup University of Naples Federico II,
Faculty of Food Science and Technology, Via Università 100,
Institute of Life Sciences, I-​80055 Portici (Napoli),
University of Agricultural Sciences and Veterinary Medicine Italy
Cluj-​Napoca,
Calea Mănăştur 3–​5, Christian Salles
400372 Cluj-​Napoca, CSGA (Centre des Sciences du Goût et de l’Alimentation),
Romania AgroSup Dijon,
CNRS,
Lara Querciagrossa INRAE,
Dipartimento di Chimica Industriale “Toso Montanari”, Université de Bourgogne Franche-​Comté,
Università di Bologna, F-​21000 Dijon,
Viale del Risorgimento 4, France
Bologna I-​40136,
Italy Pascal Schlich
Centre des Sciences du Goût et de l’Alimentation,
Andreas Rieger AgroSup Dijon, CNRS,
Restaurant einsunternull, INRAE,
Hannoversche Str. 1, Université Bourgogne Franche-Comté,
10115 Berlin, F-​21000 Dijon,
Germany France
and
Guro Helgesdotter Rognså CNRS, INRAE,
Nofima Processing Technology, ChemoSens Facility,
Måltidets Hus, PROBE Infrastructure,
Richard Johnsens gt. 4, F-21000 Dijon,
Po box 8034, France
4068 Stavanger,
Norway Charlotte Vinther Schmidt
Department of Food Science,
Yrjö H. Roos Taste for Life, Design and Consumer Behavior,
School of Food and Nutritional Sciences, University of Copenhagen,
University College Cork, 26 Rolighedsvej,
Cork, DK-​1958 Frederiksberg C,
T12 YN60, Denmark
Ireland
Christine Schreiber
Megan M. Ross Max Planck Institute for Polymer Research,
School of Food and Nutritional Sciences, Ackermannweg 10,
University College Cork, 55128 Mainz,
Cork, Germany
T12 YN60,
Linda Sellou
Ireland
National University of Singapore,
21 Lower Kent Ridge road,
Nikos Roussos
Singapore 119077,
Funky Gourmet Restaurant,
Singapore
Athens,
Greece
List of Contributors xxiii

Thomas Séon Régine Talon

Institut Jean Le Rond d’Alembert, Université Clermont Auvergne,
UMR CNRS 7190, INRAE,
Sorbonne Universités, MEDIS,
Paris, Clermont-​Ferrand,
France France

Lau Shi Yun Frédéric J. Tessier

National University of Singapore, Université de Lille,
21 Lower Kent Ridge road, Inserm U1167,
Singapore 119077, F-​59000 Lille,
Singapore France

Weon-​Sun Shin Hervé This vo Kientza

Laboratory of Food Chemistry & Molecular Gastronomy, Group of Molecular Gastronomy,
Department of Food & Nutrition, INRAE-​AgroParisTech International Centre for Molecular
Hanyang University, Gastronomy,
Republic of Korea F-​75005 Paris,
France
Guillaume Siegler and
Chef-​teacher UMR 0782 SayFood,
Le Cordon Bleu Paris, AgroParisTech (INRAE),
13–​15 Quai André Citroën, Université Paris-​Saclay,
Paris, F-​91300 Massy,
France France

Lorenzo Soprani David Toutain

Dipartimento di Chimica Industriale “Toso Montanari”, Restaurant David Toutain,
Università di Bologna, 29 rue Surcouf,
Viale del Risorgimento 4, Paris 75007,
Bologna I-​40136, France
Italy
Mark Traynor
Charles Spence Department of Nutrition, Dietetics, and Hospitality Management,
Head of the Crossmodal Research Laboratory, College of Human Sciences,
Department of Experimental Psychology, Auburn University,
Anna Watts Building, Auburn, Alabama,
University of Oxford, United States
Oxford OX2 6GG,
Alex Tsionitis
United Kingdom
CTC Restaurant,
Athens,
Benjamin L. Stottrup
Greece
Augsburg University,
Department of Physics,
Juan Valverde
2211 Riverside Ave,
Business Development and Innovation Manager,
Minneapolis, MN 55454,
O’Reilly Institute,
United States
Trinity College Dublin,
Dublin,
Elena Strocchi
Ireland
Dipartimento di Chimica Industriale “Toso Montanari”,
Università di Bologna,
César Vega
Viale del Risorgimento 4,
Barry Callebaut Americas,
Bologna I-​40136,
600 W Chicago Avenue,
Italy
Chicago, IL 60654,
United States
xxiv List of Contributors

Nicolas Vergnole Gn Ying Wei

Chef at the At-​Sunrice GlobalChef Academy, Chef at the At-​Sunrice GlobalChef Academy,
28 Tai Seng Street, 28 Tai Seng Street,
Level 5, Level 5,
Singapore 534106, Singapore 534106,
Singapore Singapore

Thomas A. Vilgis Juan C. Zambrano

Max Planck Institute for Polymer Research, Max Planck Institute for Polymer Research,
Ackermannweg 10, Ackermannweg 10,
55128 Mainz, 55128 Mainz,
Germany Germany

Dan-​Cristian Vodnar Birgitta I. Zielbauer

Faculty of Food Science and Technology, Max Planck Institute for Polymer Research,
Institute of Life Sciences, Ackermannweg 10,
University of Agricultural Sciences and Veterinary Medicine 55128 Mainz,
Cluj-​Napoca, Germany
Calea Mănăştur 3–​5,
400372 Cluj-​Napoca,
Romania

David S. Waldron
School of Food and Nutritional Sciences,
University College,
Cork,
T12 YN60,
Ireland
Foreword
One of the most exciting areas of research and experimentation in
the food and culinary areas in recent years is the emerging discip-
line of molecular and physical gastronomy (in short, “molecular
gastronomy”), the scientific discipline dedicated to the study
of phenomena that occur during the preparation and consump-
tion of dishes. Molecular gastronomy considers the chemistry,
biology and physics of food, along with the physiology of food
consumption.
The objective of this book is to provide a comprehensive over-
view of this field, based on contributions from some of the main
scientists in their areas, but also with contributions by chefs for
the application part. In this last part, the book also addresses some
of the most popular techniques of molecular cooking, a cooking
style associated with molecular gastronomy and characterized
mainly by the use of new tools and methods, often imported from
chemistry laboratories, as well as cultural and artistic aspects of
food preparation. The newer cooking trend called “note by note
cuisine” is also explored.
Considering that “gastronomy is the knowledge and
understanding of all that relates to man, as he eats” (JA Brillat-​
Savarin, Physiology of taste, 1825), it may be proposed that
molecular gastronomy is the chemical, physical and biological
part of this knowledge, i.e., the scientific activity that relates
to culinary phenomena. As natural sciences have technical
applications, molecular gastronomy can lead to better and/​or new
ways of cooking.
This book aims to fulfil the need for a comprehensive refer-
ence in molecular gastronomy along with a practical guide to
molecular cooking and more recent applications such as note
by note cuisine. Indeed, many books already exist for a general
audience, either addressing food science in a “light” way and/​
or dealing with modern cooking techniques and recipes, but
no book exists yet that encompasses the whole molecular gas-
tronomy field, providing a strong interdisciplinary background in
the physics, biology and chemistry of food and food preparation,
along with discussions on creativity and the art of cooking.
We hope that such a new resource will be very useful for food
scientists and chefs, as well as students studying food science or
food technology and all lay people interested in gastronomy. It
should be noted that the chapters in each part are organized in
what appeared to be the most logical structure, and are hence
alphabetical in Part I and in a logical sequence in Parts II and III.
The book is not intended to be read in sequence but to be used as
a reference work, which can be consulted under specific headings
or dipped into at will.
Finally, we as editors wish to thank all the authors, coming
from so many different fields, for their enthusiasm for the pro-
ject and their patience with the process of editing and producing
this very significant end product, and to thank Stephen Zollo and
colleagues at CRC Press/​Taylor and Francis for their support
for this project, guidance through the publication process, and
great patience in awaiting the end result. We also thank Nicola
Howcroft at Newgen Publishing UK for support during the pro-
duction stages. We would also like to thank Ciara Tobin of the
School of Food and Nutritional Sciences at University College
Cork for significant and invaluable assistance in the preparation
of this final manuscript, and Dylan Kelly for his assistance with
editorial work in the process.
xxv
Introduction to Molecular Gastronomy and Its Applications
Hervé This vo Kientza, Christophe Lavelle, Róisín Burke and Alan Kelly
Molecular Gastronomy was originally defined as the scientific dis-
cipline that has the goal of “looking for mechanisms of phenomena
occurring during culinary processes” (Announcement poster for the
International Workshop on Molecular and Physical Gastronomy,
Ettore Majorana Centre for Scientific Culture, Erice, Italy, 1992).
The initial name “molecular and physical gastronomy” was later
shortened to “molecular gastronomy”, but the scientific programme
did not change, except that it was progressively made clearer when
it was increasingly recognized that technology and education were
not scientific activities but rather, applications of science.
In spite of this quite clear definition by its creators, “molecular
gastronomy” has been widely referred to in the media and in
culinary circles using different definitions and terminology. These
include, for example, “Molecular Cooking/​Cuisine”, “Modernist
Cuisine” and “Scientific Cooking”, to name but a few.
Thus, some confusion has remained, because “gastronomy”
is often confused with “haute cuisine” or “fine dining”, and
also because innovative chefs were interested in the possible
applications of molecular gastronomy and hence, indeed, doing
molecular cooking!
Clarification of What Molecular Gastronomy Is
Even after years of efforts to explain the differences, two main
sources of confusion remain:
1. between molecular gastronomy, on the one hand, and
food science and technology, on the other;
2. between molecular gastronomy, molecular cooking and
molecular cuisine.
In terms of the first of these, it should be observed that
“molecular gastronomy” is certainly not intended to include all
food sciences, but only part of this field. Indeed, the emergence
of molecular gastronomy was due in part to the fact that, in the
1980s, food scientists did not study real culinary processes and
dishes prepared daily in homes or restaurants or other culinary
enterprises, but mainly industrial processes and food ingredients.
There were no scientific investigations into the physical and
chemical transformations that occurred during the preparation of
coq au vin, Irish stew or in fact, most dishes. For example, it
should be noted that cooking using wine was never considered by
science, in spite of the fact that wine is used in about half of all
the recipes for French traditional sauces!
Of course, it could be imagined that molecular gastronomy
would disappear when its duty was done, i.e., promoting the sci-
entific study of dishes, but at the same time, there is a strong argu-
ment for keeping it in the field of food science and technology,
because this is not the same as the science of food ingredients or
the sciences of processes. The work is not finished, and the field
remains a source of discoveries and technical innovation through
technological work, particularly because the newly created inter-
face lies between so many disciplines.
Molecular Gastronomy was first proposed as a scientific
activity, but it is true that at the time when it was introduced, there
was also a side goal of modernizing culinary processes, in par-
ticular with tools from labs (with a basis in chemistry and physics).
The expression “molecular cooking” was introduced to describe
the new culinary techniques using these tools. This led progres-
sively to the introduction of a new culinary style, called “molecular
cuisine”, based on these new tools. Two examples of this are dishes
made using siphons and meat cooked at a low temperature; these
are the obvious hallmark of chefs practising molecular cooking or
molecular cuisine, which was, in fact, the original intention. Let us
insist on the slight difference: molecular cooking is the name for
cooking with “modern” culinary tools, especially tools that were
not found in kitchens in the 1970s; in particular, these include tools
coming from laboratories, such as siphons, thermal circulators,
liquid nitrogen, ultrasonic probes and rotary evaporators. On the
other hand, molecular cuisine is the style of cooking based on
using such new tools, in particular for new applications. All this is
well illustrated in Part III of this Handbook.
Let us finish by observing that the “modern tools” of the 1980s
are no longer modern today, and young chefs cannot even imagine
a world without low-​temperature cooking or sous-​vide, just as chefs
at the end of the 20th century could hardly have imagined how to
cook when gas wasn’t widely available. It has even been said that
molecular cooking and molecular cuisine are already no longer
current concepts (because “note by note cooking” is a drastic mod-
ernization), but the technical transformation of culinary activity is
certainly not complete. Only today have companies begun to sell
ultrasonic probes for making emulsions, for example.
Science and Cooking
For “science”, one should first observe that this word has various
meanings, such as knowledge in general, or “natural sciences”.
However, when someone discusses the issue of precision and
1
2 Introduction to Molecular Gastronomy and Its Applications
rigour, the general meaning of knowledge is not specified, and
instead, the natural sciences are emphasized. The goal of science
is to discover mechanisms and phenomena, using a method that
classically goes through (1) identification of a phenomenon;
(2) quantitative characterization of the phenomenon; (3) grouping
the data in equations (“laws”); (4) producing theories or models
by grouping these laws and introducing new notions or concepts;
(5) drawing testable theoretical conclusions from the theory; and
(6) experimentally testing these theoretical conclusions. This has
nothing to do with making food per se, and it is clear as well that
cooking has nothing to do with the natural sciences; the goals
and methods are both completely different, and they will always
remain separate.
In another point about the relationship between science and
cooking, there has been some discussion about “science in the
kitchen”, in particular in relation to a quotation by the French
chefs Auguste Escoffier, Phileas Gilbert and Emile Fetu, who
predicted in their famous Guide Culinaire (1907) that cooking
would become “scientific”. Let’s first observe that cooking
is an activity mixing a social component, a technical compo-
nent and art (see Hervé This and Pierre Gagnaire’s discussion
in Cooking, a quintessential art, University of California Press,
2010). Indeed, there is also an element of human relationships
and sociology in the act of producing food for others or of eating
food that was prepared by others. In terms of technique, it is clear
that cutting, heating and mixing are all technical activities, but
one important component is “art”, i.e., the activity of producing
“beautiful” objects. Here, the beauty is in the food to be eaten, not
only to be looked at, which leads to the conclusion that “good”
means “beautiful to eat”.
The fact is that there will never be any sciences “in” the kit-
chen. On the other hand, one can have applications of natural
sciences in the kitchen (e.g., molecular cooking or note by note
cooking), and also, one can use culinary phenomena for scien-
tific objects, which is indeed the main principle of molecular
gastronomy. As shown in the description of the cycle of the nat-
ural scientific method, there is no end to the process, as theories
are always insufficient and can always be improved. In contrast
to molecular cooking, which will end when all the “new” hard-
ware has become old, molecular gastronomy will continue to
evolve.
Today, as molecular cooking is less trendy than it used to be
(low-​temperature cooking and making foam with siphons are
“classical” techniques nowadays), molecular gastronomy, in con-
trast, is developing regularly in universities around the world,
with new laboratories, new research groups and new educational
curricula. In this way, new aspects of molecular gastronomy are
generated, and the discipline continues to evolve.
It is a very exciting time … which is also why this Handbook
is timely! This book is divided into three parts. The first one
includes scientific information about phenomena that occur
during culinary activities, and this is, as already said, exactly what
molecular gastronomy is. The second and third parts then show
educational and technical and culinary applications, respectively.
In the case of “education”, groups all over the world have been
using results from molecular gastronomy in order to promote
scientific studies. Regarding technical applications, many new
techniques have been introduced after research in molecular gas-
tronomy, examples of which will also be given.
Applications in Schools, Colleges and Universities
We have said that molecular gastronomy is a scientific activity,
and this is true. As for any scientific field, there are applications
in both directions: educational and technical. These applications
are developing simultaneously, and even more since cooking has
become more popular, with TV shows in all countries. Indeed,
cooking, which was traditionally an activity for women, is now
routinely done by men. One can observe that in developed coun-
tries such as France (and probably also elsewhere), in the 1950s,
culinary lessons were given only to girls at school. For this
country in particular, so called “Ateliers expérimentaux du goût”,
i.e., activities mixing natural sciences and cooking (plus art),
were introduced in all primary schools in 2001. Today, “science
and cooking” activities are performed in colleges, high schools
and even universities focusing on the sciences.
In specialized colleges, such as culinary schools, more and
more molecular gastronomy activities are being performed, and
today the new French teachers for culinary schools are all being
educated in molecular gastronomy with a view to experimenting
in their classes, conducting science and cooking activities, or
activities in which the culinary teacher is working alongside
science teachers.
For chefs, there are many initiatives for continuous education;
for example, in France, every month for 18 years, seminars of
molecular gastronomy have been run in Paris. Such seminars also
exist in many other places, with various names.
The general public has through organized workshops and
other activities and through the media, many opportunities to
enjoy the delights that applying science in the kitchen can bring.
However, this is not new; as early as the very beginning of the
20th century, the microbiologist Edouard de Pomiane was a star,
in particular with Radio-Cuisine, the first French programme on
cooking, writing best-​sellers about what was not called molecular
gastronomy.
For all these educational applications, in spite of their great
success, one could make the same criticism as for science popu-
larization in general, i.e., that a discourse is given, avoiding
calculations, and this is a difficulty, because in this way, the
receiver of the popularized discourse cannot evaluate the validity
of this information but has to trust the source. Indeed, the receiver
remains at the surface, and probably the only way to circumvent
this difficulty is through experimenting. This is what is proposed in
many educational activities mentioned here, and, for sure, testing
“culinary precisions” is a good way to invite everybody to share
the excitement of science by taking the very first step and making it
useful. It is no use trying to investigate phenomena that do not exist!
Applications to the Culinary Arts
We have explained that the applications of science are not science
itself, and it is true that the fruit is not the tree. However, it is
Introduction to Molecular Gastronomy and Its Applications
also true that molecular gastronomy has had many technical
applications.
We also explain that molecular cooking was the name given (in
1999 only) to the use of “new” tools, but this cannot be considered
as a direct application of molecular gastronomy, because no
new knowledge of the mechanisms of cooking discovered by
molecular gastronomy was needed for this. For example, the
proposal to use tools from chemistry laboratories in the kitchen
was put forward just when one of us began his studies of testing
culinary precisions, well before the official creation of molecular
gastronomy.
However, this is not to say that nothing in molecular cooking
was a result of molecular gastronomy; on the contrary. For
example, the proposal of cooking eggs at 65 °C was a result of
theoretical explorations of egg coagulation, and this is only one
example among many.
At some time before the first International Workshop on
Molecular and Physical Gastronomy (1992), the lawyer and
3
gastronome Jean-​Anthelme Brillat-​Savarin commented that “the
discovery of a new dish does more for the happiness of mankind
than the discovery of a new star”. It was a goal, and a good one,
and this is why the third part of this book is devoted to innovation
in the culinary art. This is the place for molecular cooking, for
molecular cuisine and for “note by note cooking”, the new way
of cooking in which the ingredients are pure compounds. Also,
without delay, let’s say that after about two decades of devel-
opment of this “synthetic cooking” (as for synthetic music), a
style begins to appear, so that “note by note cuisine” has to be
discussed along with note by note cooking.
Finally, all this is very exciting, because there is a wide family
of initiatives all over the world, and this Handbook intends to
show some of them. The book cannot be a comprehensive descrip-
tion, because there are too many things to cover, but it should give
the reader a good idea of what has been done in recent decades.
Celebrate Knowledge! Celebrate Enlightened Gourmandise!
 Part I

Scientific Foundations
Acids in Foods and Perception of Sourness
Christian Salles
CSGA (Centre des Sciences du Goût et de l’Alimentation), AgroSup Dijon, CNRS, INRAE,
Université de Bourgogne Franche-​Comté, F-​21000 Dijon, France
Introduction
Sourness is mainly caused by the detection of protons. This per-
ception may act as a warning signalling the presence of a high
concentration of acid dangerous for the body, of unripe fruits,
or of spoiled foods contaminated by microorganisms. At low
concentrations, acidic stimuli evoke a sour taste, while at high
concentrations, the trigeminal lingual system is activated, leading
to a sensation of irritation. A low level of sour taste is attractive
to humans and animals in some foods such as citruses, while it is
aversive when present at a high level.
Numerous organic and mineral acids are responsible for
sourness. Citric acid, malic acid, and tartaric acid are generally
found in vegetal products (e.g., fruits and vegetables), while
lactic acid is found in animal products (e.g., dairy products and
meat). Compounds other than organic and mineral acids can
also be responsible for a sour note; amino acids with a lateral
acid function, such as L-​glutamic and L-​aspartic acids, have an
acid taste, while their sodium salts are responsible for umami
taste (Kato et al., 1989). Peptides with a sour taste generally
contain acidic amino acids such as aspartic and glutamic acid.
A dipeptide with a sour taste contains at least one acidic amino
acid linked to another acidic, neutral, or aromatic amino acid
(Kirimura et al., 1969).
In this chapter, I will give a non-​exhaustive overview of the
main characteristics of sour compounds from a physico-​chemical
and sensory point of view.
Influence of Food Matrix Composition
on Sourness
The release of acids in the mouth during eating and the percep-
tion of sourness may be influenced by the food matrix compos-
ition and structure. For example, fat, acid, and salt were found
to influence the temporal perception of firmness, saltiness, and
sourness of cheese analogues (Stampanoni and Noble, 1991).
Increasing salt and acid contents were both found to increase the
perception of firmness, and a higher fat content resulted in softer
but sourer cheese analogues. Whatever the fat level, analogue
cheeses with high acid and high salt levels had higher sourness
intensity and the longest sourness perception. The time to reach
maximum saltiness and sourness intensity was shortest for low-​
salt and low-​acid analogue cheeses.
The kinetic release of non-volatile compounds (leucine,
phenylalanine, glutamic acid, citric acid, lactic acid, propanoic
acid, sodium, potassium, calcium, magnesium, chloride,
and phosphates) during the eating of a model cheese and the
relationships to some oral (salivary and masticatory) parameters
have also been studied (Pionnier et al., 2004a). The maximum
concentration in saliva varied significantly according to the
compound. However, there was no significant effect of the com-
pound on the time to reach maximum concentration. A long time
to reach maximum concentration and total quantity of released
compounds could be related to high chewing time and low saliva
flow rates, low chewing rates, low masticatory performances, and
low swallowing rates. The time to reach the maximal intensity
of the sour attribute was positively related to the time to reach
the maximal concentration of citric acid and to oral parameters
(Pionnier et al., 2004b).
Significant differences among several Camembert cheeses
concerning bitterness and saltiness were reported, but no diffe-
rence was observed for sourness (Engel et al., 2001a). During
ripening, a decrease of the perceived sour note was observed in
all portions of the cheese (rind, under-​rind, and centre). The sour
note in food products may be partially due to H3O+ concentra-
tion, which progressively varies with the consumption of lactic
acid by microorganisms. However, pH and sourness may not be
correlated, and H3O+ concentration was not sufficient to fully
explain the sour taste; other chemical species, such as sodium
chloride, may act on this taste characteristic. The effect of pH
and interactions between sourness and other perceptions are
detailed later. The evolution of sourness in the cheese portions
may be explained by the migration phenomena of taste-​active
compounds such as molecules that change the pH (acids, and
phosphates in particular) or those that are responsible for some
enhancing effect (Engel et al., 2001c).
The fatty acid residue compositions of oil species affect taste
perception. For oil-​in-​water emulsions with basic taste substances
and oil species, Koriyama et al. (2002) reported that the type of
oil extended retention of perception, and did not affect sweetness
or saltiness, but suppressed sourness and bitterness. The degree
of sourness was dependent on oil species, in particular the fatty
7
8 Christian Salles
acid residue composition. Thus, oils to be added to food should
be carefully selected by the manufacturer because of their effects
on taste perception as well as on texture and aroma (retronasal
odour).
An effect of viscosity on perceived intensity of sour taste
was also reported (Sediva et al., 2004). The sourness inten-
sity decreases with increasing solution viscosity. However, the
differences depend on the concentration of thickener agents and
the acid used.
Mechanisms Leading to Sour Perception
The Acid Receptor
The molecular mechanisms of detection of sourness are not
well known yet. A wide range of cell types, receptors, and even
receptor-​independent mechanisms have been proposed to mediate
acid detection in the tongue. It has been proposed that sour taste
is mediated by a unique cell type, independent of all other taste
qualities (Huang et al., 2006). In the tongue, polycystic kidney
disease 2-​ like 1 (PKD2L1) ion channel, coexpressed with
PKD1L3, has been demonstrated to be necessary for the detec-
tion of sour compounds. Therefore, sour, salty, sweet, bitter,
and umami tastes are each mediated by a unique cell type, inde-
pendent of all other taste qualities. Salty and sour tastes directly
activate ion channels while bitter, sweet, and umami tastes are
elicited by G-​ protein-​
coupled receptors (Briand and Salles,
2016). Recently, it has been shown that the Otop gene family
encodes a family of ion channels that are unrelated structurally
to previously identified ion channels and are highly selective for
protons (Tu et al., 2018).
Particular proteins were reported to be able to convert sourness
to sweetness. Miraculin, a tetramer and dimer of a 25 kDa
protein, was found to transform a sour taste into a sweet taste
(Kurihara and Beidler, 1968). Both the tetramer and the dimer
have taste-​modifying properties. Miraculin has been estimated to
be as much as 400,000 times sweeter than sucrose on a molar
basis. The mechanisms involved in the sweetness of miraculin
are unique; at neutral pH, miraculin interacts with the sweet
taste receptor T1R2/​T1R3, but it does not activate the receptor
and partially suppresses the response to other sweeteners. At
acidic pH, miraculin changes its conformation and activates the
sweet taste receptor (Briand and Salles, 2016). Another pro-
tein named neoculin (previously named curculin) elicits a sweet
taste and also exhibits sweet taste-​modifying properties able to
convert sourness to sweetness. In addition, it is noteworthy that
peptides generated from pork loin during cooking were capable
of strongly suppressing sourness (Okumura et al., 2004). The
proposed mechanism of sourness suppression by the peptide was
an inhibition of the binding of sour taste substances at the proton-​
sensitive ion channel level.
The Role of Saliva in Sourness Perception
Sourness is related to the proton concentration and thus pH, as
sourness intensity decreases with increasing pH. However, pH
does not fully explain the intensity of acid solutions. It has been
shown that sourness perception is fully dependent on titratable
acidity: Norris et al. (1984) showed that binary acid mixtures
of equal pH and titratable acid differed significantly in sourness
intensity and in saliva-​inducing capacity. They observed signifi-
cant differences in both maximum intensity of perceived sourness
and parotid flow as a function of the specific anionic composition
(i.e., citric, tartaric, or fumaric) of the samples, since they had
equal potential (titratable) hydrogens and equal free hydrogens
(pH). For example, in tartaric–​fumaric acid mixtures varying in
pH (3.0–​3.75) at a constant titratable acidity and varying in titrat-
able acidity at a constant pH, sourness intensity and salivary flow
rate increased with acidity and decreased with pH. A large con-
trast between subjects with high and low salivary flow rate and
perceived intensity of sourness was also reported (Norris et al.,
1984). In response to variation in pH and total acid, the high-​flow
subjects demonstrated marked alteration in flow but little change
in sourness perception, while low-​flow subjects responded at
a lower absolute level of acids but showed marked changes in
sourness and little change in salivary flow rate.
Another study reported that changes in solution pH were
related to individual salivary flow rates. Greater increases
in expectorated solution pH were observed for individuals
with higher flow rates (Christensen et al., 1987). Moreover,
on measuring taste threshold and suprathreshold responses
to different volumes of acids, those authors demonstrated
that individuals with high salivary flow rates were less sensi-
tive to the taste of acids and that large volumes of acid were
more easily perceived. It was suggested that dilution by saliva
with different pH is not the correct mechanism and also that
adaptation of taste receptors to differing concentrations of free
hydrogen ions was unlikely, because the sour threshold results
were opposite to those predicted by an adaptation model. This
was interpreted as greater flow rate adding a greater quantity
of saliva to taste solutions and consequently adding a greater
total amount of salivary buffers as bicarbonate concentration
increased (Christensen et al., 1987).
Comparing low-​ flow judges and high-​ flow judges, it was
found that salivary flow rate did not affect temporal responses
for sourness, sweetness, or fruity flavour (Bonnans and Noble,
1995). Large differences in sweetness and small, insignificant
differences in sourness were produced at a constant acid con-
centration by increasing sweetener levels. However, salivary
flow rate was significantly correlated only with sourness ratings.
Changes in perceived sourness intensity influenced salivary flow
independently of acid concentration, suggesting that salivary
response is not only due to stimuli concentration and may be
mediated by the cognitively processed taste response (Bonnans
and Noble, 1995).
Saliva has been suggested to act as a buffering system affecting
the way we perceive the sourness of acids. Moreover, Bonnans and
Noble (1995) observed that the maximum intensity of sourness
and salivary flow rate decreased as the level of sweeteners was
raised at constant acid concentration. This suggested that salivary
flow rate is mediated by cognitively processed taste response
and not only the concentration of stimuli. Salivary flow rate was
found to increase when the pH of different beverages decreased
(Guinard et al., 1998).
Acids: Perception of Sourness 9

The chemical properties of the sour molecules, such as pKa,

number of carboxyl groups, hydrophobicity, and salivary flow
rate, influence sour temporal perception. High-​ flow subjects
and low-​flow subjects were submitted to a continuous stimulus
delivery flow rate of acid solutions, and the time–​intensity per-
ception and the pH on the tongue surface were continuously
measured (Lugaz et al., 2005). The results showed that the saliva
of high-​flow subjects decreased the acidity of the acid solution
more efficiently than the saliva of low-​flow subjects. However,
high-​flow subjects exhibited higher perceived intensity for acid
solutions than low-​flow subjects. The saliva of high-​flow subjects
can modify the pH of an acid solution more efficiently than the
saliva of low-​flow subjects, thanks either to a dilution effect or
to a difference in buffering capacity of saliva between high-​flow
subjects and low-​ flow subjects. Moreover, by comparing the
effect of different weak acids, the authors reported that titrat-
able acidity rather than pH has to be considered to explain sour
perception. They reported also that the difference in sensitivity
between high-​ flow subjects and low-​ flow subjects might be
due to a higher permeability of epithelial tissue to hydrophobic
molecules (Lugaz et al., 2005). Acid stimuli also influence the
composition of saliva. Annexin A1 and calgranulin A, anti-​
inflammatory compounds involved in inflammation, were over-​
represented after umami, bitter, and sour stimulations (Neyraud
et al., 2006).

Interactions of Sourness Perception with Other

Perceptions
In mixtures, the various tastes are known to interact through
binary taste–​taste interactions (Breslin, 1996; Keast and Breslin,
2002; Figure 1.1). The physiological mechanisms involved in
these interactions, while not well understood yet, occur at the
taste receptor cell level. However, the integration of taste signal at
the central level cannot be excluded. Concerning sourness, it was
reported that sour and salty mixtures symmetrically affect the
intensity of each other, with enhancement at low concentrations
and suppression or no effect at high concentrations. At low con-
centration, sourness had variable effects on sweetness, while at
higher concentrations of mixed of sour and sweet compounds,
sweetness and sourness were mutually suppressed. Mixtures
of sour and bitter compounds enhanced each other at low con-
centration; at moderate intensity, bitterness was suppressed and
sourness enhanced, while at high concentration, sourness was
suppressed, and the effect on bitterness was variable (Keast and
Breslin, 2002).
These taste–​taste interactions are functional in complex media
such as real foods. For example, in goat cheeses, sourness is
mainly due to a synergistic effect involving sodium chloride,
phosphates, and lactic acid (Engel et al., 2000). Omission tests
using model goat cheese extracts showed that the sourness of
lactic acid was enhanced by sodium chloride but lowered by the
presence of phosphates. The enhancement of sourness due to
lactic acid by sodium chloride was also reported in Camembert
cheese (Engel et al., 2001b). In tomato juices, the omission of all
the sugars led to an almost total disappearance of sweetness and
FIGURE 1.1 Binary taste interactions involving sourness perception: Low
(a), medium (b), high (c) intensity/​concentration. Enhanced (+); suppressed
(-​); variable (v), nil effect (0).
(Adapted from Keast and Breslin, 2002)
a significant increase in sourness and saltiness, which had been
partly masked by the sugars (Salles et al., 2003). Thus, sourness
is mainly due to citric and malic acids but modulated by sugars.
Such interactions between sourness and sweetness perceptions
were also reported in champagne (Martin, 2002).
Interaction between sourness and saltiness was also reported in
rice vinegar. Both detection and recognition sourness thresholds
were decreased when vinegar ingredients (salt) were added,
inducing sourness (Hatae et al., 2009). These authors studied the
interaction of saltiness and sourness at the threshold level. They
measured the detection and recognition thresholds for salt solu-
tion to which vinegar of subthreshold concentration was added,
and for vinegar solutions to which salt of subthreshold concen-
tration was added. When vinegar at a half concentration of the
detection threshold of each panellist was added to the salt solu-
tion, both the detection and the recognition threshold of salt were
reduced significantly. This phenomenon was more pronounced
with rice black vinegar than with rice vinegar. In contrast, when
10
NaCl solution was added to vinegar at half of the concentration
level of the detection threshold for each panellist, no intensifying
or weakening effect on the detection and recognition threshold
was observed. Thus, this saltiness–​ sourness interaction was
found to be asymmetrical, as no effect of sourness was observed
on saltiness with the same model. Moreover, this result shows the
potential of vinegar to partially substitute salt to give a satisfac-
tory salty taste in dishes.
Aroma was found to interact with sourness perception through
crossmodal perceptive interactions. It has been shown that no
physicochemical mechanism was involved in the odour–​taste
interactions (Pfeiffer et al., 2006). For example, the sourness of a
citric acid solution can be lowered by caramel aroma (Stevenson
et al., 1999), while an increase of the fruitiness of a water solu-
tion flavoured with orange was observed when sourness and
sweetness were perceived (Bonnans and Noble, 1993). The influ-
ence of the concentration of sucrose and citric acid on the flavour
of drinks and sherbets containing orange and lemon aromas was
studied (Stampanoni, 1993); there was a positive effect of citric
acid and sucrose on all the rated descriptors, fruitiness, freshness,
juiciness, and global impact, but not on peely note. In the case
of orange sherbets, sourness was correlated with freshness and
peeliness, while in the case of lemon sherbets, sourness was
correlated with freshness and juiciness. Citric acid and sucrose
were able to contribute to retronasal aroma intensity of a citral
solution for some panellists (Kuo et al., 1993). Lemon and straw-
berry odour were found to enhance the sourness of citric acid
(Frank, 2003).
Acceptability of Sourness
Sour taste is related to the presence of organic acids, but the asso-
ciation between nutrient content and taste varies among foods
that have diverse taste profiles (Liem and Russell, 2019). Among
these, nutrient-​rich foods such as fruits were mostly classified as
sweet/​sour in different studies.
There are several biological underpinnings to the sense of
taste. At birth, humans can already distinguish between sweet,
sour, and bitter tastes, as shown by distinct facial expressions
of newborns when exposed to tasting substances. As judged by
facial expression, their sucking responses to and to some extent,
intake of these tastants follow the positive (e.g., sweet taste)
or negative (e.g., bitter and to some extent, sour taste) facial
expressions (Steiner et al., 2001). Strong sour and bitter solutions
are both disliked by newborns, but the facial response to sour
solutions (e.g., lip pursing) is remarkably different from the facial
response to bitter solutions (Desor et al., 1975; Steiner et al.,
2001). The combination of lip pursing and sucking, seen typically
in response to tasting sour substances, contributes to stimulation
of salivary flow in the oral cavity. In adults, it has been suggested
that the increased flow and buffering capacity of saliva neutralize
sour-​tasting substances, as described previously.
Infants’ ingestive responses to sour taste do not indicate a
clear rejection; no difference in the infants’ ingestion has been
observed in response to water and water with added citric acid
(Cowart et al., 1990). Some infants and young children showing
Christian Salles
a preference for high concentrations of citric acid in a sugar solu-
tion were found to be more likely to consume fruit than others
(Blossfeld et al., 2007; Liem et al., 2006; Liem and Mennella,
2003). This suggests that sour taste preferences directly influ-
ence food consumption. Thus, dietary learning and experience
are important for the acceptance and intake of sour fruits (Liem
and Mennella, 2002). Children fed on formula containing sour-​
tasting protein hydrolysates tended to have higher acceptance
and intake of sour foods and higher levels of citrate in orange
juice (Liem and Mennella, 2002). Similarly, children tend to have
a higher preference for sourness during childhood (Liem and
Mennella, 2003).
A perception study of sourness and food behavioural data
for apple juice and fruit drinks with different dry matter levels
showed that children on average had a high preference for
versions of beverages perceived as less sour (Kildegaard et al.,
2011); however, a minor segment of children with high liking and
wanting for the apple juice perceived as most sour was observed.
To date, the application of sourness intensity has been little
studied as a stimulus in adults in relation to its impact on food
intake or the onset of satiation. Moreover, similarly to bitterness,
many sour foods tend to have low energy density and are unlikely
to contribute significantly to energy intake (Forde, 2016). It has
been reported that sourness of citric, malic, tartaric, and acetic
acid solutions could be related to an interaction of the titratable
acidity and the pH of the solution. Sourness also influenced the
overall acceptability of imitation fruit beverages prepared with
these acids (Coseteng et al., 1989). Pre-​stirred flavoured com-
mercial yogurts were evaluated for sweetness and sourness
rather than for overall liking. The results showed that overall
consumer liking was significantly correlated with sweetness
intensity, sweetness–​sourness ratio, and the cumulative impact
of sweetness and sourness for strawberry and raspberry yogurt.
Generally, it was found that the sweeter the yogurt, the higher the
acceptance of these fruit-​flavoured yogurts by consumers (Barnes
et al., 1991).
Conclusion
The sour note found in many foods is caused by the presence
of protons and by substances that lead to their production by
hydrolysis. The sour compounds are, in particular, organic acids,
the structure, size, polarity, or pKa of which influence the inten-
sity of the sour note. There is a great variety of these, because
they come from the major metabolic pathways; however, some
of them are predominant in certain foods. Due to their physico-
chemical properties, they are sensitive to the buffering capacity
of saliva. This acid perception is modulated by the presence of
other stimuli responsible for bitterness, sweetness, and salti-
ness. For example, in cheeses, the perception of the acid note
can be influenced by the presence of sodium chloride. The per-
ception of particular aromas can also modulate acid percep-
tion through perceptual interactions. Overall, there is less work
on sour perception than on bitter, salty, and sweet perceptions.
Indeed, the masking of bitterness and the strengthening of salti-
ness and sweetness are of definite economic interest, either to
Acids: Perception of Sourness
improve the taste quality of food, drink, or medicines, in the case
of bitterness, or for public health reasons, in the case of saltiness
and sweetness. However, perception of sourness is important in
some cases, as it brings a certain balance with other perceptions,
thus impacting appreciation by consumers. This sour perception
is appreciated in certain contexts but is often rejected if it is too
intense. The mechanisms at play in the perception of acid at the
peripheral level are still little known; they deserve deeper study
to better control this perception.
REFERENCES
Barnes DL, Harper SJ, Bodyfelt FW, McDaniel MR. 1991. Prediction
of consumer acceptability of yogurt by sensory and analytical
measures of sweetness and sourness. Journal of Dairy Science,
74, 3746–​3754.
Blossfeld I, Collins A, Boland S, Baixauli R, Kiely M, Delahunty C.
2007. Relationships between acceptance of sour taste and fruit
intakes in 18-​month-​old infants. British Journal of Nutrition,
98, 1084–​1091.
Bonnans S, Noble AC. 1993. Effect of sweetener type and of sweet-
ener and acid levels on temporal perception of sweetness,
sourness and fruitiness. Chemical Senses, 18, 278–​283.
Bonnans SR, Noble AC. 1995. Interaction of salivary flow with
temporal perception of sweetness, sourness, and fruitiness.
Physiology & Behavior, 57, 569–​574.
Breslin PAS. 1996. Interactions among salty, sour and bitter compounds.
Trends in Food Science and Technology, 7, 390–​399.
Briand L, Salles C. 2016. Taste perception and integration. In:
Etievant P, Guichard E, Salles C, Voilley A (eds.), Flavour:
From food to behaviors, wellbeing and health. Elsevier Ltd,
Duxford, UK, 101–​119.
Christensen CM, Brand JG, Malamud D. 1987. Salivary changes in
solution pH –​a source of individual differences in sour taste
perception. Physiology & Behavior, 40, 221–​227.
Coseteng MY, Mclellan MR, Downing DL. 1989. Influence of titrat-
able acidity and pH on intensity of sourness of citric, malic,
tartaric, lactic and acetic acids solutions and on the overall
acceptability of imitation apple juice. Canadian Institute of
Food Science and Technology Journal-​Journal de L’Institut
Canadien de Science et Technologie Alimentaires, 22, 46–​51.
Cowart B, Beauchamp G, Mcbride R, Macfie H. 1990. Early devel-
opment of taste perception. In: McBride R and MacFie H
(eds.), Psychological Basis of Sensory Evaluation. Elsevier
Applied Science, London, New York.
Desor JA, Maller O, Andrews K. 1975. Ingestive responses of
human newborns to salty, sour, and bitter stimuli. Journal of
Comparative and Physiological Psychology, 89, 966–​970.
Engel E, Nicklaus S, Septier C, Salles C, Le Quéré JL. 2000. Taste
active compounds in a goat cheese water-​ soluble extract.
2. Determination of the relative impact of water-​ soluble
extract components on its taste using omission tests. Journal
of Agricultural and Food Chemistry, 48, 4260–​4267.
Engel E, Nicklaus S, Septier C, Salles C, Le Quéré JL. 2001a.
Evolution of the taste of a bitter Camembert cheese during
ripening: Characterization of a matrix effect. Journal of
Agricultural and Food Chemistry, 49, 2930–​2939.
Engel E, Septier C, Leconte N, Salles C, Le Quéré JL. 2001b.
Determination of taste-​active compounds of a bitter Camembert
cheese by omission tests. Journal of Dairy Research, 68, 675–​688.
Engel E, Tournier C, Salles C, Le Quéré JL. 2001c. Evolution of
the composition of a selected bitter Camembert cheese during
ripening: Release and migration of taste-​active compounds.
Journal of Agricultural and Food Chemistry, 49, 2940–​2947.
11
Forde CG. 2016. Flavour perception and satiation. In: Etievant P,
Guichard E, Salles C, Voilley A. (eds.), Flavour: From Food to
Behaviors, Wellbeing and Health. Elsevier Ltd, Duxford, UK,
251–​276.
Frank RA. 2003. Response context affects judgments of fla-
vour components in foods and beverages. Food Quality and
Preference, 14, 139–​145.
Guinard JX, Zoumas-​Morse C, Walchak C. 1998. Relation between
parotid saliva flow and composition and the perception of gus-
tatory and trigeminal stimuli in foods. Physiology & Behaviour,
63, 109–​118.
Hatae K, Takeutchi F, Sakamoto M, Ogasawara Y, Akano H. 2009.
Saltiness and acidity: Detection and recognition thresholds and
their interaction near the threshold. Journal of Food Science,
74, S147–​S153.
Huang AL, Chen X, Hoon MA, Chandrashekar J, Guo W, Trankner
D, Ryba NJP, Zuker CS. 2006. The cells and logic for mamma-
lian sour taste detection. Nature, 442, 934–​938.
Kato H, Rhue MR, Nishimura T. 1989. Role of free amino acids and
peptides in food taste. In: Teranishi R, Buttery RG, Shahidi F
(eds.), Flavour Chemistry Trends and Developments. American
Chemical Society, Washington, DC, 158–​175.
Keast RSJ, Breslin PAS. 2002. An overview of binary taste-​taste
interactions. Food Quality and Preference, 14, 111–​124.
Kildegaard H, Tonning E, Thybo AK. 2011. Preference, liking and
wanting for beverages in children aged 9–​14 years: Role of
sourness perception, chemical composition and background
variables. Food Quality and Preference, 22, 620–​627.
Kirimura J, Shimizu A, Kimizuka A, Ninomiya T, Katsuya N. 1969.
Contribution of peptides and amino acids to taste of foodstuffs.
Journal of Agricultural and Food Chemistry, 17, 689–​695.
Koriyama T, Wongso S, Watanabe K, Abe H. 2002. Fatty acid
compositions of oil species affect the 5 basic taste perceptions.
Journal of Food Science, 67, 868–​873.
Kuo YL, Pangborn RM, Noble AC. 1993. Temporal patterns of nasal,
oral, and retronasal perception of citral and vanillin and inter-
action of these odourants with selected tastants. International
Journal of Food Science and Technology, 28, 127–​137.
Kurihara K, Beidler LM. 1968. Taste-​modifying protein from mir-
acle fruit. Science, 161, 1241–​1243.
Liem DG, Bogers RP, Dagnelie PC, De Graaf C. 2006. Fruit con-
sumption of boys (8–​11 years) is related to preferences for sour
taste. Appetite, 46, 93–​96.
Liem DG, Mennella JA. 2002. Sweet and sour preferences during
childhood: Role of early experiences. Developmental
Psychobiology, 41, 388–​395.
Liem DG, Mennella JA. 2003. Heightened sour preferences during
childhood. Chemical Senses, 28, 173–​180.
Liem DG, Russell CG. 2019. The influence of taste liking on the con-
sumption of nutrient rich and nutrient poor foods. Frontiers in
Nutrition, 6, 174.
Lugaz O, Pillias AM, Boireau-​Ducept N, Faurion, A. 2005. Time-​
intensity evaluation of acid taste in subjects with saliva high
flow and low flow rates for acids of various chemical proper-
ties. Chemical Senses, 30, 89–​103.
Martin N. 2002. Sweet/​sour balance in champagne wine and depend-
ence on taste/​odour interactions. Food Quality and Preference,
13, 295–​305.
Neyraud E, Sayd T, Morzel M, Dransfield E. 2006. Proteomic ana-
lysis of human whole and parotid salivas following stimula-
tion by different tastes. Journal of Proteome Research, 5,
2474–​2480.
Norris MB, Noble AC, Pangborn RM. 1984. Human-​saliva and taste
responses to acids varying in anions, titratable acidity, and pH.
Physiology & Behavior, 32, 237–​244.
12
Okumura T, Yamada R, Nishimura T. 2004. Sourness-​suppressing
peptides in cooked pork loins. Bioscience, Biotechnology, and
Biochemistry, 68, 1657–​1662.
Pfeiffer JC, Hort J, Hollowood TA, Taylor AJ. 2006. Taste-​aroma
interactions in a ternary system: A model of fruitiness percep-
tion in sucrose/​acid solutions. Perception & Psychophysics,
68, 216–​227.
Pionnier E, Chabanet C, Mioche L, Taylor AJ, Le Queré JL, Salles
C. 2004a. In vivo nonvolatile release during eating of a
model cheese: Relationships with oral parameters. Journal of
Agricultural and Food Chemistry, 52, 565–​571.
Pionnier E, Nicklaus S, Chabanet C, Mioche L, Taylor AJ, Le Quéré
JL, Salles C. 2004b. Flavour perception of a model cheese:
Relationships with oral and physico-​ chemical parameters.
Food Quality and Preference, 15, 843–​852.
Salles C, Nicklaus S, Septier C. 2003. Determination and gustatory
properties of taste-​active compounds in tomato juice. Food
Chemistry, 81, 395–​402.
Sediva A, Panovska ZK, Pokorny J. 2004. Effect of viscosity on
the perceived intensity of acid taste. Czech Journal of Food
Sciences, 22, 143–​149.
Christian Salles
Stampanoni CR. 1993. The “quantitative flavour profiling” tech-
nique. Perfumer & Flavourist, 18, 19–​24.
Stampanoni CR, Noble AC. 1991. The influence of fat, acid, and
salt on the perception of selected taste and texture attributes
of cheese analogs: A scalar study. Journal of Texture Studies,
22, 367–​380.
Steiner JE, Glaser D, Hawilo ME, Berridge KC. 2001. Comparative
expression of hedonic impact: Affective reactions to taste
by human infants and other primates. Neuroscience and
Biobehavioral Reviews, 25, 53–​74.
Stevenson RJ, Prescott J, Boakes RA. 1999. Confusing tastes and
smell: How odours can influence the perception of sweet and
sour tastes. Chemical Senses, 24, 627–​635.
Tu YH, Cooper AJ, Teng B, Chang RB, Artiga DJ, Turner HN,
Mulhall EM, Ye W, Smith AD, Liman ER. 2018. An evolu-
tionarily conserved gene family encodes proton-​selective ion
channels. Science, 359, 1047–​1050.
Anthocyanins in Food
Véronique Cheynier
SPO, INRAE, University of Montpellier, Institut Agro-Montpellier SupAgro, 2, Place Viala, 34060 Montpellier, France
Anthocyanins are a major group of natural pigments that
confer red, purple, blue, and even black colour on plants. These
compounds, belonging to the polyphenol family, are essential
components of the human diet, contributing to the attractive-
ness of fruits, vegetables, and beverages such as wine, and to the
health benefits associated with the consumption of plant-​derived
products. Anthocyanins are also widely exploited by the food
industry as natural colourants and putative bioactives.
Anthocyanins are found in numerous fruits (Wu and Prior,
2005a) as well as in some vegetables, nuts (Wu and Prior, 2005b),
and cereals (Wu and Prior, 2005b; Zhu, 2018). Most anthocyanins
in foods are based on six aglycones called anthocyanidins, usually
glycosylated in the 3-​position (Figure 2.1). Known exceptions
are 3-​deoxyanthocyanidins (e.g., luteolinidin and apigeninidin),
found in some plant sources such as sorghum, which lack the
hydroxyl group at the C-​3 position (Wu and Prior 2005b), and
cyanidin 4′-​ O-​glucoside derivatives, identified in red onions
(Fossen et al., 2003).
The large anthocyanin diversity is due to variations in the
nature of the anthocyanidin skeleton, i.e., number and positions
of hydroxyl and methoxyl groups (Figure 2.1) and in the
glycosylation and acylation patterns, i.e., nature, number, and
positions of sugar and acyl substituents (Andersen and Jordheim,
2006). Conjugation with sugars and acids can result in very com-
plex structures, e.g., in red cabbage (Wu and Prior, 2005b). Some
examples of anthocyanin structures are presented in Figure 2.1.
Moreover, further complexity and diversity result from antho-
cyanin reactions taking place during processing, cooking, and
storage of plant-​derived foods and beverages.
Colour Versatility of Anthocyanins
Anthocyanins are usually described as red flavylium cations.
The colour of these pigments shifts from orange to purple as the
number of phenolic hydroxyl groups increases (from pelargonidin
to delphinidin). Their stability is increased by methylation of the
phenolic groups, by the lack or glycosylation of the hydroxyl
group at the C-​3 position, and by acylation of the sugar, espe-
cially with phenolic acids.
However, the flavylium cation form is prevalent only under
highly acidic conditions. When dissolved in water, it readily
converts to other forms through hydration and proton transfer
reactions (Figure 2.2) (Trouillas et al., 2016). Thus, as the pH
increases, flavylium cations are deprotonated to blue quinonoidal
bases or hydrated to colourless water adducts (called hemiketals
or hemiacetals). The 2-​water adduct (the most abundant) is in
fast equilibrium with its open chain isomer (cis-​chalcone, CE),
which is in slow equilibrium with the trans-​isomer (Trouillas
et al., 2016).
These equilibria are characterized by thermodynamic
constants, K (hydration constant), relating the flavylium cation
and the mixture of hydrated forms (hemiketals and chalcones),
and K (deprotonation constant), relating the flavylium cation with
the mixture of tautomeric neutral quinonoidal bases, which can be
determined by spectrophotometry. Most common anthocyanins
(except highly acylated ones) have pKh values (i.e., the pH
value at which the anthocyanin is evenly distributed between the
flavylium and hydrated forms) in the range 2–​3, meaning that
they are present mostly in colourless hydrated forms in mildly
acidic solutions such as fruit juices and wine, unless some
pigment-​stabilizing mechanism is involved.
Like water, sulfite reversibly adds onto the flavylium cation to
form a colourless product, which is converted back to the flavylium
under acidic conditions. This explains why fumigation with
sulphur dioxide, used to control post-​harvest diseases and prevent
browning on fruits such as litchi, results in significant bleaching.
Colour Stabilization through Interactions
Colour stabilization is provided by molecular interactions of
the anthocyanin chromophore with other compounds, referred
to as copigments. A copigment can be an identical anthocyanin
molecule (self-​association), one of its aromatic acyl group
substituents (intramolecular copigmentation), or another mol-
ecule (intermolecular copigmentation). The major driving force
is π–π stacking between the planar pigmented forms (flavylium
cation or quinonoidal base) and another planar molecule, forming
complexes from which water is excluded (Brouillard et al.,
1989; Trouillas et al., 2016). Under conditions where hydrated
forms predominate, these interactions result in displacement
of the hydration equilibrium toward the pigmented forms and
hence in colour enhancement (hyperchromic effect) and often
13
14
FIGURE 2.1 Chemical structure of anthocyanidins (left) and examples of anthocyanins (right); malvidin 3-​glucoside and cyanidin 3-​(sinapoyl)diglucoside
5-​glucoside are particularly abundant in grape and red cabbage, respectively.
a small bathochromic shift (shift toward higher wavelengths,
from red to purple). Conversely, addition of copigments has
little effect on anthocyanins which are present in foods mostly
in pigmented forms such as 3-​deoxyanthocyanins (Awika, 2008),
or anthocyanins acylated with phenolic acids, which are already
stabilized by intramolecular copigmentation (Yoshida et al.,
1991). Moreover, metal ions bind with anthocyanins, showing
ortho di-​or tri-​hydroxy substitution, and may contribute to stabil-
ization of copigmentation complexes and the expression of blue
colours (Trouillas et al., 2016).
Anthocyanin Reactions
Anthocyanins are highly reactive, yielding numerous products
with extremely diverse colour properties. Discolouration and
browning can be observed following bruising or cutting of plant
tissues containing anthocyanins, due to enzymatic reactions (e.g.,
oxidation, catalysed by polyphenoloxidases and peroxidases, or
loss of sugar and acyl substituents, catalysed by glycosidases and
hydrolases, respectively). Anthocyanin reactions can also occur
chemically, especially during heating under acidic conditions
or as a result of oxygen exposure during food preparation and
storage. Anthocyanidins may also be generated by acid-​catalysed
cleavage of condensed tannins (syn. proanthocyanidins), another
major group of phenolic compounds, resulting in the appearance
of red colouration.
Anthocyanin reactions have been particularly explored in wine,
as they are responsible for the colour changes from red to purple
Véronique Cheynier
and then to tawny observed during ageing of red and rosé wines.
Several reaction mechanisms involving other food components
such as flavanols (i.e., catechins and proanthocyanidins, also called
condensed tannins), hydroxycinnamic acids, or aldehydes (e.g.,
acetaldehyde, which can form in planta, during fermentation, or
through oxidation of ethanol, furfural, and hydroxymethylfurfural
(HMF) formed from sugars during heating) have been established
in this context and later shown to occur in fruits and other food
products.
Anthocyanins can react both as electrophiles (i.e., able
to react with electron-​ rich species), because of the positive
charge of the flavylium cation, and as nucleophiles, in their
hemiketal form, showing an electron excess, in the C6 and C8
positions. This ambivalence enables a variety of reactions. The
nucleophilic hemiketal adds to electrophiles such as flavylium
cations, carbocations formed by acid-​ catalysed cleavage of
tannins, and o-​quinones resulting from enzymatic oxidation of
hydroxycinnamic acids, yielding anthocyanin dimers, tannin–​
anthocyanin adducts, or hydroxycinnamic acid–​ anthocyanin
adducts, respectively. Anthocyanins in their hemiketal form can
also undergo condensation with aldehydes (R-​ CHO) such as
acetaldehyde, glyoxylic acid, furfural, hydroxymethyl furfural,
or vanillin, yielding R-​ methine anthocyanin oligomers or R-​
methine flavanol–​anthocyanin adducts, if flavanols (monomers
or condensed tannins) are also involved. In all these products,
the anthocyanin units are initially in the hemiketal form but can
dehydrate to the flavylium cation.
The flavylium cation undergoes nucleophilic addition of
flavanols, yielding colourless flavenes, which can oxidize to the
Anthocyanins in Food 15

R3' R3 '
OH OH
OH
+ + OH
HO O HO O
R5' R5 '
OH
OR OR HO O
O O R3'
pyranoanthocyanin CH3
OH pyranoanthocyanin OH
R'
OH +
HO O HO O
OH R5'
R3'
OH O OR
OH
portisin OH
HO O
R3' R5'
OH methyl-methine
OR tannin-anthocyanin adduct
+
R3' Ka - H+ OH
HO O neutral quinoinoid base
R5'
OH

R3' OR +
HO O
HO O R5 ' OH
Kh
OR HO O
+ H2O / - H+
R5' OH
OH R3' R3'
flavylium cation
O O OH OH OH
HO
OH +
pyranoanthocyanin dimer HO O HO O
R5 ' R5'
OH
OR hemiketal OR
pyranoanthocyanin dimer OH OH

Tannin-anthocyanin adduct
R3'
R3' OH
OH O
HO OH
HO O R5 '
R5'
OR cis-chalcone
R3
OR OH ring cleavage
OH
OH cis-chalcone OH
O O
OH HO
R5
R3'
OH O
HO OH
HO OH OH OH
OH OR
O
A-type anthocyanin-tannin adduct R5'
OH
OH O
trans-chalcone

FIGURE 2.2 Structural transformations of anthocyanins in solution. (Colours of structures correspond approximately to typical colours.)

flavylium (Salas et al., 2003) or rearrange to another colourless reactions. Thus, over 160 compounds have been detected after a
form through formation of an A-​type bond (Remy-​Tanneau et al., few hours of reaction of a very simple solution consisting of an
2003). Anthocyanin oligomers based on the A-​type structures anthocyanin, a flavanol monomer, and acetaldehyde, confirming
have been detected in grapes (Vidal et al., 2004) and wine (Salas the occurrence of such reaction cascades (Vallverdu-​ Queralt
et al., 2005) and recently formally identified (Oliveira et al., et al., 2017).
2013). The flavylium cation can also react with compounds
having a polarizable double bond to form other pigments called
pyranoanthocyanins. These include phenylpyranoanthocyanins,
carboxypyranoanthocyanins (vitisins A), pyranoanthocyanins Colour Changes Induced by Anthocyanin Reactions
(vitisins B), formed by addition of vinylphenol (Fulcrand Anthocyanin reaction products show a wide range of colour
et al., 1996) or hydroxycinnamic acids (Schwarz et al., 2003), properties. Thus, tannin–​anthocyanin adducts are red and sus-
of pyruvic acid (Fulcrand et al., 1998), and of acetaldehyde ceptible to water and sulfite addition, like their anthocyanin
(Cheynier et al., 1997), respectively, onto the anthocyanin. The precursors (Salas et al., 2004), while anthocyanin–​tannin adducts
list of pyranoanthocyanins is constantly expanding. Moreover, are colourless. In contrast, pyranoanthocyanins and condensa-
pyranoanthocyanin dimers have been recently identified (Oliveira tion products with aldehydes are orange and purple pigments,
et al., 2010). Finally, nucleophilic attack of water or hydrogen respectively, while further reactions of pyranoanthocyanins yield
peroxide onto the flavylium induces opening of the C-​ring, which blue portisins and even turquoise pyranoanthocyanin dimers
can subsequently cleave (Vallverdu-​ Queralt et al., 2016a) or (Figure 2.2). Moreover, these pigments show increased resist-
rearrange (Es-​Safi et al., 2008). ance to bleaching induced by water or sulfite addition. Water
All these reactions occur simultaneously, yielding a large diver- addition on pyranoanthocyanins is negligible, so that their preva-
sity of products as intermediate products are involved in further lent form in mildly acidic solution is the neutral quinonoidal base
16
(de Freitas and Mateus, 2011). Aggregation of the red quinonoid
base of some pyranoanthocyanins yields an intense blue colour
and eventually leads to precipitation (Vallverdu-​Queralt et al.,
2016b). Similarly, ethyl-​linked anthocyanin derivatives resulting
from condensation with acetaldehyde show high resistance to
hydration and sulfite bleaching, and are present as pigments
(flavylium cation and quinonoid base) in mildly acidic beverages
(Duenas et al., 2006).
Anthocyanin reactions and hence the colour depend on the
medium composition and on processing or preparation conditions.
For example, heating increases conversion of the hemiketal
to the chalcone (Brouillard, 1982), which is easily cleaved to
smaller colourless molecules and promotes rearrangements of
the anthocyanin molecules, generating yellow pigments (Es-​Safi
et al., 2008). Oxidation steps (catalysed by metal ions such as
iron or copper) are involved in numerous reaction mechanisms,
including condensation with aldehydes and formation of
pyranoanthocyanins. Oxidation can also induce cleavage of the
anthocyanidin skeleton, yielding colourless breakdown products,
especially at higher pH values and higher temperature (Vallverdu-​
Queralt et al., 2016a).
Pigment Diffusion and Changes during Cooking
Anthocyanins are water-​ soluble pigments and thus easily
extracted with water. Most fruits contain anthocyanins only in
their skins, but some species or varieties also have pigmented
flesh. Anthocyanin extraction is easier from the vacuoles of flesh
cells than from those of skin cells, due to differences in tissue
structure, so that varieties with coloured flesh are more suitable
for the preparation of red juices. Physical processes, e.g., heating
and maceration, or the use of pectolytic enzymes that degrade
plant cell walls, are required to release pigments from skins.
Anthocyanin extraction, for instance in the preparation of
hibiscus beverages (known as bissap in West Africa, karkade
in Egypt, and agua de Jamaica in Mexico), which are highly
popular in Africa and Latin America, increases with infusion
time and water temperature. Thus, colour density (calculated as
(Absorption at 420 nm − Absorption at 700 nm) + (Absorption at
520 nm − Absorption at 700 nm), i.e., the sum of yellow and red
FIGURE 2.3 Colour of red cabbage extract with (A) or without (B) lemon juice (pH effect on reversible acid–​base equilibrium) and changes induced by
heating in a water bath for 30 minutes (C: with lemon juice, no change, D: without lemon juice, irreversible discolouration).
Véronique Cheynier
colours) increased two-​fold between 30 and 120 minutes of infu-
sion at 20 °C, while identical values were obtained after 4 minutes
at 90 °C and 1 hour at 20 °C (Ramirez-​Rodrigues et al., 2011).
Colour quality is also affected by temperature, higher infusion
temperatures resulting in a shift from red to orange reflected by
an increase in hue tint (i.e., (Absorption at 420 nm − Absorption
at 700 nm)/​(Absorption at 520 nm − Absorption at 700 nm), the
ratio of yellow to red), reflecting degradation of the anthocyanin
structure.
The colour of red cabbage extracts containing highly acylated
anthocyanins is much more resistant than that of hibiscus extracts
(Fernandez-​Lopez et al., 2013); this enhanced stability is due
to protection of the anthocyanin against nucleophilic attack by
water and hydrogen peroxide and subsequent ring opening by
intramolecular copigmentation. Addition of copigments can also
provide some protection (Trouillas et al., 2016). For example,
ellagitannin copigments increased the heat and colour stability
of pomegranate anthocyanins (Fischer et al., 2013). The best
stability was observed in juices with high initial anthocyanin
contents and, in model solution, for a copigment to pigment ratio
of 2:1.
Cooking of red cabbage pigments, especially with alkaline
water, results in a shift toward purple hints (due to conversion
to blue quinonoidal forms) and to discolouration (conversion
to hemiketal and chalcone forms followed by cleavage of the
heterocyclic C-​ring) upon heating (Mateus et al., 2003). The
former is reversible, while the latter is irreversible but can be
prevented by adding acid such as lemon juice, as illustrated in
Figure 2.3.
Interaction of extracted anthocyanins with the solid plant
material is another conspicuous phenomenon. Thus, red cabbage
anthocyanins localized in the external layers of the leaves are
released during cooking and absorbed on the whole surface,
which becomes uniformly pigmented. This may be related to their
particular structure, as it has been recently shown that acylated
anthocyanins are particularly prone to interacting with protein
(Gil et al., 2017). However, it may also reflect more complex
mechanisms, involving extraction of anthocyanins, their subse-
quent reactions with other plant components such as tannins, as
detailed earlier, and adsorption of the resulting tannin-​like poly-
meric pigments onto the plant matrix proteins and cell walls.
Anthocyanins in Food
Finally, some observations such as the reddening of pears after
cooking in acidic medium remain unexplained, although one
might speculate that this involves the release of anthocyanidins
after acid-​catalysed cleavage of tannins, which are abundant in
the flesh of some pear varieties.
REFERENCES
Andersen OM, Jordheim M. 2006. The anthocyanins. In Flavonoids:
chemistry, biochemistry and applications, ed. Andersen O
and Markham K, 471–​552. Boca Raton, CRC Press, Taylor &
Francis group.
Awika J. 2008. Behavior of 3-​deoxyanthocyanidins in the presence of
phenolic copigments. Food Research International, 41, 532–​8.
Brouillard R. 1982. Chemical structure of anthocyanins. In
Anthocyanins as food colours, ed. Markakis P, 1–​40. New York,
Academic Press.
Brouillard R, Mazza G, Saad Z, Albrecht-​Gary AM, Cheminat,A.
1989. The copigmentation reaction of anthocyanins: a micro-
probe for the structural study of aqueous solutions. Journal of
the American Chemical Society, 111, 2604–​10.
Cheynier V, Fulcrand H, Sarni P, Moutounet M. 1997. Application
des techniques analytiques à l’étude des composés phénoliques
et de leurs réactions au cours de la vinification. Analusis, 25,
M14–​M21.
De Freitas V, Mateus N. 2011. Formation of pyranoanthocyanins in
red wines: a new and diverse class of anthocyanin derivatives.
Analytical and Bioanalytical Chemistry, 401, 1463–​73.
Duenas M, Salas E, Cheynier V, Dangles O, Fulcrand H. 2006. UV-​
Visible spectroscopic investigation of the 8-​8-​methylmethine
catechin-​malvidin 3-​glucoside pigments in aqueous solution:
structural transformations and molecular complexation with
chlorogenic acid. Journal of Agricultural and Food Chemistry,
54, 189–​96.
Es-​Safi N, Meudec E, Bouchut C, Fulcrand H, Ducrot P, Herbette
G, Cheynier V. 2008. New compounds obtained by evolution
and oxidation of malvidin 3-​O-​glucoside in ethanolic medium.
Journal of Agricultural and Food Chemistry, 56, 4584–​91.
Fernandez-​Lopez J, Angosto J, Gimenez P, Leon G. 2013. Thermal
stability of selected natural red extracts used as food colourants.
Plant Foods for Human Nutrition, 68, 11–​17.
Fischer UA, Carle R, Kammerer D. 2013. Thermal stability of
anthocyanins and colourless phenolics in pomegranate (Punica
granatum L.) juices and model solutions. Food Chemistry,
138, 1800–​9.
Fossen P, Andersen OM. 2003. Anthocyanins from red onion, Allium
cepa, with novel aglycone. Phytochemistry, 62, 1217–​20.
Fulcrand H, Cameira dos Santos P, Sarni-​Manchado P, Cheynier V,
Favre-​Bonvin J. 1996. Structure of new anthocyanin-​derived
wine pigments. Journal of the Chemical Society, Perkin
Transactions, I, 735–​9.
Fulcrand H, Benabdeljalil C, Rigaud J, Cheynier V, Moutounet M.
1998. A new class of wine pigments yielded by reactions
between pyruvic acid and grape anthocyanins. Phytochemistry,
47, 1401–​7.
Gil M, Avila-​ Salas F, Santos LS, Iturmendi N, Moine V,
Cheynier V, Saucier C. 2017. Rosé wine fining using
polyvinylpolypyrrolidone: colourimetry, targeted poly­
phenomics and molecular dynamics simulations. Journal of
Agricultural and Food Chemistry, 65, 10591–​7.
Mateus N, Silva AMS, Rivas-​Gonzalo JC, Santos-​Buelga C, De
Freitas V. 2003. A new class of blue anthocyanin-​ derived
pigments isolated from red wines. Journal of Agricultural and
Food Chemistry, 51, 1919–​1923.
17
Oliveira C, Azevedo J, Silva A, Teixeira N, Cruz L, Mateus N, De
Freitas V. 2010. Pyranoanthocyanin dimers: a new family of
turquoise blue anthocyanin-​derived pigments found in port
wine. Journal of Agricultural and Food Chemistry, 58, 5154–​9.
Oliveira J, Alhinho da Silva M, Parola A, Mateus N, Brás N, Ramos
M, de Freitas V. 2013. Structural characterization of a A-​type
linked trimeric anthocyanin derived pigment occurring in a
young Port wine. Food Chemistry, 141, 1987–​96.
Remy-​Tanneau S, Le Guerneve C, Meudec E, Cheynier V. 2003.
Characterization of a colourless anthocyanin-​ flavan-​
3-​
ol
dimer containing both carbon-​carbon and ether interflavanoid
linkages by NMR and mass spectrometries. Journal of
Agricultural and Food Chemistry, 51, 3592–​7.
Salas E, Fulcrand H, Meudec E, Cheynier V. 2003. Reactions of
anthocyanins and tannins in model solutions. Journal of
Agricultural and Food Chemistry, 51, 7951–​61.
Salas E, Le Guernev C., Fulcrand H, Poncet-​Legrand C, Cheynier
V. 2004. Structure determination and colour properties of a
newly synthesized direct-​linked flavanol-​anthocyanin dimer.
Tetrahedron Letters, 45, 8725–​9.
Salas E, Dueñas M, Schwarz M, Winterhalter P, Cheynier V, Fulcrand
H. 2005. Characterization of pigments from different high
speed countercurrent chromatography wine fractions Journal
of Agricultural and Food Chemistry, 53, 4536–​46.
Schwarz M, Wabnitz TC, Winterhalter P. 2003. Pathway leading to
the formation of anthocyanin-​vinylphenol adducts and related
pigments in red wines. Journal of Agricultural and Food
Chemistry, 51, 3682–​7.
Trouillas P, Sancho-​García J, de Freitas V, Gierschner J, Otyepka
M, Dangles O. 2016. Stabilizing and modulating colour
by copigmentation: insights from theory and experiment.
Chemical Reviews, 116, 4937–​82.
Vallverdú-​Queralt A, Meudec E, Sommerer N, Dangles O,
Cheynier V, Le Guernevé C. 2016a. A comprehensive inves-
tigation of guaiacyl-​ pyranoanthocyanin synthesis by one-​ /​
two-​dimensional NMR and UPLC-​DAD-​ESI-​MSn. Food
Chemistry, 199, 902–​10.
Vallverdú-​Queralt A, Biler M, Meudec E, Le Guernevé C, Vernhet
A, Mazauric J, Legras J, Loonis M, Trouillas P, Cheynier V,
Dangles O. 2016b. p-​Hydroxyphenyl-​pyranoanthocyanins: an
experimental and theoretical investigation of their acid –​base
properties and molecular interactions. International Journal of
Molecular Sciences, 17, 1842.
Vallverdú-​Queralt A, Meudec M, Eder M, Lamuela-​Raventos R,
Sommerer N, Cheynier V. 2017. The hidden face of wine poly-
phenol polymerization highlighted by high-​resolution mass
spectrometry. Chemistry Open, 6, 336–​9.
Vidal S, Meudec E, Cheynier V, Skouroumounis G, Hayasaka Y. 2004.
Mass spectrometric evidence for the existence of oligomeric
anthocyanins in grape skins. Journal of Agricultural and Food
Chemistry, 52, 7144–​51.
Wu LC, Prior R. 2005a. Systematic identification and characteriza-
tion of anthocyanins by HPLC-​ESI-​MS/​MS in common foods
in the United States: fruits and berries. Journal of Agricultural
and Food Chemistry, 53, 2589–​99.
Wu LC, Prior R. 2005b. Identification and characterization of
anthocyanins by high-performance liquid chromatography-
electrospray ionization-tandem mass spectrometry in common
foods in the United States: vegetables, nuts, and grains. Journal
of Agricultural and Food Chemistry, 53, 3101–​13.
Yoshida K, Kondo T, Goto T. 1991. Unusually stable monoacylated
anthocyanin from purple yam Dioscorea alata. Tetrahedron
Letters, 32, 5579–​80.
Zhu F. 2018. Anthocyanins in cereals: composition and health effects.
Food Research International, 109, 232–​49.
Alcoholic Beverages: Production, Trends, Innovations
Konstantin Bellut1, Kieran M. Lynch1 and Elke K. Arendt1,2
1 School of Food and Nutritional Sciences, University College Cork, Cork, T12 YN60, Ireland
2 APC Microbiome Ireland, University College Cork, Cork, T12 YN60, Ireland
Introduction
Alcoholic beverages have a very long history of production and
consumption. Their production by humans can be traced back
to about 7,000 BC, but it is very likely that prehistoric humans
had unknowingly consumed alcohol in overripe berries and fruits
for many thousands of years before that (Philipps, 2014). For
the production of alcohol, two things are required: a reasonable
amount of sugar, and yeast. Ethanol is one of the end products
of “alcoholic” fermentation by yeast. Yeast are small single-​cell
microorganisms, which are classified as members of the fungus
kingdom. Saccharomyces cerevisiae is the most well-​ known
and commonly used yeast species for alcoholic fermentations.
It has been harnessed as a trustworthy workhorse for alco-
holic fermentations since ancient times, and has since been
domesticated for its use in brewing, distilling, winemaking and
baking. If a beverage contains alcohol, it is the product of alco-
holic fermentation by yeast, which turn sugars into ethanol and
carbon dioxide as part of their metabolism.
The three most common classifications for alcoholic beverages
are beer, wine and spirits. In beer brewing, the substrate is a
sugary extract from malted grains, mostly barley, which is boiled
with hops and then inoculated with yeast. In winemaking, grapes
must be used as the substrate, and the fermentation usually
involves several yeast strains. Spirits are an indirect product of
alcoholic fermentation; exceeding the biological upper alcohol
limit achievable through fermentation, spirits are the distilled
product of alcoholic fermentation of various substrates.
Beer
Production Method and Styles
Only four ingredients are required to produce beer: water,
barley malt, hops and yeast. Beer brewing is a multi-​step pro-
cess (Figure 3.1). First, barley malt is crushed and mixed with
water in the mash tun. During mashing, the so-​called “mash” is
led through different temperature rests to release the sugars and
nutrients from the malt via the action of its endogenous enzymes.
After mashing, a filtration step is needed to separate the liquid
sugary extract from the remaining grain particles. To achieve sep-
aration, different techniques can be applied; in a lauter tun, the
grain particles build their own filter bed by sedimentation on a
false bottom, which acts as a sieve, and the clear extract, now
called “wort”, can be separated from the leftover grain particles.
This leftover grain is the so-​called “brewers’ spent grain” (BSG).
Alternatively, a mash filter separates the wort from the BSG in
a pressurized, membrane-​based process. During boiling to ster-
ilize the wort, hops are added to give beer its distinct aroma and
bitterness and to introduce hop-​derived anti-​microbial compounds
into the wort. After boiling, hop leftovers and denatured proteins
(“hot trub”) are removed physically in a “whirlpool” and the wort
is cooled and transferred into a fermentation tank. During fermen-
tation, after the addition of yeast, the yeast metabolizes the wort
sugars into ethanol and CO2, which can take between one and
three weeks. After fermentation and subsequent maturation, the
beer is usually filtered again and is then ready for filling and sale.
Through the almost infinite combinations of ingredients,
there is no limit to the brewer’s fantasy and creativity in cre-
ating different beers. Brewers can choose from over 270 hop
varieties, over 80 malt varieties and over 200 different brewing
yeast strains (Deutscher Brauer-​Bund e.V., 2016; Stempfl, 2016;
IHGC, 2018). In addition, in compliance with their countries’
legislation, brewers can add adjuncts such as sugars, unmalted
grains, fruits and spices to further individualize their beers.
However, the brewers’ creativity is not only limited to ingredients
but also includes variations in brewing practices. All process
steps, from mashing to fermentation and even maturation, can be
altered to influence the beers’ characteristics. Generally, beers are
classified into two main styles: lagers and ales. These come with
many different sub-​styles, based on the use of ingredients and
yeast, and brewing practices. Lager beer, the most popular beer
worldwide, is brewed using Saccharomyces pastorianus, a hybrid
sister of S. cerevisiae. It is generally fermented and matured at
lower temperatures compared with ales, which are brewed using
S. cerevisiae. Commonly used parameters to define beer styles,
besides lager and ale, are original and final gravity, bitterness,
alcohol content and colour.
Home Brewing
Home brewing, the brewing of beer on a small scale for personal,
non-​commercial purposes, has its roots in the very beginnings of
beer brewing. Historians believe that in the early days of brewing
19
20 Konstantin Bellut et al.

FIGURE 3.1 Illustration of the beer brewing process.

in the Mesopotamian and Egyptian cultures, beers were first

brewed at home. Only when towns developed did brewing grow
from a household enterprise to larger scales (Papazian, 2003).
Today, home brewing is practised all over the world by beer
enthusiasts who want to live out their experimental and creative
spirit or simply save money on their beer consumption. While
the early home brewers had to build their equipment themselves,
today home brewing kits exist in all varieties and sizes. A standard
home brewing kit is shown in Figure 3.2. Alongside the standard
home brewing kits, brewing equipment manufacturers have
developed high-​quality home brewing equipment, designed for
high functionality and convenience. The degree of automation of
such equipment can reach from temperature control and program-
ming to a fully automated system. Ingredients and equipment can
easily be sourced from local or online home brewing shops.
In terms of the choice of ingredients and the level of skill,
home brewers can choose between extract brewing and all-​grain FIGURE 3.2 Typical home brewing starter kit.
brewing. While the beginner would start with liquid or dry malt
extract, the more sophisticated home brewer will opt for all-​grain
brewing. can then easily be removed and drained upon completion of the
For all-​grain brewing, mashing is performed in a heated vessel mashing process, leaving behind only the sugary extract, which
or by mixing the crushed grain with hot water. Temperature con- can subsequently be boiled in the same vessel. Extract brewing
trol of some sort during mashing is necessary to lead the mash makes the whole process of mashing and lautering redundant,
through the desired temperature rests. Subsequently, a filtering since the extract itself, in its dry or liquid form, is already the
step (lautering) is required to separate the spent grain from the concentrated product of mashing. However, whether all-​grain or
sugary extract. With the “brew-​in-​a-​bag” method, mashing is extract, a subsequent boiling step is essential to sanitize the wort.
performed in what could be considered as a large tea bag. The bag Hops can be added in hop bags to facilitate the easy removal of
Alcoholic Beverages: Production, Trends, Innovations
hop residue after the boil. Pre-​hopped malt extract eliminates the
necessity of adding hops during the boil.
To separate the clear wort from hop residue and “hot trub”,
home brewers can create the whirlpool effect applied in commer-
cial brewing by vigorously stirring the wort. Subsequently, the
clear wort needs to cool before it can be inoculated with yeast. The
sophisticated and impatient home brewer can use special cooling
equipment such as an “immersion cooler”, a hollow stainless-​steel
spiral which can be connected to a water supply, to dissipate the
heat. An option that does not require additional equipment is to
fill the hot wort into the fermenter. By tumbling slightly, the hot
wort can even be used to sanitize the fermenter surfaces before
it is left alone to cool. Upon reaching the desired fermentation
temperature, yeast is added in the form of dried or liquid yeast.
Plastic buckets or glass or plastic carboys are well suited for use as
fermentation vessels. The vessels are closed with a fermentation
lock, which allows the escape of carbon dioxide produced during
fermentation but prevents oxygen or contaminants from entering
the fermentation vessel. At the end of the fermentation process, the
beer can be filled into sanitized bottles. Small, calculated amounts
of sugar are added, and the bottles are capped. The remaining live
yeast in the beer will consume this sugar and give the homebrew
its fizz in the closed bottles.
As with commercial brewing, a high standard of hygiene is
essential to get a safe, good-​tasting product. As well as equipment
and ingredients, sanitizing agents can be sourced in any home
brewing shop.
Home brewers are engaging in local home brewing clubs or
online home brewing communities, which serve as platforms to
share recipes and experiences, equipment and, importantly, home
brewed beer!
Craft Beer
According to the Brewers Association, a craft brewery is defined
as small (with an annual production <7 Mio. hL), independent
(< 25% owned by a non-​craft brewer) and traditional (with beer
flavours derived from traditional or innovative brewing ingredients
and their fermentation) (Brewers Association, 2019b). Born in
America in the 1970s, craft beer has developed into a global phe-
nomenon. Its primary concentration is in America and Europe,
with the United States being the largest craft beer producer,
followed by the United Kingdom. Since the turn of the millen-
nium, craft beer has been growing four times faster than total beer
production, a trend that is forecast to continue (Technavio, 2017).
In 2017, the number of breweries worldwide surpassed 19,000,
with 17,732, or 94%, being defined as craft breweries (Brewers
Journal, 2017). Such strong growth, fuelled by consumers’ desire
for exclusive, small-​scale, premium products, gave rise to the
emergence of new and old rediscovered beer styles (Baker, 2018).
In recent years, craft beer has been dominated by hop-​focused
beer styles such as India pale ales (IPA) and other pale ales.
These beer styles use an elevated amount of hops during boiling
to create a more pronounced bitterness. Additionally, in a process
called “dry-​hopping”, hops are added during the cold stages of
brewing (mostly lagering/​maturation), which gives the beers their
distinctly fruity, floral hop aroma. Another beer style, which has
21
actually been around for hundreds of years, has recently attracted
increasing attention; sour beers, such as “Berliner Weisse”,
“gose”, “lambic” and “Wild ales”, are celebrating their revival
among craft brewers who are looking beyond how hoppy a beer
can taste. To brew a sour beer, brewers harness lactic acid bacteria,
which produce lactic acid, in addition to the alcohol-​producing
yeast, giving the beers their characteristically tart taste. Also,
unconventional yeast (i.e., Brettanomyces) may be used for co-​,
or secondary, fermentations. Sour beers are sometimes infused
with fruits or brewed using fruit juice as an adjunct. Craft brewers
have succeeded in establishing high levels of quality, consistency
and innovation, expanding the minds of the beer consumers and
creating the most diverse brewing culture in the world. Beer is
not just an alcoholic beverage; for many consumers, craft beer
has become a lifestyle.
More Trends and Innovations
The skyrocketing number of new craft brewery openings in the
United States has gone hand in hand with the emergence of
new brewpubs (i.e., restaurant-​breweries selling at least 25% of
volume on site) (Brewers Association, 2019a). This trend has also
been observed in Europe. Brewpubs serve as a craft beer lovers’
Mecca, where they can indulge in locally produced craft beer
while enjoying a meal or a snack, which is often uniquely fitted
to the beers (Figure 3.3). In fact, another growing trend is food
pairing, the combination of a certain beer style with a certain dish
or ingredients.
With advances in production methods and a society that is
more conscious about health and well-​being, non-​alcoholic and
low-​alcohol beers (NABLAB) are on the rise (Bellut and Arendt,
2018). While more and more big breweries and craft breweries
are adding NABLAB to their portfolio, craft breweries focusing
solely on NABLAB have started to pop up, examples being
Bravus Brewing Co. and WellBeing Brewing Co. in the United
States (Washington Post, 2018) and Nirvana Brewery in the
United Kingdom (METRO, 2017).
In light of the global gluten-​free trend, brewers have been
exploring with naturally gluten-​free pseudocereals in brewing,
such as millet, buckwheat and quinoa. Quinoa Italia has created
FIGURE 3.3 Brewpub, with the brewhouse on display in the background.
22
FIGURE 3.4 Beer from 100% quinoa.
(Quinoa Italia, 2018)
a gluten-​free beer, which is brewed using 100% locally farmed
quinoa and matured in a bottle (Figure 3.4).
The IntelligentX Brewing Co. based in London is tapping into
the trends of individualism and digitalization by brewing beer
with the help of artificial intelligence (AI). Its smart algorithm
takes customer feedback into account to produce a constantly
evolving beer (IntelligentX Brewing Co., 2018).
Wine
Production Method and Styles
As with beer, the principles of winemaking date back thousands
of years. Only in the 1800s did our understanding of the micro-
bial nature of winemaking develop. There exist major philosoph-
ical differences about whether “wine is made in the vineyard” or
whether the grapes are simply the raw material in the hands of a
skilled winemaker, who crafts the wine in the cellar.
Winemaking begins with the removal of the leaves and some-
times the stems from the grapes, before the grapes are crushed to
yield the juice. The freshly crushed grape juice that contains the
pulp, skins and seeds is called the “must”, with the solid portion
being called the “pomace”. During the subsequent maceration
process, nutrients, flavourants and colourants are extracted from
the pomace. For white wine production, maceration is kept to a
minimum, the juice and pomace are quickly separated, and the
fermentation is initiated only after a few hours. For red wines,
maceration is prolonged and occurs simultaneously with the
alcoholic fermentation, which enhances the extraction of skin
Konstantin Bellut et al.
and seed phenolics that are responsible for the red wine’s flavour.
Contrary to popular belief, red wine inherits its rich colour from
the skin of the grapes and not its pulp. The red colour is achieved
by the prolonged maceration process, leading to the extraction of
anthocyanins from the grapes’ skin. In fact, most red grapes have
light-​coloured pulp and therefore give light-​coloured juice. White
wines made from red grapes are known as “blanc de noirs”. The
most popular example is Champagne, the sparkling wine from
the Champagne region of France.
Historically, fermentation started spontaneously due to
endogenous yeasts present on the grapes and in the fermentation
vessels, but today the juice is inoculated with one or more yeast
strains of known characteristics. Yeast are responsible not only
for the formation of alcohol but also for the creation of odorant
compounds that contribute to the overall bouquet of the wine.
Upon the termination of alcoholic fermentation, an optional
secondary “malolactic fermentation” may be applied to reduce
acidity by transformation of malic acid into lactic acid with the
aid of specific lactic acid bacteria. During the maturation process,
the wine bouquet develops further and the colour is stabilized.
Before bottling, the wine is sometimes fined to soften or reduce
its astringency and remove proteins capable of haze formation,
and usually filtered to further enhance its stability and clarifica-
tion (Jackson, 2014). A flow diagram of the winemaking process
can be seen in Figure 3.5.
A generally accepted system to classify wines does not exist.
Wines may be organized by their colour, alcohol, sugar and
carbon dioxide content, or by their stylistic, varietal or geograph-
ical origin (Jackson, 2014).
Trends and Innovations
The market research organization Euromonitor analysed
spending habits of consumers around the world and identified
several trends in the wine industry (Euromonitor, 2018). Unlike
the preceding generation of baby boomers, the generation of
millennials is more interested in experiences over material
goods and chooses unusual, unique wines over well-​established
premium brands (the drinks business, 2016). More uncommon
grape varieties and styles are gaining attention, as well as wines
from new, emerging geographical regions, such as China. Just as
in the craft beer industry, where cans are big business, making
packaging cheap, eco-​friendly and stylish, canned wine is on the
rise in the United States and Europe (the drinks business, 2017).
In addition, wine on tap is becoming more popular, with com-
panies like Vagabond Wines pioneering in London (Vagabond
Wines Ltd, 2018). So-​called “natural wines”, farmed organically
and processed naturally without the addition of sulfites, are also
growing in popularity.
In recent years, more and more “urban wineries” have started
to pop up all around the globe, but mostly in the United States and
Europe. The principal concept of urban wineries is the sourcing
of grapes, or grape juice, from local or further-​afield vineyards
and processing them in the urban winery, which usually owns
a central location in a big city. The offers in urban wineries are
usually not limited to the wine produced in-​house; they may also
offer experiences and education, such as winemaking or tasting
courses (American Association of Wine Economists, 2014).
Alcoholic Beverages: Production, Trends, Innovations
FIGURE 3.5 Schematic illustration of the winemaking process.
(Adapted from Jackson, 2014)
Distilled Drinks
Principal Production Method and Styles
There is a limit to the amount of alcohol that can be formed by
alcoholic fermentation, which is defined by the yeast strain’s
maximum alcohol tolerance. The most resistant yeast strains can
reach ethanol concentrations of around 20% alcohol by volume
(ABV). In order to overcome the biologically defined limit, a
process called distillation was developed. The principle of dis-
tillation is based upon the different boiling points of ethanol (ca.
78 °C) and water (ca. 100 °C). A basic distillation setup consists
of three parts: the still pot for heating the liquid; the condenser for
cooling the vapours; and the still receiver for collecting the distil-
late. With heating above the boiling point of ethanol but staying
below the boiling point of water, the produced vapour is high in
alcohol and can be condensed to recover a concentrated distillate.
The production of whisk(e)y includes production procedures
similar to that of beer. A cereal-​based mash, called “wash”, is
fermented and distilled. The distillate is then matured in oak
casks, which impart many aroma and flavour characteristics to
the whiskey, before blending, bottling and selling. By law, the
distillate must typically be matured for a minimum number of
years before it can be called whiskey; for example, in order to
be labelled as Irish whiskey, it must be matured in casks for a
23
minimum of three years, whereas bourbon must be matured for a
minimum of two. Brandy is yielded by distilling wine with subse-
quent maturing in casks. Rum is the distillate of fermented sugar
cane juice and molasses. Tequila and mezcal are the distillates of
fermented agave juice. Vodka is the distillate of fermented neutral
grain like wheat, barley, rye or potato. For the production of gin,
a neutral grain spirit from barley, corn, rye or wheat is macerated
with botanicals like juniper berries, which contribute to its fla-
vour as the major ingredient, and then distilled. Liqueurs and
creams are drinks made from a combination of water, alcohol,
sugar, fruits, and spices or herbs.
Trends and Innovations
Due to the increasing consumer demand for premium craft
products, which are usually made locally with limited scale and
distribution, the spirits industry is experiencing a craft spirits,
trend much like craft beer. Craft spirits are forecast to have enor-
mous growth over the next few years, with whisk(e)y (whiskey in
Ireland and America, whisky in Scotland) being the most sought-​
after variety of craft spirits. It is closely followed by gin, vodka
and rum (Grand View Research, 2017).
As a part of the omnipresent health and well-​being trend,
alcohol-​free distilled spirit brands have emerged in recent years.
Seedlip, which was founded in 2014 in the United Kingdom,
now produces three alcohol-​free, distilled spirits, which are sold
in over 20 international cities in their best restaurants, cocktail
bars, luxury hotels and high-​quality retailers (Seedlip Ltd, 2018).
More companies followed in their footsteps, like Ceder’s, Stryyk,
Surendran & Bownes and Whissin. In addition, Gordon’s has
introduced a ready-​to-​drink “Ultra Low Alcohol G&T” into the
market (The Spirits Business, 2018).
Category Blurring
As a result of the craft trend, consumers now expect and demand
more creative offerings that allow them to experience new tastes
and flavours, the potential of which may well lie between cat-
egory lines. Also known as cross pollination, category blurring
is happening in the whole beverage industry, but especially the
alcoholic beverage industry.
In a collaboration, Irish whiskey distiller Jameson (Irish
Distillers Pernod-​ Ricard) teamed up with a local craft beer
brewery. From an exchange of barrels, the brand Jameson
Caskmates was born, a blended Irish whiskey aged in wooden,
stout-​seasoned barrels. After this phenomenal success, an IPA
edition of Caskmates was released. The craft brewery on the
other end was able to serve craft beer matured in whiskey barrels
(Irish Distillers Ltd, 2018).
Beer and wine hybrids are also gaining popularity. Craft brew-
eries are adding grape must to their wort to introduce more fla-
vour and complexity to their brews, but it is also happening the
other way around. With hops at its peak in the craft beer trend,
winemakers are adding hops to their wine, creating dry-​hopped
wine (Vinepair, 2018). Apart from grape must, various fruit juices
or pureés are popular ingredients in the growing sour beer trend,
and harmonize with the sour and tart character of sour beers.
24 Konstantin Bellut et al.

However, fruit juice additions are not only limited to the sour METRO. 2017. We took a look around the UK’s first non-​alcoholic
beer style but are also added to pale ale styles to go hand in hand craft beer brewery. https://​metro.co.uk/​2017/​10/​11/​we-​took-​
with their fruity hop character. a-​look-​around-​the-​uks-​first-​non-​alcoholic-​craft-​beer-​brewery-​
6982690/​, last access 17 November 2019.
Papazian C. 2003. Beer, History, America and Homebrew. In
The Complete Joy of Homebrewing (Third Ed.), Harper
REFERENCES Resource, 5–​10.
American Association of Wine Economists. 2014. The Global Urban Philipps R. 2014. Alcohol –​a History. The University of North
Winery Crush: Model, Forecast and Prospect. www.wine-​ Carolina Press.
economics.org/​wp-​content/​uploads/​2014/​05/​AAWE_​WP156. Quinoa Italia. 2018. Quinoa Products. https://​quinoaitalia.com/​shop/​
pdf, last access 4 December 2020. it/​Marca_​Quinoa/​QUIBIR75cl_​Quinoa-​Birrificata-​75-​c, last
Baker K. 2018. Key trends in the global beer market, Conference access 15 December 2020.
International Beer Strategies, May 2018, London, UK. Seedlip Ltd. 2018. Seedlip. https://​seedlipdrinks.com/​, last access 17
Bellut K, Arendt EK. 2018. Chance and Challenge: Non-​ November 2019.
Saccharomyces Yeasts in Non-​ Alcoholic and Low Alcohol Stempfl W. 2016. Beeindruckende Welt der Aromen –​Die faszinierende
Beer Brewing: A Review. J Am Soc Brew Chem 77, 77–​91. Biervielfalt. Brauindustrie, Special edition: 500 Jahre
Brewers Association. 2019a. Craft Beer Industry Market Segments. Reinheitsgebot. http://​fzarchiv.sachon.de/​Zeitschriftenarchiv/​
www.brewersassociation.org/​statistics/​market-​segments/​, last Getraenke-​Fachzeitschriften/​Brauindustrie/​2016/​500_​Jahre_​
access 4 December 2019. Reinheitsgebot/​500_​Jahre_​50-​53_​Beeindruckende_​Welt_​der_​
Brewers Association. 2019b. Craft Brewer Definition. www. Aromen.pdf, last access 15 December 2020.
brewersassociation.org/​statistics/​craft-​brewer-​defined/​, last technavio. 2017. Craft Beer Market in Europe 2017–​2021. www.
access 4 December 2019. technavio.com/​report/​craft-​beer-​market-​in-​europe, last access
Brewers Journal. 2017. Craft Beer Surge: Top Ten Countries to Get 17 November 2019.
a Beer This St Patrick’s Day. www.brewersjournal.info/​craft-​ the drinks business. 2016. How can wine producers cater to the tastes
beer-​surge-​top-​ten-​countries-​to-​get-​a-​beer-​this-​st-​patricks-​ of millennials? www.thedrinksbusiness.com/​2016/​11/​how-​
day/​, last access 4 December 2019. can-​wine-​producers-​cater-​to-​the-​tastes-​of-​millennials/​, last
Deutscher Brauer-​Bund e.V. 2016. Das Reinheitsgebot –​Fragen access 17 November 2019.
und Antworten zum Reinheitsgebot. www.reinheitsgebot. the drinks business. 2017. Cans now make up a quarter of all craft
de/​ f ileadmin/​ u ser_​ u pload/​ a rtikelbilder/​ R einheitsgebot/​ beer sold in UK. www.thedrinksbusiness.com/​2017/​11/​cans-​
Homepage_​Reinheitsgebot_​-​_​Fragen_​und_​Antworten_​zum_​ now-​make-​up-​a-​quarter-​of-​all-​craft-​beer-​sold-​in-​uk/​, last
Reinheitsgebot.pdf, last access 4 December 2019. access 17 November 2019.
Euromonitor. 2018. Three Key Trends for Wine for 2018. www. The Spirits Business. 2018. Seven non-​ alcoholic “spirits” for
euromonitor.com/​, last access 17 November 2019. abstainers. Retrieved November 18, 2018, from www.
Grand View Research. 2017. Craft Spirits Market Size Worth $80.43 thespiritsbusiness.com/ ​ 2 018/ ​ 0 8/ ​ t op- ​ s even- ​ n on- ​ a lcoholic-​
Billion by 2025. www.grandviewresearch.com/​press-​release/​ spirits-​brands/​, last access 17 November 2019.
global-​craft-​spirits-​market, last access 17 November 2019. Vagabond Wines Ltd. 2018. Vagabond Wines. www.vagabondwines.
IHGC. 2018. IHGC Hop Variety List 2017. www.hmelj-​giz.si/​ihgc/​ co.uk/​, last access 17 November 2019.
activ/​may18.htm, last access 17 November 2019. Vinepair. 2018. Beer-​wine hybrids are seducing promiscuous drinkers.
Intelligent X Brewing Co. 2018. www.f6s.com/​intelligentxbrewingco, https:// ​ v inepair.com/ ​ a rticles/ ​ b eer- ​ w ine- ​ h ybrids-​ d rinks-
last access 4 December 2020. trend/​, last access 17 November 2019.
Irish Distillers Ltd. 2018. Jameson Caskmates. www.jamesonwhiskey. Washington Post. 2018. The search for a non-​alcoholic beer that’s
com/​en-​IE/​ourwhiskeys/​jameson-​caskmates, last access 17 actually worth drinking, from www.washingtonpost.com/​
November 2019. lifestyle/​food/​the-​search-​for-​a-​nonalcoholic-​beer-​thats-​actu-
Jackson RS. 2014. Fermentation. In Wine Science (Fourth Ed.,), ally-​worth-​drinking/​2018/​04/​27/​9965860c-​496d-​11e8-​827e-​
Academic Press, 427–​534, ISBN 9780123814685. 190efaf1f1ee_​story.html, last access 17 November 2019.
Ash in the Kitchen
Marta Ghebremedhin1, Christine Schreiber1, Bhagyashri L. Joshi1, Andreas Rieger2 and Thomas A. Vilgis1
1 Max-​Planck-​Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany
2 Restaurant einsunternull, Hannoversche Str. 1, 10115 Berlin, Germany
For thousands of years, grilling on an open fire has been a
unique cooking technique to produce foods that have a smoky
flavour. This procedure provides the characteristic roast flavour.
This tradition-​rich preparation method is used in every culture
around the world, and different cultures have developed a var-
iety of methods to optimize the roasting process of their dishes.
However, little attention has been given to what remains after the
cooking process. Accordingly, ash may be hiding interesting yet
unexplored properties (Myhrvold et al., 2011).
History
In ancient times, the Incas used ashes to enhance the flavour and
baking qualities of corn (Stella and Carasco, 2009; Cookingissues,
2020). This was cooked in alkaline water, then washed, husked and
further processed in a procedure called nixtamalization. Another
effect of using ash is the release of bound niacin (pyridine-​3-​
carboxylic acid), which makes this essential nutrient accessible
for human metabolism and prevents deficiency symptoms (Reilly,
2009). For instance, if corn is a main food source and is not treated
using the above-​mentioned method, deficiency diseases such as
pellagra can occur (Hampl and Hampl, 1997; INAO, 2019). This
particular disease existed in Europe and western Africa until the
relationship between corn and the disease was discovered.
Furthermore, ashes also have antimicrobial effects, which were
used by French peasants to make their cheese non-​perishable
(Franche-​Comte, 2019; Reiger, 2018). This cheese is still well
known as Morbier, named after the village where it was prepared.
Inspired by this method, one of us (Andreas Rieger), the chef of
the restaurant einsunternull in Berlin, developed a new variation
to mature meat. He used Danish ducks with a high fat content, thus
protecting them from drying out during maturation and enhancing
the flavour with interesting results (Oldenbourg, 2013). Changes
in the ripening and texture of meat due to the covering of the meat
with ashes will be discussed later in this chapter.
Preparation Method of Leek Ash
To get a better understanding of ash and its properties, the prepar-
ation method of leek ash will be considered. A. Rieger produced
the ash first by cleaning the outer and dark green parts of leeks
and then drying them overnight to lose moisture. Afterwards, the
leeks are burned at 300 °C using hot air in a convectomat (0% air
humidity and 80% blower speed) for 35 minutes. After cooling,
the ash is pulverized in a blender and passed through a hair sieve
(Oldenbourg, 2013). The ash has a nice earthy and smoky aroma
and still a scent of leek. It also tastes like soil and smoke and a
little bit salty. In Figure 4.1, photos of the leek ash are shown.
Characterization of Leek Ash
The properties of leek ash have been studied using different
techniques. First of all, optical microscopy was used to get
a closer look at the leek ash: in photomicrographs of leek ash
(Figure 4.2), it can be observed that the morphology does not
differ much from the morphology present in bright field mode.
On the right side, a picture was captured under polarized light
mode. Polarized light microscopy is used to study crystal
structures of different kind of materials. Using the polarizer filter,
the transmitted light is polarized linearly in one direction. When
this polarized light is transmitted by a birefringent specimen, it is
blocked by another filter, called the analyser, which is oriented at
90° to the polarizer. Crystalline regions of the specimen appear
bright only if the polarization is changed by the specimen due to
its birefringence or to stress in the material (Spiess et al., 2009).
Consequently, it can be observed that leek ash contains bright
crystalline structures. In Figures 4.2 and 4.3, these ash crystals
are shown in greater detail using higher magnification.
These figures show that there could still be some crystals
present even after burning the leek at 300 °C. Hence, to ensure
the crystallinity, X-​ray diffraction (XRD) was used, which is a
common method to characterize crystal structures (Bergmann
and Schaefer, 2005; Kühl and Linnemann, 2017). Figure 4.4
shows the graph of the XRD measurement. One can see an
increase and a decrease of the baseline and some sharp peaks.
The broad peak of the baseline is an indication of an amorphous
structure due to the presence of carbon, and the distinct peaks are
evidence for crystal structures, as they are present in salts and
minerals. Therefore, the XRD analysis also confirms the results
of the polarized light microscopy pictures.
25
26 Marta Ghebremedhin, Christine Schreiber et al.

FIGURE 4.1 Appearance of leek ash.

a) b)

FIGURE 4.2 Light microscope pictures of leek ash in (a) bright field mode and (b) polarized light mode.

FIGURE 4.3 Polarized light microscopy images of leek ash with higher magnification.

In order to determine the composition of leek ash, samples
were analysed with scanning electron microscopy (SEM). This
is a method to get detailed information about the surface top-
ography and the composition of a sample by using a focused
beam of electrons to scan the surface (McMullan, 1995; Gaisford
et al., 2016). Figure 4.5 shows SEM pictures with sizes ranging
from 300 to 500 µm and finer details than the light microscopy
pictures. Furthermore, the pictures show particles with a rough
surface while maintaining the typical morphology of ashes.
The following pictures (Figure 4.6) are micrographs with even
higher magnification. Some of them demonstrate patterns that do
not look like completely burned plant structures (a and b) and
vascular tissue (c, d, e and f). Furthermore, the last two pictures
(g and h) show rough and porous surfaces resulting in a massive
Ash in the Kitchen
surface enlargement. This property leads to a high adsorption of
liquids because of capillary forces within the large numbers of
infinitesimal indentations.
Another feature of SEM measurement is the analysis of elem-
ental composition (Figure 4.7b). Table 4.1 reveals an average per-
centage of elements in the focused spots (Figure 4.7a). The main
elements are carbon and oxygen, followed by potassium and cal-
cium and other elements such as phosphorus, sulphur, chlorine,
sodium, magnesium and zinc. This composition correlates
with the composition of ash in general, which contains oxides
and carbonates of different metals, for instance CaO, Fe2O3,
MgO, MnO, P2O5, K2O, SiO2, Na2CO3, NaHCO3, etc. These
components form crystal structures, which have already been
seen. This is due to highly ordered atomic arrangement, meaning
that their atoms are arranged in an ordered geometric pattern. For
a more exact determination of the composition, more methods
and measurements are required.
The plant remains can also be investigated by another meas-
urement: in thermogravimetric analysis (TGA), substances can
be burned under controlled conditions such as temperature rates,
350
300 Leek Ash
250
200
Intensity
150
100
50
0
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80
2 Theta
FIGURE 4.4 X-ray diffraction spectrum of crystals in leek ash.
FIGURE 4.5 Scanning electron microscopy pictures of leek ash (see scale).
27
different atmospheres or a variety of pressures. The weight of
a sample is measured over time as the temperature changes
until a constant weight is reached (Coats and Redfern, 1963;
Ardö, 2011). Figure 4.8 demonstrates the TGA measurement
of leek ash under nitrogen atmosphere and air. The small drop
at the beginning is due to adsorbed water, which evaporates. In
both measurements, the weight of the ash decreases, starting at
288 °C in a nitrogen atmosphere, and 247 °C under ambient air
conditions. This decrease confirms that the leek is not incinerated
entirely during the manufacturing process. It is also obvious that,
under nitrogen conditions, more mass remains, while in con-
trast, when it is heated under air, more sample is burned. This
is because oxidation reactions can occur due to the oxygen con-
tent of air, and the compounds can be burned to carbon dioxide.
Under a nitrogen atmosphere, this does not occur because of the
absence of oxygen, which leads to the incomplete burning of the
ash under nitrogen.
“Labmade” Cheese
The porous surface of ashed materials, which could be seen in
the SEM pictures, is responsible for the high adsorption capacity
of water and other substances. This property is a good platform
to develop edible mould. The enzymes produced from the edible
mould may lead to odour intensification. This particular effect of
edible mould is used in cheese maturing, e.g., in the production
of blue cheese, Roquefort, Camembert or Brie (Hayaloglu, 2016;
Brehme, 2019).
Cheese maturing with ashes was also attempted in our labora-
tory, as described in detail in the following (Kaesereibedarf,
2019; Carroll, 2010; Vilgis, 2018).
First, 1.5 L of fresh raw milk (Gill’s Weidenhof, Mainz
Bodenheim) was heated up to 36 °C, and six drops of liquid calf
rennet (Bunte Kuh Käsereibedarf) were added. The milk mixture
was left in the oven at a constant temperature of 36 °C for about
1 hour. The cheese curd was then cut into cubes and placed in
the oven at 36 °C for another 2 hours (Figure 4.9a). The cheese
mass was sieved through a cotton cloth and further liquid was
28 Marta Ghebremedhin, Christine Schreiber et al.

a) b)

c) d)

e) f)

g) h)

FIGURE 4.6 Scanning electron microscopy pictures of leek ash (see main text for explanation of the various pictures).
Ash in the Kitchen 29

a)

b)

FIGURE 4.7 Scanning electron microscopy analysis. (a) The microscopy image of leek ash particles, in which “focused spots” highlighted in green and red
were chosen to perform further elemental analysis. (b) The superimposed graph of the spectra of the measured elements referring to the spots in (a). The char-
acteristic peaks of the elements are marked in different colours.

pressed out (Figure 9b, c and d). The mass was formed into a
ball, halved and placed in a 25% salt bath for 2 hours with fre-
quent turning (Figure 9e and f). Finally, the cheese was patted
dry and one half covered with leek ash (Figure 9g and h). Both
halves were kept under a crystallizing dish, ensuring air circula-
tion. An earlier development of edible mould on the ash-​covered
cheese was observed (Figure 9i and following pictures). Both
cheeses developed a very intense odour. The ash-​covered cheese
had more of an earthy, musty odour.
Meat Maturation
The texture and flavour development of meat depends on the
muscle fibres and their components. Figure 4.10 illustrates a
muscle fibre, which is composed of capillaries, mitochondria,
cell nucleus, blood vessels, collagen and myofibrils. These
myofibrils consist mainly of actin and myosin proteins, which
build the sarcomeres and are responsible for muscle contraction
(Vilgis, 2015; Rayment et al., 1993).
30 Marta Ghebremedhin, Christine Schreiber et al.

TABLE 4.1
Elements of Leek Ash Detected by SEM
Abs. error (%) Rel. error (%)
Element Atomic number Net Mass (%) Mass norm. (%) Atom (%) (1 sigma) (1 sigma)
Carbon 6 25005 58.18 55.71 66.55 7.31 12.57
Oxygen 8 7888 33.65 32.22 28.89 4.90 14.58
Potassium 19 2396 5.34 5.12 1.88 0.25 4.63
Calcium 20 1716 5.68 5.44 1.95 0.28 4.88
Phosphorus 15 299 0.33 0.32 0.15 0.05 15.42
Sulphur 16 290 0.34 0.33 0.15 0.05 15.07
Chlorine 17 302 0.42 0.40 0.16 0.06 13.52
Sodium 11 162 0.15 0.15 0.09 0.04 28.39
Magnesium 12 427 0.33 0.32 0.19 0.05 15.78
Zinc 30 0 0.00 0.00 0.00 0.00 4.09
Sum 104.42 100.00 100.00

With longer maturing time, bonds are systematically cut, the

FIGURE 4.8 Thermogravimetric analysis measurement of leek ash.
After the death of an animal, the aerobic metabolic processes
and the production of adenosine triphosphate (ATP), which
provides energy for the muscle movements, are interrupted.
Therefore, the actin-​myosin complexes are immobilized and
the muscle becomes stiff and rigid as rigor mortis sets in
(Vilgis, 2015; Luther, 2009). During meat maturation, enzym-
atic softening of the structure hardened by rigor mortis is there-
fore highly desirable. Enzymes located in the muscle and in the
sarcoplasm are responsible for this process; these enzymes are
cathepsins and calpains, which have different modes of action
and interfaces. Calpains are only activated when calcium
ions are released, while cathepsins are activated post mortem
when the pH value is sufficiently low. The calpains act on the
bonds of the myofibrils at the M-​discs, while the cathepsins
act predominantly on the peptide bonds in the proteins of the
myofibrils, actin and myosin (Vilgis, 2015). This is indicated
in Figure 4.11.
cohesion of the fibrils becomes weaker, and the firmness and
elasticity decrease. Hence, the various proteins of the myofibrils
are cut at different places and the meat becomes more tender.
These processes are initiated after slaughter, progressive cooling
and the associated reactions in the phosphagen system. After
slaughtering, ATP continues to be produced and degrades to
inosine monophosphate (IMP) via adenosine diphosphate (ADP)
and adenosine monophosphate (AMP). IMP and glutamic acid
are synergistically responsible for the umami taste of foods. At
pH values around 6.5, calcium ions begin to be released, which
activate calpain. Only after the onset of rigor mortis (at about
pH = 6, when there is hardly any ATP left) do cathepsins begin
to become active. The pH value drops further, and the enzymes
break down proteins (Vilgis, 2015).
These biochemical processes are altered by coating the meat
with ash. Leek ash contains potash (potassium carbonate), which
increases the pH value because of its alkalinity. We confirmed
this by measuring pH over several days (pH = 7.8 ± 0.2). The
lowering of the pH value through the accumulation of protons
from ATP degradation is decelerated (“buffered”), and ATP deg-
radation, IMP formation and calpain activity are slowed down.
The muscles are first cut primarily at the protein bonds by
cathepsins. This altered pathway in the molecular processes can
explain the extraordinarily soft texture of the meat. The faster the
ash is applied after slaughter, the more pronounced these effects
may be (Vilgis, 2015).
A nice demonstration has been recently developed in the res-
taurant einsunternull (Oldenbourg, 2013).
Recipe for Ash-​Ripened Danish Duck (by Andreas
Rieger)
Ash-​Ripened Danish Duck (144 Portions)
12 Danish ducks (approx. 3000 g), plucked, without head and feet
leek ash (1000 g)
Ash in the Kitchen 31

a) b) c)

d) e) f)

g) h)

i) j)

k) l)
FIGURE 4.9 Cheese production showing (a) renneted milk, (b) cheese curd (c, d, e) after sieving and pressing out of liquid, (f) in saltwater bath, (g, h)
finished cheese and cheese (i, j) after 12 days and (k, l) after 20 days.
32
Remove the entrails and throat from the ducks. Open the
abdominal cavity of the ducks wide, push a meat hook through
the base of the neck and let it hang for 5 days at 0 °C (change the
draining trays for blood every day and make sure that the ducks
do not touch each other). Dismantle the ducks on the fifth day: cut
off the sebaceous gland and discard; cut off the wings; cut off the
legs, bone out hollowly and cut off the excess fat up to 10 mm with
poultry scissors; separate the breast on the carcass from the torso
with the scissors close to the ribs and cut the raven bone horizon-
tally to the breast with a boning knife; remove the overlapping fat
of the breast with the scissors. Roll breast and leg meat completely
in the ashes and tap off excess. Hang the breast with a meat hook
on the inside from the small hole and place the legs on the meat
side next to each other on a 1/​1 GN (Gastronorm) plate. Let both
mature for at least 10 days at 0 °C in the refrigerator and turn them
once every 2 days so that the meat is not wet.
Duck Cooking (Results in 8 Portions per Double
Breast/2–​3 Portions per Leg)
4 matured breasts or legs of duck
10 mL rape seed oil, raw vegetable quality
Iodized salt to taste
Sunflower oil for frying
FIGURE 4.10 Schematic of a muscle fibre.
(Vilgis, 2015)
FIGURE 4.11 Calpain and cathepsins, which are located in each muscle with different points of action.
(Vilgis, 2015)
Marta Ghebremedhin, Christine Schreiber et al.
Preheat the convector to 250 °C combined steam (25% humidity/​
40% blower). Slide a 1/​1 GN 65 mm tray into the bottom rack
and a 1/​1 GN 100 mm punch insert into the third rack from below.
Both meat pieces are taken out of the refrigerator shortly before
cooking, completely rubbed with rape seed oil and salted from all
sides. (See Figures 4.12 and 4.13.)
Chest
Place on the bone in the perforated inset and apply to cook for
11–​12 minutes. Then let it rest for at least 15 minutes –​maximum
25 minutes –​in the warming drawer at 60 °C. Fill a medium
pan with sunflower oil (about 10 mm) and heat to about 200 °C.
Remove the breast from the bone and fillet, place in the pan with
the skin side and fry until crispy golden yellow. Turn once briefly
before removing and fry the meat side for 3 seconds. Let the meat
rest again for 1 minute under the heat lamp and cut lengthwise
into 10 mm slices; salt.
Haunch
Place on the meat side in the perforated inset and cook for 12–​14
minutes. Then let it rest for at least 15 minutes –​maximum 25
minutes –​in the warming drawer at 60 °C. Fill a medium pan
with sunflower oil (about 10 mm) and heat to about 200 °C. Place
the skin side of the leg in the pan and fry until crispy golden-​
yellow. Turn once briefly before removing and fry the meat side
for 3 seconds. Let the meat rest again for 1 minute under the heat
lamp and cut lengthwise into 8 mm slices; salt.
Acknowledgements
The authors would like to thank Judith Hege and Markus
Ketomäki for helping with the cheese production, and Gunnar
Glaßer for helping us with the electron microscopy and elem-
entary analysis. We would like to thank Juan Carlos Zambrano
and Trivikram Nallamilli for critical reading of this article and
providing suggestions.
Ash in the Kitchen
a) b)
FIGURE 4.12 (a) The whole plucked duck after 1 week hanging. (b) Powdered ash is applied so that the duck breast is completely blackened. (c) The
hollowed and ashed legs, ready to ripen.
(Courtesy of Andreas Rieger)
FIGURE 4.13 Served ash-​ripened duck.
(Courtesy of Rene Riis for einsunternull)
REFERENCES
Ardö YM. 2011. Blue mold cheese, Encylopedia of Dairy Sciences,
Academic Press, London, UK, 767–​772.
Bergmann L, Schaefer C. 2005. Lehrbuch der Experimentalphysik
6. Festkörper (Vol. 6). Walter de Gruyter, Berlin, Germany.
Brehme J. 2019.http://​joerg.brehme.org/​wp-​content/​uploads/​2014/​
09/​068-​Alles-​käse-​und-​dabei-​selbstgemacht.pdf, accessed on
4 December 2020.
Carroll R. 2010. Home Cheese Making: Recipes for 75 Delicious
Cheeses. Storey Publishing, North Adams, MA.
Coats AW, Redfern JP. 1963. Thermogravimetric analysis. A review.
Analyst, 88(1053), 906–​924.
Cookingissues. 2020. www.cookingissues.com/​ index.html%3Fp=
5129.html, accessed on 4 December 2020.
33
c)
Franche-​Comte. 2019. www.franche-​comte.org/​Decouvrir/​
Gastronomie-et-​recettes/​Le-​fromage-​de-​Morbier, accessed on
19 February 2019.
Gaisford S, Kett V, Haines P (Eds.). 2016. Principles of Thermal
Analysis and Calorimetry. Royal Society of Chemistry,
London, UK.
Hampl JS, Hampl WS. 1997. Pellagra and the origin of a myth:
Evidence from European literature and folklore. Journal of the
Royal Society of Medicine, 1, 11.
Hayaloglu AA. 2016. Cheese: Microbiology of Cheese. In Reference
Module in Food Science. Elsevier, Amsterdam, The Netherlands,
1–​11. http://​dx.doi.org/​10.1016/​B978-​0-​08-​100596-​5.00675-​2
INAO. 2019. www.inao.gouv.fr/​eng/​produit/​13160, accessed on 19
February 2019.
Kaesereibedarf. 2019. http://​kaesereibedarf.de/​info/​grundrezept-​
fuer-​frischkaese/​, accessed on 20 February 2019.
Kühl S, Linnemann A. 2017. Grundlagen der Licht-​ und
Elektronenmikroskopie. UTB, Stuttgart, Germany.
Luther PK. 2009. The vertebrate muscle Z-​disc: Sarcomere anchor
for structure and signalling. Journal of Muscle Research and
Cell Motility, 30(5–​6), 171–​185.
McMullan D. 1995. Scanning electron microscopy 1928–​ 1965.
Scanning, 17(3), 175–​185.
Myhrvold N, Young C, Bilet M. 2011. Modernist Cuisine. Taschen,
Cologne, Germany.
Oldenbourg R. 2013. Polarized light microscopy: Principles and
practice. Cold Spring Harbor Protocols, 11, doi:10.1101/​pdb.
top078600.
Rayment I, Holden HM, Whittaker M, Yohn CB, Lorenz M, Holmes
KC, Milligan RA. 1993. Structure of the actin-​myosin com-
plex and its implications for muscle contraction. Science,
261(5117), 58–​65.
Reilly B. 2009. Disaster and Human History: Case Studies in Nature,
Society and Catastrophe. McFarland, Jefferson, NC.
Rieger A.2018. Im Dunkel verborgen. Aschegereifte Ente, Mirabellen
und Schmalz. Reifung als Kulturleistung. Journal Culinaire,
27, 85–​94.
34 Marta Ghebremedhin, Christine Schreiber et al.

Spieß L, Teichert G, Schwarzer R, Behnken H, Genzel C. 2009. Vilgis T. 2018. Ikejime versus karashi jukusei (dry aging): vielfältige
Moderne röntgenbeugung: röntgendiffraktometrie für molekulare Umami-​ Phasen:Riefung als Kulturleistung,
materialwissenschaftler, physiker und chemiker. Springer-​ Journal Culinaire, 27, 56–​84.
Verlag, Berlin, Germany. Vilgis TA.2015. Soft matter food physics –​the physics of food and
Steller JE, Carrasco M. 2009. Pre-​Columbian Foodways: cooking. Reports on Progress in Physics, 78(12), 34–​43.
Interdisciplinary Approaches to Food, Culture, and Markets in
Acient Mesoamerica. Springer-​Verlag, Berlin, Germany, 317.
Baking: Laminated Bakery Products
Roxane Detry, Christophe Blecker and Sabine Danthine
Food Science and Formulation, Gembloux Agro-​Bio Tech, University of Liège, Avenue de la faculté d’Agronomie 2B,
5030 Gembloux, Belgium
Classic pastry, such as shortcrust, choux or filo pastry, is obtained
from a homogeneous dough consisting of flour, water and fat,
sometimes supplemented with sugar, salt or eggs. Laminated
doughs, however, form a distinct category of pastry comprising
an alternation of up to hundreds of dough and fat layers. This par-
ticular structure is obtained by successive sheeting and folding
operations, referred to as the “lamination process”. During
baking, the layered dough transforms into a generously lifted,
aerated and flaky product, with a characteristic crispiness and
mouthfeel. Laminated bakery products include puff pastries, such
as vol-​au-​vent cases, and also fermented products like Danish
pastry and croissants.
From Basic Ingredients to a Sophisticated Pastry
The preparation of puff pastry starts with elaboration of a simple
dough, usually called the pre-​dough, traditionally made of wheat
flour, cold water (about 50–​60%), in-​dough fat (about 0–​15%)
and salt (0–​1%, flour based) (Corke et al., 2008; Ooms et al.,
2016). A large proportion of lamination fat, also called roll-​in
fat, is then added to the pre-​dough in an appropriate manner.
Based on flour weight, 50%, 75% or 100% of roll-​in fat is used
to prepare half, three-​quarter and full puff pastries, respectively.
In order to obtain the desired puffing effect, alternating layers of
intact pre-​dough and roll-​in fat should be created and maintained
during the whole pastry preparation.
Ingredient dosage influences pastry development and provides
each laminated product with its own specificities. Therefore,
while a puff pastry is typically very crispy, a croissant is crispy
on the outside but tender on the inside as a consequence of its
lower (in-​dough and roll-​in) fat content and the action of the
yeast. A Danish pastry is also softer than a puff pastry but richer
and softer than a croissant due to the addition of eggs and higher
levels of fat and sugar (Corke et al., 2008; Ooms et al., 2016).
Nonetheless, different types of laminated goods share a similar
preparation procedure.
Pre-​dough ingredients are first mixed to get a homogeneous
paste. Roll-​ in fat (stored at appropriate temperature before-
hand) and pre-​dough are then separately sheeted to about 7 mm
height before being combined. Two different main methods are
dedicated to the inclusion of the roll-​in fat inside the pre-​dough:
the French and English methods (Figure 5.1a and b). The main
dough obtained is then folded according to one or a combination of
folding procedures, the most widely used being the half turn (three-
fold) and the book turn (fourfold) methods (Figure 5.1c and d).
Sheeting and folding operations followed by resting periods are
repeated a certain number of times in order to get the desired
amount of layers. Before baking, Danish pastry and croissant
doughs are proved in appropriate temperature and humidity
conditions to favour fermentation and allow the dough to rise.
In the oven, high temperatures (above 200 °C) cause the water
present in the dough to evaporate. The characteristic lifting of
puff pastry is generally attributed to the entrapment of steam
inside the pastry due to the mechanical barrier created by fat
layers (Cauvain and Young, 2001; Deligny and Lucas, 2015;
Wickramarachchi et al., 2015). Nevertheless, while water vapour
is the essential leavening agent in puff pastries, the growth of
large gas bubbles obtained by fermentation is mainly respon-
sible for the lifting of croissants and Danish pastries (Ooms
et al., 2018a; Deligny and Lucas, 2015). Pastry height is further
influenced by the amount and type of roll-​in fat incorporated, as
well as the number of layers created (Cauvain and Young, 2001).
Ingredient Functionality for Pastry Lift
The development of a gorgeous evenly lifted crispy pastry results
from the interplay between the different constitutive ingredients
of laminated products, each possessing essential physicochemical
properties. A typical structure made of well-​separated and homo-
geneous alternating pre-​dough and roll-​in fat layers able to entrap
gasses from water evaporation and fermentation has been shown
to be the key for a suitable pastry lift. Furthermore, most prob-
ably, obtaining such a high-​quality layered system essentially
relies on roll-​in fat and pre-​dough rheology (Renzetti et al., 2016;
Ooms et al., 2017).
Pre-​dough
Similarly to bread, rheological properties of the pre-​dough such as
strength and extensibility are extremely important for the quality
35
36 Roxane Detry et al.

a) b)

c) d)

FIGURE 5.1 Schematic roll-​in fat incorporation methods: (a) the French method; (b) the English method. Schematic dough folding methods: (c) half turn;
(d) book turn.

of laminated bakery products. These are mainly determined by of proteins is typically used for the preparation of puff pastry and
the protein content and composition of the flour, the amount its fermented counterparts (Ooms et al., 2016). Gluten-​forming
of water added, the temperature and the mixing procedure. proteins, glutenins and gliadins, account for about 80% of total
Medium to strong wheat flour containing between 8% and 15% wheat proteins and play a key role in the establishment of dough
Baking: Laminated Bakery Products
rheological properties (Hay, 1993; Cauvain, 2015). Glutenins
are linear proteins that can form extremely large polymerized
protein strands, while gliadins are globular proteins (Delcour
et al., 2012).
It is thought that pastry dough development is closely related
to what is observed for bread dough (Ooms et al., 2016). The
mechanisms of dough formation have been extensively studied
but are still not fully understood. Upon addition of water and
energy to the flour (i.e., during pre-​dough preparation), protein
hydration allows the formation of a gluten matrix, comprising
long and entangled glutenin molecules, including smaller globular
gliadin molecules. With additional energy supply (i.e., with fur-
ther mixing or, in the case of laminated doughs, during sheeting),
glutenin strands are stretched. In addition, covalent bonds such
as disulphide bonds and non-​covalent bonds such as hydrogen
bonds and hydrophobic interactions dissociate and reconnect
to neighbouring protein molecules that are progressively being
aligned in the sheeting direction. The role of dityrosine bonds
in dough development is unclear (Tilley et al., 2001; Hanft
and Koehler, 2005). The network therefore reorganizes during
mixing and transforms into a visco-​elastic continuous mass with
improved strength and elasticity, a phenomenon called gluten
development. The pre-​dough possesses some elasticity and cohe-
siveness, imparted by glutenin molecules. On the other hand,
gliadins present in the network act as plasticizers and are respon-
sible for dough viscosity and extensibility (Delcour et al., 2012;
Cauvain, 2015).
The dough is allowed to relax during the resting periods
observed between the different sheeting steps of the lamination
process. At rest, covalent and non-​covalent bonds may reform,
and dough elasticity and strength decrease. Furthermore, dough
viscosity and extensibility increase due to the contribution of
gliadins to the network (Delcour et al., 2012; Cauvain, 2015;
Ooms et al., 2017). Pre-​ dough should be moderately elastic
to prevent dough shrinkage, but extensible and strong to with-
stand the sheeting steps and not disturb the layering (no tearing
or breaking). A good balance between glutenin and gliadin
molecules (intrinsic flour properties), as well as sufficient energy
input and resting, are therefore essential to get the appropriate
consistency.
In addition to adequate integrity and separation of the layers,
pre-​dough rheological properties should also enable gas reten-
tion in fermented laminated products such as Danish pastries and
croissants. The gluten network has to be sufficiently strong to
retain gas bubbles that have been trapped in the dough during
folding and allow them to grow without leakage during fer-
mentation (Ooms et al., 2018a). As the layered structure may
be disrupted due to bubbles rising inside the pre-​dough during
fermentation, fewer but thicker and therefore more robust layers
are typically created for fermented laminated products (Cauvain
and Young, 2001). As an example, puff pastry dough typically
contains about 100 to 250 fat layers, while only 18 to 50 layers
are usually present in fermented doughs (Cauvain and Young,
2001; Bent, 2007).
Starch granules represent the major flour constituents (65%
flour, dry weight based) and participate in increasing pre-​dough
viscosity as they also absorb water (Cauvain, 2015). Whether
37
starch and starch gelatinization during baking play a funda-
mental role in pastry structure and lifting is, however, not clear
yet (Ooms et al., 2016; Ooms et al., 2018b).
Water is added to the pre-​dough for several purposes. In add-
ition to allowing the formation of the gluten network and acting
as a leavening agent during baking, water is also used to control
pre-​dough temperature. Cold water is usually added in order to
cool the dough and avoid melting of the roll-​in fat during incorp-
oration. Finally, a small proportion of fat can also be added to
the pre-​dough to provide lubricity and facilitate lamination (Bent,
2007; Cauvain, 2015).
Roll-​in Fat
Historically, butter was mainly used as the lamination fat, but
it is now largely replaced by manufactured fat products such as
margarine and shortening, specifically designed for this applica-
tion (Ooms et al., 2016). Margarine is a structured water-​in-​oil
emulsion consisting of water droplets stabilized by a fat network
of solid fat crystals and liquid oil (Danthine, 2014). The term
“shortening” generally refers to any fat product made of (usually
modified) vegetable and/​or animal oils and fats, which may in
some cases contain water, designed to meet a wide diversity of
applications. Both shortenings and margarines may also include
emulsifiers and other additives such as antioxidants, colourings
and flavours (O’Brien, 2004). While it is certain that water vapour
from pre-​dough is critical for puff pastry lifting, there is currently
no scientific evidence on the role of water originating from the
lamination fat in the leavening phenomenon (Ooms et al., 2016).
Also, it has been shown that anhydrous shortenings provide
similar results and appropriate pastry development as compared
with shortenings containing water (Kriz and Oszlanyi, 1976).
It is generally accepted that the primary role of roll-​in fat in
puff pastry is to act as an impervious barrier that prevents water
vapour created during baking from escaping from the dough,
which allows the latter to swell. Although the exact funda-
mental origins of the lifting process are unknown, it is thought
to result from a combination of several mechanisms. Firstly,
water vapour created inside the pre-​dough layers is trapped and
expands in the gluten network (Podmore, 2002). In addition, a
portion of steam from the pre-​dough layers may migrate into the
molten fat layers, joining water liberated from the roll-​in fat, if
it contains any. Water vapour can therefore also expand between
the pre-​dough layers as growing gas bubbles surrounded by a
thin film of molten fat (Ooms et al., 2016). For Danish pastries
and croissants, similar mechanisms probably occur, with carbon
dioxide as the main leavening gas (Deligny et al., 2017; Lucas
et al., 2018) (Figure 5.2).
The rheology of the fat layers is of upmost importance for
the integrity of the layered structure. The roll-​in fat has to be
strong and plastic enough to stay intact during the sheeting and
folding steps. A brittle fat could break in the dough, while an
excessively hard fat could damage the pre-​ dough layers due
to the shear and pressure forces applied during the lamination
process (Cavillot et al., 2009; Wickramarachchi et al., 2015).
Similarly, a too soft and molten fat may become incorporated
in the pre-​dough layers and interrupt the layering. Therefore, in
38 Roxane Detry et al.

FIGURE 5.2 Schematic hypothetical mechanisms responsible for the leavening of laminated doughs. Upper and lower lines respectively represent the evolu-
tion of an unfermented and a fermented laminated dough from the final lamination step (left) until the end of baking (right). After lamination, fat layers may be
slightly disrupted and introduce breaks or thinner regions. Once the temperature is high enough in the oven, water from the pre-​dough transforms into steam,
expands (1) and tries to find a way out, pushing the melting fat layers aside (2) (Zoom 1). In addition, water present inside the roll-​in fat evaporates, creating
gas bubbles inside the fat. Later during baking, gas from dough and fat may merge to form large elongated bubbles (4) still visible after baking (Zoom 2). For
fermented doughs, carbon dioxide produced by the yeast expands (5) and may migrate into the fat layers (6) (Zoom 3). At the end of fermentation, large gas
bubbles are present in the fat layers (Zoom 4). During baking, it is mainly the expansion of those bubbles that enables the lifting of the pastry (7) (Zoom 5).
During the late stages of baking, the pastry dries out and bakes, fixing its structure. Mechanisms are inspired by Ooms et al. (2016), Deligny et al. (2017) and
Lucas et al. (2018).

order to achieve its mission, lamination fat must possess appro- in the mouth in order to avoid the “waxy taste” resulting from
priate firmness, plasticity and melting properties (Cavillot et al., an excessive amount of solid fat at body temperature (O’Brien,
2009; Wickramarachchi et al., 2015; Ooms et al., 2016). Those 2004). Besides, the fat melting in the mouth requires energy
macroscopic characteristics result from the interaction between taken as heat, which provides a feeling of freshness, typical of
several physicochemical properties of the crystal network: the the sensory experience associated with puff pastries.
organization of the network at different levels of structure (e.g.,
size and arrangement of the crystals and crystal aggregates),
polymorphism (three-​dimensional crystalline structure) and the Conclusion
amount of solid fat present (Cavillot et al., 2009; Ooms et al.,
As a result of their particular structure made of alternating pre-​
2016). These properties themselves depend on the nature of the
dough and roll-​in fat layers, laminated doughs transform into
triglyceride molecules and the processing conditions (Miskandar
a highly aerated pastry in the oven, illustrative of the so-​called
et al., 2005; Wickramarachchi et al., 2015).
“puffing effect”. The integrity and appropriate consistency of the
High consistency is usually achieved for a roll-​in margarine
pre-​dough and lamination fat layers are essential for adequate
by incorporation of a high amount of saturated fatty acids (high
pastry lift. In addition to the importance of the physicochemical
melting temperature), which often represent at least 50% of the fat
properties of the constituents, the laborious preparation pro-
formulation (Wickramarachchi et al., 2015). In order to obtain the
cedure needs to be carried out with care. Nevertheless, obtaining
desired plasticity and melting properties, these need to be carefully
the resulting delightful and sophisticated puffed product provides
selected and processing conditions adapted to the fat composition.
a feeling of satisfaction that encourages anyone who took up the
In addition to their important role in the formation of robust
challenge to do it again.
independent impervious fat layers, it has been shown that roll-​
in fat consistency and behaviour also influence the elasticity
and strength of the whole dough (Simovic et al., 2009; Renzetti REFERENCES
et al., 2016). Bent AJ. 2007. Speciality fermented goods. In Cauvain SP, Young LS
(eds.). Technology of breadmaking (Second Edition). Springer.
Although they should stay firm and unmelted during pastry
Cauvain S, Young L. 2001. Baking problems solved. Woodhead
preparation, the thin fat layers should also melt rapidly once Publishing.
Baking: Laminated Bakery Products
Cauvain SP. 2015. Principles of dough formation. In Cauvain SP
(ed.). Technology of breadmaking (Third Edition). Springer.
Cavillot V, Pierart C, De Meerendré MK, Vincent M, Paquot M,
Wouters J, Danthine S. 2009. Physicochemical properties
of European bakery margarines with and without trans fatty
acids. Journal of Food Lipids, 16(3), 273–​286.
Corke H, De Leyn I, Nip WK, Cross NA. 2008. Bakery products:
science and technology. John Wiley & Sons.
Danthine S. 2014. Propriétés physico-​chimiques et fonctionnalités
technologiques des matières grasses végétales. Quelques
applications alimentaires. In Lavelle C. (ed.) Science culinaire:
Matière, procédés, dégustation. Belin, Paris, 89–​104.
Delcour JA, Joye IJ, Pareyt B, Wilderjans E, Brijs K, Lagrain B.
2012. Wheat gluten functionality as a quality determinant in
cereal-​based food products. Annual Review of Food Science
and Technology, 3, 469–​492.
Deligny C, Collewet G, Lucas T. 2017. Quantitative MRI study of
layers and bubbles in Danish pastry during the proving pro-
cess. Journal of Food Engineering, 203, 6–​15.
Deligny C, Lucas T. 2015. Effect of the number of fat layers on
expansion of Danish pastry during proving and baking. Journal
of Food Engineering, 158, 113–​120.
Hanft F, Koehler P. 2005. Quantitation of dityrosine in wheat flour
and dough by liquid chromatography− tandem mass spec-
trometry. Journal of Agricultural and Food Chemistry, 53(7),
2418–​2423.
Hay RL. 1993. Effect of flour quality characteristics on puff pastry
baking performance. Cereal Chemistry, 70, 392–​396.
Kriz EF, Oszlanyi AG. 1976. U.S. Patent No. 3,985,911. U.S. Patent
and Trademark Office, Washington, DC.
Lucas T, Collewet G, Bousquières J, Deligny C. 2018. The size of
eye-​shaped bubbles in Danish pastry in relation to the size of
fat fragments: A reverse engineering approach of the alveolar
structure. Journal of Food Engineering, 237, 194–​203.
Miskandar MS, Man YC, Yusoff MSA, Rahman RA. 2005. Quality
of margarine: Fats selection and processing parameters. Asia
Pacific Journal of Clinical Nutrition, 14(4), 387.
39
O’Brien RD. 2004. Fats and oils –​formulating and processing for
applications. CRC Press. Boca Raton, Florida.
Ooms N, Pareyt B, Brijs K, Delcour JA. 2016. Ingredient func-
tionality in multilayered dough-​ margarine systems and the
resultant pastry products: A review. Critical Reviews in Food
Science and Nutrition, 56(13), 2101–​2114.
Ooms N, Pareyt B, Jansens KJ, Reyniers S, Brijs K, Delcour JA.
2017. The impact of redox agents on further dough develop-
ment, relaxation and elastic recoil during lamination and fer-
mentation of multi-​layered pastry dough. Journal of Cereal
Science, 75, 84–​91.
Ooms N, Jansens KJ, Pareyt B, Reynier S., Brijs K, Delcour JA.
2018a. The impact of disulfide bond dynamics in wheat gluten
protein on the development of fermented pastry crumb. Food
Chemistry, 242, 68–​74.
Ooms N, Vandromme E, Brijs K, Delcour JA. 2018b. Intact and
damaged wheat starch and amylase functionality during multi-
layered fermented pastry making. Journal of Food Science,
83(10), 2489–​2499.
Podmore J. 2002. Bakery fats. Fats in Food Technology, 10, 30.
Renzetti S, de Harder R, Jurgens A. 2016. Puff pastry with low
saturated fat contents: The role of fat and dough physical
interactions in the development of a layered structure. Journal
of Food Engineering, 170, 24–​32.
Simovic DS, Pajin B, Seres Z, Filipovic N. 2009. Effect of low-​trans
margarine on physicochemical and sensory properties of puff
pastry 1. International Journal of Food Science & Technology,
44(6), 1235–​1244.
Tilley KA, Benjamin RE, Bagorogoza KE, Okot-​ Kotber BM,
Prakash O, Kwen H. 2001. Tyrosine cross-​links: Molecular
basis of gluten structure and function. Journal of Agricultural
and Food Chemistry, 49(5), 2627–​2632.
Wickramarachchi KS, Sissons MJ, Cauvain SP. 2015. Puff pastry
and trends in fat reduction: An update. International Journal
of Food Science & Technology, 50(5), 1065–​1075.
Baking: Chemical Leaveners
Linda A. Luck
Professor of Chemistry-​Emeritus, State University of New York at Plattsburgh, Plattsburgh, New York 12901, United States
Bread has been the staple of our society for thousands of years.
It was the “breaking of bread” that brought people together and
symbolized friendship for centuries. Each country and reli-
gious sect had its own form of bread. In France, for example,
the boulangers (formerly talmeliers) have existed since 1219,
baking and selling their own breads (TLFi, 2020). Until
breadmaking became commercialized in the 1800s, making
bread was the most loving, yet time-​consuming, task in the
household (Snell, 2014).
Our modern society has moved the baking of bread and other
pastries from the domestic home to commercial enterprises,
largely due to the innovations and understanding of the science
behind this process and the development of new methodologies
for mass production. The studies of Louis Pasteur on yeast and
the discovery of pearl ash, the precursor to baking soda obtained
by baking potash to remove impurities, started the revolution in
manufacturing a variety of baked goods on an industrial scale
(Snell, 2014).
Scientifically, our cakes, muffins, cookies, biscuits and quick
breads are aerated products, of which 80% of the total volume is
due to gas bubbles. The bubbles observed in these products are
the result of gas expansion that occurs before and during their
time in the oven. Chemical reactions from chemical leaveners,
namely baking soda and baking powder, are responsible in some
cases, rather than yeast, for producing the gas that allows baked
goods to rise and are hence responsible for the tender and fluffy
texture of these delightful foods.
Yeasts have been used for raised breads for thousands of years,
but these single-​celled fungi are not as efficient for certain baked
goods, especially in the commercial setting. Chemical leaveners
are much faster in action, and the acid/​base chemical reactions
that are the basis of these leaveners can be more controlled.
Baking soda (NaHCO3) is an alkali, or chemical base, also
known as sodium bicarbonate or bicarbonate of soda, and more
properly called sodium hydrogen carbonate. Most of the world’s
production of this chemical is done by the Solvay process, in
which calcium carbonate, sodium chloride, ammonia and carbon
dioxide react together in water. Baking soda, a key component
of baking and the culinary industry, can produce carbon dioxide
(CO2) by two means: heating and reacting with an acid.
At elevated temperatures, the reaction is faster than at room
temperature, because the gas is released in the dough and cannot
participate in the reverse reaction. Eventually, the baking soda is
consumed.
The chemical stoichiometry is that 2 moles of sodium bicar-
bonate (84 g) makes 1 mole of gas (about 22.4 L):
2 NaHCO3 (s) → Na2CO3 (s) + CO2 (g) + H2O(6.1)
When baking soda, an alkali, is reacted with an acid, such as
hydrochloric acid (HCl), in liquid form, the following reaction
takes place:
NaHCO3 (s) + HCl (l) → NaCl (s)+ CO2 (g) + H2O(6.2)
Hydrochloric acid in liquid form in this reaction is not normally
used in cooking. Any other acid can be used for this purpose, and
in the kitchen setting there are a host of acids used for culinary
purposes. Baking soda alone can be used for leavening if the
ingredients of a recipe include an acid. Buttermilk, sour cream,
yogurt, chocolate, brown sugar, vinegar, fruit juices or molasses,
to name a few, contain weaker acids, such as lactic acid in milk
products and citric acid from fruit. The acid/​base reaction takes
place with these weaker acids, but less gaseous product (CO2) is
produced than with HCl. Each weak acid has differing tendencies
to ionize, which is a measure of strength of the acid or the pKA
of the acid. As in the following discussion of baking powder, we
will see that this scientific information is critical to fine-​tuning
gas production in baking.
Baking powder is not synonymous with baking soda. Baking
powder is utilized when the recipe does not have inherent
acid ingredients and when the timing of the gas production is
important. Baking powder does, however, contain baking soda,
comprising about one-​quarter to one-​third of the total makeup.
Also included in the ingredients are a dry weak acid, such as tar-
taric acid, known as cream of tartar, and double-​dried cornstarch,
flour or potato starch. There are a variety of different mixtures
of these three ingredients, which will be discussed later (LaBaw,
1982). The non-​reactive components are included in the mixture
to prevent moisture absorption and premature gas production
during storage. The leavening reaction of baking powder is set in
motion when the dry acid becomes liquefied and reacts with the
baking soda in the mixture, as shown in equation (6.2). Because
liquefaction of the dry acid is paramount, it is imperative that
41
42
baking powder be kept dry to keep its potency. Baking powder
is known to deteriorate with age. A quick substitute for commer-
cial baking powder is ¼ teaspoon baking soda combined with ½
teaspoon of cream of tartar. This substitute must be used imme-
diately when mixed.
When one type of dry acid is combined with the baking soda
and the action of the acid and base are immediate, the baking
powder is referred to as “single-​acting”. “Double-​acting” baking
powders contain a mixture of dry acids, whose action on the
baking soda is fine-​tuned to release CO2 at specific times and
temperatures. As a rule, these provide an initial gas produc-
tion when the liquid is added, followed by a second inflation of
gas during the baking time. In addition, gas production by the
baking powder reaction in baking can be modulated by the use of
different types of dry acids or a mixture of dry acids.
Dry acids have a variety of solubilities, strengths and reaction
temperatures, so the beauty of baking powder is the ability to
carefully combine any of these dry acids in the mixture and sci-
entifically tune the gas production, resulting in the perfect texture
for your pastries.
Common dry acid components of baking powder are
monocalcium phosphate, Ca(H2PO4)2, sodium aluminum
sulfate, NaAl(SO4)2 12H2O, sodium aluminum phosphate,
NaH14Al3(PO4)8 4H2O, dimagnesium phosphate, MgHPO4, and
dicalcium phosphate dehydrate, CaHPO4. For example, Rumford
Baking Powder, a common and popular leavening agent, contains
monocalcium phosphate as a dry acid, along with baking soda
and cornstarch (American Chemical Society, 2006). Two-​thirds
of the gas is released at room temperature and the remainder is
released at 60 oC. Glucono-​delta lactone, C6H10O6, is used as the
dry acid leavening agent in refrigerated doughs that have water
present. The lactone is converted to gluconic acid, C6H12O7,
when heated, and this reacts with sodium bicarbonate to produce
CO2. Cream of tartar releases over 50% of the CO2 immediately
upon mixing, while. sodium aluminium phosphate and dicalcium
Linda A. Luck
phosphate release 100% of the CO2 during the first 15 minutes of
baking (LaBaw, 1982).
In many cases, both baking soda and baking powder are used,
because the combination can control the gas release and the alka-
linity of the mixture. Baking soda is added to keep the balance of
the acid ingredients so that the baking powder is not neutralized,
and the end result of an alkaline dough for many types of biscuits
is preserved. The alkaline environment allows better browning
via the Maillard reactions and tends to make a more tender and
porous crumb.
Baking soda and baking powder are also used in recipes for
purposes other than leavening. For example, by creating an alka-
line environment to promote Maillard reactions, one can produce
browner skin on poultry and oven-​roasted home fries. Baking
soda can also react with pectin strands in beans and weaken the
cell walls, thus reducing cooking time. The colour of black beans
is due to anthocyanins that change colour with pH change; the
more alkaline, the darker they are. A small amount of baking soda
will change the pH so that the beans will stay black (America’s
Test Kitchen and Cosby, 2012).
REFERENCES
American Chemical Society. 2006. Development of Baking Powder:
a Historical Chemical Landmark. www.acs.org/​ content/​ acs/​
en/​education/​whatischemistry/​landmarks/​bakingpowder.html,
last accessed 4 December 2020.
America’s Test Kitchen, Cosby G. 2012. The Science of Good
Cooking, America’s Test Kitchen, Brookline, Massachusetts.
LaBaw GD. 1982. Chemical leavening agents and their use in bakery
products, Bakers Digest, 56, 16–​21.
Snell R. 2014. The Recipes Project, Having Their Cake: Ingredients
and Recipe Collection in the Nineteenth Century. https://​
recipes.hypotheses.org/​4629, last accessed 14 January 2020.
TLFi. 2020. “Talmellier”. http://​stella.atilf.fr/​Dendien/​scripts/​tlfiv5/​
search.exe?23;s=985995750;cat=0;m=talmellier, last accessed
4 December 2020.
Baking: Injera –​the Multi-​Eyed Flat Bread
Mahelet Girma, Sumaya M. Abdullahi and Benjamin L. Stottrup
Augsburg University, Department of Physics, 2211 Riverside Ave, Minneapolis, MN 55454, United States
Definitions (in Amharic)
1. Injera: a thin, spongy, elastic flat bread with pores on
one side that is used to lay wot on top and scoop sauces.
2. Irsho: a small amount of a previous batter saved and
used as a starter for fermentation.
3. Wot: vegetable or meat sauces typically eaten with
injera.
4. Absit: a portion of batter taken out after the first fermen-
tation to be boiled and returned in the batter. The role
of this partially gelatinized starch is to correct the vis-
cosity of batter, which influences the formation of eyes.
Injera is a staple of East African food culture and is consumed
by Ethiopians, Eritreans, and Somalis often more than twice a
day around the world. Laid flat on a big plate, this unique chewy
and spongy bread is often the base for meat and vegetable wot.
While the variations in recipe and culture make injera distinct, for
many it is literally and metaphorically the center of table and diet.
Injera is eaten with the hands and often serves as a pliable utensil
and serving plate. In appearance, injera is light gray, tan, or
brown with distinctive open bubbles or “eyes” a few millimeters
in diameter. Injera’s distinctive sour taste is the result of the fer-
mentation (for several days) of the flour obtained by milling
the seeds of Eragrostis tef, also known as teff, an annual grass.
Multiple factors are required for acceptable injera, including the
quality of the eyes, the pliability and ability to be rolled without
tearing, a spongy texture, soft but not adhesive or sticky surfaces,
and a distinctive sour taste. The materials at hand determine
how easily acceptable injera is achieved by home bakers. For
example, for Ethiopians, ideally injera is made entirely from teff
flour. However, a home baker may use a mixture of teff and other
flours such as barley, sorghum, and all-​purpose wheat flour. Injera
may even be prepared using just yeast and only all-​purpose wheat
flour. This chapter reflects an Ethiopian perspective but respects
the diversity of how injera is prepared in other parts of the world.
Injera prepared from teff is almost exclusively preferred. Teff
is a hardy plant originating in Ethiopia and Eritrea but is now
cultivated across Africa, the United States, and the European
Union. The tiny grain comes in varieties of white, brown/​red, or
mixed with seeds just a millimeter in size (Hagos et al., 2012).
In addition to producing grain, teff straw is also used as feed for
livestock, making it a valued agricultural product. An increase in
demand has raised teff’s popularity in the United States. It is now
produced in Idaho and Oregon, two states that share a geography
and climate with Ethiopia. However, teff availability around the
world and the impact of climate change will increase demand
for acceptable substitutes. A variety of flours from grains such as
barley, sorghum, and wheat can be fermented in the same way to
yield the spongy flavorful flat bread. The global surge in popu-
larity of teff is due to two reasons. First, for immigrants from
Ethiopia, Eritrea, and Somalia, injera has a symbolic significance
in culture and remains a continuous central staple of dishes.
Second, because teff lacks gluten and contains all eight essen-
tial amino acids, it is gaining popularity as a nutritious gluten-​
free cereal (Gebremariam et al., 2014). Presently, food science
research is focused on the commercialization and production
of injera. However, it is still made in homes around the world
using ingredients available to bakers (usually women). Figure 7.1
illustrates teff and starch granules on several length scales.
The general approach to preparing injera is to knead flour and
water together to create a dough, and then let the dough ferment
for a period of time (typically ranging from one to three days).
Traditionally, fermentation in the batter is initiated by using a
small amount of batter, called irsho, from a previous batch, much
like a starter for sourdough bread. Most commonly, the starter
culture for injera is a symbiotic combination of lactic acid bac-
teria and yeast whose co-​metabolism provides a mechanism for
the breakdown of carbohydrates (Chavan and Chavan, 2010).
After this initial fermentation, a small portion (10–​20%) of the
batter is taken out to be boiled for several minutes. After boiling,
this batter of gelatinized starch (called absit) is added back to
the batter and allowed to ferment for a few hours or up to a day
as a second fermentation. The first fermentation process creates
acidity and gas in the batter. The second fermentation process
initiates additional CO2 production as both gelatinized swollen
starch granules from the absit and ungelatinized starch granules
from the original batter are fermented (Parker et al., 1989).
Injera is baked as a flat bread on a large hot plate under a well-​
sealed lid to retain steam. During the baking process, the starch
is gelatinized, and bubbles from the fermentation process evolve
into pores and ultimately form “eyes”. Because the bread is only
cooked on one side, eyes only form on one side of the bread.
Depending on the choice of flour and the fermentation method,
the batter preparation time can be reduced to 24 hours. For
43
44
FIGURE 7.1 Teff at various length scales. (A) Grains of ivory and brown teff with a coin shown in size for comparison. (B) SEM image of single teff grain.
(C) Starch granules in teff. (D) All-​purpose flour for comparison.
SEM, scanning electron microscopy.
instance, injera can be made from all-​purpose wheat and/​or rice
flour by fermenting it for a few hours using yeast, and absit is not
needed to acquire the distinctive eye formation.
Regardless of flour, recipe, or method, a simple way to quickly
judge injera quality and the baker’s skill is to assess the quality
of the eyes: uniform eye size and density are an easy proxy for a
baker’s skill. This is more than simply aesthetics. Properly made
injera will have a uniform distribution of eyes to hold stews and
sauces.
Mahelet Girma et al.
Several factors influence the distribution, size, and area fraction
of the eyes. A batter poured unevenly on the griddle, poor fer-
mentation and low CO2 levels, or too high viscosity would result
in uneven eye formation and density. As a dispersed system of gas
within batter, the rheology of the injera batter as it is baked and
gelatinized determines eye properties. The work of Attuquayefio
(2014) has explored how fermentation and batter viscosity influ-
ence eye formation and elasticity of the bread. A comprehensive
study by Assefa et al. (2018) found that kneading conditions and
Baking: Injera – the Multi-Eyed Flat Bread 45

FIGURE 7.2 (A) Image of injera. (B) Size distribution of eyes. (C) SEM image of injera illustrating an interconnected network of pores that open up into
the eyes at the surface. (D) SEM image of injera’s baked side. Note the smaller pore size and the closed nature of some cells, illustrating that some pores must
rupture against the baking surface before the bubbles can evolve into the larger network.
SEM, scanning electron microscopy.

absit preparation were important factors in the molecular profile
of flavanoids as well as sensory perception of injera. Specifically,
the ratio of batter to water during absit preparation was found to
influence the number of eyes. Because teff is gluten-​free, the absit
is believed to increase the viscosity of the batter and increase the
gas retention (Zannini et al., 2012). Reduced to physics, the for-
mation of eyes becomes a complex problem in the fluid dynamics
of bubbles passing through a dense batter as it solidifies.
Several metrics have been used to quantitatively assess the
quality of eyes in injera. Some examples include the number,
size, density, and area fraction. Figure 7.2 illustrates the eyes
from several perspectives. The importance of injera as a utensil
implies that additional quantitative analysis may be appropriate.
The volume and network structure of the pores through the flat
bread may also impact the sensory experience. Specifically,
the volume of material that may be easily absorbed and held
46
in the eyes and the weight the injera can take on before tearing
or yielding are both of interest. These considerations are only
partly accessible by previous surface quantification tools.
This network structure of the pores can be seen in Figure 2c.
Understandably, the pores, network structure, and eyes have
received attention as viewed from above. However, CO2 bubbles
also nucleate near the baking surface and likely rupture against
the baking surface before evolving and flowing up through the
batter (Figure 2d). This may provide an additional perspective
on bubble and pore formation. Finally, because eyes are such
an important and distinctive feature of injera, several strategies
are utilized when preparing injera under improvised conditions.
These include the addition of a small amount of baking powder
to increase the CO2 available just before baking or the use of
self-​rising flour.
Conclusions
This chapter conveys the importance of injera to people of East
African descent around the world. The ways injera is served
and eaten demonstrate its versatility. It is served on a plate big
enough for a whole family to share eating. Its nutritional value
and physical uses to lay wot on top and scoop sauces with make
it an important part of most meals. Although it originated in
East Africa, we highlight that injera can be made in households
around the world using materials at hand. Eyes on surface, elas-
ticity, and sour taste have been studied and determined to be its
most distinct properties.
Mahelet Girma et al.
Acknowledgments
MG and SMA dedicate this to our mothers Amezenech
Woldemeskel and Siham Hussien. BLS acknowledges the
support of Augsburg University’s Food Lab and Mr. Seth Frand
for scanning electron microscopy work.
REFERENCES
Assefa YL, Shimelis Admassu Emire SA, Abebe W, Villanueva M,
Ronda F. 2018. The effect of mechanical kneading and absit
preparation on tef injera quality. African Journal of Food
Science, 12, 246–​253.
Attuquayefio WD. 2014. Influence of processing parameters on
eye size and elasticity of tef-​ based injera. MSc Thesis,
Pennsylevania State University.
Chavan RS, Chavan SR. 2011. Sourdough technology—​a traditional
way for wholesome foods: a review. Comprehensive Reviews
in Food Science and Food Safety, 10, 170–​183.
Gebremariam MM, Zarnkow M, Becker T. 2014. Teff (Eragrostis
tef) as a raw material for malting, brewing and manufacturing
of gluten-​free foods and beverages: a review. Journal of Food
Science Technology, 51, 2881–​2895.
Hagos K, Jayanth CV, Somashekar R. 2012. Characterization of
white and red teff grains using X-​ray technique. Journal of
Scientific and Industrial Technology, 71, 534–​538.
Parker ML, Umeta M, Faulks RM. 1989. The contribution of flour
components to the structure of injera, an Ethiopian fermented
bread made from tef (Eragrostis tef). Journal of Cereal
Science, 10, 93–​104.
Zannini E, Miller Jones J, Renzetti S, Arendt EK. 2012. Functional
replacements for gluten. Annual Review of Food Science and
Technology, 3, 227–​245.
Baking: Viennoiserie –​Laminated Pastry Production

James A. Griffin
School of Culinary Arts and Food Technology, College of Arts and Tourism, Technological University Dublin,
City Campus, Dublin 1, Ireland

A Brief History of the Croissant from the 17th specialities including the Kipferl and the Vienna loaf, quickly
Century became popular and inspired French imitators (and the concept, if
not the term, of viennoiserie, a 20th-​century term for supposedly
The birth of the croissant was the result of a series of evolu-
Vienna-​style pastries). Indeed, this was at the best time, because
tionary steps. Moon-​shaped pieces of bread are documented in
falling sugar prices toward the end of the 17th century enabled
Enenchel’s 13th-​century poetry as being presented by Viennese
the rising wealthy merchant dynasties in the prosperous cities of
bakers to Duke Leopold of Austria in 1227 to celebrate the
Vienna and Paris to trade in sugar, which had once been available
Austrian victory over their Turkish invaders (Chevallier, 2009):
exclusively to royalty. Café culture, as we know it, was born in
this pastry, known as Kipferl, was a crescent/​horn-​shaped formed
1683 in Vienna (City of Vienna, 2019), and Paris was introduced
bread. Some historians maintain that the Kipferl originated in
to coffee three years later in 1669 by a Turk, Hoşsohbet Nüktedan
monastic bakeries. The Kipferl was first mentioned in records
Süleyman Ağa, who was dispatched by sultan Mehmet IV as
from the 12th century (Savic, 2019), and it was enriched using
his ambassador to the court of King Louis XIV. Records indi-
large quantities of butter/​lard, almonds, and sugar (Fiegl, 2015).
cate that Paris’s first genuine coffee house was Café de Procope,
After lamination was introduced (the story is better documented
which opened in 1686 (Arat, 2019). It soon became popular and
in François Pierre de La Varenne’s 1653 book Le Pâtissier
was frequented by famous historical figures such as Voltaire,
François), the Kipferl became the French croissant as we know it
Rousseau, and Diderot, among others. Café culture involved
today through the merging of the Austrian Kipferl and the French
meeting over tea, coffee, and cake to discuss revolutionary ideas
puff paste lamination technique (Chevallier, 2009; Goldstein and
against King Louis XIV (Bramen, 2010).
Mintz, 2015).
Other origins are proposed. According to the Culinary Institute
of America (Culinary Institute of America, 2016), the crois-
sant would have been created first by the Hungarian bakers of Viennoiserie: A Modern Twist
Budapest to signify and celebrate the liberation of their beloved
Laminated viennoiserie pastry is made by rolling and folding
city from the Turkish army in 1686. But there is another story
butter and a yeasted dough together to form a block of pastry
(Chevallier, 2009) that the creation was by the Austrian bakers
containing many layers, as seen in Figure 8.1. First, with fer-
in Vienna, who were besieged in 1683 by Turkish invaders: the
mentation, the layers of dough expand (Hartings, 2016), and this
legend recounts that the Viennese bakers, while at work early in
expansion, along with the separation of sheets of dried dough, is
the morning, heard the Turkish army digging under the walls of
increased during the baking process.
Vienna; they alerted their army commanders, and the Turks were
First, the fermentation by the yeasts generates carbon dioxide
routed. The croissant was created to celebrate the liberation of
bubbles that make an aerated dough, the bubbles being trapped by
Vienna: its shape and the name were a consequence of the cres-
the viscosity of the dough and also by the viscoelastic “gluten”
cent moon, a symbol of Turkish tyranny.
network created during the first steps of dough making, when
It is also said that the French and Italian bakers soon followed
flour is kneaded with some water (and possibly butter) (This,
their Viennese counterparts and included this pastry as a part of
2018). During baking in a hot oven, the surface of the fermented
their daily breakfast. The croissant in its original classical form
dough is heated, and the heat is transferred toward the interior
was very different from today’s creation, as it was made using
by conduction, with five main effects: (1) the expansion of CO2
puff paste with lard and milk, a laminated dough devoid of yeast
bubbles, (2) killing the yeasts when the temperature inside the
(Chevallier, 2009). Willan (2016) corroborates Chevallier’s story
dough is over 60 °C, (3) the vaporization of water from the dough,
of how the croissant arrived in Paris, at least in 1839 (some
with steam separating the dough layers, (4) drying the dough and
say 1838), when an Austrian artillery officer, August Zang,
making crisp layers, and (5) creating a colour at the surface of the
founded a Viennese bakery (“Boulangerie Viennoise”) at 92,
pastry (Figure 8.2).
rue de Richelieu in Paris. This bakery, which served Viennese

47
48 James A. Griffin

The first effect is a simple thermal expansion, due to the
increased velocity of CO2 molecules, responsible for an increased
pressure on their interface with the dough. The third effect is vis-
ible at the edges of puff pastry during baking: steam can be seen
bubbling at the edges of the sheets. This steam is formed because
the temperature inside the laminated pastry reaches 100 °C, and
the effect of this evaporation is important, because 1 g of water
generates about 1 L of steam; and during the baking of puff pastry,
as much as 20% of the weight is lost: this means that a volume of
about 20 L is produced for 100 g of dough, which is enough for a
very large expansion of the dough (This, 2020).
The degree of lamination and the care shown during the lamin-
ation process define the volume and appearance of the baked
product. It is sometimes said that egg washing the pastry before
baking moistens the surface of the pastry, preventing the dough
from forming a thick skin or crust initially; the function of the
retarding effect of the egg wash on bakery products would be
to delay the formation of crust and allows the pastry to achieve
full volume (Peterson, 2012). However, an experiment done by
H. This did not corroborate this “culinary precision”.
Of course, the colour of the upper surface is important as well:
it is due to protein degradation, as can be understood from a com-
parison of grilling (in a pan) flour and starch: only flour turns
brown, due to the presence of the proteins that are absent from
starch. Butter browning during baking contributes also. And of
course, if sugar (sucrose) is added in the dorure, caramelization
can occur, as the baking temperature is generally higher than the
140 °C needed for caramel formation (Defaye, 1994; Luna and
Aguilera, 2013).
All these effects have to be controlled by the baker and con-
tribute to the success of the participants of the Coupe du Monde
de la boulangerie (CDM), a competition held every three to four
years in Paris that is an arena where the world’s best bakers go
head to head, competing for the title of World Champion of
Bakery. The competition has led to much innovation in the field
of viennoiserie products made of brioche dough and yeasted
laminated pastry. The process that was used to develop the
product for Ireland’s participation in CDM in 2002 is explained
in the following section.

Production of Laminated Pastry

Dough Stage
500 g Strong flour
7g Salt
25 g Milk powder
20 g Butter
35 g Fresh yeast
45 g Caster sugar
50 g Egg
FIGURE 8.1 Laminated pastry cut to reveal the alternating layers of dough 210 g Water
and fat. Total 892 g

H2O

Steam
Dough

Fat
H2O H2O

Fat
Steam

H2O

FIGURE 8.2 Mechanism of pastry lift during baking.

Baking: Viennoiserie
FIGURE 8.3 Lock-​in sequence, the first “3” layers; note the direction of the seam in the third photo.
Method
• Disperse yeast, sugar, and egg in water
• Add liquid to the flour and mix to a dough
Lamination Stage
330 g chilled butter for lamination
Laminate as described later
Total pastry weight: 1222 g
Dough Stage
Sieve the flour, milk powder, and salt together; rub the butter into
the flour. The dough should be mixed on a 20-​quart Hobart-​type
mixing machine using a dough hook attachment. Mixing times
of two minutes on the first speed and six minutes on the second
speed are recommended. The mixed dough should be moulded
into a ball, placed into a container that will allow for the expan-
sion of the dough during the cold fermentation process, and
sealed tight or covered with plastic to prevent skinning.
The dough is fermented for 45 minutes at room temperature,
followed by overnight fermentation in a fridge at a temperature
of 3–​6 °C. The dough is then degassed by sheeting it through a
pastry break set to 10–​12 mm thickness, wrapped in plastic to pre-
vent skinning, and placed in a freezer to stiffen the dough and
chill it close to 0 °C for 30 minutes. The dough is then taken from
the freezer and formed into an even rectangle. The butter block
is placed in the centre of the dough rectangle as illustrated in
Figure 8.3, and the centre is sealed by pinching the dough together,
leaving the ends exposed with butter showing at each end.
49
Dough
Buer
Dough
FIGURE 8.4 The Lock-​in with three layers or a “3”. The ends are sliced to
ease elastic tension of the pastry.
At this stage, the pastry contains three layers, two dough layers
on the top and bottom, with the butter layer at the centre. The
dough to the sides of the pastry block can also be sliced to ease the
processing, which is known as the sandwich method (Figure 8.4).
In this process, the first stage is known as the “Lock-​in”, as the
butter is locked in between two layers of dough.
The following example explains how a Lock-​in 3-turn followed
by a second 4-​turn determines the number of layers developed by
lamination and emphasizes the need to factor in the dough con-
tact points. The left-​hand side of Figure 8.5 shows the dough after
folding a 4 from the Lock-​in after sheeting. During sheeting, the
shaded areas/​the dough touching points compress and form one
layer of dough, as illustrated in Figure 8.5; hence the need to
subtract one layer for each dough contact point to determine the
actual layers.
50 James A. Griffin

"4" Second turn Layers Second turn complete Layers

Dough 1 Dough 1
Dough touching point #1
Buer 2 Buer 2
Dough 3 Dough 3
points Dough 3 (-1) Buer 4
Buer 4 Dough 5 Dough touching point #2
Dough 5 Buer 6
points Dough 3 (-1) Dough 7
Buer 6 Buer 8
Dough touching point #3
Dough 7 Dough 9
points Dough 3 (-1)
Buer 8
Dough 9

FIGURE 8.5 Illustrating dough on dough layering after folding three to four times with dough touching points.

FIGURE 8.6 An offset 4-​fold noting the open and closed ends of the pastry block.

Lamination Sequencing 3 4 3 minutes. The 90° rotation of the pastry ensures that the dough is
stretched in each direction evenly throughout the process, which
The dough and the butter will have a similar consistency, which
eliminates shrinkage in the final proofing and baking stages of
allows the pastry to be sheeted with ease without breaking down
production. Reducing the pastry to a thickness of 12 mm enables
either the dough layers or the butter. The sheeting of the pastry
the freezer to reduce the core temperature of the pastry rapidly,
block followed by folding sequences allows layers of dough and
controlling the fermentation until the proving stage.
butter to build up simultaneously.
As the pastry at Lock-​in initially had three layers or a 3, it was
Following the Lock-​in, the dough now contains three layers,
sheeted and folded into four or a 4. One would imagine that 3 ×
i.e., dough, butter, dough. This is known as the first 3 of the
4 would equal 12 layers; however, where dough touches dough
lamination sequences. The dough is placed on the sheeter and
in the folding process, known as “The Dough Contact/​Touching
rotated 90° from the formation of the Lock-​in so that the seam
Point”, one layer is subtracted at each dough contact point. Under
formed is pointing horizontally towards the machine rollers. The
compression, the dough merges to form a single dough layer
thickness of the dough is gradually reduced, 3 mm at a time, to
while the butter layer remains unchanged. In the case of a 4, three
a final thickness of 6 mm on a pastry break. The pastry should
dough touching/​contact points are counted, leaving a total of nine
then be given an offset book turn or a 4-​fold, as illustrated in
layers of dough and butter.
Figure 8.6.
The chilled pastry is then removed from the freezer for its
An offset method of folding is recommended, as this reduces
final sheeting and folding. Once again, it is essential to rotate
the possibility of a bulge in the leading edge of the pastry
the pastry 90° so that the closed seam is facing the operator. The
block, which can result in uneven sheeting. At this stage, it is
pastry is then sheeted to a thickness of 8–​10 mm and given a
recommended that the pastry is rotated 90° with the closed seam
half-​turn or a 3. At this stage of the process, the pastry contains 9
facing the operator and reduced to a 12 mm thickness on the pastry
× 3 layers or 27 layers. However, as there are two dough contact
break, wrapped in plastic, and placed in a freezer at −18 °C for 20
Baking: Viennoiserie
FIGURE 8.7 Example of two dough on dough contact points in a 3-​fold.
FIGURE 8.8 The sheeted pastry, which can now be cut and processed.
points, as seen in Figure 8.7, two layers are subtracted from the
total, and the finished pastry now has a final total of 25 separate
alternating layers of dough and butter. The dough should again be
rotated 90°, with the closed end facing the operator, sheeted to a
thickness of 12 mm, wrapped in plastic, and chilled in a freezer
at −18 °C for 30 minutes.
The pastry is now ready for the preparation of many different
types of viennoiserie, such as classic croissant, pain au chocolat,
or pain au raisin (Figure 8.8).
REFERENCES
Arat E. 2019. The History of Turkish Coffee. www.turkishcoffee
world.com/​History-​of-​Coffee-​s/​60.htm, last access 4 d’Agriculture de France/​
December 2019.
Bramen L. 2010. When Food Changed History: The French
Revolution. www.smithsonianmag.com/​arts-​culture/​when-​
food-changed-history-​the-​french-​revolution-​93598442/​, last
access 4 December 2019.
51
Dough contact point #1
Dough contact point #2
Chevallier, J. 2009. August Zang and the French Croissant: How
Viennoiserie Came to France. 2nd ed. Chez Jim Books, North
Hollywood (California).
City of Vienna. 2019. 1683 –​the Beginning of Viennese Coffee House
Culture. www.wien.gv.at/​english/​culture-​history/​viennese-
coffee-​culture.html, last access 19 November 2019.
Culinary Institute of America. 2016. Baking and Pastry. John Wiley
& Sons, Hoboken (New Jersey).
Defaye J, Garcia Fernandez JM. 1994. Protonic and thermal acti-
vation of sucrose and the oligosaccharide composition of
caramel, Carbohydrate Research, 256, C1–​C4.
Fiegl A. 2015. Is the croissant really French -​A brief history of the
croissant –​from kipferl to Cronut? www.smithsonianmag.
com/​arts-​culture/​croissant-​really-​french-​180955130/​, last
access 19 November 2019.
Goldstein D, Mintz S. 2015. The Oxford Companion to Sugar and
Sweets. Oxford University Press, Oxford.
Hartings M. 2016. Chemistry in Your Kitchen. The Royal Society of
Chemistry, Cambridge.
Luna MP, Aguilera JM. 2013. Kinetics of colour development of
molten glucose, fructose and sucrose at high temperatures,
Food Biophysics, 9, 61–​68.
Peterson J. 2012. Baking. Potter/​TenSpeed/​Harmony, Berkeley
(California).
Savic O. 2019. Vienna’s Kipferl –​the croissant’s grandfather. www.
itinari.com/​vienna-​s-​kipferl-​the-​croissant-​s-​grandfather-​mf3u,
last access 4 December 2020.
This H. 2018. Who discovered the gluten and who discovered its
production by lixiviation?, Notes Académiques de l’Académie
Academic Notes from the French
Academy of Agriculture, 3, 1–​11.
Willan A. 2016. France. Oxford Reference. www.oxfordreference.
com/​ v iew/​ 1 0.1093/​ a cref/​ 9 780199313396.001.0001/​ a cref-​
9780199313396-​e-​202, last access 19 November 2019.
Baking: How Does Starch Gelatinization Influence Texture?
Anaïs Lavoisier
UMR SayFood (Paris-Saclay Food and Bioproduct Engineering Research Unit), INRAE, AgroParisTech, Université
Paris-Saclay, F-91300 Massy, France
Starch and Starch Granules
In nature, starch is mainly found in cereal grains, rhizomes, roots,
tubers, and pulses (Carvalho, 2008). Regardless of its botanical
origin, starch consists of two types of polysaccharides: amylo-
pectin (between 70% and 80%) and amylose. Both are chains
of D-​glucose residues linked by α (1,4) bonds, forming linear
segments, and α (1,6) bonds, forming branches. Amylose is
essentially linear, with a degree of polymerization (DP) between
hundreds and tens of thousands (Ai and Jane, 2018). On the con-
trary, amylopectin is a highly branched molecule, consisting of
branch chains with a DP from 6 to ~100 (Ai and Jane, 2018),
resulting in a complex molecular structure. The chains of amylo-
pectin organize as double helices and crystallize (Bertoft, 2017).
These regions of crystalline order alternate with amorphous
layers (containing the branch points of the amylopectin mol-
ecule) and form a lamellar structure. These stacks of amorphous
and crystalline lamellae are arranged in rings embedded in an
amorphous matrix and constitute structures called “blocklets”.
These blocklets organize themselves into hard (i.e., crystalline)
and soft shells (i.e., amorphous), which, in turn, form alter-
nating concentric rings around the hilum (i.e., the point where
starch biosynthesis was initiated). These “growth rings” make up
granules of 1 to 100 μm in size, depending on the origin of the
starch (Jane et al., 1994). Because of this high degree of order
inside the native starch granules, a “Maltese cross” pattern is typ-
ically observed under polarized light.
Starch granules have, therefore, a particular multi-​ scale
structure. The exact nature of the granule architecture is still a
matter of debate. For example, the actual localization of amylose
molecules inside this three-​dimensional structure remains uncer-
tain. It is also pretty unclear how amylopectin and amylose
interact with each other (Bertoft, 2017).
Starch Gelatinization
When starch granules are heated in excess of water (i.e.,
starch granules dispersed in at least three times their weight in
water), they undergo a series of changes known as gelatiniza-
tion (Figure 9.1): they readily absorb water and swell to sev-
eral times their original size (between 55 and 75 °C, depending
mainly on the starch’s botanical origin). First, water enters the
amorphous regions of the granules, which expand and conse-
quently destabilize the crystallites (Wang and Copeland, 2013).
Consequently, the crystalline order of the granules is irreversibly
lost, and the Maltese cross disappears. Then, the structure of the
hydrated granules crumbles as amylose leaches out and becomes
solubilized. Granular remnants (insoluble material), commonly
called “starch ghosts”, are finally observed within a continuous,
viscous matrix.
Upon cooling, the amylopectin and amylose molecules in
solution interact with each other and form a paste whose viscosity
depends on parameters like the botanical origin of the starch, the
concentration, the pH, and ionic force, among others (Semeijn
and Buwalda, 2018). Eventually, during a period of storage of
several days, for example, the disrupted polymers can also
reorganize and recrystallize. This complex process is known as
retrogradation and may lead to the formation of a gel. Therefore,
gelatinization directly impacts the mechanical and rheological
properties of cooked starchy foods. This feature is also the
reason why starch is widely used as a texturing agent by the food
industry (as in cheese analogs or extruded snacks) as well as in
our kitchens (to thicken sauces or bakery fillings, for example).
Although starch gelatinization has been extensively studied
since the early 1800s, no consensus has been reached yet on the
exact nature of this transition process (Schirmer et al., 2015). As
a consequence, various definitions of the term “gelatinization”
exist in the literature, and many theories have been proposed
to explain this phenomenon at a molecular level. To avoid con-
fusion, Matignon and Tecante (2017) recently proposed an
enhanced definition of starch gelatinization. They described it as
“a non-​equilibrium process dependent on high energy input that
leads to the collapse of native semi-​crystalline orders of lamellae
in starch granules”. Here, it is essential to note that they consider
the swelling of the starch granule as a separate event and a conse-
quence of the gelatinization process.
Degree of Gelatinization
Starch influences the texture of food depending on its degree of
gelatinization, and since most of our food is cooked, most of the
53
54 Anaïs Lavoisier

FIGURE 9.1 Potato starch granule heated in water at 80 °C (images extracted from a video of an isolated starch granule observed under a microscope
equipped with a hot-stage).

starch we consume is at least partially gelatinized. The degree

of gelatinization depends mainly on the type of starch (e.g.,
amylose/​amylopectin ratio or crystalline polymorph), the com-
position of the food (e.g., water availability or sugar content), and
the processing conditions used (e.g., temperature, heating rate,
or shear). We usually find in our diet a combination of fully, par-
tially, and not gelatinized starch granules. In mashed potatoes,
for example, almost all starch granules will be gelatinized, while
starch granules will almost completely remain in their native state
in products like sugar-​snap cookies. The degree of gelatinization
may also vary between different parts of the same food, as in
white wheat bread, where the amount of fully gelatinized starch
granules is not the same in the crumb and the crust.
Furthermore, starch granules are most of the time encased
in a food matrix, and this matrix may also affect the degree
of gelatinization. In foods with a dense microstructure (e.g.,
pasta products), water transport is limited, and swelling is
FIGURE 9.2 Storage modulus (G′) of gels as a function of PS content (pure
constrained by the compact matrix surrounding starch granules
WPI gel, composite gels WPI 10% + PS 1%, 3%, 5%, 7%, or 9%), at 20 °C
(Cuq et al., 2014). Gelatinization is therefore restricted during before heat treatment, at 90 °C, and at 20 °C after heat treatment up to 90 °C.
cooking. Similarly, hydrocolloid gums, like xanthan gum, also Values were extracted from temperature sweep measurements.
influence starch gelatinization and swelling during processing
(From Lavoisier and Aguilera (2019b), reprinted with permission of the
because of their very high water binding capacities (Mahmood publisher.)
et al., 2017).

function of their potato starch (PS) content at 20 °C before heat

Gelatinization and Texture: Example of a Protein
Gel Modified by Starch
Starch gelatinization inside a preformed microstructure may influ-
ence the properties of the food matrix as a whole. Gelatinization
is, thus, a critical parameter to think about in the formulation of
new food products.
Consider, for example, protein-​ based hydrogels used to
modulate starch digestibility (Lavoisier and Aguilera, 2019a).
Hydrogels are three-​dimensional polymeric structures that can
entangle a lot of liquid water inside their network. So, such
matrices have high water content (i.e., >80%), and heat treatment
will lead to starch gelatinization. The properties of these protein/​
starch gels will, therefore, be considerably modified by heating
during cooking, pasteurization, and/​or sterilization, but how?
Figure 9.2 presents the storage modulus G´ of whey protein
isolate (WPI) gels formed by calcium-​induced cold gelation as a
treatment, at 90 °C, and at 20 °C after heat treatment.
As can be observed on this graph, starch did not significantly
modify G´ before heat treatment. Starch granules in their native
state behaved as inactive filler particles, and the architecture of
the protein network alone is responsible for the viscoelastic prop-
erties of the composite gels. However, during heating from 20
to 90 °C, starch gelatinization occurred, and G´ increased. This
reinforcement effect seems to be connected to three different phe-
nomena related to gelatinization. First, starch granules encased
in the WPI gel absorbed water, swelled, and exerted pressure on
the protein network. Then, due to this increase in volume, starch
granules were able to touch and interact with each other. Finally,
starch swelling removed water from the gelled protein phase,
which induced changes in the microstructure and the properties
of the WPI matrix (Lavoisier and Aguilera, 2019b).
Interestingly, after cooling, different rheological properties
were obtained depending on the PS content of the composite gels.
Baking: Starch Gelatinization
FIGURE 9.3 Scheme representing the changes in the microstructure of the composite gels after heat treatment.
(From Lavoisier and Aguilera (2019b), reprinted with permission of the publisher.)
For small amounts of starch, a weakening of the WPI gel was
observed: gelatinized granules probably ruptured the protein net-
work, creating flaws in the microstructure (i.e., antagonist effect).
On the contrary, higher amounts of starch led to a strengthening
of the protein gel. In this case, an interpenetrating network formed
between PS and WPI (i.e., synergistic effect). These conclusions
are summarized in Figure 9.3.
This example illustrates how starch gelatinization can influ-
ence the properties of a food matrix. This process involves
multiple factors, and it must be taken into consideration when
designing new foods, either to understand changes that may
occur during thermal processing, or to improve the product by
tailoring its texture.
Note: this chapter is based on the following doctoral thesis:
Lavoisier, A. (2019). Effect of a whey protein network formed by
cold gelation on starch gelatinization and digestibility. Pontificia
Universidad Católica de Chile, Santiago, Chile (available at
https://​repositorio.uc.cl/​handle/​11534/​23344).
REFERENCES
Ai Y, Jane J. 2018. Understanding starch structure and function-
ality. In Sjöö M & Nilsson L (Eds.), Starch in food: Structure,
function and applications, second edition. Woodhead
Publishing, Cambridge, UK, 151–​178.
55
Bertoft E. 2017. Understanding starch structure: Recent progress.
Agronomy, 7(56), 129.
Carvalho A. 2008. Starch: Major sources, properties and applications
as thermoplastic materials. In Belgacem M & Gandini A
(Eds.), Monomers, polymers and composites from renewable
resources. Elsevier Science, Amsterdam, The Netherlands,
321–​342.
Cuq B, Abecassis J, Morel H. 2014. Chapitre 1, la physique et la
chimie au service de l’élaboration des pâtes alimentaires.
In Lavelle C (Ed.), Science culinaire: matière, procédés,
dégustation. Belin, Paris, 28–​51.
Jane J, Kasemsuwan T, Leas S, Ames IA, Zobel H, Il D, Robyt JF.
1994. Anthology of starch granule morphology by scanning
electron microscopy. Starch Stärke, 46(4), 121–​129.
Lavoisier A, Aguilera JM. 2019a. Effect of a whey protein net-
work formed by cold gelation on starch digestibility. Food
Biophysics, 14(2), 214–​224.
Lavoisier A, Aguilera JM. 2019b. Starch gelatinization inside a
whey protein gel formed by cold gelation. Journal of Food
Engineering, 256, 18–​27.
Mahmood K, Kamilah H, Shang PL, Sulaiman S, Ariffin F, Alias AK.
2017. A review: Interaction of starch/​non-​starch hydrocolloid
blending and the recent food applications. Food Bioscience,
19, 110–​120.
Matignon A, Tecante A. 2017. Starch retrogradation: From starch
components to cereal products. Food Hydrocolloids, 68,
43–​52.
56 Anaïs Lavoisier

Schirmer M, Jekle M, Becker T. 2015. Starch gelatinization and its Wang S, Copeland L. 2013. Molecular disassembly of starch granules
complexity for analysis. Starch Stärke, 67(1–​2), 30–​41. during gelatinization and its effect on starch digestibility: A
Semeijn C, Buwalda PL. 2018. Potato starch. In Sjöö M & Nilsson review. Food & Function, 4(11), 1564–​1580.
L (Eds.), Starch in food: Structure, function and applications,
second edition. Woodhead Publishing, Cambridge, UK,
353–​372.
Baking: Sourdough Bread
Mark Traynor1 and Imran Ahmad2
1 Department of Nutrition, Dietetics, and Hospitality Management, College of Human Sciences, Auburn University,
Auburn, Alabama, United States
2 Chaplin School of Hospitality and Tourism Management, Florida International University, North Miami, Florida,
United States
A Mixture with Many Names
If a mixture of flour and water is left at ambient temperature for
a prolonged period it transforms into a remarkably sour, aro-
matic, and gaseous mass of dough. This transformation, caused
by endogenous and exogenous microflora, is a biochemical phe-
nomenon known as fermentation. A variety of labels are given
to this mixture: pre-​ferment, levain, poolish, biga, barm, pâte
fermentée, mother, chef, and sponge, but sourdough starter is
one of the most common terms used. The specific label used is
usually dependent upon the geographic location of production,
the choice of production method, and the particular formulation
used. The breadth of terms used can often confuse and bewilder
many bakers. Nonetheless, while the label for the mixture may
differ, the fundamental scientific principles behind its production
and application do not.
Protecting the Heritage of Sourdough
The primary role of a sourdough starter is to aerate (leaven) a
bread dough (Decock and Capelle 2005). Sourdough starter typ-
ically accounts for no more than 50% of the bread dough (Hansen
and Schieberle 2005). The resulting breads are collectively known
as sourdough bread. Rye or wheat flours are the most commonly
used flours for sourdough production, although a variety of flours
can be used, such as barley and sorghum (Mariotti et al. 2014;
Sluková et al. 2016). Sourdough fermentation is considered to be
one of the oldest food biotechnological processes for fermenting
cereal foods (Gobbetti et al. 2008). Today, sourdough breads
are common in a wide range of cuisines around the globe, in
particular in Europe, the Middle East, North America, and the
Mediterranean (Spicher 1999). In Italy alone, almost all of the
approximately 200 traditional Italian breads today use sourdough
starters in their production (Minervini et al. 2012).
Traditional sourdough production is quite a labour-​intensive
and time-​consuming process, considerably more so than the con-
ventional ‘straight dough method’ that uses a commercial yeast
to leaven bread. Baker’s Yeast (Saccharomyces cerevisiae) is the
primary leavening agent used in the ‘straight dough method’
(Jayaram et al. 2013). This involves a relatively short bulk fer-
mentation, usually lasting a couple of hours at 27 °C (McGee
2004). Due to the ubiquity of these yeast breads, the ‘culture’
and craft surrounding sourdough baking has been, for the most
part, lost by many bakers. A select few specialists and enthusiast
bakers are preserving the heritage of sourdough bread production
(Catzeddu 2011).
Artisan over Convenience
In recent times, there has been a global upsurge in demand and
consumer appreciation for the flavour and taste of authentic,
artisan-​
style sourdough bread (Minervini et al. 2012). The
nutritional, technological, and shelf life properties of sour-
dough bread have been found to be superior to those of yeast
bread (Gobbetti et al. 2014). Moreover, their sensory properties
set these breads apart. The deep sour taste, pronounced com-
plex aromas and flavours, dense crumb, and thicker crust are
just some of the distinctive sensory properties that a sourdough
starter can bestow on a bread (Pétel et al. 2017). Research has
shown that sourdough bread contains a broader range of vola-
tile compounds, which contribute to bread flavour (especially
in the bread crumb), in comparison to yeast bread (Decock and
Cappelle 2005).
The baking process is an important baking step for the devel-
opment of bread crust flavour, whereas complex microbial
interactions during fermentation are responsible for the forma-
tion of the characteristic flavour of the bread crumb (Hansen
and Schieberle 2005). A multitude of environmental and eco-
logical factors govern the fermentation process: temperature, pH,
redox potential, ionic strength, dough composition, dough yield,
and microbial enzymatic reactions (Font de Valdez et al. 2010;
Spicher 1999). Thus, a deeper understanding of the sourdough
fermentation process can assist the baker in controlling the pro-
duction of consistent-​quality bread through a more straightfor-
ward process.
57
58
Sourdough Fermentation –​It Is All About the
Microbes
Traditional sourdough fermentation is a complex process. The
microflora involved in the fermentation process are composed
of stable associations of cultures of symbiotic Lactic Acid
Bacteria (LAB) and wild yeasts (Gänzle and Ripari 2016). These
are derived from natural contaminants present in the flour or as
spores throughout the environment, and convert simple sugars,
found in the flour, into carbon dioxide gas (CO2), organic acids,
and alcohols under anaerobic conditions (Gobbetti 1998).
Lactic Acid Bacteria
Lactic Acid Bacteria (LAB) are bacteria characterised by the
production of lactic acid as the primary by-​product of fermen-
tation (Monedero et al. 2017). In addition to lactic acid, many
LAB strains can produce carbon dioxide gas (CO2) as a by-​
product, which contributes to leavening. There is immense diver-
sity among the strains of LAB in a sourdough; approximately
50 different species of sourdough LAB have been isolated from
traditional sourdough starters (De Vuyst and Neysens 2005).
The majority of strains belong to the genus Lactobacillus, with
L. sanfranciscensis, L. brevis, and L. plantarum among the key
species of Lactobacillus (Gänzle et al. 2007). Growth tempera-
ture for Lactobacillus is in the range 2–​53 °C; however, the
optimum range is 30–​40 °C (Hammes and Vogel 1995).
In mature sourdough starters, the ratio of LAB to yeast ranges
from 10:1 to 100:1 (Minervini et al. 2012). The dominance of
LAB is a result of their adaption to the unique environment. LAB
possess several stress response mechanisms that are activated to
enable the bacteria to overcome the hostile environment: acidity,
oxidation, periods of starvation and dehydration, and extremes of
temperature (De Angelis et al. 2001).
Broadly speaking, LAB fall into three general categories based
on specialised carbohydrate fermentation pathways: (i) faculta-
tively homofermentative, (ii) facultatively heterofermentative,
TABLE 10.1
Examples of Common Species of Lactobacillus Strains Associated with Sourdough Fermentation
Facultatively Homofermentative Facultatively Heterofermentative
L. acidophilus L. plantarum
L. amylovorus L. pentosus
L. farciminis L. alimentarius
L. mindensis L. paralimentarius
L. crispatus L. casei
L. johnsonii
L. amylolyticus
Mark Traynor and Imran Ahmad
and (iii) obligately heterofermentative. Facultatively
homofermentative LAB almost exclusively degrade simple
sugars into lactic acid (>85%) as a sole by-​product of fermen-
tation. On the other hand, heterofermentative types produce
ethanol, acetic acid, and CO2 in addition to lactic acid (<50%)
as by-​products of fermentation (Hansen and Schieberle 2005).
The final category, obligately heterofermentative, differ from
facultatively heterofermentative as they have the ability either
to produce lactic acid solely or, depending on the sugars avail-
able, to produce ethanol and acetic acidic in addition to lactic acid
(Hammes and Vogel 1995).
Yeasts
Yeast are single-​celled facultative microbes (a type of fungi) cap-
able of converting simple sugars into ethanol (alcohol) and CO2
under anaerobic conditions (Jayaram et al. 2013). The primary
role of yeast is to leaven the dough; however, they also play a
role in flavour and aroma production. Researchers have isolated
more than 20 different species of yeasts from sourdoughs (De
Vuyst and Neysens 2005). The most representative species
belong to the genera Saccharomyces and Candida, with the
species being S. exiguus, C. humilis, and C. krusei (Corsetti and
Settanni 2007).
Sourdough Fermentation Process
In wheat and rye flours, the primary fermentable carbohydrates
(maltose, sucrose, glucose, and fructose) are within starch
granules. Starch accounts for approximately 60–​ 70% of the
carbohydrates in wheat and rye flours (Gänzle and Ripari 2016).
LAB predominantly metabolise maltose and fructose (Corsetti
and Settanni 2007). Yeasts, on the other hand, mainly metab-
olise glucose, fructose, and sucrose. Before they can digest these
sugars, endogenous flour enzymes, such as amylase, break the
starch (and polysaccharides) down into simple sugars (Brandt
2007) (Figure 10.1).
Obligately Heterofermentative
L. acidifarinae
L. brevis
L. buchneri
L. fermentum
L. frumenti
L. hilgardii
L. panis
L. pontis
L. reuteri
L. sanfranciscensis
L. siliginis
L. spicheri
Baking: Sourdough Bread
Following this, the microbes degrade simple sugars into
pyruvic acid through a process known as glycolysis (Pétel et al.
2017). Several salient compounds are subsequently generated
from pyruvic acid through one of two fermentation processes
characterised by different general metabolic paths: lactic acid
fermentation and alcohol fermentation (Figure 10.1). During
lactic acid fermentation, the enzymes secreted by the LAB reduce
pyruvic acid into lactic acid. In contrast, similarly to the brewing
of beer, alcoholic fermentation involves conversion of pyruvic
acid into ethanol by the action of yeast. Alcoholic fermentation
FIGURE 10.1 Basic metabolic pathways for lactic acid fermentation and
alcoholic fermentation
FIGURE 10.2 Schematic of traditional sourdough fermentation method
59
by yeast directly produces a number of key flavour compounds
(Hansen and Hansen 1994).
Traditional Sourdough Fermentation with a Touch of
Back-​Slopping
The underlying function of traditional sourdough fermentation
is the continuity of the established microflora. This is achieved
through the consecutive re-​inoculation of a new batch of sour-
dough with a portion of a previously fully fermented batch of
starter (10–​40% w/​w), known as an inoculum (De Vuyst and
Neysens 2005). The inoculum is mixed into an initial basic dough
of flour and water, which is then incubated at 20–​30 °C (Nionelli
et al. 2014). Continuous propagation is maintained and optimised
through a cyclic enrichment process known as back-​slopping
(Figure 10.2). This entails the scheduled refreshment of nutrients
into the dough through the addition of raw materials: fresh flour
and water (typically 5–​25% w/​w) (Figure 10.2).
The fermentation times for the microbes differ. LAB require
over 12 hours, under optimum conditions, to produce sufficient
quantities of flavour compounds. Yeasts, on the other hand, only
require a few hours to produce adequate amounts of flavour
compounds (Hansen and Schieberle 2005; Chavan and Chavan,
2011). Depending on the microflora present at the beginning of
the process and the desired qualities of the final product, each
refreshment step takes place every 8–​24 hours (Pontonio et al.
2015) and is repeated between 5 and 10 times in total (Böcker
et al. 1995; Hammes and Gänzle 1998).
Back-​slopping introduces the required nutrients for the devel-
opment of the microbiota and flavour-​contributing molecules for
the sourdough (Ercolini et al. 2013). In this way, starter micro-
flora can be continuously maintained over several decades or even
longer (Vogel et al. 2011). One commercial example of this is the
renowned Boudin Bakery in San Francisco, USA; the starter cul-
ture used in each loaf of bread at Boudin Bakery is said to be the
60
original ‘mother dough’, kept alive by regular refreshments since
1849 (De Vuyst and Neysens 2005).
Stability of the Ecosystem
The main characteristic of a mature sourdough starter is that fac-
ultatively and obligately heterofermentative LAB dominate over
homofermentative LAB (Van der Meulen et al. 2007). The common
facultative heterofermentative strains include L. fermentum,
L. brevis, and L. plantarum (Corsetti 2012; Mariotti et al. 2014;
Minervini et al. 2015). Of obligately heterofermentative LAB,
L. sanfranciscensis is the prevalent strain, and an important
microbe, which is found in over 75% of sourdoughs (Gänzle
and Ripari 2016). Heterofermentative LAB are actually called
the ‘aromatic’ microflora (Decock and Cappelle 2005). These
microbes produce considerable levels of acetic acid, which is
important for pleasant, mild, sour flavour development (Hansen
and Schieberle 2005; Kaseleht et al. 2011). Furthermore, acetic
acid-​tolerant yeast species, mainly S. cerevisiae and C. humillis,
thrive in these acidic conditions (Lacumin et al. 2009; Valmori
et al. 2010; Minervini et al. 2015). In addition, these yeast strains
are also known to provide a considerable contribution to the fla-
vour profile of the final bread (Makhoul et al. 2015).
The initial dough encompasses microbiota endogenous
to the raw material itself. These include several strains of
yeast (Candida, Pichia, Rhodotorula, Saccharomyces, and
Torulaspora), filamentous fungi (Alternaria, Cladosporium,
Fusarium, Aspergillus, and Penicillium), bacteria (Gluconobacter
sp., Micrococcaceae, Enterobacteriaceae, and Pseudomonas),
and diverse LAB (Enterococcus, Lactococcus, Lactobacillus,
Leuconostoc, Pediococcus, and Streptococcus) (Siepmann
et al. 2018). The decline of these microbes and the prevalence
of favourable microbes is not immediate. The dough is ever-​
evolving throughout the fermentation process; it can take up to
one week for microbiota to evolve toward a stabilised, mature,
and active sourdough (De Vuyst et al. 2014). Temperature, time,
acidity, and the addition of nutrients (back-​slops) are key inter-​
dependent fermentation parameters (Van der Meulen et al. 2007).
Temperature affects the acetic acid-​to-​lactic acid ratio in the
dough. Higher temperatures (30–​37 °C) tend to lead to the dom-
inance of L. fermentum (Gänzle et al. 1998). High fermentation
temperature generally results in high acidity, caused by enhanced
LAB metabolism, which is detrimental to yeast metabolism
(Vrancken et al. 2011). Low to moderate temperatures (<30 °C)
facilitate heterofermentative LAB species thriving in sourdough;
with longer fermentation times, they produce a favourable mix-
ture of lactic acid, acetic acid, and/​or ethanol (De Vuyst et al.
2014). As time and the number of back-​slops increase, the pH
of the dough decreases. During the early days of fermentation
and back-​slopping, the addition of nutrients enables the produc-
tion of organic acid, resulting in a decrease in pH to 5.0–​6.2.
At this stage, lactic acid fermentation begins, hindering the
growth of microbes that are sensitive to acidity (De Vuyst and
Neysens 2005; Ercolini et al. 2013). Between 5 and 10 days of
fermentation, with four or more back-​slops completed, a gradual
decrease in pH to approximately 4.0 is achieved. This results
Mark Traynor and Imran Ahmad
in a well-​adapted LAB and yeast community being established
(Brandt 2007; Ripari et al. 2016).
Bringing the Dough Together
The heritage, art, and skill involved in the production of high-​
quality sourdough bread are undeniable. Attention to detail,
experience, and passion for sourdough production are vital qual-
ities of the artisan baker; however, these cannot solely account
for consistent, high-​quality bread. Controlling the fermentation
process, through back-​slopping, provides favourable conditions
in the sourdough for the development of a coveted microflora
profile. Therefore, it is a firm understanding of the fundamental
scientific principles behind the metabolic activity of the sour-
dough microorganisms that is the most significant knowledge a
baker can possess.
REFERENCES
Böcker G, Stolz P, Hammes WP. 1995. Neue erkenntnisse zum
ökosystem sauerteig und zur physiologie des sauerteig-​
typischen stämme Lactobacillus sanfranciscensis und
Lactobacillus pontis. Getreide Mehl und Brot 49, 370–​374.
Brandt MJ. 2007. Sourdough products for convenient use in baking.
Food Microb. 24(2), 161–​164.
Catzeddu P. 2011. Sourdough breads. In Flour and breads and
their fortification in health and disease prevention 1st Ed.,
ed. V. Preedy, R. Watson, and V. Patel, Academic Press,
London, 37–​46.
Chavan RS, Chavan SR. 2011. Sourdough technology—​a traditional
way for wholesome foods: a review. Comp. Reviews in Food
Sci. and Food Safety 10, 170–​183.
Corsetti A. 2012. Technology of sourdough fermentation and sour-
dough applications. In Handbook on sourdough biotechnology
1st Ed., ed. S Cappelle, L Guylaine, M Gänzle and M Gobbetti,
Springer Verlag, Berlin, Germany, 85–​103.
Corsetti A, Settanni L. 2007. Lactobacilli in sourdough fermentation.
Food Res. Int. 40, 539–​558.
De Angelis M, Bini L, Pallini V, Cocconcelli PS, Gobbetti M. 2001.
The acid-​ stress response in Lactobacillus sanfranciscensis
CB1. Microbiology 147, 1863–​1873.
De Vuyst L, Neysens P. 2005. The sourdough microflora: biodiver-
sity and metabolic interactions. Trends Food Sci. Technol.
16, 43–​56.
De Vuyst L, Van Kerrebroeck S, Harth H, Huys G, Daniel H-​M,
Weckx S. 2014. Microbial ecology of sourdough fermentations:
diverse or uniform? Food Microbiol. 37, 11–​29.
Decock P, Cappelle S. 2005. Bread technology and sourdough tech-
nology. Trends Food Sci. Technol. 16(1), 113–​120.
Ercolini D, Pontonio E, De Filippis F, Minervini F, La Storia A,
Gobbetti M, Di Cagno R. 2013. Microbial ecology dynamics
during rye and wheat sourdough fermentation. Appl. and
Environ. Microbiol. 79, 7827–​7836.
Font de Valdez G, Gerez CL, Torino MI, Rollán G. 2010. New
trends in cereal-​based products using lactic acid bacteria. In
Biotechnology of lactic acid bacteria: novel applications, ed.
F Mozzi, RR Raya and GM Vignolo, Wiley-​Blackwell, Iowa,
273–​287.
Gänzle M, Ripari V. 2016. Composition and function of sourdough
microbiota: from ecological theory to bread quality. Int. Food
Microbiol. 239, 19–​25.
Baking: Sourdough Bread
Gänzle MG, Ehmann M, Hammes WP. 1998. Modeling of growth
of Lactobacillus sanfranciscensis and Candida milleri in
response to process parameters of sourdough fermentation.
Appl. Environ. Microbiol. 64(7), 616–​2623.
Gänzle MG, Vermeulen N, Vogel RF. 2007. Carbohydrate, peptide
and lipid metabolism of lactic acid bacteria in sourdough. Food
Microbiol. 24(2), 128–​138.
Gobbetti M. 1998. The sourdough microflora: interactions between
lactic acid bacteria and yeasts in sourdoughs. Trends Food Sci.
Technol. 9(7), 267–​274.
Gobbetti M, De Angelis M, Di Cagno R, Rizzello CG. 2008.
Sourdough/​lactic acid bacteria. In Gluten-​free cereal products
and beverages, ed. EK Arendt and F Dal Bello, Elsevier,
Amsterdam, The Netherlands, 267–​288.
Gobbetti M, Rizzello CG, Di Cagno R, De Angelis M. 2014. How
the sourdough may affect the functional features of leavened.
baked goods. Food Microbiol. 37, 30–​40.
Hammes WP, Gänzle MG. 1998. Sourdough bread and related
products. In Microbiology of fermented foods vol. 1 2nd ed.,
ed. BJB Wood, Blackie Academic & Professional, London,
UK, 199–​216.
Hammes WP, Vogel RF. 1995. The genus Lactobacillus. In The lactic
acid bacteria vol. 2, ed. BJB Wood and WH Holzapfel, Blackie
Academic and Professional, London, UK, 19–​54.
Hansen A, Schieberle P. 2005. Generation of aroma compounds
during sourdough fermentation: applied and fundamental
aspects. Trends in Food Sci. and Technol. 16(1–​3), 85–​94.
Hansen A, Hansen B. 1994. Influence of wheat flour type on the pro-
duction of flavour compounds in wheat sourdoughs. Cereal
Sci. 19(2), 185–​190.
Jayaram VB, Cuyvers S, Lagrain B, Verstrepen KJ, Delcour JA,
Courtin CM. 2013. Mapping of Saccharomyces cerevisiae
metabolites in fermenting wheat straight-​ dough reveals
succinic acid as pH-​ determining factor. Food Chem. 136,
301–​308.
Kaseleht K, Paalme T, Mihhalevski A, Sarand I. 2011. Analysis of
volatile compounds produced by different species of lacto-
bacilli in rye sourdough using multiple headspace extraction:
volatiles produced by LAB in sourdough. Int. Food Sci. Tech.
46(9), 1940–​1946.
Lacumin L, Cecchini F, Manzano M, Osualdini M, Boscolo D, Orlic
S, Comi G. 2009. Description of the microflora of sourdoughs
by culture-​dependent and culture-​independent methods. Food
Microbiol. 26, 128–​135.
Makhoul S, Romano A, Capozzi V, Spano G, Aprea E, Cappellin L,
Benozzi E, Sacmpicchio M, Märk TD, Gasperi F, El-​Nakat
H, Guzzo J, Biasioli F. 2015. Volatile compound produc-
tion during the bread-​making process: effect of flour, yeast
and their interaction. Food and Bioprocess Technol. 8(9),
1925–​1937.
Mariotti M, Garofalo C, Aquilanti L, Osimani A, Fongaro L, Tavoletti
S, Hager AS. 2014. Barley flour exploitation in sourdough
bread-​making: a technological nutritional and sensory evalu-
ation. LWT-​Food Sci. Technol. 59, 973–​980.
McGee H. 2004. On food and cooking: an encyclopedia of
kitchen science, history and culture. Hodder & Stoughton,
London, UK.
61
Minervini F, Di Cagno R, Lattanzi A, De Angelis M, Antonielli L,
Cardinali G, Cappelle S, Gobbetti M. 2012. Lactic acid bac-
terium and yeast microbiotas of 19 sourdoughs used for trad-
itional/​typical Italian breads: interactions between ingredients
and microbial species diversity. Appl. Environ. Microbiol.
78(4), 1251–​1264.
Minervini F, Lattanzi A, De Angelis M, Celano G, Gobbetti M. 2015.
House microbiotas as sources of lactic acid bacteria and yeasts
in traditional Italian sourdough. Food Microbiol. 52, 66–​76.
Monedero V, Revilla-​Guarinos A, Zúñiga M. 2017. Physiological
role of two-​component signal transduction systems in food-​
associated lactic acid bacteria. Adv Appl. Microbiol. 99, 1–​51.
Nionelli L, Curri N, Curiel JA, Di Cagno R, Pontonio E, Cavoski
I, Gobbetti M, Rizzello CG. 2014. Exploitation of Albanian
wheat cultivars: characterization of the flours and lactic acid
bacteria microbiota, and selection of starters for sourdough
fermentation. Food Microbiol. 44, 96–​107.
Pétel C, Onno B, Prost C. 2017. Sourdough volatile compounds
and their contribution to bread: a review. Trends in Food Sci.
Technol. 59, 105–​123.
Pontonio E, Nionelli L, Curiel JA, Sadeghi A, Di Cagno R, Gobbetti
M, Rizzello CG. 2015. Iranian wheat flours from rural and
industrial mills: exploitation of the chemical and technology
features, and selection of autochthonous sourdough starters for
making bread. Food Microbiol. 47, 99–​110.
Ripari V, Gänzle MG, Berardi E. 2016. Evolution of sourdough
microbiota in spontaneous sourdough starter. Int. J. Food
Microbiol. 232, 35–​42.
Siepmann FB, Ripari V, Waszczynskyj N. 2018. Overview of sour-
dough technology: from production to marketing. Food
Bioprocess Technol. 11, 242–​272.
Sluková M, Hinková A, Henke S, Smrž F, Lukačíková M, Pour V,
Bubnik Z. 2016. Cheese whey treated by membrane separation
as a valuable ingredient for barley sourdough preparation. J.
Food Eng. 172, 38–​47.
Spicher G. 1999. Zur Geschichte des Sauerteiges. In Handbuch
sauerteig, biologie, biochemie, technologie (5th ed.), ed. G
Spicher and H Stephan, Behr’s Verlag, Hamburg, Germany, 3–​7.
Valmori S, Tofalo R, Settanni L, Corsetti A, Suzzi G. 2010.
Yeast microbiota associated with spontaneous sourdough
fermentations in the production of traditional wheat sour-
dough bread of the Abbruzzo region (Italy). Antonie Van
Leeuwenhoek 97(2), 119–​129.
Van der Meulen R, Scheirlinck I, Van Schoor A, Huys G, Vancanneyt
M, Vandamme P, De Vuyst L. 2007. Population dynamics
and metabolite target analysis of lactic acid bacteria during
laboratory fermentations of wheat and spelt sourdoughs. Appl.
Environ. Microbiol. 73(15), 4741–​4750.
Vogel RF, Pavlovic M, Ehrmann MA, Wiexer A, Liesegang H,
Offschanka S, Voget S, Angelov A, Bocker G, Liebl W. 2011.
Genomic analysis reveals Lactobacillus sanfranciscensis as
stable element in traditional sourdoughs. Microb. Cell Fact.
10(Suppl. 1), S6:1–​11.
Vrancken G, Rimaux T, Weckx S, Leroy F, De Vuyst L. 2011.
Influence of temperature and backslopping time on the micro-
biota of a type I propagated laboratory wheat sourdough fer-
mentation. Appl. Environ. Microbiol. 77(8), 2716–​2726.
Barbecue: The Chemistry behind Cooking on a Barbecue
Florent Allais
URD Agro-​Biotechnologies Industrielles, CEBB, AgroParisTech, 51110, Pomacle, France
Summer is not only the time for everyone to enjoy the sun; it
also gets us excited about the mouthwatering smoky taste of
barbecued food. Recent data show that about 80% and 90% of
American (HPBA, 2019) and French households (France Info,
2019), respectively, own barbecue grills. In Europe, the biggest
barbecue fans are the Germans, followed by the French and the
Polish, with 19, 17 and 17 barbecues a year, respectively (Idealo,
2019). Although these barbecue grills are mostly used during the
spring and summer seasons, some people are still keen on having
barbecues in the winter (75% in Italy, for instance). Although
everyone agrees that charcoal-​and wood-​grilled food is deli-
cious, few really appreciate the chemistry behind the great smoky
flavour of barbecued meat.
Composition of Wood, Charcoal and Meat
Charcoal and wood contain three main polymer components:
cellulose, hemicelluloses and lignins. While cellulose is a linear
polymer made only of glucose, hemicelluloses are branched
polymers constituted of various sugar residues, including
residues of D-​glucose, D-​xylose, D-​arabinose, D-​galactose and
D-​rhamnose (Figure 11.1; Wertz et al., 2017). As these two poly-
saccharide components are not involved in the flavouring of the
meat, their biosynthesis and precise composition will not be fur-
ther discussed here.
Not only are lignins the main constituents of the cell walls in
plants; they also come in second place after cellulose as the most
abundant natural polymer in the world (Belitz and Grosch, 1999).
Lignins are heterogeneous cross-​linked polymers made of three
phenolic monomers called monolignols (i.e., p-​coumaryl alcohol,
coniferyl alcohol and sinapyl alcohol) that differ in the degree
of methoxylation on their aromatic ring. The monolignols are
oxidized by enzymes such as peroxidases and laccases to pro-
vide O-​centred phenoxy radicals that can then be delocalized
through conjugation to lead to C-​centred radical species at the 3,
1 and β positions. Finally, via radical–​radical coupling, oxidized
monolignols are polymerized through labile C-​ O-​C ether
bonds and strong C-​C bonds, depending on the radical species
involved (Figure 11.2) (Boerjan et al., 2003). Once inserted into
the polymer, p-​coumaryl alcohol, coniferyl alcohol and sinapyl
alcohol lead to H, G and S units, respectively. Finally, the ratio
between these three monolignols in lignins depends on the plant
and its environment.
When it comes to trees, hardwoods (e.g., acacia, birch, ebony,
eucalyptus, maple and oak) are rich in S-​and G-​lignins, whereas
softwoods (e.g., cedar, pine and spruce) are mainly composed of
G units with traces of H units.
Meat, on the other hand, is generally composed of water
(75%), proteins (19%), intramuscular fat (2.5%), sugars (1.2%)
and other non-​protein substances (2.3%; amino acids, minerals,
etc.).
Proteins are high-​ molecular-​weight natural polymers –​
polyamides –​composed of amino acids that can be modi-
fied through thermally or enzymatically induced hydrolysis
(Figure 11.3).
Flavour from the Burning Wood
When wood and charcoal burn, cellulose, hemicelluloses and
lignins are thermally degraded. Lignins are responsible for the
smoky flavour; indeed, a variety of phenolic compounds are
generated through the oxidative degradation of lignins via pyr-
olysis. Among these phenolics, the most important are two ortho-​
methoxy phenols, guaiacol and syringol, that cause the smoky
taste and the smoky smell, respectively. When present in the
smoke, these two phenolics are entrapped by the moisture of the
grilled food and infuse it.
Also known as 3-​ methoxyphenol (CAS 90-​ 05-​1, C7H8O2),
and first isolated by O. Unverdorben in 1826 (Stevens et al.,
1943), guaiacol is one of the main components of the smoke
generated when wood/​charcoal burns (Figure 11.4). It is note-
worthy to mention that guaiacol is naturally present in plants such
as Guaiacum and can be biosynthesized by certain organisms
(Duffey and Blum, 1977). Guaiacol is also found in whiskey and
roast coffee, and is commonly used to produce vanillin at the
industrial level.
Similarly to guaiacol, syringol (2,6-​dimethoxyphenol, CAS
91-​10-​1, C8H10O3) is also found in wood/​charcoal smoke and is
63
64
FIGURE 11.1 Structures of cellulose and xylan-​type hemicellulose.
produced through the pyrolysis of lignins (Figure 11.4). Synthetic
syringol is commonly used as smoke flavouring for those who
want the barbecue taste without having to grill the food (this also
avoids consumption of benzopyrenes).
Although the mechanisms of the formation of guaiacol and
syringol through the pyrolysis of lignins are still rather unclear,
homolytic breakage of the Cβ-​O and the Cα-​C1 bonds in lignins
is necessary (Figure 11.5). It is noteworthy that, as guaiacol and
syringol are derived from G and S units, respectively, the guai-
acol/​syringol ratio in the smoke therefore depends strongly on the
wood species used for the barbecue.
Flavour from the Cooked Food
Wood and charcoal have been proven to play a crucial role in the
production of the smoky taste and smell; however, the glycation
reactions sometimes called Maillard reactions also strongly
contribute to the delicious flavour of grilled food. Moreover,
these also provide the grilled meat with the characteristic brown
colour.
Discovered by Louis Camille Maillard in 1912 (Everts, 2012),
these reactions involve the amino acids (obtained through the
breakdown of the proteins) and the reducing sugars present in the
meat. Similarly to the mechanisms of formation of guaiacol and
syringol, that of Maillard products remains unclear.
However, it is generally accepted that the Maillard reaction
proceeds as shown in Figure 11.6. The reducing end of the sugars
reacts with the amine moiety of the amino acids to form an imine
(called a Schiff base). The latter undergoes an isomerization to
provide the enol, which is in equilibrium with the corresponding
Florent Allais
ketone (1,2-​ enolization) that is also known as the Amadori
product. Through a combination of reactions such as deamin-
ation, dehydration and fragmentation, the Amadori products can
then lead to a very complex mixture of a wide variety of products
(e.g., furans, thiophenes, furanones, alkyl-​and acylpyridines,
pyrazines, pyrroles and oxazoles), each having its own type of
taste and odour (Figure 11.7).
Melanoidins, polymers also produced through the Maillard
reactions, contribute to the colouration of the grilled food. As this
reaction only occurs at temperatures above 110 °C, the browning
is mainly observed on the surface of the meat.
Due to the composition of amino acids and sugars being quite
different from one kind of meat to another, the large diversity
brought by the Maillard reaction therefore leads to a wide range
of different flavour profiles.
Harmful Substances Created through Barbecuing
Although wood-​and charcoal-​grilling brings delicious smoky fla-
vour and smell to the meat, it can also generate harmful chemicals.
The most common ones are polyaromatic hydrocarbons (PAH),
the heterocyclic amines (HCA) and acrylamide.
PAH are formed when melted fat drips on hot charcoal,
and some are proven carcinogens, such as benzo[α]pyrene
(Figure 11.8) (Phillips, 1999). Production of PAH by cooking
over charcoal, being a function of both the fat content of the meat
and the proximity of the meat to the heat source, can be reduced
by cooking for longer periods at lower temperatures.
HCA and acrylamide are intimately linked to the Maillard
reaction. Indeed, HCA are formed as meat cooks and are
Barbecue: The Chemistry behind a Barbecue 65
FIGURE 11.2 Lignin biosynthesis.
66 Florent Allais

FIGURE 11.3 General structure of proteins.

FIGURE 11.4 Guaiacol and syringol.

FIGURE 11.5 Guaiacol and syringol formation via lignin pyrolysis.

 Barbecue: The Chemistry behind a Barbecue
FIGURE 11.6 Maillard reaction and related products.
(Halford et al., 2012)

67
68
FIGURE 11.7 Maillard reaction products and their odour.
FIGURE 11.8 Typical PAH found in grilled meat.
generally found in the charred parts of the cooked meat. These
compounds result from the reaction of creatine with other amino
acids and monosaccharide-​ derived aldehydes at temperatures
between 125 °C and 300 °C (Figure 11.9) (Skog et al., 1998).
It is noteworthy that such compounds can be obtained at lower
temperatures with longer cooking time. Low-​fat meats have been
shown to generate less HCA (Jägerstadt et al., 1998). Just like
PAH, these compounds are also carcinogenic.
Acrylamide, which has been identified as one of the many
compounds that can be obtained through the Maillard reaction
(Figure 11.10) (Mottram et al., 2002), is another potential car-
cinogen. Although high doses of acrylamide cause cancer in
mice and rats, considering the relatively low concentration of
acrylamide in consumed grilled meat, the risk associated with
Florent Allais
barbecuing remains quite low as long as the meat is not cooked at
too high a temperature for excessive times.
Conclusion
If it were not for lignin pyrolysis and the Maillard reaction,
we would not enjoy delicious grilled meats with their burst of
flavours and brown colours. Phenols, sugars and amino acids
are responsible for this popular cooking style, that is, barbecue
grilling. Although grilling meat over wood or charcoal seems to
be the easiest –​if not the oldest –​way humans cook food, the
chemistry that hides behind it is quite complex. Indeed, there are
various parameters that come into play, not only at the flavour
Barbecue: The Chemistry behind a Barbecue
FIGURE 11.9 Suggested pathway of HCA formation.
(Jägerstadt et al., 1998)
FIGURE 11.10 Production of acrylamide.
level but also in terms of potential toxicity. For instance, wood
and charcoal origins (i.e., tree type) will significantly impact the
overall flavour, whereas the cooking duration and temperature, as
well as the distance between the meat and the fire, can both be
tuned to limit the production of toxic compounds such as PAH,
HCA and acrylamide. With so many parameters (e.g., meat type,
wood/​charcoal, temperature and cooking duration), almost an
infinite number of recipes can be created. Finding the ones that
will fit what your taste buds crave is just a matter of experimenta-
tion. Who knew that barbecuing is so reliant on chemistry?
REFERENCES
Belitz HD, Grosch W. 1999. Food Chemistry, Springer, Heidelberg.
Boerjan W, Ralph J, Baucher M. 2003. Lignin Biosynthesis, Annu.
Rev. Plant Biol., 54, 519–​546.
69
Duffey SS, Blum MS. 1977. Phenol and guaiacol: Biosynthesis,
detoxication, and function in a polydesmid millipede, Oxidus
gracilis, Insect Biochem., 7, 57–​65.
Everts S. 2012. The Maillard reaction turns 100, Chem. Eng. News,
90, 58–​60.
France Info. 2019. www.francetvinfo.fr/​ decouverte/​
vacances/​
barbecue-​la-​star-​de-​l-​ete-​pour-​les-​francais_​1545153.html, last
access 4 December 2020.
Halford N, Curtis TY, Muttucumaru N, Postle J, Elmore JS, Mottram
DSJ. 2012. The acrylamide problem: A plant and agronomic
science issue, Exp. Bot., 8, 2841–​2851.
HPBA. 2019. www.hpba.org/​Resources/​Publications, last access 4
December 2020.
Idealo. 2019. www.idealo.co.uk/​blog/​4709-​bbq-​habits-​uk-​europe-​
idealo-​survey/​, last access 4 December 2020.
Jägerstadt MI, Skog KI, Arvidsson P, Solyakov A. 1998. Chemistry,
formation and occurrence of genotoxic heterocyclic amines
70
identified in model systems and cooked foods, Eur. Food Res.
Technol., 6, 419–​427.
Mottram DS, Wedzicha BL, Dodson AT. 2002. Acrylamide is formed
in the Maillard reactions, Nature, 419, 448–​449.
Phillips DH. 1999. Polycyclic aromatic hydrocarbons in the diet,
Mutat. Res., 443, 139–​147
Skog KI, Johansson AE, Jagerstad MI. 1998. Carcinogenic hetero-
cyclic amines in model systems and cooked foods: A review
on formation, occurrence and intake, Food Chem. Toxicol., 36,
879–​896.
Florent Allais
Stevens ME, Ronan AK, Sourkes TS, Boyd EM. 1943. On the expec-
torant action of creosote and the guaiacols, Can. Med. Assoc.
J., 48, 124–​127.
Virk-​Baker MK, Nagy TR, Barnes S, Groopman J. 2014. Dietary
acrylamide and human cancer: a systematic review of litera-
ture, Nutr. Cancer, 66, 774–​790.
Wertz JL, Deleu M, Coppée S, Richel A. 2017. Hemicelluloses and
Lignin in Biobiorefineries. CRC Press, Boca Raton, Florida.
Bioactivity and Its Measurement
Hervé This vo Kientza1, 2
1 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
2 UMR 0782 SayFood, AgroParisTech, INRAE, Université Paris-​Saclay, 91300 Massy, France
For studies in molecular and physical gastronomy, quantitative
descriptions of the exchange of compounds between the various
compartments that make up food, and also between these and the
human body (from pans to intestines), are needed. In this chapter,
the concepts of “absolute compound release” and “instant com-
pound release” are introduced, and the concept of “matrix effect”
is derived from them.
Foods are physical and chemical systems (Figure 12.1), but
they are special ones, because they have to be adapted to human
consumption. The interest in their properties (nutritional, sen-
sory, etc.) does not lie solely in their microstructure or in their
chemical composition but also in their interactions with the
human body. As described in other chapters of this book, dishes
are formulated products, often colloidal in nature (This, 2005;
Dickinson, 2006). The fact that some of their compounds can be
released and interact with biological receptors (mouth, stomach,
etc.) has been called “bioactivity” (This, 2016), but this prop-
erty needed quantification. In this chapter, we address this before
looking at applications.
For this discussion, let us define as “bioactive” any compound
that can interact with a biological receptor. In some cases, phys-
ical binding is needed to trigger physiological effects (e.g., olfac-
tion, taste perception, or trigeminal effects), but for vision, the
effect is indirect, and for receptors inside tissues, a transfer into
f(G, O, S, W) F'(G, O, S, W)
Culinary
transformation
Release Release
of of
Bioactive Bioactive
FIGURE 12.1 One important question for molecular and physical gas-
tronomy is to study how compounds are differently released by food
ingredients when food systems (as described by the DSF formalism) are
processed, starting from a system f(G,O,S,W) and producing a system F(G,
O, S, W).
the blood system is needed, sometimes after modifications during
digestion.
In order to characterize quantitatively the property of bio-
activity, parameters have to be introduced. There has been some
confusion over definitions. Two terms –​bioavailability and
bioaccessibility –​have been used with different meanings in
different scientific or technological communities. For example, in
1999, Shargel and Yu defined the bioavailability of nutrients as a
quantity measured in the blood plasma of humans (in vivo assay),
so factors such as individual variability, physiological state, dose,
and presence of other meal components come into play (Shargel
and Yu, 1999). Although the whole of a nutrient is potentially
bioaccessible, in reality, almost no nutrient is totally converted
during digestion into a potentially absorbable form.
In 2000, Oomen et al. observed that “the concept of bioavail-
ability is interpreted in the literature in various ways” (Oomen
et al., 2003). According to the general interpretation in pharma-
cology, oral bioavailability is presently defined as “the fraction of
an orally administered dose that reaches the systemic circulation”.
In 2002, however, the Food and Drug Administration (FDA)
proposed to define bioavailability as
the rate and extent to which the active ingredient or active
moiety is absorbed from a drug product and becomes available
at the site of action. For drug products that are not intended
to be absorbed into the bloodstream, bioavailability may be
assessed by measurements intended to reflect the rate and
extent to which the active ingredient or active moiety becomes
available at the site of action.
(Shargel and Yu, 1999; Shi and Le
Maguer, 2000; FDA, 2002)
In particular, “absolute bioavailability” was defined as the
quantity and rate that characterize the transfer of the drug in
blood, and “relative bioavailability” as a measurement of the
quantity and speed that characterize the transfer of a drug when
many forms are compared with a reference form, for example.
This definition was proposed to apply also to active substances
(nutrients) present in foods; in nutritional science, it is the pro-
portion of a nutrient that is absorbed by the body.
71
72
In 2002, Hedren et al. were using the term “bioaccessibility”
as the
amount of an ingested nutrient that is available for absorption
in the gut after digestion. Thus, it is not equivalent to speak of
bioavailability or bioaccessibility. If the amount of recovered
nutrient after digestion is of relevance, then the term to use is
bioaccessibility. On the other hand, bioavailability of nutrients
is usually measured in the blood plasma of humans (in vivo
assay).
(Faulks and Southon, 2005)
Although all of a nutrient is potentially bioaccessible, in
reality, almost no nutrient is totally converted during digestion
into a potentially absorbable form. In almost every case, the
bioaccessibility and bioavailability of a nutrient are governed
by the physical properties of the food matrix, which affect the
efficiency of the physical, enzymatic, and chemical digestion
processes (Boyer and Liu, 2004).
Confronted with different definitions, the International Union
of Pure and Applied Chemistry (IUPAC) proposed in 2004 to
define bioavailability as the “extent of absorption of a substance
by a living organism compared to a standard system” (IUPAC,
2004). For toxico-​or pharmacokinetics, a quantitative definition
was given as the ratio of the systemic exposure from extravas-
cular (ev) exposure to that following intravenous (iv) exposure as
described by the equation:
Aev .Div
F= (12.1)
Biv .Dev
where F (fraction of dose absorbed) is a measure of the bioavail-
ability, A and B are the areas under the (plasma) concentration
time curve following extravascular and intravenous administra-
tion, respectively, and Dev and Div are the administered extravas-
cular and intravenous doses.
FIGURE 12.2 One definition of bioaccessibility and bioavailability for the particular case of contaminants. It does not apply to all food compounds.
(Adapted from Collins et al., 2015)
Hervé This vo Kientza
However, the various authors and the various communities did
not always stick to this definition (Gregory et al., 2005). In 2007,
bioaccessibility was defined (IUPAC, 2007) as the “potential for
a substance to come in contact with a living organism and then
interact with it. This may lead to absorption”. A note was added
as follows:
A substance trapped inside an insoluble particle is not bio-​
accessible although substances on the surface of the same
particle are accessible and may also be bio-​available. Bio-​
accessibility, like bio-​availability, is a function of both chem-
ical speciation and biological properties. Even surface-​bound
substances may not be accessible to organisms which require
the substances to be in solution.
In 2012, it was observed that the release of compounds by
food ingredients during cooking, or by food in the mouth or in
the digestive tract, can be described by measuring the net dis-
appearance of the compounds from the food items, and quanti-
tative parameters called “absolute bioactivity” and “dynamic
bioactivity” were introduced (This, 2012). However, in spite of
very clear definitions of these “bioactivities”, confusion between
them and the two words “bioavailability” and “bioaccessibility”,
as well as the fact that a compound can have a bioactivity (i.e.,
an action on the body), led to a change, which will be proposed
later: we shall see that the new proposed terminology is “com-
pound release”.
This proposal should be helpful, because as said, today various
scientific communities are still using old definitions, so it is
sometimes difficult for them to exchange information. In 2015,
Collins et al. proposed the view shown in Figure 12.2, adding that
there are a number of definitions of bioaccessibility, these
can be particularly confusing because they can relate to both
human ingestion and microbial degradation. We use the
following definition: the maximal amount of contaminant
released from the test matrix in a synthetic gastrointestinal
system. This fraction represents the maximum amount of a
Bioactivity and Measurement 73

FIGURE 12.3 One nutritional definition of bioaccessibility and bioactivity.

(After Shi and Le Maguer, 2000)

contaminant that is available for absorption within the human
gastrointestinal tract.
Compound Release (CR)
With the proposed quantitative parameters associated with “com-
pound release”, we do not want to focus excusively on nutritional
properties, or flavour properties, or toxicological effects; more
generally, we want to describe the exchange of any compound
between a (possibly food) system and its environment, in par-
ticular during cooking; such activities are often based on the
thermal treatment of plant or animal tissues in an aqueous solu-
tion. As molecular and physical gastronomy seeks to characterize
and understand the effect on how food exchanges compounds
with the body, we define the first notion using quantitative indexes
such as “absolute compound release” (ACR) for the description
of the total content of bioactive compounds, and “dynamic com-
pound release” (DCR) for the time-​course release of compounds
from food systems.
Let us start from a food system F, containing a mass m of a
compound C. Concerning “absolute compound release”, a diffe-
rence is introduced between “potential absolute compound
release” (ACRp) and “actual absolute compound release” (ACRa),
because the presence of a compound in a matrix is not a guar-
antee that all the molecules of the compound will be able to leave
the matrix (and possibly interact with receptors). We propose to
define “potential absolute compound release” as the possibility
to bind to receptors and to measure this property as the mass of
the compound C present in the food matrix or at its surface. On
the other hand, “actual absolute compound release” is the real
quantity (mass) of the compound that can be released. Of course,
0 < ACRa < ACRp
These definitions go along with a definition of an “absolute
matrix effect”, E, defined in such a way that there is no matrix
effect (E = 0) when the compound is entirely released (ACRa =
ACRp), and a maximum matrix effect (E = 1) when the matrix
completely traps the compound (ACRa = 0) (Chemistry
International, 2009; Engel et al., 2001; Kördel et al., 2009).
A simple definition can be (if the matrix effect is chosen as a
dimensionless parameter):
E=
( ACR p − ACRa ) (12.2)
ACR p
In general, the food system F is an object of complex shape inside
an environment M. Let S be its surface, and let t be the contact
time of F and M. A mass flux (mass exchanged through the sur-
face S by time unit) can be defined algebraically as:
dm(t )
dt
= ∫∫ j(t ).ds
S
(12.3)
where ds is a surface element through which a quantity j(t) of
the compound C is exchanged between the food system F and its
environment, the integration being taken over the entire surface
S of F.
Using this definition, the released quantity of the compound
C is equal to:
74
+∞
∫0
dm (t )
ACRa = dt (12.4)
dt
where the integration is taken between the contact time considered
0 and +∞.
Using this definition, we can now consider a definition of
“dynamic compound release” (DCR). When the matrix effect
is physical in nature (no molecular modification of a compound
during the transfer), a dynamic compound release can be defined
for each compound being exchanged between F and M.
Let us first observe that one could calculate dynamic compound
release as the variation of the released quantity as a function of
time, but the new parameter would not have the same dimension
as absolute compound release. This is why we proposed to define
the dynamic compound release after “releasability” l(t):
dm (t )
l (t ) = (12.5)
dt
Using this function, dynamic compound release DCR(t) can be
defined as:
∫
DCR (t ) = l (t )dt
0
(12.6)
where integration is taken between 0 and t.
Using this definition, the dynamic compound release DCR for
t → +∞ is equal to the actual absolute compound release ACRa.
Whereas the dynamic compound release describes the
exchange of the compounds as a function of time, the dynamic
matrix effect e(t) describes the variation of the rate (because of
the matrix) at which a compound is released. For the same reasons
as before, the dynamic matrix effect has to be a dimensionless
function of time, and a reference has to be chosen. It is proposed
that the dynamic matrix effect is nil when no matrix is present, so
that releasibility l(t) has to be compared with releasibility without
a matrix (pure diffusion):
f (t ) − l (t )
e (t ) = (12.7)
f (t )
where f(t) is releasibility without a matrix (simple molecular
diffusion in the considered environment). Of course, in order to
determine this function, one has to consider an object having the
same shape as the food system F (same surface), with:
∫ ∫ j ( t ) .ds
f (t ) = (12.8)
the integration being taken over the entire surface S.
Whereas the physical matrix effect describes the molecular
interaction of particular compounds with the matrix, a “chem-
ical matrix effect” is used when molecular transformations
occur between a compound and the matrix during culinary
Hervé This vo Kientza
transformations (thermal treatment, i.e., “cooking”, being a
particular case of such transformations). In order to define a
chemical matrix effect, one considers a compound that is modi-
fied during the F–​M interaction. Chemical processes reduce the
released quantity of this compound. The absolute matrix effect E
describes both the chemical as well as the physical matrix effects,
but the dynamic matrix effect only indicates an influence of the
matrix on the overall release of a compound without taking into
account the molecular transformations. In order to introduce a
“chemical dynamic matrix effect”, one has to make assumptions
about the independence of the compounds and their reaction
products. When chemical transformations occur, the release
of a compound is reduced (releasibility l′(t) instead of l(t)) by
processes that make a quantity r(t) (per time unit) disappear. One
can write:
l′(t) = l(t) –​ r(t) (12.9)
Here again, the dynamic chemical matrix effect ec(t) can be
introduced as a dimensionless function:
r (t )
ec (t ) = (12.10)
l’(t )
Using this definition, the chemical dynamic matrix effect is nil
when there is no chemical transformation at any time, and equal
to 1 when the bioactive compound is entirely transformed within
the matrix (r(t) = l′(t)). Depending on the experimental situation,
the other definition can be used:
l (t ) − l’(t )
ec (t ) = (12.11)
l (t )
In situ Quantitative NMR
How can the transfers of compounds in matrices be followed and
all these parameters numerically estimated? The analytical method
called in situ quantitative nuclear magnetic resonance (is q NMR)
was proposed for this purpose, based on the fact that gels are
colloidal systems that enclose a liquid component. This method
was first studied for the analysis of saccharides, amino acids,
and organic acids of plant tissues (This et al., 2010; Weberskirch
et al., 2011), and compared with the classic method for the deter-
mination of the saccharide content of plant tissues devised by
Davis et al. (2007) as a modification of the method set up by
O’Donoghue et al. (2004) and others (Viola and Davies, 1992;
Kahane et al., 2001). In the “modified O’Donoghue” method,
plant tissues are lyophilized, heated under reflux in a mixture of
methanol and water (62.5: 37.5, w:w) for 15 min at 55 °C; after
filtration, solvent evaporation, and lyophilization, the resulting
product is analysed using various analytical techniques, such
as quantitative proton nuclear magnetic resonance spectroscopy
(q 1H NMR) (Cazor et al., 2006; Tardieu et al., 2009). However,
analyses done by this modified O’Donoghue method are long
and destructive and require manipulations with toxic solvents.
Bioactivity and Measurement
As metabolites from plant tissues, including saccharides, are
supposed to be dissolved in an aqueous cytosolic environment
(Clerg, 1984), it was important to determine how much of them
could be detected using is q 1H NMR of whole tissues.
Our studies (Weberskirch et al., 2011) were performed using
carrot (Daucus carota L.) roots, in which the mean water con-
tent is about 88% of fresh weight (Nonnecke, 1989). In these
tissues, as in most plant tissues, the cytoplasm of plant cells is a
gelled system (Turner, 1981), where metabolites and ions are in
an aqueous, liquid environment that can be described by the DSF
(see the chapter on this) formula D0(W)/​D3(S). In the cytoplasm,
the cytoskeleton forms a network, including an aqueous solu-
tion of metabolites (cytosol) in which organelles are dispersed.
Moreover, a free aqueous solution makes up the sap filling the
vessels, called xylem and phloem; saccharides are dissolved in
the elaborated sap (Campbell, 1995).
The is q NRM method was compared with extraction followed
by quantitative NMR spectroscopy of extracts (q NMR): non-​
treated samples of plant tissues were introduced into a 5 mm
glass NMR tube with enough D2O for locking (in practice, about
0.05 g). Experiments validated the assumption that plant tissues
could be directly studied by the application of liquid q 1H NMR
spectroscopy to whole, non-​treated tissues, in spite of a large
quantity of ordinary (i.e., non-​deuterated) water being present
in the tissues. In particular, even if the baseline is sometimes
deformed in some way, saccharides can be easily determined in
the aqueous environment of cells and of conducting tissues (for
the latest results, a 1/​1000 precision was obtained).
0
sucrose
%
ppm 5.4 5.2 5 4.8 4.6 4.4
FIGURE 12.4 An is q 1H NMR spectrum for a carrot (Daucus carota L.) root sample.
(Cazor et al., 2006).
75
NMR spectra obtained by direct determination have the same
general appearance as NMR spectra obtained with solutions
of saccharides, including solutions made using the modified
O’Donoghue method (O’Donoghue et al., 2004). Moreover, with
both MOD/​q 1H NMR and is q 1H NMR methods, a quantification
of the major saccharides (D-​glucose, D-​fructose, and sucrose) in
carrots was possible with minimal mathematical treatment. The
main resonance of spectra obtained by both methods was due to
water; this was surrounded by resonances of various metabolites,
including primarily saccharides and amino acids. Saccharides are
not the only metabolites that can be quantitatively determined; in
particular, many resonances associated with amino acids can also
be studied following some mathematical treatment. One diffe-
rence between the spectra in the two methods was the intensity of
the resonances; using is q NMR, the resonances are smaller than
with the MOD/​q 1H NMR method (but still more than 10 times
the noise level), because the mass of plant tissue used for the
direct determination (0.2 g) is about 50 times lower than the mass
of fresh tissue used in the MOD/​q 1H NMR method (7–​10 g).
Of course, the water resonance is huge with is q NMR, but this
does not prevent the determination of the important metabolites
(using signal decomposition). As the characteristic resonance of
D-​glucose at a chemical shift of 5.24 ppm was partly covered
by the water resonance, the other characteristic resonance at
3.25 ppm was used for the direct determination (in any case, res-
onance decomposition can be used even when a signal merges
into another, bigger signal). Figure 12.4 shows the general shape
of the spectra.
fructose glucose
4.2 4 3.8 3.6 3.4 3.2 3
76
In some spectra obtained by the in situ method, the multipli-
city of the sucrose and D-​fructose resonances was not observed,
and some analyses gave irregular spectra, where the singlet of 3-​
(trimethylsilyl)propionic-​2,2,3,3-​d4 acid, sodium salt (TSP, used
as a chemical shift reference and also for quantitative determin-
ation) was split into a doublet. These poor spectra were gener-
ally associated with a magnetic field heterogeneity that could not
be corrected easily. As the “shim” gives information about the
homogeneity of the magnetic field in the analysed sample, it is
an important factor for the quality of the spectra. The solution we
found to overcome this irregularity was to partially dry the tissues
before analysis. Thus, both the water resonance is minimized and
the D-​glucose resonance at 5.24 ppm (H1α,) is more visible in the
spectra (however, the D-​glucose resonance at 3.25 ppm was used
for quantification because it gives better results).
In general, the direct, in situ NMR determination gives a higher
saccharide content than the modified O’Donoghue extraction/​q
1H NMR method. However, this trend is statistically significant
only for D-​glucose and sucrose (one-​way analysis of variance
test, significance level of 1%). Of course, the is q NMR method
is much faster and easier than the method using extraction and
NMR analysis: whereas sample preparation takes six days for
the extraction and NMR analysis (because of double lyophiliza-
tion), only a few hours are needed with the is q NMR method.
Moreover, methanol is not used in the is q method.
Classification of Bioactivities
With a rational way to study bioactivity and methods for the
determination of compound release from colloidal systems,
let us now consider theoretically what the matrix effect could
be. Indeed, such an effect is due both to structure and to the
binding of compounds to molecular details of the matrices. Such
interactions can be ranked in order of energy (Atkins, 1990).
At the low-​energy end of the interaction energy spectrum, van
der Waals interactions can limit the autodiffusion coefficient of
compounds; this is the case for most solutions. At higher ener-
gies, hydrophobic forces or hydrogen bonds can play a role.
At even higher energies, we can envision the effect of disulfide
bridges, with the ability to rearrange, as in dynagels (This, 2016).
At the highest end of the energy scale, covalent and electrostatic
bonds are important.
One important question is whether the matrix effect in com-
plex disperse systems can or cannot be described as a result
of elementary interactions between the various phases. This
question was addressed first by analysing the processing of
“vegetable stocks”, i.e., aqueous solutions obtained by thermal
processing of plant tissues in water (Cazor et al., 2006). In spite
of extensive use of carrot stock in homes, in restaurants, and
in the food industry, very little was known about metabolites
found in the stock and the time evolution of their concentration
during the thermal processing. The release of saccharides from
carrot root samples thermally processed in water was studied for
various reasons (Varoquaux et al., 1986), such as identification
Hervé This vo Kientza
of stock composition, loss of vitamins, and nitrate release in food
products. The mechanism of saccharide release was also studied.
Solutes are present either in phloem sap or in intra-​and inter-
cellular spaces (Stapley et al., 1995). Hence, saccharides can be
released from any of these compartments.
In one study (before is q NMR was set up), q1H NMR was used
for the quantitative determination of saccharides in carrot stocks
prepared at three temperatures: 50, 75, and 100 °C. Using the
protocol of Cazor et al. (2006), carrot roots were peeled, and the
bottom and top parts were removed. Then, the samples were cut
into cylinders of a specific size and heated in demineralized water
at a determined temperature. The aqueous solution samples were
centrifuged, filtrated, double freeze-​dried, and dissolved in D2O
with pH adjustment to 7 (using solutions of DCl). For the NMR
analysis, an Ultra Shield Bruker 500 MHz spectrometer was used
at T = 21 °C (±0.1 °C); 64 scans of 65 k, spectral width 6 kHz,
aq 5.3 s, D1 25 s.
The main compounds found in carrot stocks were saccharides,
amino acids, and organic acids. Three regions were identified,
corresponding to the amino acid region (0–​3 ppm), the saccharide
region (3–​6 ppm), and the phenolic region (6–​10 ppm). Each
major metabolite of the different extracts was identified after res-
onance assignment using 1H NMR spectra from pure compounds
associated with comparison of published data (Fan, 1996; Le Gall
et al., 2003). Twenty-​one metabolites were identified on each
spectrum, including 3 saccharides, 11 amino acids, 7 organic
acids, and other compounds.
The thermal processing caused significant enrichment of the
water environment in saccharides (Figure 12.4). For all three
saccharides studied, extraction was less effective at 50 °C than at
the two other temperatures (75 °C and 100 °C). As the optimal
temperature for carrot pectin methyl esterase (PME) activity is
in the range 40–​60 °C for pH values between 4.5 and 7 (Ly-​
Nguyen et al., 2002), we proposed to explain this difference by
enzymatic activity; at 50 °C, cell membranes remain intact in
spite of some protein denaturation, and loss of nutrients occurs
mainly due to osmosis, capillarity, and diffusion (Garrote et al.,
1984; Oliveira, 1988). Carrot PME activity is responsible for
pectin linking through calcium cations, and diffusion through
a cell wall is more difficult than diffusion through a cell wall
with hydrolysed pectic material in the high-​temperature regime.
Alzamora et al. (1985) showed that, in order to minimize
leaching losses of vitamins during water blanching of peas, it is
preferable to have a high temperature in “static” water and the
biggest peas possible. On the contrary, to maximize saccharide
recovery in a water solution, higher temperatures than 60 °C are
to be chosen.
To explain the increase in D-​glucose and D-​fructose con-
centration at 100 °C, associated with a decrease in sucrose
level at the same temperature (Figure 12.5), sucrose hydrolysis
was proposed. In theory, this assumption could be checked
by correlating the sucrose with the D-​ glucose+D-​fructose
concentrations. However, glycation reactions occurring
between reducing saccharides and amino acids also have to be
taken into account.
Bioactivity and Measurement 77

FIGURE 12.5 Time-​course release of three saccharides (sucrose, D-​glucose, and D-​fructose) detected in carrot stocks for two different experimental set-​ups
(plant tissue:water, 1:2 and 1:3).
(Cazor et al., 2006)

Sc k1 k-1 Sb
k5 Gc k2 k-2 Gb k'5

Fc k3 k-3 Fb
k6 k'6
k7 k'7

Hc k4 k-4 Hb

FIGURE 12.6 Evolution of saccharides during carrot stock making. The letters S, G, F, and H stand for sucrose, D-​glucose, D-​fructose, and 5-​
hydroxymethylfurfural, respectively. The indexes c and b correspond to “carrot” and “boiling water”, respectively. Various kinetic coefficients are introduced.

Modelling Saccharides however, decrease because of hydrolysis. Although fitting experi-

The quantitative study of the time-​course evolution of saccharide mental data with the functions obtained in solving general cases
concentration in carrot or onion stocks led us to some assumptions can give constants for transfers and various chemical processes
about the mechanisms of this evolution. In Figure 12.6, the left occurring during thermal treatment, it is unlikely, due to uncer-
rectangle stands for the inside of the plant tissue. In this dia- tainties in experimental data, that any new scientific result will be
gram, the letters S, G, F, and H represent sucrose, D-​glucose, D-​ obtained using this strategy. On the other hand, using a different
fructose, and 5-​hydroxymethylfurfural, respectively; the second model provides other information, for example, where tissues
letter (c or b) indicates the carrot tissue or the stock. are considered to be made of channels (xylem and phloem) and
Given measured initial conditions, a differential system, parenchymous tissues, such as in Figure 12.8 (model based on
assuming linearity, can lead, with simplifying assumptions, to a microscopic pictures such as Figure 12.9).
solution such as that in Figure 12.7. From the initial equations, Such pictures (in this case, an onion bulb sample soaked in
sucrose, D-​glucose, and D-​fructose have to decrease in the plant methylene blue, 3,7-​bis(dimethylamino)-​phenothiazin-​5-​ium
tissue and increase in the stock. Sucrose in the stock should, chloride) indicate that saccharides and other solutes can migrate
78
1.5
0.5
0
0 2 4 6 8 10
f plant cells. The diffusion of the colourant inside the plant tissues indicates
Gc, Fc Sb Gb, Fb Sc
FIGURE 12.7 Modelling changes in saccharide levels during carrot stock
processing. Here the variations are calculated for D-​glucose (G), D-​fructose
(F), and sucrose (S). The abscissa scale is for the time, and the ordinates are
arbitrary units of saccharide quantities.
FIGURE 12.8 A model of plant tissues with various compartments allowing
saccharide exchange.
from some aqueous solutions toward the inside of plant tissue. If
this phenomenon were due to molecular diffusion, it could also
occur in the opposite way, from plant tissues toward an aqueous
environment, such as in many culinary processes.
In order to model the extraction of water-​soluble compounds
from onion bulb tissues in an aqueous environment, these
compounds have to be identified, and their concentration has
to be estimated in the aqueous solution at different times. The
Hervé This vo Kientza
FIGURE 12.9 An onion (Allium cepa L.) bulb sample soaked in methylene
blue shows some colourant inside channels and very little colourant inside
that the compounds from the elaborated sap of the tissue could migrate in the
reverse direction, toward the aqueous environment, during cooking. However,
the presence of blue colour around the channels shows that diffusion from sap
to the aqueous environment is not the sole mechanism responsible for stock
making.
phenomenon of extraction of solutes from plant tissues to the
soaking solution was investigated experimentally, and the data
were studied analytically for bulbs of onions (Allium cepa L.).
With Marcia France and Audrey Tardieu (Tardieu et al., 2011),
we explored the time-​course extraction features, considering that
the solutes could be located either in channels or in parenchyma
cells; short and long times of extraction were studied. First, only
two compartments were considered: “channels” and “parenchyma
cells”. Two main limit profiles for metabolites can be considered:
quantity in parenchyma cells >> quantity in channels; or quan-
tity in channels >> quantity in parenchyma cells. For each case,
there are two possibilities: release from parenchyma cells faster
than release from channels, or release from channels faster than
from parenchyma cells. Another case could be quantity in par-
enchyma cells ~ quantity in channels, with the two transfer rates
being equal or different (Figure 12.10).
Fitting curves of the dry matter against time measured during
experiments indicates that approximately the same quantity
of saccharides is extracted from parenchyma cells and from
channels, with a large difference in transfer rate (0.048 for the
former, 0.876 for the latter; arbitrary units). At this point, we
do not know which compartment releases bioactive compounds
faster, but, if pure molecular diffusion occurs from channels,
as was observed when plant tissues were soaked in methylene
blue, it is likely that this compartment is responsible for faster
exchanges than the other.
Of course, these studies are far from finished, in particular
because one can see from Figure 12.10 that methylene blue
migrating toward the inside of plant tissues diffuses not only in
the channels but also in the surrounding tissues. The question
of measuring the fluxes and the kinetic parameters associated
Bioactivity and Measurement
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0 Recommendations 2003), PAC. 76, 1033–​1040.
0 2 4 6 8 10
x
Curve 1 Curve 2 Curve 3
FIGURE 12.10 One particular behaviour that can be obtained for the model
of Figure 12.8 restricted to two compartments (“cells” and “channels”). Curve
1 corresponds to the compartment richest in metabolites; Curve 2 to the com-
partment with the lowest concentration in metabolites; and Curve 3 to the sum
of the two phenomena.
with them remains an important goal for understanding the
mechanisms of a simple culinary process such as a stock.
REFERENCES
Alzamora SM, Hough G, Chirife J. 1985. Mathematical prediction of
leaching losses of water soluble vitamins during blanching of
peas, J. Food Sci. Technol. 20, 251.
Atkins PW. 1990. Physical Chemistry, p. 687. W. H. Freeman & Co,
New York.
Boyer J, Liu RH. 2004. Apple phytochemicals and their health
benefits, Nutr. J. 3, 5.
Campbell N. 1995. Biologie. De Boeck-​Wesmael, Québec.
Cazor A, Deborde C, Moing A, Rolin D, This H. 2006. Sucrose, glu-
cose, and fructose extraction in aqueous carrot root extracts
prepared at different temperatures by means of direct NMR
measurements, J. Agric. Food Chem. 54, 4681.
Clegg JS. 1984. Properties and metabolism of the aqueous cytoplasm
and its boundaries, Physiol. Regul. Integr. Comp. Physiol.
246, R133.
Collins CD, Craggs M, Garcia-​Alcega S, Kademoglou K, Lowe S.
2015. Towards a unified approach for the determination of the
bioaccessibility of organic pollutants, Environ. Int. 78C, 24–​31.
Davis F, Terry LA, Chope GA, Faul CFJ. 2007. Effect of extraction
procedure on measured sugar concentrations in onion (Allium
cepa L.) bulbs, J. Agric. Food Chem. 55 (11), 4299–​4306.
Dickinson E. 2006. Colloid science of mixed ingredients, Soft
Matter. 2, 642.
Engel E, Nicklaus S, Septier C, Salles C, Le Quere JL. 2001.
Evolution of the taste of a bitter Camembert cheese during
ripening: characterization of a matrix effect, J. Agric. Food
Chem. 49, 2939.
79
FDA. 2002. Guidance for industry. Bioavailability and bioequivalence
studies for orally administered drug product. General consid-
eration, www.fda.gov/​cder/​guidance/​index.htm, last access 4
December 2019.
Fan TWM. 1996. Metabolite profiling by one-​and two-​dimensional
NMR analysis of complex mixtures, Prog. Nucl. Mag. Res.
Spec. 28, 161.
Faulks RM, Southon B. 2005. Challenges to understanding and
measuring carotenoid bioavailability, Biochim. Biophys. Acta.
1740, 95–​100.
Garrote RL, Bertone RA, Silva ER. 1984. Effect of thermal treatment
on steam peeled potatoes, Can. Int. Food Sci. Technol. J.
17(2), 111.
Gregory JF, Quinlivan EP, Davis SR. 2005. Integrating the issues of
folate bioavailability, intake and metabolism in the era of forti-
fication, Trends Food Sci. Tech. 16, 229–​40.
Hedren E, Mulokozi G, Svanberg U. 2002. In vitro accessibility of
carotenes from green leafy vegetables cooked with sunflower
oil or red palm oil, Int. J. Food Sci. Nutr. 53, 445–​53.
IUPAC. 2004. Glossary of terms used in toxicokinetics (IUPAC
IUPAC. 2007. Glossary of terms used in toxicology, 2nd edition,
IUPAC recommendations, Chem. Int. 31(4), 29. https://​
envirotoxinfo.nlm.nih.gov/​toxicology-​glossary-​b.html, last
access 14 November 2019.
Kahane R, Vialle-​Guerin E, Boukema I, Tzanoudakis D, Bellamy C,
Chamaux C, Kik C. 2001. Changes in non-​structural carbo-
hydrate composition during bulbing in sweet and high-​solid
onions in field experiments, Environ. Exp. Bot. 45, 73.
Kördel W, Peijnenburg W, Klein CL, Kuhnt G, Bussian BM, Gawlik
BM. 2009. The reference-​matrix concept applied to chemical
testing of soils, Trends Anal. Chem. 28(1), 51.
Le Gall G, Colquhoun IJ, Davies AL, Collins GJ, Verhoeyen
ME. 2003. Metabolite profiling of tomato (Lycopersicon
esculentum) using 1H NMR spectroscopy as a tool to detect
potential unintended effects following a genetic modification,
J. Agric. Food Chem. 51, 2447.
Ly-​Nguyen B, Van Loey AM, Fachin D, Verlent I, Indrawati M,
Hendickx ME. 2002. Partial purification, characteriza-
tion, and thermal and high-​ pressure inactivation of pectin
methylesterase from carrots (Daucus carota L.), J. Agric. Food
Chem. 50, 5437.
Nonnecke IL. 1989. Vegetable Production, p. 75. Van Nostrand
Reinhold, New York, NY.
Oliveira FAR. 1988. Mass transfer analysis for the leaching of water
soluble components from food, PhD thesis. University of
Leeds, Leeds.
O’Donoghue EM, Omerfield SD, Bendall M, Hedderly D, Eason
J, Sims I. 2004. Evaluation of carbohydrates in Pukekohe
Longkeeper and Grano cultivars of Allium cepa, J. Agric. Food
Chem. 52, 5383.
Oomen AG, Rompelberg CJM, Bruil MA, Dobbe CJG, Pereboom
DPKH, Sips AJAM. 2003. Development of an in vitro digestion
model for estimating the bioaccessibility of soil contaminants,
Arch. Environ. Contam. Toxicol. 44, 281–​7.
Shargel L, Yu AB. 1999. Applied Biopharmaceutics &
Pharmacokinetics (4th ed.). McGraw-​Hill, New York, NY.
Shi J, Le Maguer M. 2000. Lycopene in tomatoes: chemical and
physical properties affected by food processing, Crit. Rev.
Biotechnol. 20, 293–​334.
Stapley AGF, Sousa Gonçalves JA, Hollewand MP, Gladden LF,
Fryer PJ. 1995. An NMR pulsed field gradient study of the
electrical and conventional heating of carrot, Int. J. Food Sci.
Technol. 30, 639.
80
Tardieu A, Guerez A, Phana S, De Man W, This H. 2009. Quantitative
Nuclear Magnetic Resonance (qNMR) analysis of mono-​and
disaccharides in aqueous solutions obtained by soaking raw or
fried dice of onion bulbs (Allium cepa L.), J. Food Sci. 74(4),
C319–​325.
Tardieu A, France MB, This H. 2011. NMR Determination of a
Model of Solute Release from Plant Tissues in an Aqueous
Environment, Fruit&Veg Processing, Avignon, France (to be
published).
This H. 2005. Modelling dishes and exploring culinary “precisions”:
the two issues of molecular gastronomy. Br. J. Nutr. 93 (4),
S139–​S146.
This H, Weberskirch L, Plassais M, Luna A, His A, Skoglund S.
2010. La RMN du liquide voit le coeur des légumes et des
viandes... puisque ce sont des gels, L’Actualité chimique. 337
(1), 10.
Hervé This vo Kientza
This H. 2016. Statgels and dynagels, Notes Académiques de
l’Académie d’agriculture de France /​Academic Notes from the
French Academy of Agriculture. 12, 1–​12.
This H. 2012. Solutions are solutions, and gels are almost solutions,
Pure Appl. Chem. 85 (1), 257–​276.
Turner NC. 1981. Techniques and experimental approaches for the
measurement of plant water status, Plant and Soil. 58, 339–​366.
Varoquaux P, Varoquaux F, Tichit L. 1986. Loss of nitrate from
carrots during blanching, J. Food Technol. 21, 401–​407.
Viola R, Davies HV. 1992. A microplate reader assay for rapid
enzymatic quantification of sugars in potato tubers, Potato Res.
35, 55–​58.
Weberskirch L, Luna A, Skoglund S, This H. 2011. Comparison
of two liquid-​state NMR methods for the determination of
saccharides in carrot (Daucus carota L.) roots, Anal. Bioanal.
Chem. 399 (1), 483–​487.
Browning: The Glycation and Maillard Reactions –​Major
Non-​Enzymatic Browning Reactions in Food
Frédéric J. Tessier
Université de Lille, Inserm U1167, F-​59000 Lille, France
Glycation is the most general internationally accepted term
for adduction of a sugar to another biomolecule. In particular,
glycation of proteins by reducing sugars is considered the first
step in the “Maillard reaction”, a central group of chemical
reactions that occurs when food turns brown in cooking and is
involved in formation of flavour compounds and new textures.
For some 60 years, food chemists and other scientists have been
working to elucidate the mechanisms involved when glycation
reaction products form, and further, to interpret the impact of
the glycation reaction on the quality of the food and to assess
the physiological consequences related to the ingestion of
such neoformed compounds. When specific glycation reaction
products such as acrylamide are revealed as potentially harmful
to human health, mitigation strategies should be developed, both
at home and in commercial production, to reduce the risk of
exposure. On the other hand, the chemical pathways of beneficial
glycation products, such as melanoidins, must be promoted to
ensure their optimum content in food. The most difficult task of
all in food chemistry is to strike a balance between the formation
of desirable and undesirable glycation reaction products.
Today, most of the foods that we eat are thermally processed
either during cooking at home, in restaurants or in the course of
more or less complex processes in the food industry. The heat
treatments applied to fresh foods such as meat and eggs, and to
mixtures of raw ingredients such as flour, butter, milk and sugar,
induce thousands of chemical reactions between molecules.
These are responsible for the formation of a diverse range
of sensory-​active compounds, which usually enhance the sen-
sory qualities of food (Cerny, 2008). The new odorant and taste
compounds formed during cooking are usually derived from
glycation reactions, which were first discovered by Emil Fischer
and particularly thoroughly investigated by Louis-​ Camille
Maillard (1912). They play a crucial role not only in the synthesis
of odorant and taste compounds but also in the formation of
colours, from yellow to brown, in cooked food. Another complex
combination of chemical reactions that leads to the browning of
food is caramelisation (Kroh, 1994). These two reactions, the
Maillard reaction and caramelisation, are sometimes considered
to be one and the same, which is probably because: (1) they
can result in the formation of the same compounds, such as
hydroxymethylfurfural (HMF), (2) they often take place in par-
allel during cooking and (3) they lead to the formation of the
same brown colour. However, the main difference between cara-
melisation and the Maillard reaction is that the former needs only
simple saccharides such as sucrose, glucose or fructose, whereas
the latter requires both reducing sugars and amino groups from
proteins, which react together to form the so-​called Maillard
reaction products (MRPs). Compared with caramelisation, which
needs high temperatures, the Maillard reaction can start at low
temperatures (around 30 °C) (Tessier, 2010).
Initially, research on the Maillard reaction focused on the dis-
covery of its complex chemical pathways (Hodge, 1953), with
the major goals of controlling the formation of key odorants in
foods and producing a full range of synthetic flavourings for the
food industry (Belitz and Grosch, 1999). However, since the dis-
covery of the potential relationship between the intake of MRPs
other than the aroma compounds and potential health effects
(Tessier and Niquet, 2007; Delgado-​ Andrade and Fogliano,
2018) questions related to the toxicity of some MRPs and other
neoformed molecules have been gaining increasing attention
recently. The discovery of the formation of acrylamide by
glycation reactions in certain foods such as coffee and French
fries (Mottram et al., 2002; Tareke et al., 2002) has also attracted
the attention of health scientists and public health authorities.
In this chapter, a brief presentation of the three main chem-
ical steps of the Maillard reaction will be followed by a focus on
the browning of butter during cooking and on the formation of
acrylamide in a selection of foods, particularly starchy foods like
bread and potatoes, and roasted plant ingredients like coffee and
cocoa beans. Finally, possible beneficial or harmful health effects
associated with the consumption of food high in MRPs will also
be discussed.
The Basic Chemical Pathways of the Maillard
Reaction
The reaction between simple sugars and amino groups from
amino acids was first observed in 1884 by Emil Fischer, before
Maillard explored the reaction between reducing sugars and
81
82
amino acids (Maillard, 1912), but it was not chemically described
for another 40 years. John E. Hodge (1953) was the first to pro-
pose a general scheme of the Maillard reaction, and his scheme
remains the most accurate reference for all scientists in this field.
Initial Phase
In food matrices, the initial step in the reaction corresponds to
the formation of imine intermediates (also known as Schiff bases
or N-​substituted-​glycosylamines) between reducing sugars such
as glucose or fructose and free amino acids or protein-​bound
amines. The unstable imines can then regenerate the initial
substrates (i.e., sugar and amine), generate dicarbonyl compounds
through oxidative reaction (e.g., glucosone), or rearrange them-
selves into stable Amadori and Heyns products (Rufian-​Henares
and Pastoriza, 2016). The two types of early Maillard products
are formed from aldoses (e.g., glucose) and ketoses (e.g., fruc-
tose), respectively. They represent the first stable MRPs that can
be detected and thus quantified in foods. Neither the Amadori
nor the Heyns compounds absorb wavelengths of visible light.
Therefore, in the early step of the Maillard reaction, no browning
can be observed on the food.
The main early product found in food is fructoselysine (FL),
which is an amino-​deoxy-​ketose formed from the nucleophilic
attack of the ε-amino group of lysine (an essential amino acid) on
a molecule of glucose. The FL is either found on dietary proteins
or as a free Maillard product when the reaction occurs on free
lysine.
Different studies indicate that 1 L of UHT milk contains
between 130 and 600 mg FL (Henle, 2003; Erbersdobler and Faist,
2001). Due to the difference in the intensity of the heat treatment,
pasteurised milk always contains less FL than sterilised milk
(also known as UHT milk). The concentration of FL is usually
proportional to the heat load as long as the heat treatment of food
is moderate. When the thermal processing is more intense, such
as in baking or roasting, the degradation of FL may become more
intense than its formation, and therefore an apparent decrease
of FL can be observed over the cooking time. Overall, the daily
intake of FL has been estimated at around 100 mg per day when a
Western diet is consumed (Tessier and Birlouez, 2012).
Propagation Phase
The propagation phase of the Maillard reaction, also called the
intermediate stage, is a cascade of parallel chemical reactions
that lead to the formation of thousands of compounds, some of
which are more or less stable (Figure 13.1). Depending on the
physicochemical conditions of the food matrices (pH, water
activity, temperature, etc.), different intermediate molecules
will be formed after the rearrangement of the Amadori or Heyns
products (Rufian-​ Henares and Pastoriza, 2016). For instance,
fission reactive products such as glyoxal, methylglyoxal and other
carbonyls will react with amino groups to generate more stable
compounds, including odorants (e.g., pyridines, pyrazines and
imidazoles) and browning products. Some of the stable products
formed during the propagation phase are UV active, exhibit the
Frédéric J. Tessier
typical brown colour of the Maillard reaction, have fluorescent
properties and can be classified as flavour compounds.
Among the stable products formed during the propagation
phase, some of them, such as Nε-​carboxymethyllysine (CML),
glucosepane, pentosidine and pyrraline, were discovered almost
at the same time in human tissues and in foods. It must be noted
that the Maillard reaction also occurs at 37 °C in the human body
(Tessier, 2010). In this case, the reaction is called “glycation” and
the products formed are named the AGEs, or advanced glycation
end products (Henning and Glomb, 2016). Thus, AGEs and
MRPs may define the same molecules, and it is simply the case
that biologists and physicians use the term AGEs whereas food
chemists use the term MRPs.
In physiological conditions, CML, pentosidine and other
AGEs are the most advanced products that can be found in living
organisms, and it can be said that they “finish” the reaction. In
food, the situation is quite different, especially when the dur-
ation of the treatment is long and the temperature level is high
(e.g., roasting, baking, frying or grilling). In such cases, the final
compounds are called “melanoidins”.
Termination Phase
The termination phase in food occurs during prolonged heating
and at temperatures usually above 100 °C. In such conditions,
the water content and activity are usually low, at least at the sur-
face of the food matrices (e.g., the high dehydration of the bread
crust, the roasting of coffee), and the intermediate MRPs undergo
interactions between themselves and other compounds to pro-
mote the formation of brown polymers named melanoidins (Wang
et al., 2011) (Figure 13.1). Although the term “melanoidins” was
very likely created more than a century ago by Schmiedeberg
(1897) and used by Maillard to define the end products of his
chemical reaction (1912), no precise chemical structure for them
has been fully characterised so far. The complexity of isolating
and solubilising them makes their identification and quantifi-
cation difficult, if not impossible (Helou et al., 2016). What is
known so far is that their structures vary greatly depending on
the composition of the raw ingredients and the processing of
the foodstuff (Martins and van Boekel, 2003). For instance, the
chemical structures of melanoidins found in coffee are consid-
erably different from those in cereal-​based products. Even in a
single food type, such as bread, a complex variety of melanoidins
can be observed.
At least four hypotheses have been proposed concerning the
formation of melanoidins: (1) a polymerisation of furan and pyr-
role units (Tressl et al., 1998); (2) a polymerisation of sugar deg-
radation products formed during the Maillard reaction and linked
by amino compounds (Cämmerer el al., 2002); (3) a protein skel-
eton cross-​linked by carbohydrate-​derived structures (Hofmann,
1998) or MRPs (Hofmann et al., 1999); and (4) ketal and acetal
linkages between phenolic compounds, hydroxyl acids and
sugars (Moreira et al., 2017).
Overall, melanoidins can be defined as high-​molecular-​weight
nitrogen-​containing brown-​coloured pseudo-​polymers (Wang
et al., 2011).
Browning: Glycation and Maillard Reactions
FIGURE 13.1 Schematic representation of the three phases of the Maillard reaction.
Yield of Different Maillard Reaction Products
The yield of formation of the different MRPs at each stage of the
reaction has been estimated through different tests with model
systems for the Maillard reaction (e.g., the reaction between
sugars and amino acids in test tubes) and more importantly,
through the analysis of a variety of food products over recent
decades (Cerny, 2008). Depending on the food products, 1% to
10% of the initial reactants, both amino acids and sugars, can be
transformed into Amadori or Heyns products, 0.1% to 1% are
modified into stable AGEs such as CML, deoxyglucosone and
other reactive intermediate compounds, 0.01% to 0.1% lead to
the formation of fragmentation products such as glyoxal, and,
lastly, odorant compounds represent less than 0.1% of the initial
reactants.
The yield of formation for the most chemically reactive, most
thermodynamically unstable or most volatile MRPs may have
been underestimated, since they do not accumulate over time in
food matrices but either react with other compounds or are partly
eliminated in the atmosphere.
83
What Is behind the Browning of Butter?
Traditional butter contains fat (81% w/​ w), proteins (0.85%),
sugars (0.06%) and water (16%). We are all familiar with the fact
that the colour of butter changes from yellow to brown when it is
heated. One might believe that the formation of a brown colour
is due to lipid peroxidation, but our data indicate that the oxi-
dation of fat from butter is always very low and cannot explain
the browning after heating (Niquet-​Léridon et al., 2015). The
absence of browning in clarified butter (100% fat) indicates that
the fatty acid residues present in the triglycerides of butter are
stable even at temperatures above 150 °C. The high proportion
of saturated fatty acid residues and low proportion of unsaturated
fatty acid residues in butter account for this stability.
The lack of browning in heated clarified butter also suggests
that proteins and saccharides, alone or together, are the key players
of the browning reaction. In order to confirm the involvement of
the Maillard reaction in the process of browning, CML and HMF,
two markers of this reaction, have been quantified in raw and
cooked butter. No detectable HMF and only traces of CML were
84
found in raw butter. But after an exaggerated heat treatment of
butter (25 min at 150 °C), CML and HMF were found in sig-
nificant amounts (2 to 3 µg/​g and 51 to 58 µg/​g, respectively).
CML and HMF were not detected in either raw or heated clari-
fied butter. While CML is indeed a unique marker of the Maillard
reaction, HMF is not fully specific for this reaction and can also
be a marker of caramelisation. This implies that carmelisation, in
addition to the Maillard reaction, is a potential agent in the for-
mation of browning in cooked butter.
CML and HMF are currently under investigation for their
potential deleterious effects on health (Tessier et al., 2016;
Murkovic and Pichler, 2006). However, a reasonable daily intake
of cooked butter (20 g) would correspond to approximately 1%
and 6% of the total exposure to CML and HMF, respectively,
when foods commonly consumed are taken into account. A recent
study also showed that the concentration of volatile α-​dicarbonyl
compounds found in cooked butter was much lower than that in
heated beef fat, margarine and safflower oil (Jiang et al., 2013).
Based on current scientific knowledge, we can conclude that
there is no evidence that heated butter is potentially toxic and,
therefore, that it is, in fact, perfectly suitable for cooking.
Acrylamide, the Dark Side of Glycation
Acrylamide is a colourless and odourless amide (CH2CHCONH2)
of low molecular weight, 71.08 g/​mol. In 1994, it was classified
FIGURE 13.2 Schematic major pathway for the formation of acrylamide in food.
Frédéric J. Tessier
by the International Agency for Research on Cancer (IARC) as
probably carcinogenic to humans (Group 2A), and its status has
remained unchanged. The IARC originally stated that acrylamide
was not known to occur as a natural product and that it was chem-
ically produced for various industrial practices such as water
treatment (IARC, 1994). Exposure to it was believed for that
reason to be essentially occupational (and related to smoking)
until the discovery by Tareke and her colleagues in 2002 that
acrylamide is present in heated foods (Tareke et al., 2002). This
was rapidly confirmed by other groups who found acrylamide in
grilled, baked, fried and toasted foods but not in raw or boiled
foods (Ahn et al., 2002).
Several mechanisms may be involved in the formation of acryl-
amide during the cooking process, but the predominant reaction
is the Maillard reaction (Mottram et al., 2002). It is, in fact, the
glycation between asparagine in its free form (not protein-​bound)
and reducing sugars or dicarbonyl compounds that leads to acryl-
amide forming at temperatures usually higher than 120 °C and
where the water content is low (Figure 13.2). Free asparagine
is a natural constituent of many plant-​derived foods, but it is
hardly found in animal-​derived foods. Consequently, acrylamide
is found, in significant amounts, mainly in plant-​based foods
such as cereals, potatoes, chicory and coffee beans, where the
raw materials contain its two precursors, asparagine and reducing
sugars, and when the foods are processed at high temperatures.
After more than a decade of investigation by public and pri-
vate scientific organisations, the European Food Safety Authority
Browning: Glycation and Maillard Reactions
(EFSA) has now taken a stance on acrylamide in foodstuffs
(European Food Safety Authority, 2015). Based on animal
studies, EFSA has confirmed the previous evaluations, which
stated that acrylamide in food potentially increases the risk for
consumers in all age groups of developing cancer.
In 2017, the European Commission adopted a regulation
asking European food companies to apply mitigation measures
with a view to achieving levels of acrylamide as low as reason-
ably achievable below the benchmark levels defined by the EFSA
for each category of “at risk” food (Commission Regulation
2017/​2158).
While the most concerned food processors have already put
mitigation measures in place by changing their raw ingredients
and recipes (e.g., selecting potato varieties with low quantities of
reducing sugars for the production of French fries or chips) or by
adjusting their processes (e.g., using new baking techniques), the
same cannot be said for home cooking. Indeed, the relative con-
tribution to exposure to acrylamide between foods processed by
the food industry and those processed at home is difficult, if not
impossible, to estimate.
Bender’s (1978) view from 40 years ago remains true; he
commented on the widespread belief that only commercially
processed foods posed a danger to health and emphasised that
even fresh food prepared at home could be equally damaging
when cooked at high temperatures, such as by the common
methods of frying, roasting and grilling.
The same can be applied to acrylamide, for which some non-​
governmental organisations have demanded a total ban in food.
While it is possible to limit the formation of acrylamide in food,
and this is what the European regulation is demanding of food
producers, it is not possible to “ban” acrylamide, since it is
formed naturally in various foodstuffs that require cooking.
Rather than trying to ban acrylamide in industrially processed
food, private and public organisations should contribute to a gen-
eral awareness and understanding by the population of the risks
associated with a high intake of starchy foods such as potatoes and
bread cooked and toasted at home at high temperatures for long
periods. Some national organizations like the UK Food Standards
Agency have sent out a message, “Go for Gold”, to encourage
people preparing food at home to make small changes in their
cooking habits (UK Food Standards Agency, 2018). Moderate
cooking temperatures and slow cooking are already among the alter-
native techniques used by chefs in restaurants, and those methods
should be extended to domestic practice. The risk associated with
the dietary exposure to acrylamide must not, however, be used as
an argument to encourage “raw food diets”, which will eventually
lead to an unbalanced diet and most likely reduce food diversity.
Concluding Remarks and Perspectives
The glycation reactions are key reactions that influence not only
the sensory quality of cooked food but also its nutritional quality.
They can, though, be harmful to health when high temperatures
are used in cooking. People who cook, whether chefs, those in the
food industry or people at home, are fortunately, however, in con-
trol of the Maillard process to the extent that they have evidence
85
of the process when the food turns brown. As long as they are
aware of the potential dangers of cooking at high temperatures,
they are fully able to strike a balance between making the food
taste good and putting health at risk.
However, the recent discovery of acrylamide in food, together
with the potential physiopathological role of other MRPs, has
raised new questions about the safety of foods cooked at high
temperatures and brought new concerns for consumers already
aware of the problems.
There are conflicting opinions about the ultimate effects
of MRPs on human health. Some studies show that MRPs
contribute to the development of non-​ communicable and
age-​related pathologies (Cai et al., 2008; Cai et al., 2014;
Grossin et al., 2015). In this state, the MRPs are referred to
as “glycotoxins”. Another study indicates that CML, a major
Maillard product, accumulates in the body over time (Tessier
et al., 2015). However, there are voices challenging the
conclusions drawn from these and similar studies because of the
methods underlying the evidence on which these conclusions
are based (Delgado-​Andrade and Fogliano, 2018). The fact that
experiments used in some of those studies are animal-​rather
than human-​based is one important argument. There are other
concerns in terms of the methodologies of the experiments,
chiefly about the specificity and the accuracy of the dietary
exposure to glycotoxins and other MRPs.
As well as MRPs being detrimental to human health, certain
MRPs, such as melanoidins, are potentially beneficial for human
health. Taking all this into account, it is the overall composition
of the diet, in addition to isolated molecules or ingredients, that
must be focused on in future studies.
Until the toxic effects of some MRPs have been conclu-
sively refuted or confirmed, the best recommendation is to use
a variety of cooking methods (e.g., boiling, blanching, steaming,
poaching, simmering, searing, braising, roasting and grilling) as
well as eating a variety of foods. Our message is not “Cook Less
to Age Better” but “Cook Better to Age Better”.
REFERENCES
Ahn JS, Castle L, Clarke DB, Lloyd AS, Philo MR, Speck DR. 2002.
Verification of the findings of acrylamide in heated foods. Food
Additives & Contaminants, 19,1116–​1124.
Belitz HD, Grosch W. 1999. Aroma substances. In Food Chemistry,
Springer-​Verlag, Berlin Heidelberg New York.
Bender AE. 1978. Beneficial effects of food processing. In Bechtel
PJ (ed.), Food Processing and Nutrition, Academic Press,
London, 19–​23.
Cai W, He JC, Zhu L, Chen X, Zheng F, Striker GE, Vlassara H.
2008. Oral glycotoxins determine the effects of calorie restric-
tion on oxidant stress, age-​ related diseases, and lifespan.
American Journal of Pathology,173, 327–​336.
Cai W, Uribarri J, Zhu L, Chen X, Swamy S, Zhao Z, Grosjean F,
Simonaro C, Kuchel GA, Schnaider-​Beeri M, Woodward M,
Striker GE, Vlassara H. 2014. Oral glycotoxins are a modifi-
able cause of dementia and the metabolic syndrome in mice
and humans. Proceedings of the National Academy of Sciences,
111, 4940–​4945.
Cämmerer B, Jalyschko W, Kroh LW. 2002. Intact carbohydrate
structures as part of the melanoidin skeleton. Journal of
Agricultural and Food Chemistry, 50, 2083–​2087.
86
Cerny C. 2008. The aroma side of the Maillard reaction. Annals of
the New-​York Academy of Sciences, 1126, 66–​71.
Commission Regulation (EU) 2017/​ 2158 of 20 November 2017
establishing mitigation measures and benchmark levels for
the reduction of the presence of acrylamide in food. Official
Journal of the European Union. L 304/​24–​44.
Delgado-​ Andrade C, Fogliano V. 2018. Dietary advanced
glycosylation end-​products (dAGEs) and melanoidins formed
through Maillard reaction: physiological consequences of their
intake. Annual Review of Food Science and Technology, doi.
org/​10.1146/​annurev-​food-​030117-​012441.
European Food Safety Authority. 2015. Scientific opinion on acryl-
amide in food. EFSA Journal, 13(6), 4104.
Erbersdobler HF, Faist V. 2001. Metabolic transit of Amadori
products. Nahrung, 45, 177–​181.
Grossin N, Auger F, Niquet-​ Léridon C, Jacolot P, Durieu N,
Montaigne D, Schmidt AM, Susen S, Jacolot P, Beuscart JB,
Tessier FJ, Boulanger E. 2015. Dietary CML-​enriched protein
induces functional arterial aging in a RAGE-​dependent manner
in mice. Molecular Nutrition & Food Research, 59, 927–​938.
Helou C, Jacolot P, Niquet-​Léridon C, Gadonna-​Widehem P, Tessier
FJ. 2016. Maillard reaction products in bread: a novel semi-​
quantitative method for evaluating melanoidins in bread. Food
Chemistry, 190, 904–​911.
Henle T. 2003. AGEs in foods: do they play a role in uremia? Kidney
International, 84, S145–​147.
Henning C, Glomb MA. 2016. Pathways of the Maillard reaction
under physiological conditions. Glycoconjugate Journal, 33,
499–​512.
Hodges JE. 1953. Dehydrated foods, chemistry of browning reactions
in model systems. Journal of Agricultural Food Chemistry, 1,
928–​943.
Hofmann T. 1998. Studies on the relationship between molecular
weight and the colour potency of fractions obtained by thermal
treatment of glucose/​amino acid and glucose/​protein solutions
by using ultracentrifugation and colour dilution techniques.
Journal of Agricultural Food Chemistry, 46, 3891–​3895.
Hofmann T, Bors W, Stettmaier K. 1999. Radical-​assisted melanoidin
formation during thermal processing of foods as well as
under physiological conditions. Journal of Agricultural Food
Chemistry, 47, 391–​396.
International Agency for Research on Cancer (IARC). 1994.
Acrylamide. In IARC Monographs on the Evaluation of
Carcinogenic Risks to Humans, ed. World Health Organization,
60, 389–​433.
Jiang Y, Hengel M, Pan C, Seiber JN, Shibamoto T. 2013.
Determination of toxic a-​ dicarbonyl compounds, glyoxal,
methylglyoxal, and diacetyl, released to the headspace of lipid
commodities upon heat treatment. Journal of Agricultural
Food Chemistry, 61, 1067–​1071.
Kroh LW. 1994. Caramelisation in food and beverages. Food
Chemistry, 51, 373–​379.
Maillard LC. 1912. Action des acides aminés sur les sucres : forma-
tion des mélanoïdines par voie méthodique. Comptes Rendus
de l’Académie des Sciences, 54, 66–​68.
Frédéric J. Tessier
Martins SIFS, van Boekel MAJS. 2003. Melanoidins extinction
coefficient in the glucose/​ glycine Maillard reaction. Food
Chemistry, 83, 135–​142.
Moreira ASP, Nunes FM, Simões C, Maciel E, Domingues P,
Domingues MRM, Coimbra MA. 2017. Transglycosylation
reactions, a main mechanism of phenolics incorporation in
coffee melanoidins: inhibition by Maillard reaction. Food
Chemistry, 227, 422–​431.
Mottram DS, Wedzicha BL, Dodson AT. 2002. Acrylamide is formed
in the Maillard reaction. Nature, 419, 448–​449.
Murkovic M, Pichler N. 2006. Analysis of 5-​hydroxymethylfurfural
in coffee, dried fruits and urine. Molecular Nutrition and Food
Research, 50, 842–​846.
Niquet-​Léridon C, Jacolot P, Niamba CN, Grossin N, Boulanger E,
Tessier F.J. 2015. The rehabilitation of raw and brown butters
by the measurement of two of the major Maillard products,
Nε-​carboxymethyl-​lysine and 5-​hydroxymethylfurfural, with
validated chromatographic methods. Food Chemistry, 177,
361–​368.
Rufian-​Henares JA, Pastoriza S. 2016. Maillard reaction. In
Caballero B, Finglas PM, Toldra F (eds) Encyclopedia of Food
and Health, Elsevier, Oxford, 593–​600.
Schmiedebergs O. 1897. Über die Elementarformeln einiger
Eiweißkörper und über die Zusammensetzung und die Natur
der Melanine. Naunyn-​Schmiedebergs Archives, 31, 1.
Tareke E, Rydberg P, Karlsson P, Eriksson S, Törnqvist M. 2002.
Analysis of acrylamide, a carcinogen formed in heated
foodstuffs. Journal of Agricultural and Food Chemistry, 50,
4998–​5006.
Tessier FJ. 2010. The Maillard reaction in the human body. The
main discoveries and factors that affect glycation. Pathologie
Biologie, 58, 214–​219.
Tessier FJ, Birlouez-​ Aragon I. 2012. Health effects of dietary
Maillard reaction products: the results of ICARE and other
studies. Amino Acids, 42, 1119–​1131.
Tessier FJ, Niquet C. 2007. État des connaissances sur la
biodisponibilité et la toxicité des produits de Maillard issus
de l’alimentation. Journal de la Société de Biologie, 201,
199–​207.
Tessier FJ, Niquet-​Léridon C, Jacolot P, Jouquand C, Genin M,
Schmidt AM, Grossin N, Boulanger E. 2016. Quantitative
assessment of organ distribution of dietary protein-​ bound
13C-​labeled Nɛ-​ carboxymethyllysine after a chronic oral
exposure in mice. Molecular Nutrition & Food Research, 60,
2446–​2456.
Tress lR, Wondrak GT, Garbe L, Kruger RP, Rewicki D. 1998.
Pentoses and hexoses as sources of new melanoidin-​ like
Maillard polymers. Journal of Agricultural and Food
Chemistry, 3, 1765–​1776.
UK Food Standards Agency. 2018. Go for Gold. www.food.gov.uk/​
science/​acrylamide-​0, last access 4 December 2019.
Wang H-​Y, Qian H, Yao W-​R. 2011. Melanoidins produced by the
Maillard reaction: Structure and biological activity. Food
Chem., 128, 573–​584.
Canning: Appert and Food Canning
Jean-​Christophe Augustin
Ecole Nationale Vétérinaire d’Alfort, 7 avenue du Général de Gaulle, 94704 Maisons-​Alfort, France
Appertization: The Art of Preserving Animal and
Vegetal Substances
At the end of the 18th century, Nicolas Appert (1749–​1841), a
French confectioner, developed an original method of preserving
large amounts of foods for long periods. He found that the appli-
cation of heat to food in sealed glass bottles preserved the food
from deterioration. The “appertization” industry was thus born in
1795 at Ivry-​sur-​Seine, near Paris, where Appert began to elab-
orate heat-​preserved food products (Barbier, 1994). In 1802, he
relocated his activity to a bigger factory in Massy and produced
preserved meat, vegetables and fruits.
At that time, foods were preserved with the addition of
salt, alcohol, vinegar, sugar or fat or by drying and smoking.
Unfortunately, all these processes deeply modified the intrinsic
properties of food, and in particular their flavour. The preser-
vation method proposed by Appert had the advantage of not
altering the food. Gourmets and marine officers were the first
to praise his invention. The former emphasized the “freshness”
of preserved foods, allowing them to “enclose seasons in glass
bottles”. The latter underlined the remarkable capacity of the
process to preserve a variety of foods and the lack of spoilage
of bottled perishable food products after durability studies
performed aboard ships. The Navy was moreover particularly
interested in this process, since it could prevent scurvy among
ship crew members, which was attributed to the consumption
of salted meats. In 1809, a committee of three members (Louis-​
Bernard Guyton-​ Morveau, Antoine Parmentier and Bernard-​
Felix Bouriat) of the Society supporting the National Industry
published a report after testing ten animal and vegetable food
products preserved for more than eight months. The committee
reported that all foods were perfectly edible and expressed a
positive opinion of Appert’s process.
The recognition of his work came in 1810, when Nicolas
Appert chose to publish and disseminate his findings and received
a 12,000 Franc grant from the Ministry of the Interior for his
work. He published a book entitled Le livre de tous les ménages
ou l’art de conserver pendant plusieurs années toutes les
substances animales et végétales (The book for all households or
the art of preserving animal and vegetable substances for many
years) (Appert, 1810). He described the following successive
steps in the process:
• enclose, in bottles or jars, the food to be preserved;
• cork the bottle carefully;
• submit the bottles to the action of boiling water for
various lengths of time, depending on the food;
• remove the bottles after the prescribed time.
Food canning was empirically invented. The process was soon
improved in England, where a method of sealing food into
unbreakable cylindrical tin or wrought-​iron canisters (“cans”)
was developed. The efficiency of the thermal process was also
improved, and the first pressure retort was built during the mid-​
19th century, allowing steam temperatures higher than 100 °C to
be used. This provided the benefit of reduced thermal process
times, which in turn improved the flavour, texture and nutritional
value of the food.
Despite not understanding why the heat process prevented
the food from going bad, Appert was convinced that his method
depended on the action of heat inhibiting food spoilage.
The French chemist Louis-​ Joseph Gay-​ Lussac (1778–​ 1850)
thought that oxygen present in bottles was modified by heating,
allowing the preservation of foods. It was not until 50 years
later that Louis Pasteur (1822–​1895) provided the explanation
for the effectiveness of canning when he demonstrated the role
of microorganisms in food spoilage. In 1862, he performed
experiments that proved that heat-​treating beverages (milk, beer
and wine), a process that became known as pasteurization, could
stop them from spoiling.
The Botulinum Cook
Although the heating times prescribed by Appert were short,
between one and two hours in boiling water (Appert, 1810),
food spoilage and foodborne illnesses involving canned foods
were not formally reported before the end of the 19th century.
After 1895, the William Underwood Company in the USA,
which experienced tin cans swelling, decided to launch studies
to fix this problem. Many cases of canned food spoilage were
studied by William Underwood and Samuel Prescott, who
demonstrated that heat-​resistant bacterial spores were the causa-
tive agents of spoilage and that the proper time–​temperature
combinations could prevent it (Wanucha, 2009). They were
87
88
the first to perform time–​ temperature studies and published
time–​temperature requirements of canned foods to control their
spoilage.
More dramatically, at the same period, canned foods were also
implicated in botulism outbreaks. Foodborne botulism is a very
serious form of food poisoning, a flaccid paralysis with a high
case-​fatality rate caused by ingesting preformed Clostridium
botulinum neurotoxin. The name “botulism” is derived from the
Latin word “botulus” meaning sausage and came to be used in
Europe in the 18th century to describe a disease associated with
muscle paralysis, breathing difficulties without loss of cognition
and a sensory system remaining intact, which was frequently
linked to the consumption of blood sausage. At the end of the 18th
century, cases of fatal poisoning were observed in Württemberg in
southern Germany following the consumption of smoked blood-​
sausage (Erbguth and Naumann, 1999). In 1820 and 1822, the
District Medical Officer Justinus Kerner published monographs
detailing more than 200 cases of “sausage poisoning”, describing
accurately the clinical presentation of foodborne illness and
speculating on the mechanism of “the fat poison”. The causative
agent, named Bacillus botulinus, was isolated in 1896 by Emile
van Ermengem from inadequately cured ham in Belgium (Van
Ermengem, 1979). He established that foodborne botulism was
an intoxication, not an infection, and that the toxin was produced
by a spore-​forming anaerobic bacterium.
Spores of C. botulinum are ubiquitous in the environment.
Spore germination and bacterial growth allowing the elaboration
of toxin occur in anaerobic, low-​salt (<10%), low-​acid (pH >
4.6) environments at temperatures above 10 °C for classical pro-
teolytic strains and 3 °C for nonproteolytic strains (Peck et al.,
2011). The canning of foods is conducive to creating anaerobic
conditions; hence, low-​acid foods even slightly contaminated
with C. botulinum spores receiving inadequate heat treatment,
allowed to stand for a time and eaten undercooked (the toxin is
heat-​labile) may cause botulism.
In November 1913, an outbreak involved 12 young people
at Stanford University who had consumed a string bean
salad. Dickson (1918) was the first to observe that spores of
C. botulinum could survive for two hours in boiling water and
recommended that an educational campaign be instituted so
that all who practised the home-​canning of fruits and vegetables
might be informed of the danger. The first outbreak of foodborne
botulism recorded in the UK happened in August 1922 at Loch
Maree, in Scotland. The consumption of sandwiches containing
wild-​duck paste preserved in a glass jar caused the death of eight
persons (Leighton, 1923). There was little suspicion of commer-
cial canning at that time. In the USA, from 1910 to 1919, there
were 48 outbreaks attributed to home-​processed food and 14 to
commercially processed food. However, between August and
November 1919, commercially canned ripe olives caused a total
of 28 intoxications with 17 deaths.
Hence, in December 1919, the Canners League of California,
the California Olive Association and the National Canners
Association proposed and financed a detailed investigation by
the University of California and the Stanford University. This
project involved assessing the dangers of botulism in commer-
cially prepared food, how it originates, and how it could best
Jean-Christophe Augustin
be eliminated. Karl Friedrich Meyer and his co-​workers were
entrusted with the investigation, and studies were carried out
between 1919 and 1926 (Meyer, 1973).
Sterilization standards were scientifically established, derived
from experiments exposing the spores of 109 different strains
of C. botulinum to different heating temperatures and demon-
strating that thermal destruction of the most resistant pathogens
required a four-​minute exposure at 120 °C and 330 minutes at
100 °C (Esty and Meyer, 1922). Sanitary standards of packing
plants and inspection services regulating the processing of low-​
acid foods improved the safety of canned foods (Meyer, 1973).
From this period, botulism did not disappear, but home-​canned
rather than commercial-​canned foods became the leading cause
of foodborne botulism (Hall, 1943; Sobel et al., 2004). In the
USA, commercial-​canned foods accounted for 32% of the 125
outbreaks of known source from 1910 to 1929 but for only 3.3%
of the 305 outbreaks of known source reported during the period
1930–​1959 (Sabin, 1980).
Thermobacteriology: Modelling the Thermal
Processing of Foods
Early time–​temperature studies conducted at the beginning of
the 20th century were carried out, giving rise to what was named
“thermobacteriology” (Stumbo, 1949). Thermobacteriology
is dedicated, on the one hand, to studying undesirable heat-​
resistant microorganisms potentially present in canned foods,
and, on the other hand, to studying heat transfers in canned
foods during thermal processing. Mathematical models were
developed to describe the destruction of microorganisms during
heat treatment. These models are very simple and allow the heat
resistance of microorganisms to be characterized with only two
parameters: D, the decimal reduction time (required heating
time for a survival ratio of 10% of the microbial population)
(Katzin et al., 1943), and z, the thermal death-​time curve par-
ameter (interval in temperature yielding a ten-​fold change in D)
(Bigelow, 1921).
Measurement of temperature during the heating and cooling
of foods in hermetically sealed containers was performed and
allowed models to be developed that described the transfer of
heat to the food’s thermal centre for conduction or convection
heating (Ball, 1923; Ball and Olson, 1957). Parameters character-
izing bacterial resistance to heat were integrated with parameters
describing heat transfer and heat intensity to estimate the amount
of heating required to produce safe and “commercially sterile”
food products (Stumbo et al., 1975). The lethality requirement
depends on the acceptable statistical probability of observing
surviving microorganisms. This includes both pathogenic and
spoilage organisms that are capable of growth under the intended
storage conditions. The classical acceptable survival probability
of spores of C. botulinum is no greater than 1 viable spore in 1012
containers. Since the approximate maximum heat resistance of
C. botulinum spores may be represented by a D-​value of 0.21 min
at 121.1 °C, the lethality requirement at 121.1 °C was rounded up
to three minutes, known as the standard F03 sterilization process
(Stumbo et al., 1975).
Canning: Appert and Food Canning
An F03 process is safe with respect to C. botulinum spores, but
products may contain a small number of surviving spores of more
heat-​resistant spoilage microorganisms, jeopardizing the “com-
mercial sterility” of canned foods. For spoilage, spore-​forming
mesophilic bacteria more heat resistant than C. botulinum, the
acceptable probability of survival to ensure reasonably minor
economic losses should be no greater than 1 viable spore per
about 104 containers (Stumbo et al., 1975). Quality degradation
calculations must therefore be performed in order to select the
thermal process that results in the highest product quality reten-
tion. Indeed, while destruction of spoilage agents is achieved
during the thermal process, quality degradation (nutrients,
vitamins, colour, texture and flavour loss) also occurs. Therefore,
from all the time–​temperature combinations that comply with the
minimum conditions necessary to free foods of microorganisms
that might spoil the foods or endanger the health of consumers,
those less deleterious to organoleptic and nutritive properties
must be chosen. This optimization can be achieved by knowing
the kinetics of quality degradation as a function of the time–​
temperature profiles (Holdsworth, 1985). Risk–​benefit models
can therefore be developed according to the time–​temperature
profile of the canning process and the intrinsic parameters of
foods (pH, oxygen content) to manage the microbial spoilage risk
versus the nutritional benefit of preserved foods (Rigaux et al.,
2012). These studies allow a compromise on process parameters
to be identified, optimizing nutritional and organoleptic quality
while keeping microbial risk at an acceptable level. These kinds
of studies emphasize the never-​ending need for greater precision
in calculating and assessing heat processes for food preservation
(Mafart et al., 2010).
REFERENCES
Appert N. 1810. L’art de conserver pendant plusieurs années toutes
les substances animales et végétales. Patris et Cie, Paris.
Bibliothèque Nationale de France; http://​gallica.bnf.fr/​, last
access 23 November 2019.
Ball CO (1923) Thermal process time for canned food. Bulletin of the
National Research Council, 7, 10.
Ball CO, Olson FCW. 1957. Sterilization in Food Technology.
McGraw-​Hill, New York, p. 654.
Barbier JP. 1994. Nicolas Appert: Inventeur et humaniste. Royer,
Paris, France.
89
Bigelow WD. 1921. The logarithmic nature of thermal death time
curves. The Journal of Infectious Diseases 29, 528–​536.
Dickson EC. 1918. Botulism –​A clinical and experimental
study –​Monograph No. 8, Rockfeller Institute for Medical
Research.
Erbguth FJ, Naumann M. 1999. Historical aspects of botulinum
toxin: Justinus Kerner (1786–​1862) and the “sausage poison”.
Neurology 53, 1850–​1853.
Esty JR, Meyer KF. 1922. The heat resistance of the spores of
B. botulinus and allied anaerobes. XI. The Journal of Infectious
Diseases 31, 650–​663.
Hall IC. 1943. The danger of botulism. American Journal of Public
Health 33, 818–​820.
Holdsworth SD. 1985. Optimisation of thermal processing –​a
review. Journal of Food Engineering 4, 89–​116.
Katzin L, Sandholzer L, Strong E. 1943. Application of the decimal
reduction time principle to a study of the resistance of coli-
form bacteria to pasteurization. Journal of Bacteriology 45,
265–​272.
Leighton GR. 1923. Botulism in Scotland. Nature 111, 415.
Mafart P, Leguerinel I, Couvert O, Coroller L. 2010. Quantification
of spore resistance for assessment and optimization of heating
processes: a never-​ ending story. Food Microbiology 27,
568–​572.
Meyer KF. 1973. The rise and fall of botulism. California Medicine,
The Western Journal of Medicine 118, 63–​64.
Peck MW, Stringer SC, Carter AT. 2011. Clostridium botulinum in
the post-​genomic era. Food Microbiology 28, 183–​191.
Rigaux C, Renard CMGC, Nguyen-​The C, Albert I, Carlin F. 2012.
Modeling risk-​ benefit in a food chain: nutritional benefit
versus microbial spoilage risk in canned green beans. AgroStat
2012, Paris, France.
Sabin AD. 1980. Karl Friedrich Meyer 1884–​1974. Biographical
memoirs. National Academy of Sciences 42, 269–​332.
Sobel J, Tucker N, Sulka A, McLaughlin J, Maslanka S. 2004.
Foodborne botulism in the United States, 1990–​ 2000.
Emerging Infectious Diseases 10, 1606–​1611.
Stumbo CR. 1949. Thermobacteriology as applied to food pro-
cessing. Advances in Food Research 2, 47–​115.
Stumbo CR, Purohit KS, Ramakrishnan TV. 1975. Thermal process
lethality guide for low acid foods in metal containers. Journal
of Food Science 40, 1316–​1323.
Van Ermengem E. 1979. A new anaerobic bacillus and its relation to
botulism. Reviews of Infectious Diseases 1, 701–​719.
Wanucha G. 2009. Two happy clams: The friendship that forged food
science. Food Technology 63, 11.
Capillarity in Action
Hervé This vo Kientza
1 INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy, F-​75005,
Paris, France
The phenomenon of capillarity can occur during culinary
processes because the thermal treatment of animal or plant
tissues weakens the material (whether collagenic tissue or plant
cell walls) holding together the cells making up the tissues. This
weakening is mainly the result of the hydrolysis of the polymers
(collagen and pectin, respectively) that are responsible for coher-
ence (Sila et al., 2006). When pores or cracks are created as a
result of cell separation, a surrounding liquid (e.g., a sauce in
which the plant or animal tissues are thermally treated) can enter
the tissues by capillarity (De Gennes et al., 2004).
In this chapter, we discuss this mechanism of capillarity in
the context of culinary processes, but we mainly observe that,
whereas this mechanism certainly plays a role in exchanges of
matter between food ingredients and their liquid environment,
the question is not so much to admit the possible involvement of
the mechanism as to assess its quantitative importance. We shall
finish by proposing an easy way of introducing a flavour into a
piece of plant or animal tissue.
Diffusion versus Capillarity
Before analysing particular phenomena that can be observed in
the kitchen and investigated from a physical and chemical (i.e.,
molecular gastronomy) point of view, it is interesting to observe
that, as we show here, capillarity was sometimes underestimated;
Aguilera et al. (2004) discussed the fact that it occurs in choc-
olate, for which the mechanism is often said, incorrectly, to be
“diffusion”. Such a confusion can be observed both in culinary
and in scientific circles, as discussed in the chapter in this book
about osmosis (“Osmosis in the Kitchen”).
For all the phenomena for which capillarity, diffusion, osmosis
or “imbibition” is considered as the interpretation, the core
question is the exchange of matter between food and its (often
liquid) environment (This, 2019). For example, when roots of
Daucus carota L. are heated in water with a view to producing a
“carrot stock”, the aqueous environment is progressively enriched
with many compounds, the most abundant being saccharides
(D-​glucose, D-​fructose and sucrose) and amino acids (Cazor
et al., 2006). The same holds for tomato sauces containing pieces
of the bulb of Allium cepa L. (Tardieu et al., 2009), or when
producing the beverage called “coffee” from ground, thermally
processed seeds of Coffea (Febvay et al., 2019).
Water loss from the interior of foods during thermal treatments
has sometimes been attributed solely to “diffusion”, and it is
possible that some molecular diffusion occurs when the open
channels of xylem and phloem of a cut plant tissue are in direct
contact with an aqueous environment (Bauchard and This, 2015).
However, other mechanisms can occur, such as the escape of
water vapour when high temperature causes water to evaporate,
either in the “crust” or inside the tissue if microwaves are used
for heating (see chapter in this book on evaporation). Fu et al.
(2003) studied moisture movement from the interior of a porous
and moist food matrix (for example, bread dough) during micro-
wave heating; the loss of water was due not only to moisture
diffusion but also to bulk vapour flow (convection) caused by the
positive pressure built up inside the product. Osmosis can also
come into play when plant and animal tissues are in an aqueous
environment, and the contraction of denaturated collagenic tissue
can lead to “juice” release when animal tissues are thermally
processed (Kopp et al., 1977).
It is not the purpose of this chapter to repeat why the word
“diffusion” is sometimes overused in food science and tech-
nology, as this was very well explained by Aguilera et al. (2004),
but it is useful to reiterate that the different velocities for material
transfer obey different laws, even if diffusion and capillarity
cannot always be readily distinguished after the solely kinetic
determination of some compounds.
For molecular diffusion, the German physiologist Adolf Eugen
Fick (1829–​1901) stated that, in the presence of a concentration
gradient, the net migration of molecules due to their random
motion occurs from a region of high concentration to one of
lower concentration; as seen in the chapter about osmosis, this
can be expressed using the chemical potential. The second Fick’s
law states that the rate at which this process proceeds at a point
M(x,y,z) in space, for a diluted binary system, is proportional to
the variation of the slope of the concentration gradient (Fick,
1855). The so-​called diffusion equation reads:
91
92
∂c ( x,y,z,t )
= D∆c ( x,y,z,t )
∂t
where c(x,y,z,t) is concentration at the point M, at time t, D is
the diffusion coefficient, or diffusivity, and ∆ is the Laplacian
operator.
Aguilera et al. (2004) showed that this model is widely used
(and abused) by food engineers as a general model for mass
transfer. The approach has the advantage that, by plotting experi-
mental data in the form of log (unaccomplished ratio of mass
transferred) against time, an apparent or effective diffusion coef-
ficient, Deff , can be determined from the straight portion of the
curve (Schwartzberg, 1987). The parameter Deff is to be correctly
redefined as a mass transfer coefficient, because it encompasses
all possible forms of mass transfer involved in the process, not
only diffusional ones.
A simple analysis used to adopt the Fickean diffusion model
is to check whether the ratio of the mass m(t) transported as
measured experimentally by the mass m(+∞) when the equi-
librium is reached varies proportionally to t (Geurtz and
Oortwijn, 1975).
For capillarity, on the other hand, the most common expres-
sion for capillary rise is the so-​called Lucas–​Washburn equation,
which assumes that the capillary pressure in a cylindrical capil-
lary in contact with an infinite liquid reservoir is compensated by
viscous drag and gravity (Krotov and Rusanov, 1999):
2 8 dh
γ cos (θ ) = 2 µh + ρgh
r r dt
where h is the distance the fluid is drawn into the capillary, γ
the surface tension of the fluid, θ the contact angle between the
fluid and the capillary wall, r the radius of the capillary, ρ and
μ the density and viscosity of the liquid, respectively, and g the
acceleration of gravity. The asymptotic solutions to this for short
and long times were discussed by Quéré (1997) and by Zhmud
et al. (2000). The short time limit (t→0) predicts that the liquid
movement in a horizontal capillary should be proportional to the
square root of time according to the expression:
r γ cos (θ )
h= t circumstances, salt can readily enter the meat (Figure 15.2) when
2µ
A plot of h against t is a straight line. Overall, curves for
diffusion as well as for capillary flow show a square-​root-​of-​time
dependence for mass transport at short times.
Sometimes, There Is No Entrance; Sometimes,
There Is One
As stated earlier, the issue is generally not so much to know
whether some capillarity can take place as how much of the
matter transfer is due to capillarity at various times after the
Hervé This vo Kientza
FIGURE 15.1 In this picture of the inside of a muscle that was cooked for
2 h in a solution of fluorescein, it can be seen that the fluorescent marker
diffused into the gelatinized collagenic tissue (green colour at the bottom and
on top left) by about 1 cm. A lower amount of fluorescein also entered the
meat by capillarity (top) between the muscular fibres. No fluorescein could be
found in the inside of the meat.
initial contact between food and its liquid (aqueous solution
or oil) environment. Regarding marination of red meat (Aktas
et al., 2003; Vlahova-​Vangelova and Dragoev, 2014), one can
observe that some publications discuss how the composition of
wines determines the tenderizing of cubes of meat marinated
in wines, but the mechanism is seldom discussed. Figure 15.1
shows the inner part of a muscle (longissimus dorsi) of Bos
taurus that was thermally treated for 2 h in a solution of “fluor-
escein” (sodium;3-​oxospiro[2-​benzofuran-​1,9′-​xanthene]-​3′,6′-​
diolate). Fluorescence spectroscopy was performed on ethanolic
extracts of samples from this muscle, but the green fluoresence
is enough to assess that some fluorescein diffusion occurred in
the gelatinized thick collagenic tissue and that some fluorescein
also penetrated the meat by capillarity when the collagenic tissue
was destroyed between muscular fibres. It it noteworthy that this
entry of the marker, through diffusion and capillarity, occurred in
spite of the strong meat contraction (by about one-​third in mass).
The study of grilled meat (see chapter on salt and meat by
This et al. in this handbook) corroborates this result; as shown
by analysis of sodium chloride in meat being grilled, no salt was
found inside the meat (processing times <10 min), but, in other
salt crystals are deposited on the upper surface of meat being
grilled; after “juices” appear at the upper surface, they dissolve
the sodium chloride, and the salty solution moves inward if the
collagenic tissue is weak and slits opens during cooking.
Capillarity occurs not only in muscular tissues but also in
some dishes made of plant tissues. In particular, the French
“gastronome” (actually a lawyer) Jean Anthelme Brillat-​Savarin
(1755–​1826) told the story of a clergyman who would eat spinach
(leaves of Spinacia oleracea) only if it had been cooked for sev-
eral days with the regular addition of cream (Brillat-​Savarin,
1825). Microscopic analysis of the spinach leaves did not show
any absorption (swelling), but capillarity was important because
the tissue disruption through pectin beta elimination created the
Capillarity in Action
FIGURE 15.2 In the first example (a), a piece of beef meat (Bos taurus,
longissimus dorsi, cut perpendicularly to fibres) was grilled. Because of the
low content of collagenic tissue, the bundles of muscular fibres separated,
creating crevices into which salted water could be attracted by capillarity. In
the second picture (b), the sauce in which the meat was cooked entered deeply
into the muscular tissue by capillarity, between the fibres, which separated
due to the partial dissolution of the collagenic tissue.
possibility of much capillarity between softened fragments of
spinach leaves. Of course, the phenomenon can also occur in
many other plant tissues commonly consumed by humans, such
as the stems of leeks (Allium porrum), bulbs of onions (Allium
cepa L.) or shallots (Allium ascalonicum).
The Mathematics behind the Phenomenon
In order to control matter transfers through capillarity, a mathem-
atical description of the phenomemon is useful. This begins by
calculating the difference of pressure between the inside and the
outside of a small structure dispersed in a medium.
Let us begin by observing that, if a fluid can dissolve into
another, energy is released, because bonds are created between
molecules of the first fluid and molecules of the second fluid. On
93
the other hand, if two fluids do not mix, it means that they need
energy to be in contact. In particular, drops of one liquid dispersed
in another organize so that their energy is minimized, i.e., the
area A of their contact surface is minimized. One can express this
using the free energy, F (here, for a simple treatment).
For example, we can write, at constant pressure:
dF = − PdV − SdT + γ dA
where P is the pressure, V the volume, S the entropy, T the abso-
lute temperature, and γ the “surface tension”, expressed in J/​m²
(but as 1 J = 1 N.m, it is often expressed in N/​m).
Let us, for example, consider the gas inside a bubble in a
liquid. The pressure inside the bubble is Pint and the pressure
outside is Pext. Let the radius of the bubble (assumed spherical)
be r.
If the radius of the sphere were increased from r to r+dr, the
change in the area dA would be equal to:
dA = 4π (r + dr ) − 4πr 2 = 8πrdr
2
The work needed to increase the area by this value would be:
dF = − Pint dVint − Pext dVext − SdT + γ int / ext dA
If we neglect the change in the entropic term, the equilibrium is
reached for dF = 0. And because:
dV = 4πr 2dr
At equilibrium, dF = 0, so that:
γ
Pext − Pint = −2
r
This is the Laplace equation (De Gennes et al., 2004).
Using this relationship, one can now calculate the “capil-
lary rise”, i.e., the tendency of liquids to move upward in thin
(this should be characterized using the “capillary length”) tubes
(Daoud and Williams, 1999). If we consider a tube of which the
lower part is in a liquid, this liquid moves upward by a distance
h, and a meniscus is formed. P being the atmospheric pressure,
and r the radius of the tube, the distance h can be calculated
(Figure 15.3).
The pressure inside the liquid, at the middle of the air–​liquid
interface M can be calculated using the Laplace law.
If Pair is the pressure (ambient) of the air, then:
γ
PM − Pair = −2
r
Because of the weight of the liquid, we can also express the
hydrostatic pressure:
PM − Pair = −2ρgh
94
FIGURE 15.3 When the lower end of a tube is in a liquid, the liquid rises
by capillarity.
FIGURE 15.4 The energies of various intermolecular interactions.
Equating the two, we get:
γ Bauchard E, This H. 2015. Investigating the performance of in situ
h=2
ρgr
From this formula, we see the importance of not only the surface
tension but also the density and the radius. In culinary processes,
when the density cannot be changed, one can improve capillarity
by increasing the surface tension or reducing the radius.
The Example of “Shitao”
Finally, let us observe that capillarity occurs generally for both
aqueous solutions and oils in most plant and animal tissues,
because culinary systems are full of both hydrophobic and
hydrophilic compounds, establishing contacts by a diversity of
intermolecular forces (Figure 15.4). However, the penetration of
liquids inside such tissues does not always occur, and the problem
of giving a specific flavour to the inside of three-​dimensional
ingredients does not always have a solution.
However, if a series of cuts are made on the sides of a plant
or animal tissue, dipping this cut tissue into a sauce triggers the
Hervé This vo Kientza
FIGURE 15.5 A “Shitao” system produced by the French chef Pierre
Gagnaire, served in Hong Kong in 2005. Cuts made in a plant tissue
facilitated the absorption of a thick red sauce by capillarity, and the chef used
this “pencil” to paint the plate.
introduction of the sauce into the tissue by capillarity. In 2007,
I proposed calling such systems “Shitao”, from the 17th-​century
Chinese painter having this name. A dish based on such principles
was served at the gala dinner during which the first note by note
dish was served, in the Mandarin Oriental Hong Kong, by the
French chef Pierre Gagnaire (Figure 15.5).
REFERENCES
Aguilera JM, Michel M, Mayor G. 2004. Fat migration in
chocolate:diffusion or capillary flow in particulate solid?
A hypothesis paper, Journal of Food Science, 69(7), R167–​174.
Aktas N, Aksu MI, Kaya N. 2003. The effect of organic acid marin-
ation on tenderness, cooking loss and bound water content of
beef, Journal of Muscle Foods, 14, 181–​194.
quantitative nuclear magnetic resonance analysis and applying
the method to determine the distribution of saccharides in various
parts of carrot roots (Daucus carota L.), Talanta, 335–​341.
Brillat-​Savarin JA. 1825. La physiologie du goût, Chez l’Auteur, Paris.
Cazor A, Deborde C, Moing A, Rolin D, This H. 2006. Sucrose, glucose
and fructose extraction in aqueous carrot root extracts prepared at
different temperatures by means of direct NMR measurements,
Journal of Agricultural and Food Chemistry, 54, 4681–​4686.
Daoud M, Williams C (eds). 1999. Soft matter physics, Springer
Verlag, Berlin-​Heidelberg.
De Gennes PG, Brochard-​Wyart F, Quéré D. 2004. Capillarity and
wetting phenomena, Springer, Heidelberg.
Febvay L, Hamon E, Werner D, This H. 2019. Identification of
markers of thermal processing (“roasting”) in aqueous extracts
of Coffea arabica L. seeds through NMR fingerprinting and
chemometrics, Magnetic Resonance in Chemistry, DOI:
10.1002/​mrc.4834.
Fick A. 1855. Ueber Diffusion, Annalen der Physik, 94, 59–​86.
Fu Y-​C, Tong C-​H, Lund DB. 2003. Moisture migration in solid food
matrices. Journal of Food Science, 68, 2497–​503.
Geurtz TG, Oortwijn H. 1975. Transport phenomena in butter, in
relation to its structure, Netherland Milk Dairy Journal, 29,
253–​262.
Capillarity in Action
Kopp J, Sale P, Bonnet Y. 1977. Contractomètre pour l’étude des
propriétés physique des fibres conjonctives: tension isométrique,
degré de réticulation, relaxation, Canadian International Food
Science and Technology Journal, 10(1), 69–​72.
Krotov VV, Rusanov AI. 1999. Physical hydrodynamics of capillary
systems, Imperial College Press, London.
Quéré D. 1997. Inertial capillarity, Europhysics Letters, 39(5),
533–​538.
Schwartzberg HG. 1987. Leaching ​organic materials. In Rousseau
RW, ed., Handbook of separation process technology, Wiley,
Hoboken, New Jersey, 540–​577.
Sila DN, Smout C, Van Loey A, Hendrickx M. 2006. Non-​
enzymatic depolymerization of carrot pectin: toward a better
understanding of carrot texture during thermal processing,
Journal of Food Science, 71(1), E1–​9.
95
Tardieu A, Guerez A, Phana S, de Man W, This H. 2009. Quantitative
Nuclear Magnetic Resonance (qNMR) analysis of mono-​and
disaccharides in aqueous solutions obtained by soaking raw
or fried dice of onion bulbs (Allium cepa L.), Journal of Food
Science, 74(4), C319–​325.
This H. 2009. Cours de gastronomie moléculaire N°1, Belin/​Quae,
Paris.
This H. 2019. The science of molecular gastronomy and the art of
innovative cooking, FEBS Letters, 411(9), 1677–​1678.
Vlahova-​Vangelova D, Dragoev S. 2014. Marination: effect on meat
safety and human health. A review. Bulgarian Journal of
Agricultural Science, 20(3), 503–​509.
Zhmud BV, Tiberg F, Hallstensson K. 2000. Dynamics of capil-
lary rise. Journal of Colloid and Interface Science, 228,
263–​269.
Champagne Tasting from a Scientific Perspective

Gérard Liger-​Belair1, Clara Cilindre1, Daniel Cordier1, Guillaume Polidori2, Fabien Beaumont2 and Thomas Séon3
1 Equipe Effervescence, Champagne et Applications (GSMA), UMR CNRS 7331, Université de Reims Champagne-​ Ardenne,
Reims, France
2 Laboratoire de Thermomécanique (GRESPI), Université de Reims Champagne-​Ardenne, Reims 51100, France
3 Institut Jean Le Rond d’Alembert, UMR CNRS 7190, Sorbonne Universités, Paris, France

Champagne tasting may be seen as the pinnacle of glamour and bottle stored at 20 °C is very high, close to 8 bars. Under such
frivolity to most people, but it should also be considered as a pressure, adiabatic expansion allows the temperature of the
fantastic opportunity for physicists to explore the subtle science escaping gas to plummet to a glacial temperature of minus 90 °C
of everyday life. Champagne making is a three-​centuries-​old (minus 130 °F). Because this frigid temperature lies below the
art, but the pursuit of this art can still benefit from the latest freezing point for carbon dioxide, it should be emphasized that
advances in science. With the invaluable help of top-​level sci- the blue haze is the signature of a partial and transient heteroge-
entific instruments, we have learned still more about the subtle neous freezing of gas-​phase CO2 on ice water clusters homoge-
processes that give this legendary wine its sparkle to the eye, its neously nucleated in the bottleneck. Blue haze is indeed typical
tingle to the tongue. Moreover, the careful study of bubbles in of the Rayleigh scattering of light by clusters much smaller
champagne and sparkling wines could also have implications for than the wavelengths of ambient light, ranging from 0.4 µm to
other areas of science where bubbles and dissolved gases play a 0.8 µm. This blue flash is finally caused by the same process that
crucial role, such as in marine science, for example, where the colours the sky with its blue hues.
production of sea spray aerosols is very similar, in principle, to
the fizz in champagne.
What Is the Best Way to Pour Champagne?
Have you ever wondered how best to pour champagne into a flute
Champagne Cork Popping Revisited to preserve the precious fizz?
Champagne and sparkling wines elaborated through the same
traditional method are under a high pressure of carbon dioxide
(CO2) close to 5–​6 bars (at 10–​12 °C). Actually, gas-​phase CO2
forms together with ethanol during a second in-​bottle fermen-
tation process promoted by adding yeasts and a certain amount
of sugar in bottles hermetically sealed with a crown cap or a
cork stopper (Liger-​Belair, 2013). This is a basic application of
Henry’s law, which states that the concentration of dissolved CO2
in the liquid phase is proportional to the partial pressure of gas-​
phase CO2 in the bottleneck. Therefore, during the cork popping
process, a plume mainly composed of gas-​phase CO2 (with traces
of water and ethanol vapours) freely expands out of the bottle-
neck through ambient air. Through what we call adiabatic expan-
sion in physics, a corresponding drop in temperature of the order
of several tens of degrees causes the immediate condensation of
water vapour in ambient air into the form of a grey-​white cloud
of fog, as seen in Figure 16.1a.
Recently, a new study (Liger-​Belair et al., 2017a) showed
that this mini-​cloud is even cooler if the bottle is warm, turning FIGURE 16.1 Cork popping process as seen through high-​speed imaging
briefly blue as the bottle is uncorked around 20 °C (i.e., 68 °F), for a bottle stored at 12 °C (a) and for a bottle stored at 20 °C (b).
as exemplified in Figure 16.1b. Actually, the pressure inside a (Credit: Equipe Effervescence)

97
98
The concentration of dissolved CO2 in champagne and spark-
ling wines is the real key to the production of bubbles. But the act
of pouring champagne into a glass is far from inconsequential in
terms of its effect on the level of dissolved CO2 remaining in the
glass. We investigated the effect of two different pouring methods
(Liger-​Belair et al., 2010). One method involved pouring cham-
pagne straight down the middle of a vertically oriented flute. The
other one involved pouring champagne down the side of a tilted
flute. Tilting the flute was found to have significantly less impact
on the concentration of dissolved CO2 than the former method
because the “beer-​like” way of serving champagne is gentler.
Pouring champagne straight down the middle of a vertically
oriented glass produces turbulence and traps air bubbles in the
liquid, both of which force dissolved CO2 to escape more rap-
idly from champagne. These findings were corroborated through
infrared imaging in order to visualize the cloud of CO2 escaping
during the pouring process (see Figure 16.2). We should therefore
treat champagne a little more like beer –​at least, when it comes
to serving it.
How Many Bubbles in Your Glass of Bubbly?
The issue of how many carbon dioxide bubbles are likely to
nucleate in a glass of bubbly is of concern for sommeliers, wine
journalists, experienced tasters, and any open-​minded physical
chemist wondering about the complex phenomena at play in a
glass of bubbly. Thermodynamically speaking, bubble formation
in a liquid phase is limited by an energy barrier that needs to
FIGURE 16.2 Losses of dissolved CO2 during the pouring of champagne, as seen through infrared imaging, in a vertically oriented flute (a) and in a tilted
flute (b).
(Credit: Reims University)
Gérard Liger-Belair et al.
be overcome. In weakly supersaturated liquids such as cham-
pagne, and carbonated beverages in general, bubbles do not
just pop into existence ex nihilo. In order to nucleate and grow
freely, bubbles need pre-​existing gas cavities with radii larger
than a critical radius of the order of several tenths of a micro-
metre. Closer inspection of glasses poured with champagne, or
any other sparkling beverage, revealed that most of the bubble
nucleation sites were found to be located on pre-​existing gas
cavities trapped within hollow and roughly cylindrical cellulose
fibers with a cavity mouth of several micrometres (Lehuédé and
Robillard 1996; Liger-​Belair, 2015), therefore exceeding the crit-
ical radius thermodynamically required to enable bubble growth
(see Figure 16.3).
We have recently attempted to provide an accurate scientific
answer to the issue of how many bubbles are likely to form in
a glass by using models that combine the dynamics of bubble
ascent and mass transfer equations (Liger-​Belair, 2014). As
one might expect, the number of bubbles likely to form per
glass depends on both the wine and the glass itself. A theoret-
ical relationship was derived, which provides the total number
of bubbles likely to form per glass depending on various
parameters such as temperature, level of champagne, pressure,
etc. If 100 mL of champagne is poured straight down the
middle of a vertically oriented flute, about one million bubbles
are likely to nucleate if you resist drinking from your flute.
Otherwise, champagne served more gently by pouring down
the wall of a tilted flute –​a technique that better preserves the
dissolved carbon dioxide –​will yield tens of thousands more
bubbles before it goes flat.
Champagne Tasting
Flute versus Coupe: And the Winner Is …
Advantages and disadvantages of both glass shapes –​flute versus
coupe –​have long been debated in bars, clubs, restaurants, and
popular wine magazines. But, to date, little to no analytical data
has ever been brought to bear on the age-​old dilemma regarding
FIGURE 16.3 High-​speed photograph showing the repetitive bubble nucle-
ation process from the tip of a tiny cellulose fibre stuck on the wall of a glass
poured with champagne.
(Credit: Gérard Liger-​Belair)
FIGURE 16.4 Glasses as seen through infrared imaging, immediately after the pouring step, whether champagne is served in a flute (a) or in a coupe (b).
Zones highly concentrated in gas-​phase CO2 appear in black and dark blue, whereas zones with a low concentration in gas-​phase CO2 appear in red. The con-
centration of gas-​phase CO2 found above the flute (close to the edge) is always higher than that found above the coupe (c).
(Credit: Reims University)
99
the effect of each type of drinking vessel on champagne tasting.
Let’s take our discussion beyond aesthetic and subjective consid-
erations and consider some science. When one tastes champagne
from a glass, gaseous CO2 and volatile aromatic compounds pro-
gressively invade the “headspace” above the glass, thus modi-
fying the taster’s global perception of aromas.
To study the effect of the glass on the perception of gaseous
CO2 and aromas, we measured the levels of gaseous CO2 and
ethanol via gas chromatography (a technique that allows us to
accurately measure the various chemical components in a mix-
ture; Liger-​Belair et al., 2012). Our results suggest that the
narrow flute funnels the gaseous CO2, and therefore the aromas,
more effectively, whereas the broader coupe “dilutes” them.
Nevertheless, gaseous CO2 is well known to irritate the nose
if too concentrated (this is the very unpleasant carbonic bite).
Gaseous CO2 was found in nearly twice the concentration above
the flute as above the coupe. Our results are consistent with sen-
sory ­analyses of champagne wines conducted by human tasters.
Actually, it is generally accepted that the smell of champagne,
and especially its first nose, is more irritating when champagne is
served in a flute than when it is served in a coupe. We also used
infrared imaging to visualize gaseous CO2 escaping from both
glass types (Figure 16.4) and confirmed the tendency of flutes to
hang on to concentrated quantities of CO2.
A tulip-​shaped wine glass (a bit shorter than a traditional flute
and curved slightly inwards at the top) would be the best com-
promise and would provide the taster with a better overall sensory
experience than either the tall flute or the shallow coupe.
Flow Patterns in Champagne Glasses Increase the
Perception of Aromas
Effervescence in your glass, the very much sought-​after bubbling
process, goes far beyond the sole aesthetic point of view in cham-
pagne and sparkling wine tasting. Rising bubbles displace the
100
fluid around them, which in turn disturbs the neighbouring fluid
layers. In champagne and sparkling wines, ascending bubbles are
no exception to this rule. Bubbles drag along fluid particles in their
wake. Contrary to non-​effervescent flat wines, at rest in your glass
(unless you swirl the glass), sparkling wines are in constant motion.
A technique called laser tomography was used to reveal what
the naked eye could not (see Figure 16.5) (Polidori et al., 2009).
The photographs displayed in Figure 16.6 show the beauty of
the flow patterns found in champagne glasses. By increasing the
diffusion of aromas above the champagne surface, flow patterns
driven by ascending bubbles are indeed a wonderful gift to the
champagne taster. Flow patterns continuously renew the air/​wine
FIGURE 16.5 Optical workbench used to capture champagne flow patterns
through laser tomography.
(Credit: Hubert Raguet)
FIGURE 16.6 Laser tomography reveals the flow patterns found inside a glass, whether champagne is served in a flute (a) or in a coupe (b).
(Credit: Reims University)
Gérard Liger-Belair et al.
interface, allowing volatile aromas to escape from the liquid
wine surface much more efficiently over a longer period of time,
whereas the surface of a wine at rest progressively becomes
odourless. There is no reason to swirl a glass of champagne or
sparkling wine to enjoy the subtle mix of scents and flavours,
since the bubbles simply do the job for you. Moreover, exam-
ining the photographs displayed in Figure 16.6 leaves no doubt
that different glasses show different flow patterns and there-
fore release aromas and gas-​phase CO2 at different rates, thus
offering a different overall global sensation for the taster. This
might explain why some consumers get different aromas from a
critic whose note they have read: they were maybe using different
glasses and therefore experienced different aromas.
Unravelling Two-​Dimensional Vortices at the
Champagne Surface
A collection of several two-​dimensional vortices, which formed
at the air/​
champagne interface, were captured through long-​
exposure photography (Beaumont et al., 2016). Depending on
the intensity of the ascending bubbly flow, and therefore on the
time elapsed since champagne was poured, various regimes were
evidenced. Early after pouring, the situation is highly unstable,
with a large number of rather small but rapidly counter-​rotating
cells changing from a very unstable eight-​cell regime (Figure 16.7)
to a usually more stable six-​cell regime (Figure 16.8). Shortly
afterward, the system reorganizes itself in a long-​lasting and
definitely stable regime with four large counter-​rotating cells
(Figure 16.9). These vortices are the result of the fine interplay
between ascending bubbles, which continuously drive some fluid
Champagne Tasting 101

FIGURE 16.9 Laser tomography showing the highly stable, long-​lasting

FIGURE 16.7 Laser tomography reveals visually appealing two-​
four-​cell regime, in which four counter-​rotating cells self-​organize at the air/​
dimensional flow patterns at the champagne surface, with a highly unstable
champagne interface.
eight-​cell regime in which eight counter-​rotating cells self-​organize at the air/​
champagne interface. (Credit: Reims University)
(Credit: Reims University)

FIGURE 16.10 Satellite image showing the sea surface currents in the
southern Indian Ocean.
(Credit: NASA)

across the champagne surface, and the circular glass edge, which
confines the fluid circulation around its interior boundary. It is
worth noting that floating bubbles on the surface of champagne
FIGURE 16.8 Laser tomography showing the poorly stable six-​ cell freely follow the two-​dimensional movement of the flat air/​cham-
regime, in which six counter-​rotating cells self-​organize at the air/​champagne pagne interface and help us to track the trajectories of champagne
interface. surface currents (exactly as drifting buoys allow us to track sea
(Credit: Reims University)
surface currents). By the way, this parallel between the cham-
pagne vortices and the sea surface currents is evidenced by the
fantastic satellite image displayed in Figure 16.10.
102 Gérard Liger-Belair et al.

FIGURE 16.11 High-​speed time sequences showing millimetric floating bubbles bursting at the surface of champagne and propelling tiny droplets of wine
several centimetres above the liquid surface.
(Credit: Institut Jean Le Rond d’Alembert/​Paris 6)

The lesson from these observations is that the tasters should 2016; Séon and Liger-​Belair, 2017). Typical high-​speed video
not smell the champagne immediately after it is poured, because sequences displayed in Figure 16.11 show various steps of a
too many vortices will release too much carbon dioxide, which single bubble bursting at the champagne surface, including the
is likely to irritate the tasters’ nostrils. The best way to do so is
to wait a little bit, so that the more stable six-​cell and four-​cell
regimes have taken over, to get a better, fuller, truer sense of the
wine’s aromas. It is hard to wait, but it can be truly worthwhile
in the end. Besides that, Dom Pérignon’s famous “chef de cave”
Richard Geoffroy is reported as saying that the best moment to
start sniffing a glass of his champagne is 11 minutes after pouring
(Liger-​Belair et al., 2017b).

Bursting Bubbles Provide an Aromatic Boost to

Champagne
Bursting bubbles are very common in our everyday life. They
play a crucial role in many natural as well as industrial processes
(in physics, chemical and mechanical engineering, oceanography,
geophysics, technology, and even medicine). The top of a flute
poured with champagne proved to be a fantastic playground to
explore the physics behind collapsing bubbles. As it reaches the
champagne surface, a bubble floats for a while before it bursts.
Actually, a floating bubble is just like a tiny iceberg, which only
slightly emerges from the champagne surface. After the disinte-
gration of the emerging liquid film, a very complex hydrodynamic
process ensues, causing the collapse of the submerged part of the FIGURE 16.12 Photograph revealing the myriad of tiny champagne
bubble. A combined approach through high-​speed imaging and droplets propelled by bursting bubbles above the surface of a flute.
numerical modelling was conducted recently (Ghabache et al., (Credit: Alain Cornu/​Collection CIVC)
Champagne Tasting 103

production of a high-​speed jet, which quickly breaks up into sev-

Ghabache E, Liger-​ Belair G, Antkowiak A, Séon T. 2016.
eral tiny droplets. “Evaporation of droplets in a Champagne wine aerosol”,
Actually, during champagne tasting, the myriad of ascending Scientific Reports, 6, 25148.
Lehuédé P, Robillard B. 1996. “Le champagne dans la flûte”, Pour
bubbles collapse and therefore radiate a multitude of tiny droplets
la Science, 229, 14.
above the free surface, into the form of refreshing aerosols, as Liger-​Belair G. 2013. Uncorked: The Science of Champagne (revised
shown in the photograph displayed in Figure 16.12. This very edition), Princeton University Press, Princeton, MA.
characteristic fizz considerably enhances the flavour release Liger-​Belair G. 2014. “How many bubbles in your glass of bubbly?”
in comparison with that from a flat wine. Based on a research Journal of Physical Chemistry B, 118, 3156–​3163.
finding that air bubbles trapped in seawater carry surfactants Liger-​Belair G. 2015. “The science of bubbly”, Scientific American
that are released at the ocean surface when the bubbles burst, (Special Issue: The Science of Food), 16–​21.
it was hypothesized that champagne aerosols should also be Liger-​Belair G, Bourget M, Villaume S, Jeandet P, Pron H, Polidori
G. 2010. “On the losses of dissolved CO2 during champagne
over-​concentrated with surface-​ active compounds. Ultra-​ high-​ serving”, Journal of Agricultural and Food Chemistry, 58,
resolution mass spectrometry was used to analyse the chemical 8768–​8775.
composition of champagne droplets (Liger-​Belair et al., 2009). Liger-​Belair G, Bourget M, Pron H, Polidori G, Cilindre C. 2012.
We prove that hundreds of surfactants preferentially divide them- “Monitoring gaseous CO2 and ethanol above champagne
selves up into champagne aerosols, which gives the bulk of glasses: flute versus coupe, and the role of temperature”, PLoS
the champagne liquid a different chemical fingerprint from its ONE, 7, e30628.
aerosols. Moreover, tens of these compounds, concentrated in Liger-​Belair G, Cordier D, Honvault J, Cilindre C. 2017a. “Unveiling
CO2 heterogenous freezing plumes during champagne cork
the champagne aerosols, were identified as being the chemical
popping”, Scientific Reports, 7, 10938.
precursors to aromas. By drawing a parallel between the fizz of Liger-​Belair G, Cilindre C, Gougeon R, Lucio M, Gebefügi I, Jeandet
the ocean and the fizz of champagne, our study revealed a rela- P, Schmitt-​Kopplin P. 2009. “Unraveling different chemical
tionship between bursting bubbles and the aromatic boost often fingerprints between a champagne wine and its aerosols”,
attributed to champagne and sparkling wines, thus supporting the Proceedings of the National Academy of Sciences of the United
idea that rising and bursting bubbles act as a continuous elevator States of America, 106, 16545–​16549.
for aromas in every glass of champagne. Liger-​Belair G, Polidori G, Beaumont F. 2017b. “Champagne physics:
Lessons from the lab”, The World of Fine Wine, 58, 34–​38.
Polidori G, Jeandet P, Liger-​Belair G. 2009. “Bubbles and flow
REFERENCES patterns in champagne”, American Scientist, 97, 294–​301.
Beaumont F, Liger-​Belair G, Polidori G. 2016. “Unveiling self-​ Séon S, Liger-​Belair G. 2017. “Effervescence in Champagne and
organized two-​dimensional (2D) convective cells in cham- sparkling wines: From bubble bursting to droplet evaporation”,
pagne glasses”, Journal of Food Engineering, 188, 58–​65. European Physical Journal -​Special Topics, 226, 117–​156.
Chantillys: The Cousins of Whipped Cream

Hervé This vo Kientza

1 INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​
AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France

Whipped cream is a traditional preparation obtained through the If it is true that milk includes fat dispersed in water, one has
introduction of air bubbles in “cream”, i.e., the emulsified system to recognize that this fat can be solid or liquid depending on the
obtained by the creaming of milk from Bos taurus. After October temperature, with melting beginning at −10 °C and finishing at
1995, I introduced a series of aerated food preparations of the 55 °C (Bouteille et al., 2013). In cream as well, part of the fat
same kind, the first one being “chocolate Chantilly” (This, 1996). is solid, and this fraction depends on the temperature; the exact
microstructure depends on the particular processes that milk
underwent while being transformed into cream. This explains
why milk is actually not an emulsion at room temperature, but
Chocolate Chantilly
probably has the DSF formula [D0(O) × D0(S)]/​D3(W) (Moens
In the early years of molecular and physical gastronomy, there et al., 2018; This, 2007; This, 2009). Also, the fact that whipped
was a time when an (inappropriate) programme for this scientific cream cannot be obtained when the temperature is too high (in
discipline was to focus on “culinary definitions” and “culinary the summer in hot countries) shows that a solid network is needed
precisions” (see the chapter “ Culinary Precisions and Robustness for making the final product, which can probably be described
of Recipes”), but it also included looking for applications of by a simplified formula such as D0(G)/​D3(S), in which liquid fat
molecular gastronomy to the improvement of the culinary (O) has to be dispersed.
art, what was named later “molecular cooking” or “molecular
cuisine” (This, 2002; Burke et al., 2016). Chocolate Chantilly, in
particular, was proposed as a generalization of the classic process
of whipping cream: first, the application was made using choc-
olate emulsions.
At that time (1995), it was considered that, as cream is obtained
from milk (said then to be a “dilute oil in water emulsion”)
by creaming (because fat has a lower density than water),
and whipped cream is obtained by foaming this concentrated
emulsion, other emulsions could be foamed as well, making
“foamy emulsions”. Chocolate was first considered as a can-
didate for fat replacement of milk fat, and it was proposed to
make an emulsion by simply heating chocolate in water and then
whipping the obtained emulsion (Figure 17.1).
It is useful to recall now that such a proposal then seemed
heretical, because water added to melted chocolate can lead to
“seizing”, i.e., the formation of a clumpy mass. However, heating
chocolate in water can make an emulsion, and indeed, the predic-
tion that one would get a foam by whipping a chocolate emulsion
was a success; I demonstrated the “chocolate Chantilly” during a
lecture at the Chocolate Fair in Paris in December 1995.
At that time, a mechanism for this process was proposed
(Figure 17.2), but the use of the disperse system formalism (DSF, FIGURE 17.1 A chocolate in water emulsion (left) and chocolate Chantilly
see the chapter on this) shows why this picture is inaccurate. obtained by whipping this emulsion while cooling (right).

105
106
FIGURE 17.2 The initial (inaccurate) description of the making of
chocolate Chantilly. Here, the picture shows a dilute emulsion (top), the
formation of a more concentrated emulsion by creaming (middle), and the
aerated emulsion (bottom). This picture is inaccurate because it considers fat
(yellow dots) as liquid droplets, but at the low temperatures used for this pro-
cess, the fat is more solid than liquid. This does not prevent one from making
a successful Chantilly cream, but the mechanism is not as simple as the O/​W
+ G → (G+O)/​W equation.
FIGURE 17.3 A bouchée including roasted bread, foie gras Chantilly, and
oysters, by Pierre Gagnaire (2002).
Hervé This vo Kientza
Other Chantillys
The successful production of chocolate Chantilly led me imme-
diately to test other emulsions made from other food ingredients
containing fat, such as cheese, foie gras, butter, brown butter, and
even olive oil, producing products such as cheese Chantilly, foie
gras Chantilly, butter Chantilly, olive oil Chantilly, etc.
For example, I first demonstrated the “Roquefort cheese
Chantilly” on a TV show (Toque à la loupe, France 5), and then
I helped the French chef Didier Clément to produce a “crottin
de Chavignol Chantilly” for a public lecture in Orleans, France.
Also, I first showed a “foie gras Chantilly” on German TV: this
food system was used by the chef Pierre Gagnaire in many dishes
(Figure 17.3).
Finally, it can be observed that experiments show clearly that
a sufficient solid fat network is needed, which is particularly true
for the production of olive oil Chantilly, which needs a very effi-
cient cooling system to be used (we used liquid nitrogen in a
public experiment).
REFERENCES
Bouteille R, Gaudet M, Lecanu B, This H. 2013. Monitoring lactic
acid production during milk fermentation by in situ quantita-
tive proton nuclear magnetic resonance spectroscopy, Journal
of Dairy Science, 96(4), 2071–​2080.
Burke R, This H, Kelly A. 2016. Molecular Gastronomy, Reference
Module in Food Sciences, FDSC 03302, http://​ dx.doi.org/​
10.1016/​B978-​0-​08-​100596-​5.03302-​3.
Moens K, Tavernier I, Dewettinck K. 2018. Crystallization behavior
of emulsified fats influences shear-​induced partial coalescence,
Food Research International, 113, 362–​370.
This H. 1996. Le chocolat Chantilly, Pour la Science, 230, 20.
This H. 2002. Molecular gastronomy. Angewandte Chemie,
International Edition in English, 41(1), 83–​88.
This H. 2007. Formal descriptions for formulation. International
Journal of Pharmaceutics, 344(1–​2), 4–​8.
This H. 2009. Molecular gastronomy, a chemical look to cooking.
Accounts of Chemical Research, 42(5), 575–​583.
Cheese: Hot Culinary Uses of Cheese
Sébastien Roustel1 and John A. Hannon2
1 Chr-​Hansen, Cheese application, Boge Allé 10–​12, 2970 Horsholm, Denmark
2 Glanbia Ireland DAC, Ballyragget, Co. Kilkenny, R95 W3KD, Ireland
Globally, cheese is mainly consumed in two forms (Roche, 2018,
Guinee, 2016): as a table cheese that does not require any special
preparation, such as in a cheese platter consumed at the end of the
meal or as cheese slices consumed at breakfast, and as an ingre-
dient cheese that is used in culinary applications (Figure 18.1).
Currently, the use of cheese as an ingredient represents the
majority of cheese use, and this trajectory is set to continue
(Kerjean, 2018).
For example, the use of cheese as a pizza topping represents
more than 50% of the culinary use of pasta filata cheese (Mintel,
2017). Culinary applications cover its use both as a cold ingre-
dient (as in salads and sandwiches) and in hot applications (e.g.,
pizzas, gratins, pasta, ready meals and toasted sandwiches)
(Guinee, 2016; Roustel, 2018).
Depending on the culinary use, the requirements with regard
to the quality of cheeses are very variable. The properties that a
cheese will develop when heated are dependent not only on the
cheese itself but also on the nature of the application (type of
dish) and the heating parameters (e.g., duration and temperature).
Thus, these are often referred to as the hot functional properties
of the cheese or as thermo-​functionalities (Richoux et al., 2002;
Roset et al., 2004)
Understanding the mechanisms that underpin the expression
of these thermo-​functionalities can be very useful to control and,
more importantly, to optimize them in order to guarantee a high
level of quality for the consumer in the final application.
Cheese
If we consider that there are more than a thousand varieties of
cheeses around the world, with some estimates mentioning
a figure of 1800 varieties (Profession fromager, 2015), these
cheeses differ according to the technology used and their com-
position, appearance, flavour, texture and cooking properties. The
variation in the thermo-​functionalities is due to manufacturing
differences, different degrees of emulsification of the fat, com-
position and levels of maturity (e.g., extent of proteolysis and
lipolysis) (Guinee, 2016). However, some varieties have better
attributes than others with regard to thermo-​functionalities. For
example, Mozzarella, Provolone and Kachkaval (which belong
to the pasta filata category), generally have a superior stretching
ability compared with many other cheese varieties. If the ability
to flow is desired, ripened Cheddar and Raclette are better suited.
In contrast, Queso Fresco and Paneer have a low ability to flow,
which is sometimes sought in certain culinary preparations where
the preservation of the cheese appearance is desired.
All these differences have their origin in the cheese making
process and the ripening regime employed (Figure 18.2). There
are five key steps in the transformation of milk into cheese, as
shown in Figure 18.3 (Mietton and Chablain, 2018):
• milk preparation: control of the microbial flora and
adjustment of fat, protein, and possibly the lactose and
mineral content;
• coagulation: which transforms the milk in the liquid
state to a solid gel. There are several possible ways to
transform milk into a gel, with the vast majority being
produced by coagulation using proteases (coagulants)
with or without the use of lactic fermentation by starter
cultures (Roustel and Boutonnier, 2015; Sperat-​Czar
et al., 2018);
• drainage: this step allows the differential concentra-
tion of the milk elements. Caseins combine to form the
fine-​protein matrix, which entraps the fat globules. The
level of drainage depends mainly on the intensity of
thermo-​mechanical work and the kinetics of acidifica-
tion. Depending on the kinetics of acidification in com-
parison to the drainage kinetics, the resultant cheese will
have different levels of mineralization. Thus, the cal-
cium content ranges from 0.7 to 13.0 g kg−1, depending
on the cheese variety (Figure 18.4). This characteristic
is fundamental in controlling the thermo-​functionalities
of cheese (Guinee and Kilcayley, 2004; Kerjean, 2018;
Roustel, 2018);
• salting: which is performed either in brine or by dry salt
addition;
• ripening: during which a series of enzymatic biochem-
ical reactions gradually transform the constituents of
the young cheese (proteins, fat and carbohydrates) into
a multitude of compounds, making the cheese more or
less smooth and supple, and conferring its character-
istic taste and appearance (Goudédrenche et al., 2007;
Sperat-​Czar et al., 2018).
107
108 Sébastien Roustel, John A. Hannon

FIGURE 18.1 Uses of cheeses.

(From Roche, 2018 and Guinee, 2016)

Coagulation Ripening Variety Examples

(c = cow, E = ewe, G = goat)

Without rind • Fresh lactic cheese G. Rove des Garrigues

E. Caillebote d’Aunis
Lactic curds,
• Lactic cheese with white C. Chaource, Neufchâtel, St-Marcelin
slow, spontaneous Lactic White rind rind Brillat-Savarin, Saint-Félicien
coagulation curds G. Saint-maure, Selles, Pouligny

Washed or mixed rind • Lactic cheese with C. Epoisses, Soumaintrain, Langres

washed rind G. Picodon Dieulefit

Soft Without rind • Fresh soft cheese G/E. Féta

C. Crescenza
cheese
• Soft cheese with C. Camembert, Brie, Coulommiers
With white rind white rind G. Brique du Forez, Bougon
G. Pérail
C. Munster, Maroilles, Livarot, Mont-
Washed or mixed rind • Soft cheese with d’Or, Pont l’évêque
washed rind E. Niolo – G. Chevrotin
Rennet and C. Bleu d’auvergne , Fourme, Stilton
mixed curds, quick • Blue cheese with
cavities E. Roquefort
coagulation Blue Inner moulds C. Bleu de gex, Gorgonzola, Bel de
cheese • Blue veined or Termignon, Bleu du Vercors
marbled cheese G. Cabralès, Persillé des Aravis

Semi-soft C. St-Nectaire, Tomme, St-Paulin,

• Supple to firm Reblochon, Morbier, Raclette, Fontina
cheese E. Manchego, Ossau, Lavort

Milling • With milled curds C. Cantal, Laguiole, Salers, Cheddar

Semi-hard
C. Abondance, Asiago, Masdaam
cheese Leerdamer, Appenzel

C. Beaufort, Comté, Parmesan

Hard- • Without holes E. Pecorino
pressed
cheese • With holes C. Emmental

FIGURE 18.2 Different cheese varieties classified according to cheese technology.

Cheese: Hot Culinary Uses of Cheese 109

FIGURE 18.3 Different steps of cheese making.

(Mietton and Chablain, 2018)

All these steps involved in the processing of milk into cheese lead melting and emulsifying cheeses(s) using heat and one or more
to the production of the great range of different cheese varieties, monovalent salts of acids and water (both these last elements are
whose composition of water, fat, proteins, lactose/​​galactose and provided by white wine). In physico-​chemical terms, the emulsion
minerals is very different (Figure 18.4). Similarly, the resultant is obtained by three key steps: ion exchange, proteolysis/​hydra-
cheese microstructure is also highly variable (Figure 18.5) tion and emulsification, the outcome of which is a thickening of
(Rowney et al., 2004; Lopez et al., 2007). From a physico-​ the product (Figure 18.6) (Roustel and Boutonnier, 2015).
chemical point of view, a cheese can be defined as an oil-​in-​water
emulsion dispersed in a solid: (O/​W)/​S.
All these elements lead to the creation of very different thermo-​
Heated Cheese
functionalities from one cheese type to another. Depending on On heating a young cheese in which the fat globules still have their
the desired culinary use, it will be necessary to choose which native membrane intact, i.e., unaltered by enzymatic reactions
thermo-​functional property is required. (proteolysis and lipolysis), the changes that occur are limited to
the protein/​water system and to the substances that are dissolved
in this moisture. As the solubility of solutes (lactose, salts and
lactic acid) increases with increasing temperature, it follows that,
Using the Emulsifying Property of Cheeses: during heating, a certain amount of the solvent water will be
The Fondue free. In addition, this is increased because the hydrogen bonds
The emulsifying property of cheese, which is involved in of the ionic groups of the proteins and amides require less water
many culinary applications and interacts with other thermo-​ of hydration. At the same time, this change is accompanied by a
functionalities (e.g., stretching and browning) (Richoux et al., strengthening of hydrophobic interactions. If high temperatures
2002), is essential to the success of a good cheese fondue and, by are used, then a partial separation of the protein mass (including
extrapolation, to the production of processed cheese and sauces the fat phase) from the aqueous phase is observed (Lee et al.,
containing cheeses. Cheese fondue is a product obtained by 1979; Caric and Kalab, 1993; Roustel and Boutonnier, 2015).
110 Sébastien Roustel, John A. Hannon

FIGURE 18.4 Different cheese varieties according to their composition.

FIGURE 18.5 Microstructure of different cheeses.

(Lopez, 2007; Rowney et al., 2004)
Cheese: Hot Culinary Uses of Cheese 111

FIGURE 18.6 The biochemistry of the “cheese fondue”.

FIGURE 18.7 Evolution of the protein/​mineral/​fat organization during the preparation of the fondue.

On the other hand, if an aged, ripened cheese is used
(evaluated by the ratio of soluble nitrogen over total nitrogen
at pH 4.6: SN/TN) (Roustel, 2018), in which the membranes of
the fat globules have been altered, then the results obtained are
very different. In this case, the increase in energy (due to heat)
increases the fluidity of the liquid phases (fat phase and protein/​
water phase). Thus, localized areas are observed with coalesced
fat globules that are freed from their membranes, which gather
as pools of fat (Figure 18.7). This results in syneresis and sep-
aration of the liquefied fat phase, as can be seen during the
heating of mature cheeses such as Cheddar, Emmental or old
Dutch cheeses.
112 Sébastien Roustel, John A. Hannon

Physico-​Chemical Mechanisms within the Fondue
For making a cheese fondue, a homogeneous melted product is
desired. This is difficult to achieve due to the insolubility of the
cheese in the water, as well as the immiscibility of the water and
the fat phases. To obtain a visible homogeneous melt without
phase separation, it is necessary to use emulsifiers. However,
it is known that native caseins (which are the main proteins in
milk and cheese), which are very good emulsifiers themselves,
lose their surfactant power in cheese due to the bridges that are
formed with divalent cations such as calcium and magnesium, the
protein-​bridging agents. Therefore, in order for casein to play its
role in the lowering of the surface tension at the water–​oil inter-
face, it must first be solubilized in water. To solubilize the casein,
there must be an exchange of the divalent cations (Figure 18.8)
(Boutonnier and Roustel, 2014) with stronger monovalent cations
such as sodium, such that the calcium-​binding sites are ionized
and hydrated at the more hydrophilic zones of the casein mol-
ecule (Lee et al., 1979; Caric and Kalab, 1993; Boutonnier and
Roustel, 2014).
These monovalent cations are provided in white wine in
the form of potassium and sodium tartrates (mostly potassium
bitartrate).
The wine also provides:
• water, which allows the dispersion and hydration of
proteins;
• ethanol, which is very useful for the dispersion of the
fat in the liquid phase;
• acidity (the pH of the white wine is between 2 and 3),
which facilitates the exchange of ions.
The first step in the physicochemical processes that take place
in the fondue is characterized by the conversion of the insoluble
calcium caseinate to soluble sodium or potassium caseinate.
This results from the displacement of the calcium equilibrium of
the mixture in the presence of melting acid salts (tartrates). It is
also possible to use other melting acid salts, such as sodium or
potassium citrates, orthophosphates or polyphosphates, the ion
exchange capacity of which is higher (Caric and Kalab, 1993;
Roustel and Boutonnier, 2015); this ion exchange process is
shown schematically in Figure 18.9 (Boutonnier and Roustel,
2014). The solubilization of the proteins allows the subsequent
emulsification of the fat phase. This ion exchange process is
shown schematically in Figure 18.9 (Boutonnier and Roustel,
2014). Given the relationship between pH and salt dissociation,
the ion exchange mechanism is pH-​ dependent (Marchesseau
et al.,1997).
The chemical mechanism of this phase is as follows: the
dissolved polyanion melting salt (mainly potassium tartrate in
the case of wine) enters into the inter-​micellar spaces, which are
filled with colloidal calcium phosphate. There, it lowers the cal-
cium concentration (Ca2+ + HPO42− → CaHPO4) and neutralizes
the negative charges of calcium caseinate with sodium ions,
which are thus released.
The exchange of the cross-​linking polyvalent cations (Ca2+)
with monovalent (Na+) cations triggers the separation of the
paracasein peptide chains. Each of the sodium ions is surrounded
by water molecules as well as carboxylic groups.
This ion exchange process results in a dispersion of the
proteins and their hydration. During this ion-​exchange phase, and
depending on the properties of the melting salt or wine used (tar-
trate concentration, pH, etc.), the initial structure of the cheese
protein network begins to gradually break down (Lee et al., 1979;
Gupta et al., 1983; Dimitreli et al., 2005). Hence, the potassium
or sodium tartrates act as dispersing agents of the protein system;
the original (cross-​linked) cheese structure is transformed into a

FIGURE 18.8 Schematization of the mechanism of calcium sequestration, initially bound to phosphoserine residues of caseins.
Cheese: Hot Culinary Uses of Cheese 113

Practical Considerations
Cheeses are the main ingredient used in the preparation of a
fondue. According to the manufacturing technology used, the
level of calcium is different (Table 18.1), which will influence
the ability of the cheese to melt. Numerous studies have shown
the importance of cheese characteristics for the thermo-​
functionalities of cheeses, so it is good to choose the characteristics
wisely. Hence, it is often advisable to use a mixture of several
cheese types in order to average out the quality differences of the
cheeses used. For example, mixing a young cheese containing
more “intact” casein and a more ripened cheese with more flavour
is recommended. Moreover, to make a successful fondue, it is
recommended to first heat the white wine and then add the cheese
(which has been cut into thin slices) while stirring to facilitate
melting (Figure 18.9). Depending on the mechanical intensity of
the stirring and the quality of the cheese (level of proteolysis),
the protein structure will be more or less altered and the fondue
will be more or less stringy. The addition of a small portion of a
spreadable cheese, which is actually a cooled fat emulsion in a
protein network, will facilitate the three key steps of the physico-
chemical process of the fondue.

FIGURE 18.9 Simplified representation of the effect of emulsifying salt.

Non-​Enzymatic Browning of Cheeses: the Gratin

new network (Figure 18.7). The unfolding of the protein chains
The development of cheese as an ingredient topping on a variety
and the increase in the negative charges conferred on the caseins
of cooked dishes (e.g., pizzas, pasta, gratins and other cooked
by the ion exchange process increase the hydrophilic character
dishes) requires, in addition to the flavour aspect, a greater or
of the polar side groups. These phenomena allow a significant
lesser degree of browning. This browning can be described as
hydration of the proteins, which results in a thickening of the
the result of a colouration of different surface intensities of the
product. In parallel, the free water content, which decreased
cheese during cooking. With regard to pizzas, restaurant owners
initially, increases again during heating due to a decrease in the
estimate that about 50% of quality defects come from browning
water-​binding power of the hydrophilic groups when they are
(Pilcher and Kindstedt, 1990). Indeed, for this particular appli-
subjected to high temperatures. This phenomenon is accom-
cation, excessively intense browning is not desired. On the other
panied, at first, by an increase in the viscosity of the product and
hand, for other applications such as pasta gratins, a more intense
then by a decrease when the temperature exceeds 60 °C. At the
browning may be considered desirable (Caric and Kalab, 1993;
end of this phase, the initial mixture of cheeses is converted into
Xixiu et al., 2013).
a homogeneous colloidal solution.
The rate of “destructuration” depends on many factors (Lee
et al., 1979; Caric and Kalab, 1993; Roustel and Boutonnier, Physico-​Chemical Mechanisms of Cheese Browning
2015): the nature and dose of the ion exchange salts (and thus the The phenomenon behind this thermo-​ functionality is known
nature of the wine), the level of cheese mineralization, the degree to cooks and chefs under the name of the “Maillard reaction”
of maturity of the cheese (proteolysis), pH and the intensity of the
thermo-​mechanical treatment applied. For example, the higher
the calcium content of the cheese, the more difficult the melting
is. After hydration and dispersion of the proteins in the aqueous TABLE 18.1
phase, the initial structure of the cheeses is transformed from a Examples of the Thermo-​Functionalities of Cheeses
gel to a sol (i.e., colloidal solution). When the proteins are suffi- Thermo-​Functionality Description
ciently solubilized under the effect of melting salts and shearing
Spreadability/​Flowability Ability of cheese to cover a surface
(stirring) exerted during the cooking phase, the emulsification of Stretchability Ability of cheese to form fibrous strands
the fat phase will occur (Lee et al., 1979; Caric and Kalab, 1993; that deform under the effect of vertical
Shirashoji et al., 2005). stretching
This emulsification results in a significant change in the vis- Meltability Tendency of the cheese to melt during a
cosity of the product. heat treatment
Since hydrophobic interactions are very important in the Oiling Off A measure of the stability of the emulsion
state of the fat in a cheese
manufacture of a fondue, the structures and viscosities will vary
depending on the temperature. (Data from Kerjean, 2018)
114 Sébastien Roustel, John A. Hannon

TABLE 18.2
Composition of Different Cheeses at De-​Molding
% Industrial Camembert Edam Cheddar Industrial Emmental Grana
Dry Matter 42 50 63 60.5 64
Fat 19.75 20 33 27.5 22.5
Fat in dry matter 47 40 52 45 35
Calcium (Ca) 0.375 0.7–​0.8 0.8 0.92–​0.96 1.15–​1.30
Ca/​solid no fat 1.6–​1.7 2.5–​2.7 2.7–​2.8 2.8–​2.9 2.8–​2.9
Oligosaccharides 1.4–​2.0 ≈0 0.2–​0.5 ≈0 ≈0
Lactates 1.2–​1.4 1.2–​1.4 1.3–​1.5 1.2–​1.5 1.2–​1.5
pH 4.75–​4.80 5.15–​5.25 5.20–​5.30 5.25–​5.35 5.20–​5.30

(Mietton, 2005)

(Thomas, 1969; Johnson and Olson, 1985; Xixiu et al., 2013).

This is also the cause of the development of colour on the crust of
bread during baking as well as during the cooking and grilling of
meat. In fact, cheese browning involves a succession of reactions
leading to non-​enzymatic browning, which are dependent on the
heat treatment applied (optimum temperature between 140 and
165 °C) and the reaction between reducing sugars and free amino
acid groups (Thomas, 1969). In addition to the appearance of a
brown colour, this reaction can lead to the formation of odorant
compounds (Bertrand et al., 2018).
In cheese, the main reducing sugar involved is D-​galactose
(Johnson and Olson, 1985). Lactose, another sugar that may be
present in some cheeses and the main sugar in milk, can also play
a role but is of less importance; this disaccharide is considered
to be 8 to 10 times less reactive than D-​galactose (a monosac-
charide) in the Maillard reaction. In cheese, the D-​galactose con-
tent comes essentially from the type of lactic acid bacteria used
(some strains of the thermophilic Streptococcus thermophilus)
during cheese making (Milesi et al., 2011) and the degree of
draining of the curd grains in the vats with respect to the rate
of acidification (Mietton and Chablain, 2018). It should also be
noted that some strains of lactic acid bacteria have the ability to
consume D-​galactose, while others do not.
On the amino acid side, these are rarely the limiting factor in FIGURE 18.10 Impact of two different coagulants on browning. Coagulant
A is less proteolytic than coagulant B.
mature cheeses. When they are present in smaller quantities (as in
young cheeses, for example), the phenomenon of browning can
be less pronounced. The presence of free amino acids results from formation of steam, which rises to form pockets of hot air
proteolysis (enzymatic degradation of proteins during ripening), under the cheese surface and interferes with browning, since
which is due to the interaction between the lactic acid bacteria these bubbles will lead to the flow of free fat from the top of the
present, the coagulant used in cheese production (McSweeney bubble, resulting in higher dehydration at the top of the blister
et al., 2017) and the ripening conditions (duration, temperature (Figure 18.11).
and relative humidity). The more proteolytic the coagulant, the
greater the amount of amino acids that will be released and poten-
tially the higher the browning (Figure 18.10). Practical Considerations
The presence of free fat also plays a role in regulating the dehy- Depending on the type of cheese used for toppings, the browning
dration of the cheese surface (Richoux et al., 2008). The presence result can be very different. In fact, the residual sugar content
of a free fat film on the surface has a protective effect and reduces (Table 18.2), the free amino acids, the fat content released during
the rate of browning. The level of free fat is, therefore, a factor heating, and the blister formation (Figure 18.12) will be very
that can explain differences in cheese browning. However, too variable (Xixiu et al., 2014).
much free fat often leads to appearance defects. For example, a cheese whose production includes a lactose-​
Moreover, the formation of bubbles on the surface of melting reducing step (e.g., washing the curd grains with water during
cheese (known as blisters) is a phenomenon produced by the manufacture), such as Raclette or Gouda, leads to a dilution of
Cheese: Hot Culinary Uses of Cheese
FIGURE 18.11 How blisters form on pizzas.
FIGURE 18.12 Schematic diagrams of blister formation and pizza baking performance of different cheeses. The number of arrows reflects the amount of
moisture.
(Xixiu et al., 2014)
the residual sugars contained in the cheese and consequently
decreases the browning potential.
Pizza cheese, which is often made using strains of lactic
acid bacteria that do not consume D-​galactose and has a micro-​
structure that can promote the formation of blisters, browns more
easily (even with a low level of proteolysis). Emmental and
Gruyère, on the other hand, will give a crunch as well as taste to
the crust.
It should be noted that ready-​to-​use gratin cheeses are often a
mixture of different cheeses, which, when combined, may have
bespoke cooking properties.
The Spreadability of Cheeses
In some culinary applications, the cheese is required to spread and
flow extensively to cover the dish with melted cheese; this is the
case, for example, for pizzas and gratins. In these applications,
115
this thermo-​functionality aims to produce a certain visual aspect
while at the same time using less actual cheese to achieve the
same visual aspect in order to limit the costs associated with
this ingredient (the cheese may represent more than 50% of the
material costs of a pizza). For other uses, such as hamburgers or
cordon bleu, the cheese must melt in the mouth but not spread in
order to maintain its geometric shape and/​or remain in its coating.
In both situations, this thermo-​functionality is due to the ability of
the cheese to flow and spread over the surface of the application.
Many methods have been developed to study this property,
but the Schreiber test and its adaptations are currently most
widely used. This test involves placing a pre-​grated and slightly
compressed cheese cylinder of defined size and mass in an oven
for a constant defined time and temperature and then measuring
the area occupied by the melted cheese. The result is expressed
as a coefficient of spreading with respect to the initial surface
area covered by the grated cheese. Depending on the cheese
used, the results can be very different, as shown in Figure 18.13.
116
FIGURE 18.13 Example of different cheese meltability and browning after 5 min at 250 °C.
Some cheeses have a very low level of spreading (coefficient less
than 1.5), while others spread a lot (coefficient greater than 4).
This test also makes it possible to visually evaluate the degree of
browning in the cheese.
Physico-​Chemical Mechanisms of Cheese Spreading
The capacity of a cheese to flow and spread depends on several
parameters, as shown in Figure 18.14, but the fat (free fat content
and size of the fat globules) and the level of proteolysis are the
main factors involved.
The fat containing fatty acids that are partially crystallized
(20% to 50% at room temperature) melts at temperatures above
40–​45 °C (Lopez et al., 2006). Depending on the level of break-
down of the fat globule membrane (a rate closely dependent
on the cheese technology used) (Guinee et al., 2004; Lopez
et al., 2007), fat will form fused clusters of coalesced fat and
fat pockets that will exude from the protein matrix. This flow
of free fat leaving the protein network is considered to be the
driving force behind the spreading of cheese (Lefevere et al.,
2000), since a minimum level of mobility of the fat is indispens-
able for spreading (Kerjean, 2018). Thus, if the fat is very well
emulsified, there will be little exudation of fat during cooking,
the surface of the cheese will dry and become stiff, and this will
have the effect of immobilizing the mass of cheese, which will
not spread (Rudan et al., 1999). This property is used in the pro-
duction of cheeses where little or no flow or spread of cheese is
required (e.g., soup).
Sébastien Roustel, John A. Hannon
The second key factor is the rate of protein degradation
(level of proteolysis). The higher the level, the more the cheese
structure will be weakened and the more it will tend to spread
(Kerjean, 2018). Conversely, a very young cheese (fresh Tome,
for example) with the breakdown of fat identical to a ripened
cheese will tend to spread less. A cheese produced from highly
heated milk (more than 80 °C) in which whey proteins are
denatured and have interacted with the caseins will also have less
ability to spread.
In addition to these two key factors, the moisture content of
the cheese will affect the ability of the cheese to flow and spread,
as well as the level of mineralization (Mietton and Chablain,
2018). A high degree of mineralization in the cheese will result
in a lower ability to spread. This mineralization is a function of
the cheese technology used and therefore a function of the type
of cheese.
Practical Considerations
The flow and spread of a cheese increase with age (due to the
level of proteolysis and the loss of insoluble calcium) as well as
with increased moisture or fat. In fresh cheeses, cottage cheese or
Feta (cheeses with high moisture contents), where the fat globules
are poorly broken down and where the pH is low (of the order of
pH 5), the ability of the cheese to flow and spread is limited.
In cooked pressed-​curd cheeses (Comté, Beaufort or Emmental),
the fat globules are significantly broken down (Lopez, 2005), and
this favours the flow and spreading of the cheese.
Cheese: Hot Culinary Uses of Cheese
FIGURE 18.14 Factors affecting cheese meltability.
If a high level of spreading is desired, it is therefore neces-
sary to choose cooked or uncooked pressed curd cheeses. For
uncooked pressed curd cheeses (Raclette, Cheddar and Gouda),
the level of breakdown of the fat globule is a function of the
intensity of the thermo-​ mechanical work applied during the
manufacture of the cheese (e.g., cutting of the curd, pressing and
stretching). Therefore, for the same cheese type, it is possible to
have different spreading ability depending on the technological
route taken.
It should also be noted that the balance of the individual
components in the cheese will lead to a thermal inertia. The more
important it is, the higher the spread.
Cheese Stretching: Pizzas and “Aligot”
Stretching is the ability of a heated cheese to form fibrous
strands that elongate without breaking when a vertical force is
applied (Kerjean, 2018). This thermo-​functionality is important
for many culinary applications. For pizza applications, many
consumers do not want “strings” that are too long, but they still
want “strings” of a certain length. In other culinary applications,
such as aligot (a dish made from cheese blended into mashed
potatoes that is made in the Aubrac region in the southern
Massif Central of France), a very stretchable cheese producing
long strings is required. The quality of the strings is also an
important attribute; ideally, they should be thin strings with a
fibrous appearance. Many young cheeses have a good ability
to stretch (Cheddar, Tomes and Comté) but, during ripening,
this ability deteriorates. For example, during the ageing of
Mozzarella, the stretching length increases during the first
117
few days after manufacture, but after 3 or 4 weeks, the strings
become short and fragile.
Physico-​Chemical Mechanisms of Stretching
The stretch results from the interaction between caseins (major
protein of cheeses), minerals (phosphate and calcium) and fat
(Figure 18.16). When the cheese is heated at a temperature above
45 °C, the fat will liquefy and act as a lubricant (Roset et al.,
2004). When the temperature is increased further and reaches
70 °C, the fat melts and forms pockets of fat, which separate
from the protein network, leading to an increase in the hydro-
phobic interactions between the proteins. During agitation and
stretching, the protein matrix rearranges and forms regions dense
in protein threads (Figure 18.17). The length of the “strings” is
strongly correlated with the level of proteolysis of the cheese (in
particular β-​ and αs1-​casein). The higher the level of αs1-​casein,
the shorter the protein “strings” (Guinee et al., 2004; Kerjean,
2018; Roustel, 2018). Richoux and Gagnaire (2013) and Guinee
et al. (2004) specified that the stretching of cheese is strongly
correlated with the content of hydrophobic peptides.
In addition, the resistance of the protein network to stretching
is closely related to the mineralization of the cheese, and more
particularly to the equilibrium between the mineral content in the
colloidal phase (insoluble proteins) and in the soluble phase. If
the cheese is too mineralized, the mechanical strength of the pro-
tein network is important and the threads will tend to break under
the application of traction. This is why some cheeses have a better
aptitude for stretching than others, because, during their manu-
facture, the kinetics of acidification with regard to the draining
of the curd makes it possible to partially demineralize the cheese.
118 Sébastien Roustel, John A. Hannon

FIGURE 18.15 Evolution of the fat/​protein matrix during melting.

FIGURE 18.16 Factors affecting cheese stretch-​ability.

Cheese: Hot Culinary Uses of Cheese 119

and widely usable measurement methods. From a cheese point

of view, understanding the technological factors impacting the
thermo-​functionalities of the cheeses, modelling them and put-
ting them into practice during the manufacturing process remains
delicate, but it is possible to intervene during cheese making to
respond to these functional properties. Finally, on the culinary
side, the objectives can be very varied and sometimes antagon-
istic, such as, for example, obtaining a melting cheese that has
no spread.
Moreover, in addition to cheese, the conditions of use and the
type of oven can have a major impact on the thermo-​functionality
of cheese. Convection or forced air ovens are common in catering
because they allow faster heating. In these ovens, where the hot
air is forced to move, rapid heating results, but the ventilated hot
air will dry the surface of the cheese, which will accentuate the
risk of browning and the formation of bubbles on the surface
of the cheese. In microwave ovens, cheese put on pizzas tends
to become hard. Today, cheese makers can provide cheese that
can behave as an ingredient in very specific types of oven and
heating.
FIGURE 18.17 Schematization of mechanism of the stretching of Pasta-​
filata and Aligot.
REFERENCES
Bertrand E, El Boustany P, Faulds CB, Berdagué JL. 2018. The
A calcium content of between 1.7% and 2.1% in relation to Maillard Reaction in Food: An Introduction. Reference Module
the solid non-​fat content (mainly composed of proteins) seems in Food Science. Elsevier, Amsterdam, The Netherlands.
optimal for stretching (Mietton and Chablain, 2018). Boutonnier J, Roustel S. 2014. Fromage fondu : physico-​chimie du
processus de fonte, Techniques de l’ingénieur, F6310, 2, 1–​15.
Available at: www.techniques-​ingenieur.fr/​base-​documentaire/​
Practical Considerations: Aligot procedes-​chimie-​bio-​agro-​th2/​filiere-​de-​production-​produits-​
When making “aligot”, a mixture of fresh Tome (about 30%), d-​origine-​animale-​42432210/​fromage-​fondu-​physico-​chimie-​
du-​processus-​de-​fonte-​f6310/​, last access 4 December 2020.
mashed potato (about 60%) and cream (about 10%) is heated Carić M, Kaláb M. 1993. Processed cheese products. In Fox PF
and mechanically stretched, whereupon the texture of the mix- (ed.) Cheese: Chemistry, Physics and Microbiology (2nd ed.),
ture will change. The fat will liquefy, the proteins will form Chapman and Hall, New York, NY, 467–​505.
fairly long strings (the level of proteolysis of fresh Tome is very Dimitreli G, Thomareis AS, Smith PG. 2005. Effect of emulsifying
low) and the starch contained in the mashed potatoes will form salts on casein peptization and apparent viscosity of processed
a starch paste, leading to an increase in the viscosity. This vis- cheese. International Journal of Food Engineering, 1(4), 1–​17.
cosity will slow the flow rate of the product during stretching and Goudédrenche H, Camier-​Caudron M, Gassi JY, Schuck P. 2007.
Procédés de transformation fromagère, Techniques de
thus allow long and stable strings. By modifying the nature of the
l’ingénieur, F6305. Available at: www.techniques-​ingenieur.fr/​
starch (amylose/​amylopectin ratio), it is possible to change the base-​documentaire/​procedes-​chimie-​bio-​agro-​th2/​filiere-​de-​
viscoelastic properties of the aligot and consequently its quality. production-​produits-​d-​origine-​animale-​42432210/​procedes-​
Similarly, depending on the shelf-​life of the potatoes before use, de-​transformation-​fromagere-​partie-​1-​f6305/​, last access 4
the starch of the latter will evolve and give different results. December 2020.
Guinee TP, Kilcawley KN. 2004. Cheese as an ingredient. In Fox FF,
McSweeney PLH, Cogan TM (eds) Cheese Chemistry, Physics
and Microbiology, Elsevier, Amsterdam, The Netherlands,
Conclusions 395–​428.
Guinee TP. 2016. Cheese as a food ingredient. Elsevier, Amsterdam,
Today, cheese is increasingly used as a hot ingredient in a wide
The Netherlands.
variety of prepared foods, such as soups, sauces, pizzas, lasagna, Gupta SK, Karahadian C, Lindsay RC. 1983. Effect of emulsifier
hamburgers and pies. It must therefore have specific heating salts on textural and flavour properties of processed cheeses.
properties, known as thermo-​ functionalities. These thermo-​ Journal of Dairy Science, 67, 764–​778.
functionalities are related to the composition of the cheese, the Johnson ME, Olson NF. 1985. Nonenzymatic browning of Mozzarella
level of ripening, the structure of the fat and the parameters cheese. Journal of Dairy Science, 68(12), 3143–​3147.
used during cooking (support, e.g., pasta, tomato sauce, bread Kerjean JR. 2018. Le fromage. In Mesure, compréhension et maîtrise
des thermofonctionnalités des fromages affinés (4th ed.),
and intensity of the heat treatments). Depending on these
Lavoisier Tec&Doc, Paris, France, ­chapter 19.
characteristics, it is possible to promote or slow down the expres- Lee BO, Paquet D, Alais Ch. 1979. Etude biochimique de la fonte
sion of these properties. From a scientific point of view, the major des fromages –​I. mesure de la “peptisation”. Le lait (1979),
problem lies in the development of reproducible, significant 589–​590.
120
Lefevere I, Dewettinck K, Huyghebaert A. 2000. Cheese fat as a
driving force in cheese flow upon melting. Milchwissenschaft,
55, 563–​566.
Lopez C. 2005. Focus on the supramolecular structure of milk fat
in dairy product. Reproduction Nutrition Development, 45,
497–​511.
Lopez C, Briard-​Bion V, Camier B, Gassi JY. 2006. Milk fat thermal
properties and solid fat content in Emmental cheese. Journal of
Dairy Science, 89, 2894–​2910.
Lopez C, Camier B, Gassi J. 2007. Development of the milk fat
microstructure during the manufacture and ripening of
Emmental cheese observed by confocal laser scanning micros-
copy. International Dairy Journal,17, 235–​247.
Marchesseau S, Gastaldi E, Lagaude A, Cuq JL. 1997. Influence of
pH on protein interactions and microstructure of processed
cheese. Journal of Dairy Science, 80, 1483–​1489.
McSweeney PLH, Fox PF, Cotter PD, Everett W. 2017. Cheese:
Chemistry, Physics & Microbiology (4th ed.). Academic Press,
London, UK.
Mietton B. 2005. Minéraux et fromage fondu. In Minéraux et
produits laitiers, Lavoisier Tec&Doc, Paris, ­chapter 16.
Mietton B, Chablain I. 2018. Le fromage. In Pratiques et techniques
fromagères des différentes familles de fromages (4th ed.).
Lavoisier Tec&Doc, Paris, ­chapter 8.
Milesi MM, Bergamini CV, Hynes E. 2011. Production of peptides
and free amino acids in a sterile extract describes peptidolysis
in hard-​cooked cheeses. Food Research International, 44(3),
765–​773.
Mintel M. 2017. Etude Innovation laitière, https://​fr.mintel.com/​, last
access 4 December 2020.
Pilcher SW, Kindstedt PS. 1990. Survey of Mozzarella cheese
quality at restaurant end use. Journal of Dairy Science, 73,
1644–​1647.
Profession fromager. 2015. Le Guide des fromages au lait cru.
Profession fromager, Lille, France.
Richoux R, Aubert L, Kerjean JR. 2002. Outils de mesures des
propriétés thermo-​ fonctionnelles de l’emmental français,
Actalia, France.
Richoux R, Aubert L, Roset G, Briard-​Bion V, Kerjean JR, Lopez
C. 2008. Combined temperature–​time parameters during the
Sébastien Roustel, John A. Hannon
pressing of curd as a tool to modulate the oiling-​off of Swiss
cheese. Food Research International, 41(10), 1058–​1064.
Richoux R, Gagnaire V. 2013. Protéases de lactobacillus helveticuas
et leurs consequences sur les propriétés culinaires de
l’emmental. Revue des Enils, 324, 8.
Roche M. 2018. Connaissance, notorité et usages du fromage par le
consommateur. In Gillis JC and Ayerbe A (eds.) Le fromage
(4th ed), Lavoisier Tec&Doc, Paris, ­chapter 18.
Roset G, Richoux R, Aubert-​Frogerais A, Kerjean JR. 2004. Etude
des propriétés culinaires des fromages. Etude Actalia O049C.
Available at: www.actalia.eu
Roustel S, Boutonnier JL. 2015. Fromage fondu: technologie de
fabrication et contrôle qualité, Techniques de l’ingénieur,
F6311.
Roustel S, Chambre M, Meyer-​Caron H. 2018. Fromages fondus. In
Le fromage (4th ed.), Lavoisier Tec&Doc, Paris, ­chapter 11.
Rowney MK, Roupas P, Hickey MW, Everett DW. 2004. Salt-​
induced structural changes in 1-​day old Mozzarella cheese
and the impact upon free oil formation. International Dairy
Journal, 14(9), 809–​816.
Rudan MA, Barbano DM, Yun JJ, Kindstedt PS. 1999. Effect of fat
reduction on chemical composition, proteolysis, functionality
and yield of mozzarella cheese. Journal of Dairy Science, 82,
661–​672.
Shirashoji N, Jaeggi JJ, Lucey JA. 2005. Effects of various emul-
sifying salts on the rheological and texture properties of
pasteurized process Cheddar cheese. Journal Dairy Science,
88(Suppl 1), 19.
Sperat-​Czar A, Pereira D, Roustel S. 2018. Cheese ripening guide.
Edition Profession Fromager, Lille, France.
Thomas MA.1969. Browning reaction in Cheddar cheese. Australian
Journal of Dairy Technology, 24(4), 185–​189.
Xixiu M, James B, Balaban MO, Zhang L, Emanuelsson-​Patterson
EAC. 2013. Quantifying blistering and browning properties of
Mozzarella cheese. Part I: Cheese made with different starter
cultures. Food Research International, 54, 912–​916.
Xixiu M, Balaban OM, Zhang L, Emanuelsson-​ Patterson EAC,
James B. 2014. Quantification of pizza baking properties of
different cheeses, and their correlation with cheese function-
ality. Journal of Food Science, 79(8), E1528–​1534.
Chocolate: Chocolates from around the World, Simple Physics,
Complex Flavour
Bhagyashri L. Joshi, Sarah Gindra and Thomas A. Vilgis
Max-​Planck-​Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany
The unique characteristic of chocolate is that it is solid at room
temperature, yet it melts in the mouth rapidly and flows smoothly,
resulting in a sensational mouthfeel. This property depends on
the ingredients of chocolate, which consists mainly of cocoa
butter, cocoa solids, and sugar, and even milk powder in so called
“milk chocolates”. Although chocolate is known today as a sweet
and an attractive food, it used to be an astringent and fatty drink
(Beckett, 2018). Over time, lots of different techniques have
been developed to produce the finest product, which we consume
nowadays.
Hierarchy of Chocolate Production
Before discussing the current advances in some questions of choc-
olate science, let us give a brief history of this product. Chocolate
production dates back to 600 AD, which marks the onset of the
first known cocoa plantation (Figure 19.1) (Whymper, 1912).
Aztecs and Incas used cocoa beans in the production of chocolate
drinks by fermenting the beans in banana leaves for several days
and then making a paste by pressing and milling. This paste was
highly fatty and astringent (Beckett, 2018).
Soon after Christopher Columbus brought cocoa beans to
Europe, sugar was introduced to the chocolate drink, which
added new flavour to the drink (Beckett, 2018). In 1727, Nicholas
Sanders added milk; however, the fat content was still high, as
cocoa beans consist mainly of cocoa butter. The Dutch engineer
Coenraad Johannes Van Houten developed a cocoa press machine
in 1828, which helped to remove half of the fat.
Later, it was possible to successfully extract cocoa butter from
cocoa beans. However, chocolate remained a drink until 1847,
when Joseph Fry first prepared a plain block of chocolate in the
UK (Beckett, 2018). It consisted only of sugar, cocoa butter, and
cocoa solids. Cocoa butter has the ability to hold the cocoa solids
and sugar particles together to form firm blocks.
Soon after the discovery of this brown plain chocolate block,
it was modified by Daniel Peter into milk chocolate (Beckett,
2018). This milk chocolate mainly consisted of milk powder
with a regulated moisture content, which helps to control sugar
reactivity towards the moisture. Later, in 1930, white chocolate
was invented, which consisted only of cocoa butter and sugar.
As the cocoa solids were not present in the white chocolate, it
was difficult to convince some people to call this type chocolate
(Beckett, 2018).
In 2017, the Belgian-​Swiss company Barry Callebaut released
the fourth type of chocolate, which is known as “Ruby chocolate”.
This basically consists of sugar, cocoa solids, and cocoa butter;
however, it has a pink colour, which is not due to any genetic
modification (Nieburg, 2017). This chocolate is made from Ruby
cocoa beans, and it releases sweet and sour taste compounds in
the mouth. Studies of fermentation show that the colour of cocoa
solids depends on the fermentation process of cocoa beans. Each
step of fermentation involves changes in colour; if the fermen-
tation process is terminated early, a violet-​pink or red-​pinkish
colour occurs. Therefore, this chocolate has a pink colour and a
fruity taste (Sulaiman and Yang, 2015).
In order to produce such a variety of chocolates, the main
ingredients play very important roles. The quality and flavour of
cocoa butter and cocoa solids depend on the climate, soil, and
environment where the cocoa trees are planted (Beckett, 2018).
Figure 19.2 represents the statistical analysis of the world’s cocoa
bean production. Today, Côte d’Ivoire is the main producer of
cocoa beans, with half of the total production. The second lar-
gest producer is Ghana, with 20% of production, followed by
Indonesia (7%). Other countries such as Ecuador, Cameroon, and
Nigeria, produce about 6% of cocoa beans. Next, total produc-
tion in Brazil and Papua New Guinea is 4% and 1%, respectively
(Statista, 2018).
The Science behind Chocolate
After this short history of chocolate, let us shed a light on the
science of the production of this product. Any food product is
intended to convey “good” taste and mouthfeel. This feeling is
essential with chocolate. What is the basic science behind it?
The answer lies in the crystallization process of the ingredients.
The main ingredients of chocolate are sugar crystals, cocoa
solids, and cocoa butter. It “tastes good” only when the fats of
the cocoa butter are in a particular crystal form (Ziegleder and
Danzl, 2016). We shall discuss this point in detail in the next
section.
121
122
FIGURE 19.1 Hierarchy of chocolate invention.
(Copyright MPIP Mainz)
FIGURE 19.2 Statistical analysis of highest cocoa-​bean-​producing countries in the world from the year 2017/​2018.
The composition of cocoa butter differs according to the region
of cultivation, climate conditions, etc., and due to such differences
in triacylglycerol (TAG) composition (Beckett, 2018), it is a difficult
task for industrial staff to achieve good food product quality regu-
larly. Let us briefly discuss the main components of cocoa butter.
Specifically, cocoa butter consists mainly of TAGs with
residues of oleic acid (O) C 18:1, stearic acid (S) C18:0, and
Bhagyashri L. Joshi et al.
palmitic acid (P), C16:0. The nomenclature C n:s means that
the fatty acids consist of n carbon atoms with s saturations;
when s = 1, this is called a mono-​unsaturated fatty acid. Most
triglycerides in cocoa butter are in the form POS, as shown in
Figure 19.3; the oleic acid residue is in the middle position on the
glycerol residue. This structure therefore essentially determines
the crystallization behaviour of the cocoa butter.
Chocolate: Simple Physics, Complex Flavour
Six different polymorphs are formed after crystallization of
cocoa butter, and depending upon their stability, the quality
of chocolate is determined. In Figure 19.4, the polymorphic
forms and their melting temperatures are given (Wille and
Lutton, 1966).
The phase γ (I) (T = 17.3 °C) is formed only with rapid cooling
of the molten chocolate, e.g., in the freezer compartment of a
refrigerator. The mouthfeel is soft to crumbly; the chocolate
quickly becomes shiny and, even with very short storage times,
shows strong fat bloom formation (Hiroyuki, 2009). Fat bloom
can be defined as a white appearance on the chocolate surface
due to either phase separation or crystal phase transition. Due to
the rapid cooling rate, the triglyceride molecules, which diffuse
more and more slowly during cooling, do not succeed in reaching
a stable structure (He and Voronine, 2016). Thus, the structure
is far from thermodynamic equilibrium, which is attained by
solid–​solid transition of crystal forms during storage. As a result,
the fat bloom occurs in a very short time. Despite the negative
mouthfeel, this phase has one advantage: the aromas are appar-
ently released very quickly.
The phase α (II) melts at 23.3 °C, and this phase is achieved
at cooling rates of 2 °C per minute (Bresson et al., 2011). This
phase is already somewhat more stable; however, it has a similar
mouthfeel to phase I. Phase II can also be obtained by storage of
phase I at room temperature. The energy barrier between them is
FIGURE 19.3 Molecular structure of a triglyceride whose fatty acid
residues are oleic, stearic, and palmitic acids.
FIGURE 19.4 Melting temperature and stability of six polymorphs of cocoa butter after crystallization.
123
approximately 50 kT; therefore, a “recrystallization” from phase
I to phase II can take place within hours.
Phase β′ (III) and phase β′ (IV) are very similar in their proper-
ties, as their melting temperatures are 24.5 and 27.3 °C, respect-
ively, very close to each other. Crystals of these phases can be
produced with cooling rates of 5 °C to 10 °C per minute. Because
of such high cooling rates, many “defects” in crystal structure
are formed, causing quick melting in the mouth. These defects
are also responsible for formation of fat bloom during storage
because of high mobility of fat molecules. Therefore, due to the
high number of irregularities in the crystal structure, defects,
fitting, and stacking faults (Vilgis, 2016), the fracture appears
brittle, and the fracture surfaces are less smooth.
Phase β (V) is the most appealing phase for culinary purposes.
With sufficient stability and a melting point of 33.8 °C, this
crystal structure is perfect for the dental enamel in the mouth;
the odorant compounds are released gradually during melting
due to oral coating with the cocoa butter, providing long-​lasting
retronasal detection. Breakage (by hand) or biting in the mouth is
smooth and instantaneous. Physically, however, this phase is not
in the thermodynamic equilibrium state. Therefore, cocoa butter
has to be “forced” into this crystal phase, which changes into
even more stable phase VI when stored for a long time.
Phase β (VI) is the most stable of all polymorphs of cocoa
butter, with the highest density and highest melting temperature
(Ziegleder and Danzl, 2016). As a result, the mouthfeel is not very
good. As the melting temperature and enthalpy are higher than
for other crystal forms, the chocolate hardly melts in the mouth, it
has to be bitten, and the aromas are released very slowly. In add-
ition, the low-​molecular weight-​volatile odorant compounds with
herbal notes are released slowly, which imparts a sweaty, greasy
flavour. The full chocolate odour is therefore not perceived; the
odour is withheld by this crystal structure, although all odorant
compounds are present. The physical properties prevent and
delay their release.
124
The Forced, Non-​Equilibrium Phase β (V)
From a physical point of view, this diversity of structures can
be illustrated using free energy (Helmholtz potential). In add-
ition to energy, free energy takes into account the entropy of
molecules and is therefore a possibility to describe crystal-
lization. During crystallization from the disordered melt (high
entropy), the order increases (low entropy). Thermodynamically,
this decrease in entropy can only take place if it is compensated
by a “gain” in energy, in this case the released crystallization
energy. Therefore, the free energy in a thermodynamically stable
state always assumes a minimum. In the crystallization of cocoa
butter, depending on the process, various minima of free energy
belong to each of the crystal phases described in this section.
However, these are only local minima, as schematically shown
in Figure 19.5.
Phase VI is the most stable form, as can be seen from the abso-
lute minimum of free energy. Phases I and II are energetically
very “unfavourable”. The aim is always the absolute minimum
in the free energy landscape, i.e., the crystal phase VI. To get
there, the process has to be designed accordingly. If one does not
give the liquid cocoa butter enough time to cool down, the solidi-
fying cocoa butter “hangs” in one of the upper minima. Depending
on the storage temperature, the system can no longer reach the
next minimum, because it is relatively unlikely that it will be able
to jump over the potential mountains between the minima. Only
if these mountains are not too high, e.g., between phases I and
II, is a transition from one phase to the other possible after some
time. The transition probability p from one phase to the other is
determined by the Boltzmann factor, which compares the thermal
energy kBT with the height of the potential mountain U B; the mean
time τ can be estimated with the inverse of the Boltzmann factor:
 −U B  U 
p = exp   τ = exp  B 
k TB k T B
These simple equations show that the higher the potential
mountain U B and the lower the temperature T , the smaller the
FIGURE 19.5 Relation of free energy to the different phases of cocoa butter
crystals.
Bhagyashri L. Joshi et al.
probability and the longer the mean transition period τ becomes.
This makes it clear that the cooling rate of chocolate is an
important parameter in process control, because the resulting
cooling time between the melting temperature and the target tem-
perature must always be comparable with the transition time of
the respective minimum. Rapid cooling speeds therefore always
lead to undesirable phases I to IV (Vilgis and Bhagyashri, 2017).
The area of the desired phase V and its neighbourhood to phase
VI is interesting. The latter has the lowest minimum, so it is the
most stable phase. However, the unstable phase V is desired from
a sensory point of view. Physical tricks must therefore be used to
“force” the cocoa butter into this phase. During the crystalliza-
tion process, before crystal growth, very small, nanometre-​sized
crystal nuclei are formed. These must have a critical size so that
they can grow into crystals. Seed crystals that are too small quickly
melt again because their surface energy is excessively high.
In the conventional cooling process, the cocoa butter melt is
cooled down to 25 °C–​27 °C in order to form polymorphic nuclei
of phase IV and V, which then grow into crystals. Afterwards,
the sample is heated again to 33 °C and kept at this tempera-
ture for about 5 to 7 minutes. During this time, the crystals of
the undesired phase IV melt, while the crystals of the form V
are just forming. Those crystals act as “seed crystals” that can
grow slowly during the tempering time. This sufficient amount
of crystals is then further cooled down in order to remain in the
desired phase V for the most part (Stapley et al., 1999; Afoakwa,
2016). This particular process for obtaining a certain crystal
phase is called the tempering process, described in Figure 19.6.
Thermodynamics of Chocolate
Thermodynamics is the science that deals with heat and tem-
perature and their relation to energy. This section emphasizes the
thermodynamics involved in different kinds of chocolate. When
chocolate is consumed, its melting process requires energy,
which is transferred from the oral cavity to the chocolate. The
energy required depends on the crystal shape and the stability
itself. Melting energies for different phases of cocoa butter are
shown in Table 19.1 (Fessas et al., 2005). This energy is taken
from the tongue and palate during the oral process and must be
replenished through the blood circulation to keep the temperature
in the oral cavity as constant as possible at 37 °C.
The melting energy, or enthalpy, of different chocolates can
be measured by using an analytical technique called differential
scanning calorimetry (DSC). This technique works on the prin-
ciple of difference in the energy required or given out during
the melting or crystallization process, between a control and a
sample, as a function of temperature and time. The difference in
heat flow is described by the following formula:
ФS – ​ФR ~ΔT
where ФS is the heat flow required for the sample (W/​g), ФR the
heat flow required for the reference (W/​g), and ΔT the change in
temperature (°C).
Chocolate: Simple Physics, Complex Flavour 125

FIGURE 19.6 Tempering process used for formation of phase V in cocoa butter.

TABLE 19.1
Melting Enthalpy of Various Crystalline Phases of Cocoa Butter
Crystal phases Melting enthalpy (J/​g)
Phase II 86
Phase III 112
Phase IV 118
Phase V 137
Phase VI 148

The quality of chocolate is studied using this technique in sev-

eral industries to determine the desired melting temperature of
chocolate after the tempering process or also during the settling
process.

Differential Scanning Calorimetry of Chocolate
In Figure 19.7, the DSC results of four types of chocolates are
shown along with pure cocoa butter. In the vertical axis, there
is a factor of 0.2 J/​g between sample measurements. Pure cocoa
butter was heated to 90 °C and kept at 22 °C for 24 h prior to
DSC analysis.
All samples were heated at the rate of 2 °C/​minute in the dif-
ferential scanning calorimeter. As the melting temperature of
the sample is dependent on the heating rate and composition of
the chocolate, the values may be different from some that were
previously published. Thus, the DSC of cocoa butter detected
one main peak (31 °C) along with a very small shoulder peak
at ~27 °C. This result suggests that it crystallized into a mix-
ture of two crystal phases. However, as we discussed earlier, the
system always goes to the state of thermodynamic equilibrium,
which is phase VI, and so, after some days, these phases convert
to phase VI and lead to fat bloom. As an example of fat bloom,
Figure 19.8 shows a microscopic picture of cocoa butter phase
transition, which ultimately leads to fat bloom. This picture was
FIGURE 19.7 DSC endothermic peaks of four types of chocolate and their
comparison with pure cocoa butter.
captured by polarized light microscopy after 8 days of storage
of cocoa butter, which was melted at 90 °C and kept at room
temperature.
Next in the hierarchy of chocolate, dark chocolate with 65%
of cocoa solids was recorded with two consecutive peaks at 30°C
and 34°C. This result suggests that dark chocolate consists of
two types of crystalline phases, and therefore, on the tongue, the
lower-​melting fraction melts first. This already molten portion
wets the tongue and can be heated, while the remaining solid
portion continues to melt. Both for heating and for melting,
energy must be supplied from the mouth. This total energy, or
amount of heat, is the sum of the latent heat of fusion ΔH and the
amount of heat necessary for heating the already liquid phase of
the already molten fats.
Q = m ∆H + mc ∆T
126 Bhagyashri L. Joshi et al.

FIGURE 19.8 Crystal structure of pure cocoa butter after 8 days at room temperature, showing an example of fat bloom.
(Copyright MPIP Mainz)

Using the values ΔH = 46 J/​g, c = 2 J/​gK and ΔT = 15 °C, the for-

mula gives a heat equal to 76 J/​g for this chocolate. This heat is
drawn from the mouth, giving a cooling sensation.
The milk chocolate (Peru) also showed two peaks at 29 °C and
35 °C. In this case, the two peaks are because of the milk fat
and cocoa butter. The energy needed for melting this chocolate is
lower than for dark chocolate. Similarly, for white chocolate, two
peaks were detected; however, the higher-​melting fraction of fat
was much less than the total fat of the chocolate. Hence, the choc-
olate appears soft in the mouth. Interestingly, the heat needed to
melt the milk and white chocolate is 16 and 17 J/​g, respectively:
this indicates that the amount of cocoa butter was less than in
dark chocolate.
The new type of chocolate is Ruby chocolate: the DSC
measurements on this form of chocolate show three melting
peaks from the lower-​ melting, medium-​ melting, and higher-​
melting fraction (melting temperatures of 26 °C, 30 °C, and FIGURE 19.9 Effect of different origins of cocoa beans on chocolate
38 °C, respectively). However, the heat required for Ruby choc- characteristics.
olate to melt is ~4 J/​g, which is really much lower than for other
chocolates. The results indicate that the cocoa butter percentage 33 °C. This tempered chocolate could be prepared using the pro-
in Ruby chocolate was lower as compared with conventional cess explained earlier. Therefore, the energy needed for melting
chocolate. of these three chocolates is 44, 39, and 47 J/​g, respectively, which
As the Latin American countries produce more cocoa beans as is higher compared with other chocolates.
compared with the rest of the world, we therefore shed light on As Ecuador has only one type of cocoa tree production
six different chocolates prepared from each different country’s (Beckett, 2018), we were curious to know the effect of different
cocoa beans (Figure 19.9). fruits on the Ecuadorian chocolates. Therefore, we bought local
All the chocolates except the Cuban one were bought from the chocolates from Ecuador containing two fruits called Higo (fig
Puyricard company, whereas the Cuban one was produced by local fruit) and Maracuja, and Cedron leaves. They were all from the
people having a limited amount of machinery. In Table 19.2, the Pacari company. Cedron is also called lemon verbena, or Aloysia
results of DSC in terms of melting temperature and mouthfeel are citrodora scientifically. This particular plant has a lemon-​like
shown for all chocolates. Interestingly, chocolate from Jamaica, aroma; hence, the chocolate has a nice combination of sweet
Peru, and Madagascar melted at 33 °C, which is the desired and lemon-​like flavour. Although the tastes of all the chocolates
crystal phase V, melting in the mouth immediately. This type of were different from each other due to the different flavours, the
chocolate is called “tempered chocolate”, as it melts sharply at DSC results did not show any significant difference in melting
Chocolate: Simple Physics, Complex Flavour
TABLE 19.2
Differences in Melting Temperature and Various Taste Diversity of Latin American Chocolates
Number Name of chocolate Melting temperature
1 Brazil, 62% cocoa 28°C, 35°C
2 Jamaica, 63% cocoa 33°C
3 Peru, 63% cocoa 33°C
4 Madagascar, 64% cocoa 33°C
5 Ecuador, 66% cocoa 28°C, 36°C
6 Cuba 30°C, 34°C
FIGURE 19.10 Effect of different flavours on Ecuadorian chocolate.
behaviour (Figure 19.10). All these three, and one other con-
ventional (unflavoured) chocolate, showed two peaks at around
30 °C and 34 °C, so the melting in the mouth was quite slow. This
result suggests that the addition of different flavours was not able
to affect the melting characteristics of chocolate.
Effect of Cocoa Percentage on Chocolate
In this section, the focus is on the effect of amount of cocoa added
on the chocolate melting process. All these chocolates are made
from Peru-​origin cocoa beans. The DSC results depict that the
higher the cocoa content, the higher is the melting temperature of
the chocolate (Figure 19.11).
However, the cocoa consists of cocoa solids and cocoa butter
together; hence, it would be difficult to interpret whether the
change in melting temperature is due to fat content or the particle
size distribution of cocoa solids. Afoakwa et al. (2008) showed
that the viscosity of dark chocolate is more affected by fat per-
centage than by the particle size distribution of cocoa solids. This
study gives a deep insight into the results that we revealed from
DSC: although 70% cocoa chocolate has more cocoa, the energy
needed to melt it was less as compared with 63% cocoa choc-
olate. The reason is that either the cocoa butter content is lower
or the particle size is very variable.
127
Tastes
Slightly bitter, fruity
Bitter
Fruity
Spicy, aromatic, acidic, sour
Spicy
Earthy, sandy
FIGURE 19.11 Effect of cocoa percentage on chocolate.
Effect of Sugar Replacers on Thermal Properties of
Chocolate
Chocolates or chocolate confectioneries have high calorie values
due to sugar usage; however, if the sugar is partly or completely
replaced by different sugar substitutes like sugar alcohols, the
ultimate calories will be decreased. In addition, this would be
beneficial for diabetic patients and for tooth decay problems.
Although it is beneficial to health, the physical and thermal prop-
erties of chocolate may differ. As an example, here are the DSC
results on chocolates with maltitol, xylitol, and erythritol + stevia
(Figure 19.12).
The 38% and 65% cocoa + maltitol chocolates were gifted, and
the 54% cocoa was a Hungarian chocolate from the ChocoMe
Company. The 70% cocoa (PaleoBon, Hungarian), 60% cocoa,
and milk chocolate (Jungbunzlauer Company) contain eryth-
ritol + stevia, and the 60% cocoa from Xucker Company
contains xylitol. The melting of these chocolates occurred at the
higher temperature of 35 °C; therefore, they are slow-​melting
chocolates. In addition, the xylitol chocolate showed one peak
at 84 °C; this peak represented the xylitol. While melting in the
mouth, the chocolates give a cooling effect, mainly due to xylitol
and erythritol. Xylitol is known as “birch sugar” and it is used in
128 Bhagyashri L. Joshi et al.

Maltitol Erythritol+Stevia and Xylitol

0.8

0.6 70% cocoa_Erythritol+Stevia

0.6
35°C
0.4 Milk Chocolate_Erythritol+Stevia
0.4 65% cocoa
Heat flow (W/g)

Heat flow (W/g)

35°C
0.2 60% cocoa_Erythritol+Stevia

30°C
0.2 35°C 54% cocoa
0.0
35°C

60% cocoa_Xylitol

0.0 38% cocoa –0.2

–0.4
–0.2 28°C
34°C 35°C
–0.6

20 30 40 50 60 20 30 40 50 60 70 80 90
Temperature (°C) Temperature (°C)

FIGURE 19.12 Effect of sugar alcohols like maltitol, erythritol + stevia, and xylitol on chocolate.

many plants, such as cauliflower and broccoli, as an antifreeze

agent to protect cells from damage and chilling injuries. Xylitol
has about the same sweetness as the commonly used sucrose.
Dissolving xylitol in water requires 153.2 J/​g, which must also
come from the environment. Due to good binding capacity with
water molecules, the heat needed for dissolution is attained in a
short time; therefore, the cooling effect of the sugar alcohol is
enormous. The same applies to erythritol, which has only about
half the sweetening power of sucrose (or xylitol) but exhibits a
comparable endothermic solution enthalpy to that of xylitol.

Salmiakki or “Salty Liquorice Chocolates”

Salmiakki is a Finnish word for “salty liquorice candies” or
confections from Finland. These chocolates contain ammonium
chloride (E510) as a flavouring agent. It is also known as “Sal
ammoniac” and “salmiac”, which is a white, crystalline salt of
ammonia. This salt is a product of chemical reactions of volcanic
gases, and it is also commercially made by reacting ammonia
with hydrogen chloride (Salmiakki, 2019). In Figure 19.13, the
DSC of dark and white chocolate containing salmiac, the main
melting peak was detected at 30 °C in both types, however, the
later-​melting fraction is melting somewhat differently for the two
chocolates. For dark chocolate, two smooth peaks were detected
after the 30 °C peak, whereas the white chocolate has more
“structure” just before 36 °C.
Therefore, the tastes of the two chocolates differ due to
differences in the insertion of the ammonia salt. In white choc-
olate, the salt is large in size; hence, it dissolves immediately after
coming into contact with saliva, also resulting in fat melting. On
the other hand, the dark chocolate contained the crushed salt
form, which dissolved immediately on the tongue. Thus, it could
be possible that the presence of different forms of salts causes
changes in the melting behaviour and so, eventually, the sensation
of chocolate in the mouth.
FIGURE 19.13 Effect of salty liquorice on chocolate melting.
Conclusions
The purpose of this chapter was to show how the different
ingredients influence the melting of chocolates from all around
the world and the scientific view on the understanding of
such behaviour. Four conventional types of chocolate showed
differences in melting temperature and heat of fusion due to the
presence of different ingredients, from dark chocolate to Ruby
chocolate. Similarly, the effect of tempering chocolate would
lead to one melting temperature at 33 °C, which releases the sen-
sational mouthfeel while melting. Increasing cocoa percentage
would lead to a higher melting point. In contrast, different
flavours do not show any significant effect on the melting tem-
perature of chocolate; they only help to enhance the taste without
changing the thermal properties. Likewise, the effect of different
Chocolate: Simple Physics, Complex Flavour
sugar replacers on chocolates is unique, releasing a cooling effect
in the mouth. In contrast to sugar replacers, salt crystals add the
sensation of dissolving salt crystals on the tongue, which releases
a different combination of salty sweet taste in the mouth. Thus,
one can understand from this chapter that different ingredients
and processes of making chocolate are very important for making
the best chocolate, and to achieve such quality, the “physics”
must be studied.
Acknowledgements
We would like to express our deepest gratitude to all the
colleagues who brought us such a variety of chocolates for
doing measurements: Markus Ketomäki for bringing Salmiakki
chocolates from Finland; Mathias Schmitt for Cuban chocolate;
Juan Zambrano for all the Ecuadorian chocolates of different
flavours; and, last but not least, we would like to thank Hannah
Hartge for always bringing us various kinds of chocolates, such
as Peruvian chocolates with different cocoa percentages, and
different sugar replacers, and for always discussing the results.
We would also like to thank “Der kleine Luxus”, a local manu-
facture in Guntersblum (Th. und B. Haertel, GbR) for Ruby
chocolates, and Puyricard Company and Jungbunzlauer Company
for providing different types of chocolates. Finally, we thank
Prasad Chikte for critical reading of this chapter and providing
suggestions.
REFERENCES
Afoakwa EO. 2016. Chocolate Science and Technology, Second
Edition, John Wiley & Sons, Hoboken, New Jersey.
Afoakwa EO, Paterson A, Fowler M. 2008. Effects of particle size
distribution and composition on rheological properties of
dark chocolate, European Food Research Technology, 226(6),
1259–​1268.
Beckett ST. 1999. Industrial Chocolate Manufacture and Use.
Blackwell, Oxford, UK.
Beckett ST. 2018. The Science of Chocolate, Royal Society of
Chemistry, London, UK.
129
Bresson S, Rousseau D, Ghosh S, Marssi M, Faivre V. 2011.
Raman spectroscopy of the polymorphic forms and liquid
state of cocoa butter, European Journal of Lipid Science and
Technology, 113, 992–​1004.
Fessas D, Signorelli M, Schiraldi A. 2005. Polymorphous transitions
in cocoa butter: A quantitative DSC study, Journal of Thermal
Analysis and Calorimetry, 82(3), 691–​702.
He S, Voronine DV. 2016. Raman spectroscopy of chocolate bloom.
arXiv preprint, arXiv:1612.00741.1-​5.
Hiroyuki P. 2009. Blooming theory of tristearin, Soft Matter, 5(4),
856–859.
Nieburg O. 2017. Ruby chocolate: New gem in confectionery crown
or pink misfit? Confectionery News, 14 September. Available
at: www.confectionerynews.com/​Article/​2017/​09/​15/​Ruby-​
chocolate-​New-​gem-​in-​confectionery-​crown-​or-​pink-​misfit,
last access 4 December 2020
Salmiakki. 2019. www.dlc.fi/​~marian1/​gourmet/​salmiakk.htm, last
access 4 December 2020.
Stapley AGF, Tewkesbury H, Fryer PJ. 1999. The effects of shear and
temperature history on the crystallization of chocolate, Journal
of the American Oil Chemical Society, 76(6), 677–​685.
Statista. 2018. www.statista.com/​statistics/​263855/​cocoa-​bean-​
production-​worldwide-​ by-​
region, last accessed 25 January
2019.
Sulaiman KB, Yang TA. 2015. Colour characteristics of dried cocoa
using shallow box fermentation technique, International
Scholarly and Scientific Research and Innovation, 9, 1277–​1281.
Vilgis T, Joshi B. 2017. Noch einmal Schokolade: Physikalisch-​
chemische Aspekte der Sensorik am Beispiel von Genussfetten,
Journal Culinaire, 24, 69–​85.
Vilgis TA. 2016. Schokoladengenuss unter molekularer Kontrolle:
ein Spiel zwischen sensorischen und physikalisch-​chemischen
Parametern, Journal Culinaire, 23, 92–​103.
Whymper R. 1912. Cocoa and Chocolate: Their Chemistry and
Manufacture, J & A Churchill, London, UK.
Wille RL, Lutton ES. 1966. Polymorphism of cocoa butter, Journal
of the American Oil Chemical Society, 43(8), 491–​496.
Ziegleder G, Danzl W. 2016. Die Kakaobutter. Perfekte Kristallisation
und genussvolles Schmelzen von Schokolade, Culinaire JC,
23, 86–​91.
Chocolate: Oral Processing of Chocolate – Successive Interplay
of Sensory and Physicochemical Parameters
Thomas A. Vilgis
Max-Planck-Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany
Chocolates seem to be a highlight for many physical reasons: the
melting, the odorant compounds, the fine bitter notes, the touch
of sweetness, the astringency … How chocolate triggers physical-
chemical sensations in the mouth during oral processing is very
different from individual to individual, but nevertheless it clearly
defines the pleasure process in the physical-chemical “sense”.
In fact, the sensory properties of chocolate are a combination of
multiple time and length scales under a molecular control.
The consumption of pure “dark chocolate” with high cocoa
content is one of few examples on which many facets of a sen-
sory system can be systematically studied, because the “chocolate
system” is, from a physical point of view, relatively simple. In
the simplest case, it consists of fat with a sufficiently simple fatty
acid composition, sucrose, finely milled powder from fermented
and roasted cocoa beans, and emulsifiers, usually lecithin (mainly
phospholipids) from soybeans or sunflower seeds (Beckett, 2002).
Thus, chocolate is a solid dispersion of cocoa butter, cocoa particles,
and sugar, as presented in Figure 20.1.
The crystal structures and the associated melting behaviour
of the cocoa butter can be more or less precisely adjusted using
various processes, as elaborated by Ziegleder and Danzl (2016a).
This allows very precise control of the mouthfeel by temperature
and melting kinetics of the crystalline fat phase (see the previous
chapter by Joshi et al. on chocolates from around the world).
FIGURE 20.1 Model of dark chocolate. At room temperature, sugar
crystals (shown as light, shaded cubes) and cocoa particles (symbolized as
black irregular shapes) are dispersed in a solid, partially crystalline fat matrix
(grey background).
Generally, chocolate, a solid at temperatures between 10 °C
and 20 °C, is placed into the mouth at temperatures around 37 °C.
The solid chocolate heats up and successively starts to melt
from the sides (where it is in contact with tongue and palate).
The melting of the fat matrix is, however, only one aspect of the
enjoyment of chocolate. Besides that, there are two main phys-
ical steps as well as the desired odour and taste release. The first
is the melting of the fat matrix, i.e., the cocoa butter when it turns
from solid fat to a liquid oil, and the second when the liquid oil is
emulsified in saliva by the shear between tongue and palate. The
oral processing under melting and shearing must be subject to
physical laws that make enjoyment and flavour release possible in
the first place. After the cocoa butter has slowly melted, a phase
inversion must inevitably occur in the mouth, since fat does not
properly dissolve in the saliva (Figure 20.2).
While cocoa butter is the “continuous phase” in chocolate,
i.e., fat encloses all water-soluble particles (sucrose or polyols)
and solids (cocoa particles), a reversal occurs in the mouth
during oral processing of the chocolate; melting fat droplets are
FIGURE 20.2 The tongue applies pressure and shear forces on the food in
the mouth, both of which are, in addition to chewing, fundamental physical
processes during oral processing.
131
132
emulsified by saliva, and an oil-​in-​water emulsion is formed,
with important effects for the flavour. In the following sections,
a physical model is systematically developed that takes most
components of chocolate into account and explains the behav-
iour of chocolate in the mouth, and consequently captures a large
part of the perceptions.
Fatty Phase
The fatty phase and its physical properties were described by
Ziegleder and Danzl (2016a) and also in the chapter by Joshi
et al. in this book, so there is no need for a detailed description
here. In the following sections, we use, however, the more gen-
erally understandable presentation for nomenclature and physical
terms, as introduced earlier by Vilgis (2010; 2015).
“Fat molecules”, triacylglycerols, are classic examples of
structure selective crystallizable systems. The molecular mass
(i.e., number of carbon atoms n), the degree of saturation s of
the fatty acids, and the composition of the triacylglycerols are
responsible for the crystallization kinetics and the melting points
of the fat. A triacylglycerol molecule consists of three fatty acid
residues, which are not necessarily identical.
At low temperature, triacylglycerols (e.g., tristearin, made of
three identical saturated fatty acid residues, C 18:0) form almost
perfect crystals, with a clearly defined melting point, which can
be allocated to the corresponding crystal structure. The larger
the number of unsaturated fatty acid residues and the broader
the chain-​length distribution, the more defect-​rich becomes the
crystal structure and the lower the melting point. Moreover, the
distribution of chain lengths and the variation in the saturation
degrees broaden the melting process.
FIGURE 20.3 Possible crystal defects (voids, grain boundaries and stacking error) in cocoa butter (highly simplified and exaggerated presentation in two
dimensions).
Thomas A. Vilgis
Most fatty acid residues in triglycerides of cocoa butter are:
• stearic acid (C 18:0), a saturated fatty acid made of 18
carbon atoms,
• mono-​unsaturated linoleic acid (C 18:1),
• palmitic acid (C 16:0) (Beckett, 2002).
The distribution of the variables n (chain length) and s (degree of
saturation) is thus “narrow”; defects are kept to a limit, and poly-
morphic crystals are formed with clearly defined melting points,
as described by Ziegleder and Danzl (2016a).
In addition, the following consideration is of importance.
Crystal formation is in the thermic equilibrium a first-​order phase
transition (Hartel, 2013). A nucleus is formed; first, triglycerides
molecules start to order themselves and co-​exist with random,
“melted” molecules. Only if the crystal nuclei reach a critical
size, which is a balance of the interface free energy and the
bulk free energy of solid state, can the nucleus grow into bigger
crystals. Since the process of nucleation is randomly distributed
in space, the crystals grow in random orientations. Consequently,
they collide after some time. Because of the random orienta-
tion of the growing nuclei, crystal imperfections, such as grain
boundaries, line defects, and dislocation errors (Hannay, 2012),
are formed. In addition, voids due to mismatches in spacefilling
are most likely (Figure 20.3).
Other “defects” are possible: stacking errors, fit inaccuracies
and cavities. This is desirable and necessary, as small molecules,
emulsifiers, cocoa particles and sugar must find space and arrange
themselves in the fat crystal matrix. Finding the optimal spatial
fits is facilitated by conching (processing of melted chocolate
with sugar), temperature treatment and tempering (Ziegleder and
Danzl, 2016b).
Chocolate: Oral Processing of Chocolate 133

sugar crystals are organized in the chocolate fat during crys-

tallization. Sucrose crystals have a hydrophilic surface, which
is thus capable of adsorbing phospholipids such as lecithins.
Besides taste development, sugar crystals in combination with
emulsifiers possess structure-​forming properties (Stortz et al.,
2014); hence, sugar particles are more homogeneously dispersed
during conching and during crystallization of the fatty phase,
since the emulsifiers adsorb on the hydrophilic crystal surface
(Saska and Myerson, 1983) and serve as “molecular dispersing
agents”. Consequently, the viscosity of the chocolate becomes
significantly lowered by adding surfactants (Figure 20.4).
In this simplified, intuitively developed model, the influence
of different phospholipid types is disregarded. Nevertheless, the
exact atomic structure of the polar head as well as the degree of
saturation and chain length of the hydrophilic tails plays a role
FIGURE 20.4 The chemical structure of limonene. It has two enantiomers
with the same physical properties but different odours: orange type (left) and in the structure-​building properties of cocoa butter, as shown
lemon-​pine type (right). with atomic force microscopy by Middendorf et al. (2016).
The differences in ability of soy lecithin and sunflower leci-
thin to adsorb on sugar crystals in the cocoa butter matrix were
Even volatile odorant compounds, taste-​relevant phenols and highlighted among other factors in that study.
trigeminal tannins are released from the cocoa particles. Most
odour, taste and trigeminal compounds are not bound into the
crystals, as this would cost too much energy and additionally
cause point defects.
The precise location of fat-​soluble compounds has not yet been
solved in detail and has only been studied in model experiments
(Rigolle et al., 2016). Such studies demonstrated, for example,
an altered crystallization behaviour when pure cocoa butter was
mixed with the simple fat-​soluble odour-​active substance lim-
onene, a cyclic terpene that might create additional defects in the
crystallization of the cocoa butter.
Furthermore, phenolic compounds are not bound into the fat
phase to any great extent due to their hydrophilic hydroxyl (OH)
groups. Therefore, these need to fit into the matrix. They either
absorb onto the cocoa particles consisting of cell materials or accu-
mulate into the voids or crystal defects. The crystal defects, grain
boundaries and voids also have a positive effect in the macroscopic
range (Parsons and Goodall, 2011). They ensure the desired frac- FIGURE 20.5 Phosopholipids, in particular lecithins (shown schematic-
ture behaviour of the chocolate, as is the case for all solids (Zhao ally on the left), not only help to improve the flow properties of chocolate
et al., 2018). Sugars are added to all chocolate, including dark (Afoakwa et al., 2009) but also improve the dispersion the fat-​insoluble
chocolate, to soften the bitter taste of roasted cocoa. sucrose crystallites (Arnold et al., 2013). The surfactant molecules absorb
with their hydrophilic head onto the polar (OH group dominating) sucrose
surface of the micrometre-​sized crystals, while lecithin forms “brush-​like”
layers.

Taste: Bitter-​Sweet
Dark chocolate contains more than 60% by weight finely ground
particles of roasted cocoa husks. In dark chocolate, the cocoa
particles are reduced by conching to about 10 micrometres
in size (Ziegleder and Danzl, 2016b). Albak and Tekin (2016)
determined the flavour composition and the total concentration
of polyphenols. The taste-​forming compounds epicatechin and
catechin (non-​volatile phenols; see Figure 20.7) as well as theo-
bromine and caffeine (alkaloids; see Figure 20.8) are important
for the additional effects during oral processing of chocolate, as
will be shown later. Theobromine Caffeine
In addition, the added sugar plays a significant role in softening
the intensity of the bitter taste. Physically interesting is how the FIGURE 20.6 Theobromine and caffeine: bitter-​tasting compounds.
134
FIGURE 20.7 Phenols such as quercetin or catechin (left) contribute to the bitter taste of chocolate. The molecular structures contains two phenol rings that
make the molecules soluble in fat and ethanol. The weak water solubility is generated by the (polar) OH groups capable of forming hydrogen bonds. For the
physical effects discussed in the following, it is sufficient to consider the molecules as hydrophobic ellipsoids, with hydroxyl groups at the outside (right).
Phenols: Bitterness, Astringency and Structure
Formation
Dark chocolate has a distinct bitter taste, which is due to fermen-
tation and roasting processes. The best-​known bitter triggers are
the chemically structurally closely related alkaloids theobromine
and caffeine (Le Berre et al., 2013) (Figure 20.5).
However, more important for understanding the oral pro-
cessing and additional effects, such as the astringent sensation
from dark chocolate and other phenolic compounds, are typical
non-​volatile phenolic compounds (Figure 20.6) (Hoskin and
Dimick, 1994). They also contribute considerably to the bitter
taste in the cases of vegetables, caffeine-​containing beverages
and, of course, chocolate. Non-​volatile phenols and bitter-​tasting
polyphenols are molecules consisting of one or more phenolic
rings (benzene rings with one hydroxyl group or more) but are
(weakly) water-​soluble due to a high number of OH groups (and
a higher molecular mass). The molecular structures are decisive
for sensory properties, both in wine and in chocolate, as both
stimulants are bitter as well as astringent. Both sensory proper-
ties are triggered by non-​volatile phenols, but with significantly
different chemical properties.
Astringency, a “contraction, or dryness in the mouth” during
the consumption of much-​infused tea, some red wine or choc-
olate, is not a taste stimulus. This sensation is triggered chem-
ically by various chemical species, including polyphenols of
the tannins category, via the stimulation of the trigeminal nerve
(Schöbel et al., 2014). The number of neighbouring OH groups
on the phenol rings expresses the difference between bitter taste
and trigeminal astringency: whereas only two OH groups are
necessary for bitter taste, for stimulation of the trigeminal nerve,
three OH groups must be present on at least one benzene ring
(Figure 20.8).
Hence, bitterness and astringency can be clearly differentiated
from a molecular point of view (Schöbel et al., 2014), or even
Thomas A. Vilgis
FIGURE 20.8 Polyphenols with groups consisting of a benzene ring with
three OH groups are responsible for astringency. Gallic acid is most astrin-
gent; as an additional example, myricetin is shown.
physiologically, because the stimuli are triggered by different
receptors and different physiological systems. The diffe-
rence between bitterness and astringency can be most easily
experienced with red wine that has been stored on mash and
matured on barrique. One can feel that the bitter taste is only pre-
sent on the tongue, but astringency is felt in other mouth regions
as well.
Besides this “chemical-​physiological” perceptible astringency,
there is an important second physical aspect of astringency,
which causes the contractive feeling in the mouth: polyphenols
and tannins interacting directly with the proteins of the saliva
(Jakobek, 2015). The egg-​white-​like and viscoelastic consist-
ency of saliva has its origin in saliva proteins, which act as a
“thickening agent” because of their strong water binding. This
keeps the saliva on the tongue for a longer time and reduces the
friction between tongue and palate. Hence, saliva acts as a “lubri-
cant” in the mouth.
The physical part of astringency can be ascribed to a significant
change of the flow properties of saliva by friction (Vilgis, 2014).
If the saliva consisted of pure water, the lubricating effect would
Chocolate: Oral Processing of Chocolate 135

OH OH
O prote
in ma
HO in ch
ain
O NH O

gly
O

c
N

osi
Thr

dic
O H

s
ide
cha
ins
FIGURE 20.9 Highly simplified presentation of a section of mucin. The long, amphiphilic main chain consists of amino acid residues, and the side chains of
glycosidic groups, which are bound to the main chain via the amino acid threonine.

FIGURE 20.10 The proteins of saliva (mucins) interact with “astringent” phenolic compounds and form clusters, which can be verified with confocal laser
microscopy (CLSM) (left). The grey background shows the aqueous phase of the saliva. A schematic model (Vilgis, 2014) is shown on the right.
(CLSM, MPI for Polymer Research)

not be sufficient, as the thermal motion of the water molecules
would be too fast and the viscosity too low. Physiology maintains
pleasant friction mainly by the aid of two protein types, so-​called
mucins and proline-​rich proteins, which maintain increased vis-
cosity. Mucins are long glycoproteins (Round et al., 2002). They
consist of a long and stiff main chain made of amino acid residues,
with side chains consisting of saccharides (Figure 20.9).
The relatively long saccharidic side chains (which are also
branched) render mucins water-​ soluble. Water molecules are
strongly bound to the sugar chains, and the viscosity of water is
greatly increased. If mucins are equally distributed in the saliva,
they guarantee pleasant friction in the mouth.
The polar saccharide side chains can, however, interact and
form hydrogen bonds with phenols and tannins (Soares et al.,
2012; Gibbins and Carpenter, 2013). Thus, different mucin
chains bind to non-​volatile phenolic compounds. Clusters are
readily formed, which result in bigger, 10–​ 30-​micrometre-​
sized, molecular assemblies. These heavy micro-​ networks
are significantly slower in diffusion, as the diffusion constant
decreases with increasing (hydrodynamic) radius (Doi and
Edwards, 1988).
This can be directly verified in experiments. When tannins
or phenols are added to the saliva, cluster formation can be
shown by confocal laser microscopy, as shown in Figure 20.10
(Muthiramaglingam, 2014).
Because of the cluster formation of the proteins in saliva and
the resulting change in water binding, the friction in the mouth
changes significantly, resulting in the physical contribution of
136
FIGURE 20.11 Two examples (displayed one below the other) of proline-rich proteins of saliva in human saliva (from the protein data base www.uniprot.
org). The blue shaded amino acids are the lipophilic (hydrophobic) sections.
astringency in wine, tea, chocolate and other tannin- and phenol-
rich foodstuffs. Saliva loses its shear-stable lubrication proper-
ties through aggregation of the proteins, and the tongue becomes
“physically rough”. This effect plays a major role in the enjoy-
ment of chocolate but is significantly altered by the liquefied
cocoa butter.
To understand the oral processing during the enjoyment of
chocolate, other protein types besides mucins are important. In
particular, so-called proline-rich proteins (PRP) have primary
structures made of longer sequences of the highly hydrophobic
amino acid proline and other hydrophobic amino acids (Canon
et al., 2013). This results in protein chains consisting of hydro-
phobic sequences (Figure 20.11).
Figure 20.11 shows the primary structure of PRP protein of
human saliva. The exact details of the information are not of sig-
nificance for the following section, but it is important to know
that these types of proteins have long hydrophobic, and thus fat-
soluble, sections. Hence, they act as “amphiphilic emulsifiers”,
which organize themselves accordingly at fat–water interfaces
(Vilgis, 2010; 2015). The hydrophobic sections reach into the oil,
whereas the hydrophilic sections stick into the aqueous phase.
Thus, the proteins coil along the water–oil interface and decrease
the surface tension, and the emulsion therefore becomes more
stable.
Colloidal Precipitation in the Mouth
However, the slow fat crystal melting is not the only process that
dominates the chocolate sensation. The tongue-induced shear
causes a significant phase change of the liquid fat, the solid
cocoa particles and the saliva: the molten cocoa butter becomes
emulsified with the saliva and results in an oil-in-water emulsion
(Le Révérend et al., 2010; Carvalho-da-Silva et al., 2013).
This process then enables sensory perception. It is important to
Thomas A. Vilgis
understand the close link and successive processes during choc-
olate enjoyment: the entire time range from “melting”, through
the associated release of taste and odorant compounds, to wetting
of the tongue and oral cavity (oral coating), which is responsible
for the long-lasting pleasure when flavours are still released in the
mouth minutes after swallowing the emulsion. During these oral
processes, all receptor types are involved (Chen, 2014).
If the chocolate is taken between tongue and palate (and not
chewed quickly), the edges of the chocolate pieces start to melt
as soon as they have reached the temperature of 33.8 °C. The
melted cocoa butter forms droplets, minimizing the surface area
and surface tension. At the same time, all particles and molecules
involved in the melting phase are distributed. Sugar crystals
coated with surfactant molecules are released, as well as the cocoa
particles previously bound in the solid phase. Weakly hydrophilic
cocoa particles are considerably smaller in size than the droplets
formed by the melting cocoa butter. These can thus accumulate
on the fat–oil interface and enclose some oil droplets, and a solid
particle-based “Pickerin-Ramsden” emulsion is formed. On the
other hand, the saliva proteins, PRP and mucin, instantaneously
organize themselves between the liquid droplets formed by the
cocoa butter. The interface-active, glycolized saliva proteins can
also bind to the oil droplets. A simplified presentation is shown
in Figure 20.12.
The melting of milk chocolate (which contains additional
proteins that contribute to the emulsifying behaviour) in the
mouth results in complex emulsions with a highly heterogeneous
size distribution of the droplets (Carvalho-da-Silva et al., 2013).
Taste and odorant compounds are released during redistribution
processes and emulsion formation. The odorant compounds are
perceived via retronasal (and trigeminal) pathways, whereas
the non-volatile compounds are responsible for the bitter taste
and trigeminal sensations. Phenolic compounds and tannins of
the cocoa husk dissolve in saliva. Therefore, similar processes
that lead to the physical contributions of astringency in wine can
Chocolate: Oral Processing of Chocolate
FIGURE 20.12 Highly simplified and idealistic model of a liquid cocoa
butter droplet with enclosed sugar particles (as symbolically indicated). After
chewing, the droplets have a diameter of 150 micrometres (Carvalho-​da-​Silva
et al., 2013). After conching, the cocoa particles have dimensions well below
20 micrometres.
FIGURE 20.13 A simplified model of shear-​and phenolic-​induced cluster
formation in the mouth during enjoyment of chocolate. Inset: The droplets of
an emulsion can be seen in microscopic images. They form bigger aggregates,
for example, with phenolic compounds, as shown in the model. The clusters
reach sizes, depending on chewing movements (shear rate) and saliva flow,
between of 500 and 800 micrometres.
(Microscopic picture from Carvalho-​da-​Silva et al., 2013)
result in a strong aggregation of the oil droplets in chocolate. The
aggregates are held together by free phenolic compounds, by
hydrogen bond formation via the saliva proteins and by favour-
able van der Waals forces between the cocoa particles, as shown
in Figure 20.13.
137
FIGURE 20.14 Schematic and simplified presentation of the solution
of sugars during the enjoyment of chocolate. Left: only sugar very close
to the surface of the liquid fat droplets contributes to the sweetness. The
phospholipids are at first in an unfavourable position. Contact of the fatty
acids with saliva is energetically not possible. Middle: phospholipids reorient
themselves. The hydrophilic heads turn to saliva. Sugar at the surface is
exposed to the aqueous saliva film. Right: sugar crystals dissolve, and the
sweet stimulus can occur.
The exact form of the aggregates, their size and the kinetics of
aggregate formation in the mouth depend on several factors: type
of chocolate, strength of individual processing in the mouth by
shear movements of the tongue in the oral cavity, possible chewing
and saliva flow, and hence the concentration of released saliva
proteins. Saliva flow is a critical factor; it increases during choc-
olate enjoyment and, therefore, the concentration of saliva proteins
increases during the oral processing. An unconscious simultaneous
intensification of shear/​chewing movements changes the size and
structure of the aggregates and thus the lining of the oral cavity
(oral coating) with a phase rich in fat and thus overall flavour.
The structure and size of the aggregates are also responsible
for the friction in the mouth and partly for the “mouthfeel”.
The friction is strongly dependent on the speed of the tongue
movement and on the size and “irregularities” of the aggregates.
Large aggregates get entangled and obstruct each other strongly
at first. Only if they are torn apart at high shear forces does
the friction decrease in the medium shear rate range. Smaller
aggregates oppose a smaller resistance at low speeds, and thus
friction is minimal.
The behaviour of the sugar and the associated perception of
sweetness are also interesting in this context. Because sucose
crystals are surrounded by phospholipids, a major part of the
sugar remains in the oil droplets of the emulsion in the mouth.
Only sugar–​lecithin complexes found directly at the surface of
the oil droplets contribute to the sweet stimulus. In this case, the
fatty acid residues of the lecithin that are exposed to the aqueous
saliva surface will orient themselves in the direction of the cocoa
butter, thereby releasing the sugar crystals, which in turn will
dissolve in the saliva. The released sucrose molecules trigger the
sweet stimulus, and all others are swallowed with the oil droplets
without contributing to the taste (although they do contribute to
the nutritive energy input). The process is schematically depicted
in Figure 20.14.
An exact analysis of the sugar release from chocolate, its pos-
ition in the chocolate dispersion and the choice of emulsifiers
may allow sugar reduction in chocolate without noticeable losses
of taste.
Conclusions
A major part of the pleasant mouthfeel resulting from the consump-
tion of simple (bitter) chocolate is due to a cascade of complicated
138
physical processes that take place in the mouth. Aroma and flavour
release are closely related to the precise procedure of oral pro-
cessing. The highly simplified approaches described here (partly
hypothetical, but physically plausible) show the complexity of the
system and also approaches as to how the unfolding mouthfeel
can be controlled, at least to some extent. Besides the controlled
melting temperature of cocoa butter, particle size, the physical
properties of the emulsifiers and the concentration of non-​volatile
phenolic compounds contribute to the formation of clusters and
aggregates of the colloidal phase in the mouth.
REFERENCES
Afoakwa EO, Paterson A, Fowler M, Vieira J. 2009. Comparison of
rheological models for determining dark chocolate viscosity.
International Journal of Food Science & Technology, 44(1),
162–​167.
Albak F, Tekin AR. 2016. Variation of total aroma and polyphenol
content of dark chocolate during three phase of conching.
Journal of Food Science and Technology, 53, 848–​855.
Arnold G, Schuldt S, Schneider Y, Friedrichs J, Babick F, Werner C,
Rohm H. 2013. The impact of lecithin on rheology, sedimenta-
tion and particle interactions in oil-​based dispersions. Colloids
and Surfaces A: Physicochemical and Engineering Aspects,
418, 147–​156.
Beckett ST. 2002. The science of chocolate. RSC Publication. The
Royal Society of Chemistry, Cambridge, UK.
Canon F, Paté F, Cheynier V, Sarni-​Marchado P, Giuliani A, Pérez J,
Durand D, Li J, Cabane B. 2013. Aggregation of the salivary
proline-​rich protein IB5 in the presence of the tannin EgCG.
Langmuir, 29, 1926–​1937.
Carvalho-​da-​Silva AM, Van Damme I, Taylor W, Hort J, Wolf B.
2013. Oral processing of two milk chocolate samples. Food
and Function, 4, 461–​469.
Chen J. 2014. Food oral processing: Some important underpinning
principles of eating and sensory perception. Food Structure,
1, 91–​105.
Doi M, Edwards SF. 1988. The theory of polymer dynamics. Oxford
University Press.
Gibbins HL, Carpenter GH. 2013. Alternative mechanisms of astrin-
gency –​what is the role of saliva? Journal of Texture Studies,
44, 364–​375.
Hannay N (Ed.). 2012. Defects in solids. Springer Science &
Business Media, Berlin, Germany.
Hartel RW. 2013. Advances in food crystallization. Annual Review of
Food Science and Technology, 4, 277–​292.
Hoskin JC, Dimick PS. 1994. Chemistry of flavour development
in chocolate. In Industrial chocolate manufacture and use,
Springer Verlag, Boston, MA, 102–​116.
Jakobek L. 2015. Interactions of polyphenols with carbohydrates,
lipids and proteins. Food Chemistry, 175, 556–​567.
Thomas A. Vilgis
Le Berre E, Boucon C, Knoop M, Dijksterhuis G. 2013. Reducing
bitter taste through perceptual constancy created by an expect-
ation. Food Quality and Preference, 28, 370–​374.
Le Révérend BJ, Norton IT, Cox PW, Spyropoulos F. 2010. Colloidal
aspects of eating. Current Opinion in Colloid & Interface
Science, 15, 84–​89.
Middendorf D, Bindrich U, Mischnick P, Juadjur A, Franke K,
Heinz V. 2016. Atomic force microscopy study on the effect of
different lecithins in cocoa-​butter based suspensions. Colloids
and Surfaces A: Physicochemical and Engineering Aspects,
499, 60–​68.
Muthiramaglingam S. 2014. Thesis, unpublished, Max-​ Planck-​
Institut für Polymerforschung.
Parsons LB, Goodall R. 2011. Testing the fracture behaviour of choc-
olate. Physics Education, 46(1), 50.
Rigolle A, Goderis B, van den Abeele K, Foubert I. 2016. Isothermal
crystallization behavior of cocoa butter at 17 and 20 °C with
and without limonene, Journal of Agricultural and Food
Chemistry, 64, 3405−3416.
Round AN, Berry M, McMaster TJ, Stoll S, Gowers D, Corfield
AP, Miles MJ. 2002. Heterogeneity and persistence length in
human ocular mucins. Biophysical Journal, 83, 1661–​1670.
Saska M, Myerson AS. 1983. The theoretical shape of sucrose
crystals from energy calculations. Journal of Crystal Growth,
61, 546–​555.
Schöbel N, Radtke D, Kyereme J, Wollmann N, Cichy A, Obst
K, Kallweit K, Kletke O, Minovi A, Dazert S, Wetzel CH,
Vogt-​Eisele A, Gisselmann G, Ley JP, Bartoshuk LM, Spehr
J, Hofmann T, Hatt H. 2014. Astringency is a trigeminal
sensation that involves the activation of G protein–​coupled
signaling by phenolic compounds. Chemical Senses, 39,
471–487.
Soares S, Mateus N, de Freitas V. 2012. Interaction of different
classes of salivary proteins with food tannins. Food Research
International, 49, 807–​813.
Stortz TA, De Moura DC, Laredo T, Marangoni AG. 2014. Molecular
interactions of ethylcellulose with sucrose particles. RSC
Advances, 4, 55048–​55061.
Vilgis T. 2010. Das Molekül-​Menü –​molekulares Wissen für kreative
Köche, S. Hirzel, Stuttgart.
Vilgis T. 2014. Spucke aus dem Gleichgewicht. Physik in unserer
Zeit, 45, 50–​50.
Vilgis T. 2015. Soft matter food physics –​the physics of food and
cooking, Reports on Progress in Physics, 78, 124602.
Zhao H, Li B, James BJ. 2018. Structure-​fracture relationships in
chocolate systems. LWT, 96, 281–​287.
Ziegleder G, Danzl W. 2016a. Kakaobutter –​Sie lässt Schokolade
perfekt kristallisieren und genussvoll schmelzen, Journal
Culinaire, 23, 86–​91.
Ziegleder G, Danzl W. 2016b. Das Conchieren-​ Die Enstehung
des feinen Schokoladengeschmack, Journal Culinaire, 23,
104–​109.
Coffee Preparation –​from Roasted Beans to Beverage

Laura Febvay1 and Hervé This vo Kientza2,3

1 Aerial, 250 Rue Laurent Fries, Parc d’innovation, 67412 Illkirch, France
2 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
3 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France

The drink commonly named “coffee” is produced after the
roasting of beans of Coffea (see the chapter on roasting). The
roasted beans are ground, and the powder obtained is extracted
with hot water in order to make the beverage. This chapter
describes the grinding and extraction processes; the resulting
beverage is then characterized chemically.
Grinding
In coffee industry jargon, grinding refers to the process of
breaking down the coffee beans into a powder (Illy and Viani,
2005). This is carried out by a device called a coffee grinder, and
the product obtained is called ground coffee. The main objective
of grinding is to increase the exchange surface of coffee material
with water, facilitating the transfer of soluble substances present
in the coffee bean to the aqueous “extract”, which will become
the so-​called “coffee” beverage (Andueza et al., 2003).
Secondly, it has been shown that the finer the average particle
size, the faster the rate of extraction by water of soluble solids
and volatile components of the coffee (Clarke et al., 1987). Thus,
small and irregular particles can release their soluble substances
too fast and lead to a beverage that is often considered “too
concentrated”. Also, these particles can pass through the pores
and end up in the cup (Folmer, 2016).
On the other hand, Puhlmann and Habel (1989) showed that
coarsely ground coffee results in the appearance of large channels
between particles. During the extraction, water flows in these
channels at high speed (in particular with percolation methods);
the bigger the particles, the easier and the faster the water flow
will be (Folmer, 2016). For some brewing techniques, the contact
time between the water and the ground coffee may be too short
for the compounds to be appropriately extracted with coarsely
ground coffee.

10
Influence of Particle Size
The average particle size distribution (usually abbreviated to
PSD) of the powder influences the efficiency of the extraction: 8

the smaller the particle, the bigger the specific surface, and the
better the extractability of the compounds that give to coffee 6
Volume (%)

the properties for which it is appreciated (von Blittersdorff

and Klatt, 2017). An example of a particle size distribution
4
obtained by laser diffraction is given in Figure 21.1 (Primavera
et al., 2014).
However, there can be two drawbacks if the particles are too 2
small. Firstly, this can increase the percolation time or even block
percolators or filters. Indeed, the size of the particles will also
0
determine how easily water will flow through the coffee powder 1 10 100 1000
during extraction, and very fine particles (usually below 100 μm) Size/Microns
will block the pathways for water and impede the separation of
Expresso French Roast
the spent ground coffee from the extract by clogging the filter
pores. In this case, the contact time between the water and the
ground coffee will be increased, resulting in an over-​extraction FIGURE 21.1 Particle size distribution for two different commercially
available coffee blends.
of some compounds, imparting bitterness and astringency in par-
ticular (Baggenstoss et al., 2008a; Folmer, 2016). (Primavera et al., 2014)

139
140
All this explains why the grinding process aims to create
an appropriate PSD, balancing the water flow, neither too fast
nor too slow, for which the particles should not be too large or
too small, and diffusion, which increases with the decreasing
size of the particles. Each coffee preparation method requires
a particular PSD (Clarke, 1987; Puhlmann and Habel, 1989;
Baggenstoss et al., 2008b; Folmer, 2016; von Blittersdorff and
Klatt, 2017). The “appropriate” grind size can be obtained by
operators and baristas in part by choosing the appropriate mill
for the intended brewing method. The different main grinding
methods are described in the following section.
Different Grinding Methods
There are two main types of grinding techniques, which differ
according to their application: (1) “impact grinding”, designed
for on-​demand grinding, and (2) “gap grinding”, designed to
work 24 hours a day, 7 days a week, for industrial application
(von Blittersdorff and Klatt, 2017).
Impact grinding is carried out by rotating blades colliding
at high speed (between 1000 and 9000 rpm) with the particles
encountered in their path. Since the ground coffee remains
between the blades even after being ground and therefore under-
goes many impacts, this technique does not easily control the par-
ticle size, and the size distribution of the powder is dependent
on the duration of the operation. That is why it is only used for
small-​scale grinders, for commercial use at the point of sale, or in
grinders for private use (Illy and Viani, 2005).
Gap grinding is based on the passage of the beans through
a space between mobile tools, called “cutting tools” (von
Blittersdorff and Klatt, 2017). The geometry of the blades results
in a gradual reduction in the width of the space during rotation,
forcing the particles to come into contact with the two blades.
Depending on the shape of the cutting surfaces, a compression or
a shear strain is applied. There are usually a number of grinding
FIGURE 21.2 Schematic drawings of a type of gap grinding: a pair of rolls
in a roller grinder.
(Folmer, 2016)
Laura Febvay, Hervé This vo Kientza
stages; Figure 21.2 shows one pair of rolls but there are generally
many in succession in a roller grinder (between five and ten).
With this technique, the average particle size is more homoge-
neous and easier to control than with impact grinding. Moreover,
the process can be done continuously, hence lending itself to
industrial application (Illy and Viani, 2005; Baggenstoss et al.,
2008a).
Beside the type of grinder used, other parameters influence the
grinding process; the main ones are described in the following
section.
Parameters Influencing the Grinding Process
One of the important variables influencing the grinding process is
the quality of the coffee beans, as this raw material can come from
different botanical varieties and different producing countries,
where it undergoes different drying and processing methods. As
a result, the miller is confronted with beans with a variable chem-
ical composition, and this results in differences of resistance to
grinding. In addition, as coffee is an agricultural product subject
to the variability of climate and natural evolution, the hardness
and water content of the beans can differ between batches. It has
been shown that coarser particles are obtained from coffee beans
with high moisture content upon grinding (Baggenstoss et al.,
2008b).
During the roasting process, coffee beans change their tex-
tural properties, losing strength and toughness, and becoming
progressively more brittle, due to chemical, physical and struc-
tural modifications (see chapter on roasting). Pittia et al. (2001)
showed that the typical brittleness of roasted coffee beans seems
to be related both to the decrease of density and to the water loss.
Indeed, the density of green coffee beans is between 550 and
700 g/​L, while the density of roasted beans is reduced to 300–​
450 g/​L (Schenker et al., 2000). According to grinding tests with
beans with different water contents (Puhlmann and Habel, 1989;
Andueza et al., 2003), the proportion of fine particles is larger
when the water content is lower. So, beans become more and
more brittle during the roasting process, and depending on the
roasting conditions (temperature, time and roasting speed), the
coffee will be more or less hard to grind. In conclusion, grinding
natural products such as coffee beans generates many particles
of different sizes and shapes. The PSD can be characterized after
grinding by different methods; techniques such as sieving, image
analysis and laser diffractometry can be used to measure the
fineness of grains (von Blittersdorff and Klatt, 2017).
Extraction and Characteristics of the Resulting
Beverage
The goal of extraction is to bring into contact the solid particles
and the solvent (water) with a view to performing a mass transfer
of soluble compounds into the solvent. Then the resulting solution
Coffee Preparation
is separated from the residual solid, often by filtration. The mech-
anism of extraction is favored by an increased specific surface
(surface per unit volume of solids), and thus a decreased radial
distance that must be traversed within the solids; this can be con-
trolled by the grind size, as discussed in the grinding section.
Numerous coffee extraction methods exist. The so-​ called
brewing methods are generally characterized by the extraction
pressure of the water, the extraction process and tool, and the
volume of the extract obtained. The pressure of water used for
coffee extraction, except for coffees made by infusion (because
the soluble coffee particles will only diffuse freely in the solu-
tion), is not an independent variable; it is the result of the equilib-
rium between the applied force on top of the coffee bed (through
the water) and the resistance of the coffee bed against water per-
colation. Each extraction method has its own driving force and
its typical coffee bed properties. As a result, each technique is
characterized by its own range of extraction pressures.
The most common methods are: expresso (or espresso), per-
colation, filtered, moka, French press, and boiled coffee (Illy and
Viani, 2005):
• Expresso is a concentrated beverage of 20 to 40 mL,
brewed by forcing hot water (temperature 90–​100 °C)
at high pressure (from 9 up to 19 bars) through finely
ground coffee, with a contact time in the order of
seconds. It is generally prepared immediately before
consumption (Farah, 2012; Gloess et al., 2013; Folmer,
2016). To prepare an expresso, the external force is
delivered by a pump, allowing higher pressures to build
up than with other methods.
• Percolation is an infusion method, which consists of
distributing the ground coffee evenly in a filter placed
on a support; hot water is then poured over the coffee
in a circular motion towards the center of the filter
(Farah, 2012).
• For boiled coffee (also called Greek coffee or Turkish
coffee), the water is poured over finely ground coffee
or sprayed into a saucepan and heated; when the water
starts to boil, the mixture is poured, unfiltered, directly
into the cup, and only the upper part of the product is
consumed (Farah, 2012).
• For the Italian press, or moka pot, water is placed at
the base of the kettle, which contains a pressure valve.
When the kettle is heated, the water flow through the
ground coffee is continuously pressurized to the top
compartment (Farah et al., 2012). For moka prepar-
ation, lower pressures than for expresso are generated
by a steam/​vapor pressurized chamber.
• The French press consists of mixing coarsely ground
coffee and hot water in a special machine equipped
with a mesh piston. After a few minutes of infusion,
the piston is pressed to trap the coffee grounds at the
bottom of the cup, and the upper infusion is poured into
the cup (Farah, 2012).
141
• Filter coffee is made by slowly pouring hot water (tem-
perature 90–​100 °C) through crushed coffee beans in a
filter (often paper), the contact time between water and
coffee being approximately 3 min (Folmer, 2016). In
a filter brewer, almost no pressure is built up, as water
flows through the coffee bed by gravity.
The type of extraction method used depends on geographic,
cultural and social context, as well as on personal preferences
(Petracco, 2008).
The following paragraphs describe the general chemical com-
position of the resulting beverage and then the impact of the
coffee preparation method (the brewing method, pressure, tem-
perature and coffee/​water ratio used) on the composition of this
resulting coffee extract. The final section is devoted to the impact
of this chemical composition on the organoleptic properties.
Chemical Composition of the Extract
Different factors affect the brew’s composition, including not
only the ground roast coffee composition but also the brewing
method, the proportion of coffee to water, the temperature and
composition (hardness) of water, the time of contact between
coffee and water, and the filter material. Here we present only the
compounds that are generally present in a coffee extract.
The brewed coffee is mainly composed of water, non-​volatile
acids, soluble saccharides, protein, caffeine and melanoidins
(Barter, 2004). The amount of soluble solids in the brewed coffee
is generally between 2 and 6 g/​100 mL (Farah, 2012; Chu, 2012).
A more detailed description of the different main compounds
is as follows:
Saccharides: polysaccharides comprise up to 15% of the total
solids of the coffee brew (Díaz-​Rubio and Saura-​Calixto, 2007).
Galactomannan [1]‌and arabinogalactan type II [2] are the pre-
dominant polysaccharides of coffee brews (Nunes and Coimbra,
2001; Gniechwitz et al., 2008; Bekedam et al., 2008). Single-​
dose coffee capsules have been shown to contain galactomannans
as the predominant polysaccharides over arabinogalactan (Lopes
et al., 2016).
142 Laura Febvay, Hervé This vo Kientza

According to Petracco et al. (2008), a typical amount of sol- D-​Mannose [3]‌, followed by D-​galactose [4], are the main
uble fibers in expresso coffee is 800 mg/​100 mL, and a regular residues of polysaccharides from single-​dose expresso coffee.
percolation method produces approximately 200 mg/​100 mL of L-​Arabinose [5], D-​glucose [6] and L-​rhamnose [7] are present in
these. Similarly, Díaz-​Rubio and Saura-​Calixto reported 470–​ lower amounts (Lopes et al., 2016).
750 mg soluble fiber in 100 mL brewed coffees (Díaz-​Rubio and
Saura-​Calixto, 2007).

           

     
Coffee Preparation
FIGURE 21.3 Molecular structure of the main chlorogenic acids present in green beans.
(Fujioka and Shibamoto, 2008)
Organic acids: approximately 80–​ 100% of chlorogenic
acids (Figure 21.3) from the beans are extracted in home coffee
brewing, resulting in between 35 and 175 mg chlorogenic acids/​
100 mL cup of coffee (Clifford, 1997).
In addition to free chlorogenic acids and those incorporated
by melanoidins, Díaz-​Rubio and Saura-​Calixto (2007) reported
that 8.7–​
10.5 mg chlorogenic acids and their derivatives are
associated with soluble fiber in 100 mL brewed coffee.
The acidity of coffee is also due to non-​aromatic organic acids
such as acetic [8]‌, formic [9], malic [10], citric [11] and lactic
[12] acids, as well as chlorogenic and quinic acids [13]. The pH
143
varies from approximately 5.2 in a brew made from light roast to
5.8 in a brew made from dark roast (Kurt and Speer, 1999).
Nitrogenous compounds: caffeine [14], trigonelline [15]
and nicotinic acid [16] are also compounds from the roasted
beans that are soluble in hot water. Typical amounts in brewed
coffee prepared with medium roasted coffee are 50–​100 mg for
caffeine, 40–​50 mg for trigonelline and about 10 mg nicotinic
acid. Trigonelline tends to be completely degraded in dark roasts,
whereas nicotinic acid content is formed during roasting (Perrone
et al., 2008).
144 Laura Febvay, Hervé This vo Kientza

H2C R1
Rx= HO C C C CH3
H2 H2 H2
HC R2 n
O
H2 H2
H2C R3 Ry= HO C C C C C C C CH3
H2 H2 H2 H H
n
n
R1, R2, R3 = Rx or Ry

FIGURE 21.4 Triglycerides.

(From Barison et al., 2010)

               

Lipids: the lipid fraction of coffee brew is mainly composed
of triglycerides (Figure 21.4) and bioactive diterpenes.
Triacylglycerols account for approximately 75% (w/​w) of total
coffee lipids in freshly brewed coffee, whereas free fatty acids
account for only approximately 1% (Trugo, 2003).
The major diterpenes present in coffee are cafestol [17],
kahweol [18] (Ratnayake et al., 1993; Urgert et al., 1995;
Gross et al., 1997) and sterols. The major category of sterol,
accounting for approximately 93% of total sterols in coffee,
is 4-​desmethylsterols (Farah, 2012; Speer, 2006), including
compounds such as campesterol [19], stigmasterol [20] and
sitosterol [21] (Itoh et al., 1973; Farah, 2012; Ratnayake et al.,
1993; Chu, 2012).

    

         
Coffee Preparation 145

Proteins: the protein content of coffee ranges from 16.93 to
29.70 mg per cup of expresso coffee (Lopes et al., 2016).
Melanoidins: melanoidins are water-​soluble, high-​molecular-​
weight polymers (Ledl and Schleicher, 1990; Nunes et al.,
2012). Gniechwitz et al. (2008) isolated different fractions of
coffee, including those containing melanoidins, and estimated
their molecular weights to be between 3 and 22 kDa. The com-
position and structure of melanoidins from food sources are still
being discussed. Melanoidin molecules have been shown to be
made up of many different residues, including sugars, proteins
and phenolic compounds (Nunes and Coimbra, 2001; Moreira
et al., 2015). The high structural diversity of melanoidins makes
their analysis and quantification difficult. Since melanoidins
are composed of very diverse material, their level is usually
determined by calculating the difference between the amount
of high-​molecular-​weight material (HMWM) and the amount of
polysaccharides and protein. The melanoidin content of regular
expresso coffees analyzed ranged from 69.8 to 145.6 mg per cup,
calculated as the difference in the mass between the total HMWM
in the fraction and the mass of protein and polysaccharides
(Lopes et al., 2016).
Minerals: coffee beans contain about 4% minerals, of which
40% is potassium (K). The other metals found in coffee are
sodium (Na), calcium (Ca), magnesium (Mg), iron (Fe) and
manganese (Mn). These elements are mainly present in the
form of cations and can therefore be extracted in coffee brew
(Antonio et al., 2011; Valentin and Watling, 2013; Stelmach
et al., 2015).
Impact of the Coffee Preparation Method
Impact of the Brewing Method
The impact of the brewing method on the chemical compos-
ition, in particular on the quantity of caffeine [14], chlorogenic
acids and non-​aromatic acids, has been the subject of numerous
studies (Peters, 1991; López-​Galilea et al., 2007; Fujioka and
Shibamoto, 2008; Pérez-​Martínez et al., 2010; Gloess et al.,
2013). Figure 21.5 shows the distribution of fatty acid content,
caffeine and chlorogenic acids according to different extraction
methods obtained by Gloess et al. (2013).
According to these studies, it appears that the levels of caffeine
[14] and chlorogenic acids vary with the brewing method. In
general, the concentration of the extracted compounds in the
brew was highest for expresso, followed by the moka extraction
(Gloess et al., 2013; López-​Galilea et al., 2007; Peters, 1991).
Indeed, the highest concentration of caffeine was measured in
expressos (21.0 ± 0.4 mg/​10 mL), and the lowest was for filter
coffee (4.7 ± 0.1 mg/​10 mL) (Gloess et al., 2013). Lower or
higher values can be found depending on the type of coffee used:
different variety or origin, for example (Peters, 1991, López-​
Galilea et al., 2007). In the case of the chlorogenic acids, again
expressos had the highest concentrations (Balakrishnan et al.,
1961; Gloess et al., 2013).
Moreover, when compared with other common coffee beverages,
the acrylamide [22] concentration was higher in expressos (Alves
et al., 2010b). However, comparing the different brewing methods

FIGURE 21.5 Content per 10 mL extract of coffee prepared according to the different extraction methods.
DE stands for Espresso from semi-​automatic machine, SE for Espresso from fully automatic machine, NE for Espresso—​Nespresso, Bia for Espresso—​
Bialetti, DL for Lungo from semi-​automatic machine, SL for Lungo from fully automatic machine, Bo for Lungo—​French Press, KK for Lungo—​Karlsbader
Kanne and F for Lungo—​Filter Coffee of (A) fatty acids in weight percent (fatty acids/​w. %), (B) caffeine in mg, (C) 3-​O-​caffeoyl quinic acid in mg (3-​CQA/​
mg) and (D) 5-​O-​caffeoyl quinic acid in mg (5-​CQA/​mg). The error bars correspond to the standard deviation of the mean (95%).
(Gloess et al., 2013)
146
showed that expresso contained more isoflavones [23] (∼170 μg/​
30 mL) than a cup of press-​pot coffee (∼130 μg/​60 mL), but
less than a moka coffee (∼360 μg/​60 mL), and amounts similar
to those of a filtered coffee cup (∼180 μg/​120 mL) (Alves et al.,
2010a). Costa et al. (2010) observed that the total mineral extrac-
tion achieved by expresso machines and electric coffee makers
was higher than that for all other percolation methods.
Usually, the extraction of water-​soluble components, including
chlorogenic acids, caffeine [14], nicotinic acid [16], soluble
melanoidins and hydrophilic volatile compounds, is greater at
higher temperatures and pressures (Trugo and Macrae, 1984).
More specifically, the effectiveness of extraction process scales
up with percolation temperature (Albanese et al., 2009).
    
The brewing method also impacts the lipid content, even though
the content is very low (below 0.2%, w in all cases (Ratnayake
et al., 1993, Gloess et al., 2013)), contrary to what is sometimes
published in popular literature about coffee.
Although the lipid content can vary depending on the brewing
method, the method of preparation of the brew and filtration
had no important influence on the lipid composition (Ratnayake
et al., 1993).
During paper filtration, lipids remain mainly in spent coffee
grounds, and the brew and filter paper retain only 0.4% and 9.4%,
respectively, of the total lipids recovered (Ratnayake et al., 1993).
Oil droplets are likely to be retained in filters made of paper
or similar types of materials. The high pressure used to make
expresso and the absence of a filter made of paper or another
lipophilic material to retain the lipids facilitates their extraction
into the brew. Thus, unfiltered coffee (such as Turkish coffee) and
expresso coffee brews contain higher amounts of lipids, including
diterpenes, compared with regular percolation methods. Coffee
prepared by boiling without filtering and expresso coffee reached
60–​160 mg lipids/​150 mL cup (Ratnayake et al., 1993), while
coffee brews filtered through filter paper contained less than
7 mg lipid in a similar volume. Unfiltered coffee also contained
Laura Febvay, Hervé This vo Kientza
the highest amounts of diterpenes, equivalent to 7.2 and 5.3 mg
cafestol [17] per cup and 7.2 and 5.4 mg kahweol [18] per cup,
respectively. In contrast, instant and drip-​filtered coffee brews
contained negligible amounts of these diterpenes, and expresso
coffee contained intermediate amounts, about 1 mg cafestol and
1 mg kahweol per cup (Gross et al., 1997).
The brewing method also has an impact on the antioxidant
capacity of coffee (López-​Galilea et al., 2007; Ludwig et al.,
2012). The main antioxidants in coffee are chlorogenic acids,
melanoidins, caffeine [14], phenolic acids and tocopherols [24],
among several others (Ludwig et al., 2012; Alves et al., 2010a);
the amounts of these compounds in the coffee brew are dependent
on the brewing method.
In addition, the acidity can vary with the brewing method
(Peters, 1991; Verardo et al., 2002; López-​Galilea et al., 2007).
The pH of regular and decaffeinated coffees ranged from 4.95 to
5.99 and from 5.14 to 5.80, respectively (Fujioka and Shibamoto,
2008). Expresso is the most acidic brew (pH 5.51) and French
press coffee is the least acidic brew (pH 5.92) (Gloess et al., 2013).
The extraction method therefore has a significant impact on
the resulting chemical composition of the coffee beverage. Other
factors play a role, such as pressure, temperature used and water
quantity.
Impact of Coffee/​Water Ratio
The impact of the coffee/​water ratio on the extraction of the
constituents of coffee has been the subject of numerous studies
(Blumberg et al., 2010; Alves et al., 2010a; Gloess et al., 2013),
which have shown that some of the water-​extractable components,
such as 4-​ O-​caffeoyl quinic acid and 3-​ O-​ γ-​quinide
caffeoyl-​
(Figure 21.3), are washed out in the first few seconds of the
extraction process under high pressure, while other compounds,
such as 4,5-​O-​dicaffeoyl-​muco-​γ-​
quinide or tocopherol, are
released rather slowly, and less water-​soluble compounds, like
chlorogenic acids or isoflavones [23], show strong retention by
the ground coffee material. This type of compound can therefore
be found in larger or smaller quantities in extracts depending on
the quantity of water used.
The extraction of some water-​ soluble compounds, like
caffeine, tocopherols and chlorogenic acids, increases when the
coffee/​water ratio is higher (Andueza et al., 2003; Alves et al.,
2010b). As an example, the extraction efficacy of acrylamide [22]
for standard expressos of 30 mL was nearly 80%, being affected
only by brew volume, with long expressos (70 mL) containing
practically all the acrylamide of the coffee cake (99%), almost
double the content of short expressos (20 mL) (Alves et al.,
2010b). The content of β-​carboline [24] in expresso coffee is
dependent primarily on the coffee species, followed by brew
length (Alves et al., 2007).
Coffee Preparation 147

Furthermore, factors such as the method of brew preparation, It has been shown that there is only a moderate correlation
the degree of roast, brew volume and temperature of extraction between pH and acidity perception (Balzer, 2008). Furthermore,
affect the saccharide composition of expresso coffee. Coffees some of these acids, such as chlorogenic, together with caffeine
with low water content were extracted more effectively than and other compounds, can also contribute to bitterness, modi-
high-​moisture coffees, and percolation was slower (Baggenstoss fying the typical bitterness–​acidity balance of espresso coffees
et al., 2008a). (Illy and Viani, 2005).
The fraction of HMWM recovered from coffee is responsible
for the characteristic color of the brews and also has an impact
Organoleptic Properties of the Resulting Beverage on mouthfeel, flavor and aroma due to the interaction with other
compounds. It is composed of saccharides (52%), proteins (10%),
The flavor of coffee is based primarily on odorant and taste
polyphenols (5%) and also a large number of brown compounds,
compounds, but some trigeminal and proprioceptive effects also
the melanoidins (Coelho et al., 2014).
occur. Over 1000 volatile compounds have been identified in
Blending is used to optimize odor, mouthfeel and flavor; it
roasted coffee, but not all of these compounds have been found
allows production of a coffee that provides higher cup quality
to be relevant in terms of the odor (Akiyama et al., 2005). Only
than can be obtained with any of the ingredients used indi-
5% of these compounds (~50) may be responsible for the odor
vidually and helps to maintain consistency in the final roasted
of coffee. These compounds include pyrazines [25], furans,
product.
aldehydes, ketones, phenols and sulfur compounds, among others
There is still a poor understanding of the relationships between
(Toci and Boldrin, 2018).
the odorant compounds and the various factors that affect the
odorant fraction, despite the considerable number of articles
that have been published on this subject. The main difficul-
ties are deficiency of adequate analytical quantification, lack of
comparisons involving the various parameters that influence the
odor (Toledo et al., 2017), and a lack of uniformity in the data
presented in different studies, which hinders satisfactory com-
A series of studies focused on the identification of the bitter-​ parison. For example, filtered and expresso coffee beverages
tasting molecules in coffee (Chen, 1979; Andueza et al., 2003; have been most extensively studied, mainly because these types
Charalambous et al., 2012). They suggested that the alkaloids of drinks are more widely consumed in many countries. The con-
caffeine [14] and trigonelline [15], as well as thermally generated sumption of moka, French, and Greek or Turkish coffee drinks
compounds such as furfuryl alcohol [26], 5-​hydroxymethyl-​2-​ is more geographically restricted, so these types have been less
furanaldehyde [27] and pyrazines [25] (Clarke and Vitzthum, studied. The different beverages are produced using different
2017), contribute to the bitter taste of coffee. Caffeine [14] and aqueous extraction systems that extract compounds with medium
chlorogenic acid are compounds related not only to bitterness but and high polarity, and coffee powder is rich in many classes of
also to astringency. compounds with such characteristics (Toci and Boldrin, 2018).

Conclusions
Many studies focus on the question of making a “good” coffee,
and they try to identify extraction parameters that can be applied
to make the “best” possible coffee. However, it should be
observed that this question is not a good one, because since the
      goal is imprecisely defined, the way to reach it cannot be found
with certainty. The definition of “good coffee” depends on culture
Other studies revealed that, during coffee roasting, the major and individual taste, so it is useless to try to find “the best pro-
polyphenols in raw coffee are thermally transformed into the cess”. One needs a clear definition of the properties of the brew
bitter-​tasting caffeoyl quinic acid lactones, such as 5-​O-​caffeoyl-​ in order to select the right parameters.
muco-​γ-​quinide, 3-​O-​caffeoyl-​γ-​quinide, 4-​O-​caffeoyl-​muco-​ The real question is: if you want a profile of soluble and
γ-​quinide, 5-​O-​caffeoyl-​epi-​δ-​quinide, 4-​O-​caffeoyl-​γ-​quinide, odorous compounds, what are the methods of grinding and pre-
3,4-​O-​dicaffeoyl-​γ-​quinide, 4,5-​O-​dicaffeoyl-​muco-​γ-​quinide and paring the drink to be chosen?
3,5-​O-​dicaffeoyl-​epi-​δ-​quinide (Chen, 1979; Blank et al., 1992;
Deibler and Delwiche, 2003; Adriana et al., 2005; Frank et al., REFERENCES
2006; Charalambous et al., 2012; Clarke and Vitzthum, 2017). Albanese D, Di Matteo M, Poiana M, Spagnamusso S. 2009.
The organoleptic properties of expresso coffee, particularly its Espresso coffee (EC) by POD: study of thermal profile during
body, are affected by saccharide content and composition, since extraction process and influence of water temperature on
saccharides act as viscosity enhancers, guaranteeing foam sta- chemical–​ physical and sensorial properties. Food Research
bility (Nunes et al., 1998). International, 42(5), 727–​32.
148
Adriana F, de Paulis T, Trugo LC, Martin PR. 2005. Effect of roasting
on the formation of chlorogenic acid lactones in coffee. Journal
of Agricultural and Food Chemistry, 53(5), 1505–​13.
Akiyama M, Murakami K, Ikeda M. 2005. Characterization of
flavor compounds released during grinding of roasted robusta
coffee bean. Food Science and Technology Research, 11(3),
298–​307.
Alves RC, Almeida IMC, Casal S, Oliveira MBPP. 2010a. Isoflavones
in coffee: influence of species, roast degree, and brewing
method. Journal of Agricultural and Food Chemistry, 58(5),
3002–​7.
Alves RC, Casal S, Oliveira BPP. 2007. Factors influencing the
Norharman and Harman contents in espresso coffee. Journal
of Agricultural and Food Chemistry, 55(5), 1832–​8.
Alves RC, Soares C, Casal S, Fernandes JO, Oliveira MBPP. 2010b.
Acrylamide in espresso coffee: influence of species, roast
degree and brew length. Food Chemistry,119(3), 929–​34.
Andueza S, Paz de Peña M, Cid C. 2003. Chemical and sensorial
characteristics of espresso coffee as affected by grinding and
torrefacto roast. Journal of Agricultural and Food Chemistry,
51(24), 7034–​9.
Antonio AG, Iorio NLP, Pierro VSS, Candreva MS, Farah A, Dos
Santos KRN, Maia LC. 2011. Inhibitory properties of Coffea
canephora extract against oral bacteria and its effect on demin-
eralisation of deciduous teeth. Archives of Oral Biology, 56(6),
556–​64.
Baggenstoss J, Perren R, Escher F. 2008b. Water content of roasted
coffee: impact on grinding behaviour, extraction, and aroma
retention. European Food Research and Technology, 227(5),
1357–​65.
Baggenstoss J, Poisson L, Kaegi R, Perren R, Escher F. 2008a.
Coffee roasting and aroma formation: application of different
time−temperature conditions. Journal of Agricultural and
Food Chemistry, 56(14), 5836–​46.
Balakrishnan N, Viraktamath R, Balachandran A, Natarajan CP,
Bhatia DS. 1961. Studies on extraction of coffee by pressure
extractors. Food Science, 10, 86–​88.
Balzer HH. 2008. Chemistry I: non-​volatile compounds. In Coffee
(Clarke RJ, Vitzthum OG eds.), 18–​32. John Wiley & Sons,
Hoboken, New Jersey.
Barison A, Pereira-​da-​silva CW, Campos FR, Simonelli F, Lenz CA.
2010. A simple methodology for the determination of fatty
acid composition in edible oils through 1H NMR spectroscopy.
Magnetic Resonance in Chemistry, 48(8), 642–​650.
Barter R. 2004. A short introduction to the theory and practice of pro-
file roasting. Tea and Coffee Trade Journal, 68, 34–​7.
Bekedam EK, Schols HA, Van Boekel MAJS, Smit G. 2008.
Incorporation of chlorogenic acids in coffee brew
melanoidins. Journal of Agricultural and Food Chemistry,
56(6), 2055–​63.
Blank I, Sen A, Grosch W. 1992. Potent odourants of the roasted
powder and brew of Arabica coffee. Zeitschrift fur Lebensmittel-​
Untersuchung und -​Forschung, 195, 239–​45.
Blumberg S, Frank O, Hofmann T. 2010. Quantitative studies on the
influence of the bean roasting parameters and hot water per-
colation on the concentrations of bitter compounds in coffee
brew. Journal of Agricultural and Food Chemistry, 58(6),
3720–​28.
Charalambous G. 2012. The Quality of Foods and Beverages
V2: Chemistry and Technology. Elsevier, Amsterdam, The
Netherlands.
Chen WC. 1979. Studies on the bitter taste of roasted coffee.
Relationship between structure and bitter taste of some organic
compounds (in German). PhD Thesis, University of Munich,
Munich, Germany.
Laura Febvay, Hervé This vo Kientza
Chu YF. 2012. Coffee: Emerging Health Effects and Disease
Prevention. John Wiley & Sons.
Clarke R, Vitzthum OG. 2017. Coffee: Recent Developments,
John Wiley & Son, Hoboken, New Jersey. www.wiley.com/​
WileyCDA/​WileyTitle/​productCd-​0632055537.html.
Clarke RJ (ed.). 1987. Coffee, Volume 2: Technology. Springer,
Netherlands. //​www.springer.com/​la/​book/​9789401080286.
Clifford MN. 1997. The nature of chlorogenic acids. Are they
advantageous compounds in coffee? In Colloque Scientifique
International sur le Café, ASIC, 17, 79–​91.
Coelho C, Ribeiro R, Cruz ACS, Domingues MRM, Coimbra MA,
Bunzel M, Nunes FM. 2014. Nature of phenolic compounds
in coffee melanoidins. Journal of Agricultural and Food
Chemistry, 62(31), 7843–​53.
Costa LL, Donangelo CM, Silveira CLP, Farah A. 2010. Mineral con-
tent in Brazilian Arabica coffee brewed by different methods.
23rd Int. Conf. Coffee Sci. ASIC/​Prospero, Trieste, Italy.
De Toledo PRAB, de Melo MMR, Pezza HR, Toci AT, Pezza L,
Silva CM. 2017. Discriminant analysis for unveiling the origin
of roasted coffee samples: a tool for quality control of coffee
related products. Food Control, 73(Part B), 164–​74.
Deibler KD, Delwiche J. 2003. Handbook of Flavour Characterization:
Sensory Analysis, Chemistry, and Physiology. CRC Press.
Díaz-​Rubio ME, Saura-​Calixto F. 2007. Dietary fiber in brewed
coffee. Journal of Agricultural and Food Chemistry, 55(5),
1999–​2003.
Farah A. 2012. Coffee constituents. In Coffee: Emerging Health
Effects and Disease Prevention, Wiley-​Blackwell, Hoboken,
New Jersey, 1, 22–​58.
Folmer B (ed.). 2016. The Craft and Science of Coffee. Academic
Press.
Frank O, Zehentbauer G, Hofmann T. 2006. Bioresponse-​guided
decomposition of roast coffee beverage and identification of
key bitter taste compounds. European Food Research and
Technology, 222(5–​6), 492.
Fujioka K, Shibamoto T. 2008. Chlorogenic acid and caffeine
contents in various commercial brewed coffees. Food
Chemistry, 106(1), 217–​21,
Gloess AN, Schönbächler B, Klopprogge B, D’Ambrosio L,
Chatelain K, Bongartz B, Strittmatter A, Rast M, Yeretzian C.
2013. Comparison of nine common coffee extraction methods:
instrumental and sensory analysis. European Food Research
and Technology, 236(4), 607–​27.
Gniechwitz D, Reichardt N, Ralph J, Blaut M, Steinhart H, Bunzel
M. 2008. Isolation and characterisation of a coffee melanoidin
fraction. Journal of the Science of Food and Agriculture,
88(12), 2153–​60.
Gross G, Jaccaud E, Huggett AC. 1997. Analysis of the content of
the diterpenes cafestol and kahweol in coffee brews. Food and
Chemical Toxicology, 35(6), 547–​54.
Illy A, Viani R. 2005. Espresso Coffee: The Science of Quality.
Academic Press.
Itoh T, Tamura T, Matsumoto T. 1973. Sterol composition of 19
vegetable oils. Journal of the American Oil Chemists Society,
50(4), 122–​5.
Kurt A, Speer K. 1999. A new component in the lipid fraction of
coffee. Proceedings of Euro Food Chem, 3, 882–​336.
Ledl F, Schleicher E. 1990. New aspects of the Maillard reac-
tion in foods and in the human body. Angewandte Chemie
International Edition in English, 29(6), 565–​94.
Lopes GR, Ferreira AS, Pinto M, Passos CP, Coelho E, Rodrigues
C, Figueira C, Rocha SM, Nunes FM, Coimbra MA. 2016.
Carbohydrate content, dietary fibre and melanoidins:
Composition of espresso from single-​dose coffee capsules.
Food Research International, 89, 989–​996.
Coffee Preparation
López-​ Galilea I, Paz De Peña M, Cid C. 2007. Correlation of
selected constituents with the total antioxidant capacity of
coffee beverages: influence of the brewing procedure. Journal
of Agricultural and Food Chemistry, 55(15), 6110–​17.
Ludwig IA, Sanchez L, Caemmerer B, Kroh LW, Paz De Peña M, Cid
C. 2012. Extraction of coffee antioxidants: impact of brewing
time and method. Food Research International, 48(1), 57–​64.
Moreira ASP, Coimbra M, Nunes FM, Passos CP, Santos SO,
Silvestre AJD, Silva AMN, Rangel M, Domingues MRM.
2015. Chlorogenic acid–​arabinose hybrid domains in coffee
melanoidins: evidences from a model system. Food Chemistry,
185(10), 135–​44.
Nunes FM, Coimbra MA. 2001. Chemical characterization of the
high molecular weight material extracted with hot water from
green and roasted Arabica coffee. Journal of Agricultural and
Food Chemistry, 49(4), 1773–​82.
Nunes FM, Cruz ACS, Coimbra MA. 2012. Insight into the mech-
anism of coffee melanoidin formation using modified “in
bean” models. Journal of Agricultural and Food Chemistry,
60, 8710–​19.
Pérez-​Martínez M, Caemmerer B, Paz De Peña M, Cid C, Kroh
LW. 2010. Influence of brewing method and acidity regulators
on the antioxidant capacity of coffee brews. Journal of
Agricultural and Food Chemistry, 58(5), 2958–​65.
Perrone D, Donangelo CM, Farah A. 2008. Fast simultaneous ana-
lysis of caffeine, trigonelline, nicotinic acid and sucrose in
coffee by liquid chromatography–​ mass spectrometry. Food
Chemistry, 110(4), 1030–​5.
Peters A. 1991. Brewing makes the difference. Fourteenth
International Conference on Coffee Science. San Francisco,
July 14–​19. ASIC, 97–​106.
Petracco M. 2008. Technology IV: beverage preparation: brewing
trends for the new millennium. In Coffee (Clarke RJ, Vitzthum
OG eds.), 140–​64. John Wiley & Sons, Hoboken, New Jersey.
Pittia P, Dalla Rosa M, Lerici CR. 2001. Textural changes of coffee
beans as affected by roasting conditions. LWT -​Food Science
and Technology, 34(3), 168–​75.
Primavera R, Barbacane RC, Congia M, Locatelli M, Celia C. 2014.
Laser diffraction and light scattering techniques for the ana-
lysis of food matrices. AFSH, 6(1), 40–​60.
Puhlmann R, Habel G. 1989. Examination to test the grinding behavior
of roasted coffee. Lebensmittelindustrie, 36(4), 161–​3.
149
Ratnayake WMNR, Hollywood R, O’Grady E, Stavric B. 1993. Lipid
content and composition of coffee brews prepared by different
methods. Food and Chemical Toxicology, 31(4), 263–​9.
Schenker S, Handschin S, Frey B, Perren R, Escher F. 2000. Pore
structure of coffee beans affected by roasting conditions.
Journal of Food Science, 65(3), 452–​7.
Speer K. 2006. The lipid fraction of the coffee bean. Brazilian
Journal of Plant Physiology, 18(1), 201–​ 216, http://​
www.
scielo.br/​scielo.php?pid=S1677-​http://​www.scielo.br/​scielo.
php?pid=S1677-​04202006000100014&script=sci_​arttext&tl
ng=es04202006000100014&script=sci_​arttext&tlng=es, last
access 4 December 2020.
Stelmach E, Pohl P, Szymczycha-​Madeja A. 2015. The content of
Ca, Cu, Fe, Mg and Mn and antioxidant activity of green
coffee brews. Food Chemistry, 182(September), 302–​8.
Toci AT, Boldrin MVZ. 2018. Chapter 12 -​Coffee beverages and
their aroma compounds. In Natural and Artificial Flavouring
Agents and Food Dyes Handbook of Food Bioengineering
(Grumezescu AM, Holban AM eds.), 397–​ 425. Academic
Press, New York.
Trugo LC, Macrae R. 1984. A study of the effect of roasting on the
chlorogenic acid composition of coffee using HPLC. Food
Chemistry, 15(3), 219–​27.
Trugo LC. 2003. Coffee analysis. In Encyclopedia of Food Science
and Nutrition, 2nd edition (Caballero B, Trugo LC, Finglas
PM eds.), 498. Oxford Academic Press, Oxford.
Urgert R, van der Weg G, Kosmeijer-​Schuil TG, van de Bovenkamp
P, Hovenier R, Katan MB. 1995. Levels of the cholesterol-​
elevating diterpenes cafestol and kahweol in various coffee
brews. Journal of Agricultural and Food Chemistry, 43,
2167–​72.
Valentin JL, Watling RJ. 2013. Provenance establishment of coffee
using solution ICP-​MS and ICP-​AES. Food Chemistry, 141(1),
98–​104.
Verardo G, Cecconi F, Geatti P, Giumanini AG. 2002. New procedures
for determination of acids in coffee extracts, and observations
on the development of acidity upon ageing. Analytical and
Bioanalytical Chemistry, 374(5), 879–​85.
Von Blittersdorff M, Klatt C. 2017. Chapter 13 –​The grind—​Particles
and particularities. In The Craft and Science of Coffee (Folmer
B ed.), 311–​28. Academic Press, New York.
Colour: Natural Pigments in Foods and Their Technical Uses
Juan Valverde
Trinity College Dublin, Dublin, Ireland
Pigments provide foods with colours. However, thermal and
other culinary processes may affect their properties and there-
fore the perceived quality of foods. Knowing the chemical
structure of pigments and their mechanisms of reaction under
culinary processes is essential for ensuring a stable food product
from a quality point of view, regardless of whether it is created
through traditional culinary processes or through Note by Note
cooking.
All fruits and vegetables contain substances that provide them
with colour (natural pigments). Often the colour of the food is
strongly associated by consumers with its general organoleptic
properties, for example, green peas, orange carrots or purple
beetroot. These colours may not be stable during food processing,
and in many cases the use of heat or acids generates a dramatic
change in colour. There are four main colouring substances in
the plant kingdom: anthocyanins, betalains, chlorophylls and
carotenoids (Belitz et al., 2004).
Anthocyanins
Anthocyanins (Figure 22.1), a class of flavonoids derived ultim-
ately from phenylalanine [1]‌, are water-​soluble (Valavanidis et al.,
2013). In nature, anthocyanins provide a large range of colours,
ranging from orange/​red to violet/​blue. It has been estimated that
there are around 1000 anthocyanins and 15 anthocyanidins (non-​
glycosylated anthocyanins) characterised from the vegetable
FIGURE 22.1 Anthocyanidins are defined by IUPAC as oxygenated
derivatives of flavylium (2-​phenylchromenylium) salts.
kingdom. The difference in colour is due to differences in their
primary or secondary chemical structure, and this structure may
change during the chemical processes due to changes in the
molecular environment (i.e., pH). Anthocyanins are known to be
particularly sensitive to modifications of their molecular struc-
ture, leading to significant colour changes, and they are sensi-
tive to (intramolecular or intermolecular) co-​pigmentation, metal
ions and pH (Dangles and Brouillard, 1992). The change is par-
ticularly sensitive under different environmental pH values, pro-
viding a large gamut of colours depending on the value of the pH.
Therefore, they can be used as chemical indicators, as different
colours will indicate different pH values.
This property of anthocyanins needs to be taken into account
when using anthocyanins for culinary preparations, as the ori-
ginal colour of the ingredient may change over time due to pro-
cessing conditions. For example, if a preparation includes acids
or bases, the colour of the preparation will potentially change in
the direction of undesirable results.
Anthocyanins are also known to be chemically reactive towards
proteins, particularly those with electron donor residues such as
proline and histamine. These chemical interactions depend on
the protein distribution within the complex disperse food system.
This interaction has, in fact, been used in some cases to preserve
the colour of anthocyanins in beverages. A review on this subject
is well documented in Dangles et al. (2018).
Anthocyanins may be industrially isolated by aqueous extrac-
tion of by-​products such as fruit skins and peels for use as food
ingredients (Schieber et al., 2001).
Betalains
Betalains are nitrogen-​ containing water-​ soluble compounds,
derived from tyrosine [2]‌, that are found only in a limited number
of plant lineages and give yellow-​to-​red colours. Some betalains
have a stronger colouring capacity than anthocyanins, and their
colour exhibits higher pH stability (Stintzing and Carle, 2007;
Tanaka et al., 2008; Azeredo, 2009). There are about 75 betalains
that have been structurally unambiguously identified from plants
of about 17 families (all within the order Caryophyllales) and
are generally obtained by simple aqueous extraction (Kahn and
Giridhar, 2015).
151
152
FIGURE 22.2 Chemical structure of betanin.
Betalains are divided into two groups depending on their
chemical structure:
• betacyanins, which contain an indole residue, leading
towards a red colour;
• betaxanthines, which do not have an indole residue and
tend to have a yellow colour. Betanins (Figure 22.2)
represent between 75% and 95% of the main colouring
principles.
Interestingly, anthocyanins and betalains are mutually exclusive
in plants. Betalains and anthocyanins are used as colouring agents
for fruit preparations, dairy products, ice creams, confectionery,
pet foods, soups, sauces, beverages and drinks.
Carotenes and Carotenoids
Carotenes and carotenoids (Figure 22.3) are nature’s most wide-
spread pigments; the earth’s annual biomass production is estimated
at 100 million tons (Fennema, 1996). A large majority of these
pigments are biosynthesized by the ocean algae population. In
higher plants, carotenoids are often masked by the more dominant
chlorophylls. The yellow-​orange colour of carotenoids becomes evi-
dent in the autumn when fruit or plants ripen (Heaton et al., 1996).
Carotenes and carotenoids have colours ranging from yellow to
red and are mostly lipid soluble (there are some exceptions, e.g.,
saffron’s pigments). Carotenes and carotenoids are a subclass of
terpenoids, often obtained by organic solvent extraction of carrots
(Daucus carota L.), oil palm fruit (Elaeis guinensis Jacq.), sweet
potato (Ipomea batatas (L.) Lam.), marigold (Tagetes erecta L.)
and some microalgae (Arthrospira platensis L.).
Animals contain carotenoids due to consumption of caro-
tene/​
carotenoid-​containing plant materials. For example, the
pink colour of salmon flesh is mainly due to the presence of
red pigment astaxanthin [3]‌, which is obtained by ingestion of
seaweed-​ containing carotenoids (Ducauze, 2006). The same
pigment present in shrimp and lobster exoskeletons is blue in
Juan Valverde
FIGURE 22.3 Chemical structure of beta-carotene and lutein.
FIGURE 22.4 Chemical structure of curcumins.
colour when complexed with proteins (Rodriguez-​Amaya, 2001;
Wrolstad, 2005). Heating denatures the complex and alters the
spectroscopic and visual properties of the pigment, thus changing
the colour from blue to red (Cianci et al., 2002).
Carotenes and carotenoids are used in multiple industrial
applications, such as dairy products (beverages, cream and dairy
desserts), oils, fats and emulsions, fruit-​based products (spreads,
desserts, canned fruits and pastry fillings), mustards, egg-​based
products, soups and broths, edible coatings, and cooked fish and
fish products.
Because of their chemical structure (C40 long hydrocarbons),
carotenoids are relatively stable in most culinary conditions (see
the case of lobster as an exception). They are not water-soluble
(except for saffron carotenoids), so they are regularly contained
in the lipid fraction of preparations, which means that the inten-
sity of their colour may “fade away” if they are included in
preparations such as emulsions. At high temperatures (>100 °C)
and in dry conditions in the presence of oxygen, carotenoids may
undergo pyrolysis, leading to a number of oxygenated derivates
with aromatic properties. This oxidative process can also take
place through culinary fermentation processes such as wine
making or green tea leaf drying (Rodriguez-​Amaya, 2001).
Curcumin (Figure 22.4), or turmeric yellow, is obtained by
solvent extraction of turmeric (ground rhizomes of Curcuma
longa L.); the extract is purified by crystallisation. This process
eliminates the pungent and aromatic essential oil in turmeric,
Colour: Natural Pigments in Foods
leaving deodourised turmeric, which is used in dairy products
and baked goods. Curcumin is relatively inexpensive and heat
stable but has poor light stability.
Chlorophylls and their Derivatives
Chlorophylls (Figure 22.5) and their derivatives are another
group of lipid-​soluble pigments, industrially obtained by organic
solvent extraction of grasses, alfalfa (Medicago sativa L.), nettles
(Urticaceaes) and other plants (Spinacia oleracea L.) or algae
materials. During the extraction of chlorophylls and the subse-
quent solvent removal, the naturally present coordinated mag-
nesium may be wholly or partly removed from the chlorophyll
molecules (pheophytinisation), causing a change of colour
towards dark olive green pheophytins.
FIGURE 22.5 Chlorophylls (the numbering is the IUPAC one).
FIGURE 22.6 Green beans after cooking for 15 min at 100 °C in buffer solutions at pH = 5.0 (left) or at pH = 8.0 ± 0.5 (right).
153
In higher plants, only chlorophylls a and b are present. The
a/b ratio is normally between 1 and 3, depending on a multitude
of factors, both genetic (species, variety, etc.) and environmental
(luminosity, water stress, mineral nutrition, etc.) (Lichtenthaler,
1987). For instance, plants exposed to the sun tend to have higher
chlorophyll a / c​ hlorophyll b ratios than plants in the shade.
Chlorophylls generally contribute 0.6% to 1.2% of the
dry weight of plants (Scheer, 1991). Some marine organisms
such as seaweeds and bacteria have completely different
chlorophylls but are still sensitive to colour degradation due to
the pheophytinisation process. This dramatic change of colour
has been studied for centuries, as many chefs and scientists tried
to understand why this was happening and how to avoid it. It is
quite common to find recommendations in cookbooks on how to
retain the green colour of vegetables. It must be said that most of
them have no scientific basis and are of limited effect (Valverde,
2008; Figure 22.6).
154
Therefore, it is not unusual to find scientific literature
dedicated to this phenomenon and technological proposals to
avoid it. There are four main technological pathways that the
food industry has explored and used commercially to obtain
higher-​quality products:
1. neutralisation of acids released during thermal
processing;
2. reduction of the processing time by increasing the pro-
cessing temperature;
3. enzymatic conversion of chlorophylls into more stable
products;
4. transformation of chlorophylls into more stable
metallo-​complexes.
Neutralisation of Acids
The addition of alkalising agents to boiling water can result in
improved retention of chlorophylls during thermal processing.
Techniques have involved the addition of CaO and NaH2PO4 into
blanching water to maintain the pH of the product or even to raise
it to 7.0. When vegetables are cooked, their biological structures
break down, liberating acids retained in the vacuoles and thus
reducing the pH of the solution. Chlorophylls and other pigments
(anthocyanins) are sensitive to these pH changes.
Compounds such as MgCO3 or Na2CO3 in combination with
Na3PO4 were proposed in the past (Fennema, 1996). However,
this process has a trade-​off, resulting also in tissue softening and
a soapy flavour. In 1940, Blair recognised the toughening effect
of Ca2+ and Mg2+ when added to vegetables. This observation
led to the use of Ca(OH)2 or Mg(OH)2 for the purpose of raising
pH and keeping the consistency. This combination of treatments
became known as the “Blair process” (Blair, 1937). The commer-
cial application of these processes was not successful because
of the inability of the alkalising agents to effectively neutralise
interior tissue acids over a long period of time, resulting in
substantial colour loss after less than two months of storage
(Fennema, 1996). However, this process is generally useful in
culinary and catering preparations, as they are able to maintain
bright green colour generally for a number of days. In industrially
canned peas, an elevated pH (8.0 or above) can cause the forma-
tion of struvite, i.e., glass-​like crystals consisting of a magnesium
and ammonium phosphate complex. Struvite is believed to result
from the reaction of magnesium with ammonium generated from
the protein in peas during heating (Fennema, 1996).
High-​Temperature Short-​Time Processing
Commercially sterilised foods processed at a higher temperature
for short times (HTST) are processed at a higher temperature
than 100 °C for several minutes (Fennema, 1996). These products
often exhibit better retention of vitamins, flavour and colour than
conventionally processed foods. The greater retention of these
constituents in HTST foods is because their destruction is more
temperature dependent than the inactivation of Clostridium
Juan Valverde
botulinum spores (Canjura et al., 1999). Temperature dependence
can be expressed in terms of Z-​value. The Z-​value is the change
in °C required to generate a 10-​fold change in the destruction
rate of bacteria. The Z-​values for the formation of pheophytins a
and b in heated spinach purée have been determined to be 51 and
98 °C, respectively (Fennema, 1996). The high values for both
as compared with that for the inactivation of C. botulinum spores
(10 °C) result in greater colour retention when HTST processing
is used (Schwartz et al., 1981). In this way, by increasing the
temperature of the treatment, the processing time can be reduced.
Other studies of plant tissue combined HTST processing with pH
adjustment. Samples treated in this manner were initially greener
and contained more chlorophylls than control samples (typical
processing and pH). However, the improvement in colour, as pre-
viously mentioned, was generally lost during storage (Buckle
et al., 1969).
Enzymatic Conversion of Chlorophylls to
Chlorophyllides
Blanching at lower temperatures than those conventionally
used to inactivate enzymes has been suggested as a means of
achieving better retention of colour in green vegetables, in the
belief that chlorophylls can degrade to chlorophyllides, which are
precursors/​derivates of chlorophylls but without the phytol chain.
Chlorophyllides are known to have greater thermal stability than
chlorophylls.
Early studies showed that, when spinach was blanched for
canning at 71 °C for 20 min, better colour retention resulted.
This occurred as long as the blanching temperature was kept
between 54 °C and 76 °C. It was concluded that the better colour
of processed spinach blanched under low-​temperature conditions
(65 °C for up to 45 min) was caused by heat-​induced conversion
of chlorophylls to chlorophyllides by chlorophyllases.
However, the improvement in colour retention achieved by this
approach was insufficient to warrant commercialisation of the
process (Clydesdale et al., 1970). Activation of chlorophyllases
takes place in spinach when it is blanched at 71 °C; however,
when it is blanched above 88 °C, the enzymatic process is
stopped due to the inactivation of the enzyme. This process is not
commercially viable, as the temperatures used in this process are
not high enough to inactivate micro-​organisms that may spoil the
food over the product’s shelf-​life. However, it can be useful in
culinary preparations.
Transformation of Chlorophylls into More Stable
Metallo-​Complexes
In order to keep the bright green colour of chlorophylls, copper
and/​or zinc complexes of chlorophyll can be synthesised. These
metallo-​complexes of chlorophylls can be obtained by addition
of an organic salt of copper and/​or zinc. However, due to the tox-
icity of copper salts, their use is limited, and maximum levels
recommended in Codex Alimentarius of the GSFA (General
Standard for Food Additives) are rarely above 500 mg/​kg.
Colour: Natural Pigments in Foods
Current efforts to improve the colour of green processed
vegetables and to prepare chlorophylls that might be used as
food colourants have involved the use of either zinc or copper
complexes of chlorophyll derivatives. Copper complexes of
pheophytins are available commercially under the name Cu-​
chlorophylls. Their use in canned foods, soups, candy and dairy
products is permitted in most European countries under regula-
tory control of the European Economic Community. The Food
and Agriculture Organization (FAO) of the United Nations certi-
fied their use as safe in foods provided that no more than 200 ppm
of free ionisable Cu is present. Chlorophylls are extracted from
dried grass or alfalfa with acetone or chlorinated hydrocarbons.
Sufficient water is added, depending on the moisture content of
the plant material, to aid penetration of the solvent while avoiding
activation of chlorophyllases. Some pheophytins form spontan-
eously during extraction. Copper acetate is added to form oil-​
soluble Cu-​chlorophyll. Alternatively, pheophytins can be acid
hydrolysed before Cu2+ ions are added, resulting in the formation
of water-​soluble Cu-​chlorophyllide (Fennema, 1996). The Cu
complexes have greater stability than comparable Mg complexes;
for example, after 25 h at 25 °C, 97% of chlorophyll degrades,
while only 44% of the Cu-​chlorophylls degrades.
The complex industrial process can also be carried out in a
simple kitchen. The process involves cooking the vegetables
to ensure complete pheophytinisation of chlorophylls in the
presence of copper/​zinc salts or copper/​zinc-​rich foods (i.e., sea-
food, offal or dried yeast). In the latter case, the process is less
efficient because copper and zinc ions are generally bound to
proteins and are therefore less readily available for the reaction.
It has been observed that, when vegetable purées are commer-
cially sterilised, small bright-​green areas occasionally appear
(Von Elbe et al., 1986). It was determined that pigments in the
bright-​green areas contained Zn2+ and Cu2+, and the formation of
bright-​green areas in vegetable purées was termed “regreening”.
Regreening of commercially processed vegetables takes place
when Zn2+ and/​or Cu2+ ions are present in process solutions.
Okra’s retention of bright green colour when processed in a brine
solution containing ZnCl2 is attributed to the formation of zinc
complexes of chlorophyll derivatives. This process (known as
the Veri-​Green process) involves blanching vegetables in water
containing sufficient amounts of Zn2+ or Cu2+ salts to raise the
tissue concentration of the metal ions to between 50 and 500 ppm
(depending on the vegetable). Green vegetables processed
in modified blanching water are greener than conventionally
processed vegetables (Canjura et al., 1999).
Conclusions
Pigments provide organoleptic properties often related to product
quality. Thermal processing and other culinary processes have an
impact on colour due to alterations in the chemical structure of
the pigments. Knowing the mechanism of these chemical struc-
ture changes can lead to technological proposals to reduce the
impact of these culinary processes. When these pigments are
used in pure form in culinary formulations (such as Note by Note
cooking), one must pay attention to these.
155
REFERENCES
Azeredo HMC. 2009. Betalains: properties, sources, applications,
and stability –​a review, International Journal of Food Science
and Technology, 44(12), 2365–​2376.
Belitz HD, Grosch W, Schieberle P (eds). 2004. Food Chemistry,
Springer-​Verlag, Berlin-​Heidelberg.
Blair JS. 1940. Patent US2189774A, March 1937, 1940.
Buckle KA, Edwards RA. 1969. Chlorophyll degradation products
from processed pea puree, Phytochemistry, 8, 1901–​1906.
Canjura FL, Watkins RH, Schwartz SJ. 1999. Colour improvement
and metallo-​chlorophyll complexes in continuous flow aseptic-
ally processed peas, Journal of Food Science, 64(6), 987.
Cianci M, Rizkallah PJ, Olczak A, Raftery J, Chayen N, Zagalsky
PF. 2002. The molecular basis of the colouration mech-
anism in lobster shell: β-​crustacyanin at 3.2-​Ả resolution,
Proceedings of the National Academy of Sciences USA,
99(15), 9795–​9800.
Clydesdale FM, Fleischman DL, Francis FJ. 1970. Maintenance
of colour in processed green vegetable, Food Product
Development, 4(5), 127–​130.
Dangles O, Brouillard R. 1992. Polyphenol interactions. The
copigmentation case: thermodynamic data from temperature
variation and relaxation kinetics. Medium effect, Canadian
Journal of Chemistry, 70, 2174–​2189.
Dangles O., Fenger JA. 2018. The chemical reactivity of anthocyanins
and its consequences in food science and nutrition, Molecules,
23(8), 1970.
Ducauze CJ. 2006. Fraudes Alimentarios, Acribia, Zaragoza,
Spain.
Fennema OR. 1996. Food Chemistry. 3rd ed., Marcel Dekker Inc.,
New York.
Heaton JW, Marangoni AG. 1996. Chlorophyll degradation in
processed foods and senescent plant tissues, Trends in Food
Science & Technology, 7, 8–​15.
Kahn MI, Giridhar P. 2015. Plant betalains. Chemistry and biochem-
istry, Phytochemistry, 117, 267–​295.
Lichtenthaler HK. 1987. Chlorophylls and carotenoids: pigments of
photosynthetic biomembranes, Methods in Enzymology, 148,
350–​382.
Rodriguez-​ Amaya DB. 2001. A Guide to Carotenoid Analysis
in Food, International Life Sciences Institute (ILSI),
Washington, DC.
Scheer H. 1991. Chlorophylls and Chlorophyll Derivatives, CRC
Press, Boca Raton, Florida.
Schieber A, Stintzing FC, Carle R. 2001, By-​products of plant food
processing as a source of functional compounds –​recent
developments, Trends in Food Science & Technology, 12(11),
401–​413.
Stintzing FC, Carle R. 2007. Review betalains –​emerging prospects
for food scientists, Trends in Food Science & Technology,
18(10), 514–​525.
Schwartz SJ, Woo SL, Von Elbe JH. 1981. High-​performance liquid
chromatography of chlorophylls and their derivatives in fresh
and processed spinach, Journal of Agricultural and Food
Chemistry, 29, 533–​537.
Tanaka Y, Sasaki N, Ohmiya A. 2008. Biosynthesis of plant pigments:
anthocyanins, betalains and carotenoids, The Plant Journal,
54(4), 733–​749.
Valavanidis A, Vlachogianni T. 2013 Plant polyphenols: recent
advances in epidemiological research and other studies on
cancer prevention in studies in natural products chemistry. In
Atta-​ur-​Rahman (ed.) Studies of Natural Products Chemistry.
Elsevier, Amsterdam, The Netherlands, 269–​295.
156
Valverde J. 2008. Study of the modifications induced by various
culinary and industrial treatments of pigment systems from
immature pods of green beans (Phaseolus vulgaris L.): intro-
duction of new analytical methods for the study of these
systems, PhD dissertation of the University Paris 6, www.
theses.fr/​2008PA066677, last access 5 December 2020.
Juan Valverde
Von Elbe JH, Huang AS, Attoe EL, Nank WK. 1986. Pigment com-
position and colour of conventional and Veri-​Green canned
beans, Journal of Agricultural and Food Chemistry, 34,
52–​54.
Wrolstad RE. 2005. Handbook of Food Analytical Chemistry (1st
ed.), Vol. 2, John Wiley & Sons, Inc., Hoboken, New Jersey.
Cooking
Hervé This vo Kientza1,2
1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
Cooking? If it is “the activity or skill of preparing food” or
“the process of preparing food by heating it” (Lexico, 2020;
Stevenson, 2010), this whole handbook deals with it, and no par-
ticular description is needed. However, here, I shall envision a
general description of some thermal processes based on thermal
transfer, before introducing an innovative method of cooking.
Then I shall discuss the chemical basis behind changes triggered
by the various thermal processes.
A Table for Innovation
Let us first observe that the definition of cooking that includes
heating is debatable, as we would certainly not consider as
“cooked” a frozen chicken that was taken out from the deep
freeze and put at room temperature: it would be heated, certainly,
but the temperature for physical and chemical changes character-
istic of what we consider “cooked” would not occur. Also, macer-
ating some animal tissues in an acidic medium (lime juice, for
example) gives a result that looks as if these tissues have been
heated, but here, there would be no temperature increase (fish
à la Tahitienne, ceviche, etc., for which the word “coction” was
proposed) (This, 2001).
Before considering the chemical changes occurring during
thermal processing of plant or animal tissues, let us begin by
observing that the various traditional methods for the processing
of food include either thermal processing or another way of chan-
ging the chemical state of the tissues (This, 1997). Indeed, we
can distinguish between giving energy to the inside of the food
ingredients by conduction, and giving energy by radiation. In the
former, the hot source can be a solid, a liquid (aqueous solution
or oil) or a gas (air, steam). In the latter, it is traditionally infrared,
with a limited depth being reached because of radiation absorp-
tion by the tissues, but of course, microwave radiation is now
also almost traditional (and other sources producing energy in
the form of visible light, such as lasers, could be used). For non-​
thermal processing, one can trigger some coagulation of proteins
using acids and bases or ethanol, and also by the application of
a high pressure, in “pascalization” or high-​pressure processing
(Chauhan, 2019).
Of course, a classification such as the one given in the previous
paragraph is very crude, because “hot” only means something in
relationship with the various phenomena that can occur, as shown
in the chapter dealing with eggs at 6X°C of this book, “Let Us
Have an Egg” (This, 2009). But it is already useful, as it allows
analysis of the various traditional cooking methods (This, 2014).
For example:
• grilling means heat transfer by conduction from a
hot solid;
• braising means heat transfer from a mildly hot aqueous
solution;
• boiling means heat transfer from a hot aqueous solution
at 100 °C;
• frying means heat transfer from a hot oil (with
differences between deep frying and flat frying, the
latter creating a unidirectional gradient);
• baking means heat transfer from a hot gas (air);
• steaming means heat transfer from a hot gas (steam);
• roasting means heat transfer from infrared radiation;
• microwaving means heat transfer from microwave
radiation;
• and, as said, transformations of tissues can occur
after contact with acids (fruit juices and vinegar, for
example), bases (lime in longevity eggs, for example),
brandies (as in the “beaumés”; see the chapter on egg
coagulation), sugar and salt, all creating “coctions”.
All these processes could be discussed in more detail, but it
is proposed now to observe that some of these “elementary”,
or “unit”, processes can be applied in a row. For example, in
braising, there is (traditionally) first the contact with hot air, and
then a processing in a closed vessel, for which the heat transfer
occurs from hot humid air at a temperature below 100 °C. Or for
some frying processes, potatoes can be boiled before being fried,
i.e., receiving heat from hot oil.
What about envisioning a systematic table with the elementary
processes in rows and in columns? This creates more than a hun-
dred possibilities with very different results (Table 23.1).
Let us observe that this general idea of coding processes was
already put into action when considering the cooking of eggs (see
157
158 Hervé This vo Kientza

TABLE 23.1
Double Cooking: 144 Different Results Are Obtained When a Process Given in a Column Is Applied Before a Second Process from a Row
Simmering Boiling
Heat transferred Hot aqueous aqueous
from solid solution solution Hot oil Hot air Steam Infrared Microwaves High pressure Acids Bases Ethanol
Hot solid
Simmering
aqueous solution
Boiling aqueous
solution
Hot oil
Hot air
Steam
Infrared
Microwaves
High pressure
Acids
Bases
Ethanol

TABLE 23.2
The Tree of Doughs Is Horizontal, Growing from Left to Right
Flour and water With fat Not kneaded, but egg added With yeast Boiled
Steamed Dampnudel
Without yeast Boiled
Steamed
Kneaded, no egg With yeast Boiled
Steamed
Without yeast Boiled Echaudes
Steamed
Without fat Not kneaded, but egg added With yeast Boiled
Steamed
Without yeast Boiled Knepfle
Steamed
Kneaded, no egg With yeast Boiled Dry noodles
Steamed
Without yeast Boiled Cornuau
Steamed

this chapter) in a different manner. More abstractly, this means
always having a number for a particular process, creating “for-
mulas” that correspond to particular processes. Alternatively, the
ingredients can also be described in this way, as in the “dough
tree” that we are going to examine now (Table 23.2).
The idea is to observe that doughs are generally made of
flour and water. Some include fat (oil, butter), and some don’t.
Some are leavened, and some are not. Some contain eggs, and
some don’t. Some are steamed, and some are boiled. This can be
arranged systematically as a “tree of doughs” (here from left to
right) showing many possibilities that are not practised tradition-
ally. Some names are given in the last column.
In More Depth: Some Chemistry of Cooking
The diversity of plant and animal tissues, and the diversity of
processes applied to them, can appear infinite, but the processes
are used because they trigger some particular changes, based
on the molecular and physical characteristics of the food
ingredients. Animal tissues, on the one hand, and plant tissues,
on the other, belong to categories sharing physical and chemical
characteristics.
Most animal tissues are muscular tissues, made of grouped
bundles of muscular fibres, chemically based primarily on water,
proteins and fat (Girard, 1990). For plant tissues, the plant cell
wall (with polysaccharides such as cellulose, hemicelluloses and
Cooking 159

pectins) surrounds the inside of cells, which are mainly composed
of water and sometimes starch and fat (Bowes, 1988).
For plant tissues, first, low temperature can cause hardening
of the tissues through the release of calcium ions that will bridge
pectins (Anthon and Barrett, 2006). But at higher temperatures:
• starch will gelatinize (Véchambre et al., 2010);
• amylose and amylopectin will be hydrolysed (Utrilla-​
Coello et al., 2014);
• pectins will be hydrolysed (Figure 23.1);
• some small saccharides can undergo intramolecular
dehydration (Wunderlin et al., 1998; Lewkowski, 2001).
In particular, pectin degradation is a very important phenom-
enon because of its contribution to the softening of plant tissues.
Many studies have shown that this elimination process (E1 mech-
anism) occurs in two steps involving a carbocation intermediate
(Neukom and Deuel, 1958). The process seems to be accelerated
by the presence of a methyl group on the C5 of galacturonic acid
residues (Albersheim et al., 1960). The more methylated the
pectins, the more hydrolysed they are during thermal treatment
(Sajjaanantakul, 1989). More generally, β-​elimination is the pri-
mary process responsible for pectin degradation during thermal
treatment at pH 6.1. However pectins can also be degraded by
acidic hydrolysis (Krall and McFeeters, 1998). Methyl groups are
needed for β-​elimination but not for acid hydrolysis. Polypectates
(degree of methylation (DM) < 5%) can be degraded in acidic
medium. For pH higher than 3.5, β-​elimination would be the pri-
mary process.
Experimentally, the importance of β-​elimination for the texture
of plant tissues has been studied, but with contradictory results.
During thermal treatment, the softening of carrot (Daucus carota
L.) tissues was described as a first-​ order kinetic mechanism
(Paulus and Saguy, 1980) or due to two different mechanisms
(Huang and Bourne, 1983), the first being the cause of pectin
transformation in the middle lamella (responsible for 95–​97% of
firmness loss). The two-​step loss of firmness was later observed
again (Huang and Bourne, 1983), with a loss of firmness observed
during the first 5–​8 min (60%) (Greve et al., 1994). β-​elimination
is increased by higher temperatures: 1% solutions of lemon

O HO H OH– O
OMe H OH
H H
H O H α OH O O H

O OH β R
O O
O OH
R H H H H
H
O HO
H HO H OMe H

α β +H2O
O Ö

R O

O
O OH
OMe H
H
H OH O O H
H O
R
O OH O
OH H 4 OH
H H H
R H HO
OH H H
H 5 O
H HO

O OMe

FIGURE 23.1 Cooking plant tissues (and softening them) is associated, for example, with degradation of pectin (mostly polymers of galacturonic acid, or
(2S,3R,4S,5R)-​2,3,4,5-​tetrahydroxy-​6-​oxohexanoic acid) through β-​elimination, a hydrolysis process due to breaking chemical bonds between galacturonic
residues.
(Keijbets and Pilnik, 1974)
160
(Citrus citrus) pectins have a viscosity that decreases from 14
to 13 or from 14 to 1 for thermal treatments at 35°C and 95°C,
respectively. Moreover, the length of galacturonic acid chains
drops from 350 units to 16 or 63 units after 1 h at 95 and 80 °C,
respectively.
Of course, pectin is not the only polysaccharide that can
be hydrolysed during thermal processing. Whereas cellulose
is highly heat resistant, starch (amylose and amylopectin) or
proteins dissociate slowly with time, in particular when the envir-
onment is acidic, such as in meat or in most dishes (Belitz and
Grosch, 1999a). These processes generate saccharides or amino
acids, which can then react by processes such as dehydration of
hexoses or Streker degradation (Belitz and Grosch, 1999b). For
example, 5-​(hydroxymethyl)-​2-​furaldehyde (HMF) is formed by
hexose dehydration (Wunderlin et al., 1998; Lewkowsi, 2001).
For animal tissues:
• water is not chemically transformed, but it can
evaporate;
• some proteins (actins, myosins) can “coagulate”, i.e., be
denatured and link through intermolecular bonds such
as disulfide bridges (see the chapter on “uncooking
an egg”);
• some other proteins are hydrolysed, such as collagen,
which makes up the envelope of the muscular fibres and
also creates bundles and superbundles of such fibres;
this creates peptides and amino acids;
• fat melts and can be oxidized (see the chapter on fat
oxidation in this book).
Of course, such a short description does not include all processes,
and many culinary phenomena remain unexplained. For example,
the formation of hydrogen sulfide during egg processing is readily
observed (from the egg white, or from the yolk) using a piece of
filter paper impregnated with a solution of lead acetate; it has
been studied (Germs, 1973), but no publication explains why this
process does not occur when eggs are thermally processed for
more than 12 h at 65 °C.
Also, “pyrolysis” mechanisms seem to be very important during
culinary processes, as very high temperatures are obtained at the
surface or even inside food products when water is evaporated.
For example, the temperature under cubes of meat 5 cm wide
was measured to be about 100 °C when the heating power was
low, so that the flow of juice expelled because of collagenic con-
traction (Kopp et al., 1977) is enough to keep the lower surface
moist; but the temperature under the meat can reach very high
temperatures when the heating power overcomes water evap-
oration (temperatures as high as 290 °C were measured). Also,
during the first step of shrimp bisque production, when shrimp
shells are heated in oil, temperatures as high as 320 °C have been
measured in a professional kitchen, which is much higher than the
temperatures studied in model systems (Mar’in and Shlyapnikov,
1980). We invite the reader to consult the chapter on glycation
reactions in this book.
As a conclusion to this chapter, it should also be observed that
the “most important” changes in terms of mass are not always the
Hervé This vo Kientza
most important in terms of flavour, as exemplified by the evapor-
ation of odorant compounds from a solution.
REFERENCES
Albersheim P, Neukom H, Deuel H. 1960. Splitting of pectin chain
molecules in neutral solutions, Archives of Biochemistry and
Biophysics, 90, 46–​51.
Anthon GE, Barrett DM. 2006. Characterization of the tempera-
ture activation of pectin methylesterase in green beans and
tomatoes, Journal of Agricultural and Food Chemistry, 54,
204–​211.
Belitz HD, Grosch W. 1999a. Food Chemistry, Springer Verlag,
Heidelberg, Germany, 81.
Belitz HD, Grosch W. 1999b. Food Chemistry, Springer Verlag,
Heidelberg, Germany, 22.
Bowes BG. 1988. Structure des plantes, INRA Editions, Paris.
Chauhan OP (ed.). 2019. Non-​Thermal Processing of Food, CRC
Press, Boca Raton, USA.
Germs AC. 1973. Hydrogen sulfide production in eggs and egg-​
products as a result of heating, Journal of the Science of Food
Agriculture, 24, 7–​16.
Girard JP (ed.). 1990. Technologie de la viande et des produits
carnés, Lavoisier Tec et Doc, Paris.
Greve LC, McArdle RN, Gohlke JR, Labavitch JM. 1994. Impact of
heating on carrot firmness: changes in cell wall components,
Journal of Agricultural and Food Chemistry, 42, 2900–​2906.
Huang J, Bourne MC. 1983. Kinetics of thermal softening of vege-
table, Journal of Texture Studies, 14, 1–​9.
Keijbets MJH, Pilnik W. 1974. Beta-​Elimination of pectin in the
presence of anions and cations, Carbohydrate Research, 33,
359–​362.
Kopp J, Sale P, Bonnet Y. 1977. Contractomètre pour l’étude
des propriétés physiques des fibres conjonctives: tension
iométrique, degré de réticulation, relaxation, Canadian Institute
of Food Science and Technology Journal, 10(1), 69–​72.
Krall SM, McFeeters FR. 1998. Pectin hydrolysis: effect of tempera-
ture, degree of methylation, pH, and calcium on hydrolysis
rates, Journal of Agricultural and Food Chemistry, 46,
1311–​1315.
Lewkowski J. 2001. Synthesis, chemistry and applications of 5-​
hydroxymethyl-​furfural and its derivates, Arkivoc, 1, 17–​54.
Lexico. 2020. www.lexico.com/​ definition/​
cookery, last access 5
December 2020.
Mar’in AP, Shlyapnikov YA. 1980 Thermal and oxidative thermal
degradation of chitin, Vysokomol Soedin, Ser A, 22(3),
589–​594.
Neukom H, Deuel H. 1958. Alkaline degradation of pectin, Chemistry
and Industry, 683.
Paulus K, Saguy I. 1980. Effect of heat treatment on the quality of
cooked carrots, Journal of Food Science, 45, 239–​245.
Sajjaanantakul T. 1989. Effect of methyl ester content on heat deg-
radation of chelator-​soluble carrot pectin, Journal of Food
Science, 54, 1272–​1277.
Stevenson A. 2010. Oxford Dictionary of English, Oxford University
Press, Oxford.
This H. 1997. La cuisson, Pour la Science, 235, 14.
This H. 2001. Lettre à la Secrétaire perpétuel de l’Académie
française.
This H. 2009. Molecular Gastronomy, a chemical look to cooking,
Accounts of Chemical Research, 42(5), 575–​583.
This H. 2014. Mon histoire de cuisine, Belin, Paris.
Cooking 161

Utrilla-​Coello RG, Hernández-​Jaimes C, Carrillo-​Navas H, González Wunderlin DA, Pesce SF, Ame MV, Faye PF. 1998. Decomposition
F, Rodríguez E, Bello-​Pérez LA, Vernon-​Carter EJ, Alvarez-​ of hydroxymethylfurfural in solution and protective effect of
Ramirez J. 2014. Acid hydrolysis of native corn starch: fructose, Journal of Agricultural and Food Chemistry, 46,
morphology, crystallinity, rheological and thermal properties, 1855–​1863.
Carbohydrate Polymers, 103(2014), 596–​602.
Véchambre C, Chaunier L, Lourdin D. 2010. Novel shape-​memory
materials based on potato starch, Macromolecular Materials
and Engineering, 295, 115–​122.
Cooking: Culinary Precisions and Robustness of Recipes
Hervé This vo Kientza1,2
1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
Culinary books include technical descriptions of dishes, artistic
prescriptions and “social” advices. Within the technical part, one
can distinguish a “definition” and added information, which were
named “culinary precisions” (This, 2010). For French cuisine
only, more than 25,000 of them were collected, and many were
tested over the last decades. A proportion of them are conveying
incorrect technical information, and it would be an improvement
for culinary education to get rid of them. Because of their huge
number, an international collaboration is needed.
This recognition did not occur immediately after the creation
of molecular and physical gastronomy, and it calls for an analysis
of the history of both culinary practices as well as food science
and technology. Indeed, in culinary books of the past (and still
today), there are remnants of old magical ideas, such as when
it is written that women having periods prevents the making
of emulsions such as mayonnaise sauce (Thiebaut, 2017). And
in food science and technology, one can remember that, some
decades ago, it was suggested that mayonnaise (again!) could be
made better in a copper vessel using an iron whisk, because there
was a “battery effect” (This, 2010). The fact is that molecular
and physical gastronomy was partly based on the investigation
of such ideas: culinary circles have been distributing incorrect
explanations of phenomena and using out-​ of-​
date scientific
information.
This view goes back several centuries: since the beginning
of the modern natural sciences, their application to phenonema
occurring during food preparation (“cooking”) has been
considered. For example, in 1560, Ambroise Paré introduced
the word “emulsion” (from the Latine word emulgere, “to draw
milk”) to describe “thick white products”, as he worked on what
we would call today drug formulations (TLFi, 2018). In 1783,
the French chemist Antoine-​Laurent de Lavoisier published the
result of his studies of meat stock (Lavoisier, 1783), quoting the
French pharmacist Claude Joseph Geoffroy, also called Geoffroy
le Cadet, who scientifically studied such preparations as early as
1730 (Cadet de Vaux, 1818; Geoffroy, 1733; This et al., 2006).
Studies of fat by Michel-​Eugène Chevreul led him to the dis-
covery of the chemical constitution of triglycerides (Chevreul,
1823). Discussing whether all this was science or technology
is not beyond the scope of this chapter (This and Kurti, 1994),
because it is related to what is called “culinary precisions”.
Unquestionably, many chemical and physical phenomena
occurring during culinary transformations were studied by food
science or food technology before 1988, when the concept of
molecular and physical gastronomy was introduced (sometimes
shortened to “molecular gastronomy”) as a discussion during the
preparation of the first International Workshop on Molecular and
Physical Gastronomy. However, it is a fact that, in the 1980s,
research in food science and technology generally neglected
culinary processes.
In France, the microbiologist Edouard Pojersky de Pomiane
interpreted (without scientific experiments) culinary practices in
the second half of the 20th century, but he also promoted incorrect
ideas, such as that you could avoid crying when you peel onions
if you bite on a wooden spoon. Japan was also an exception, with
articles published (in Japanese) in the Japanese Journal of Home
Economics and in the Japanese Journal of Cookery Science, but
this work was mostly ignored by the English-​speaking scientific
community.
In the 1980s, “food chemistry” textbooks, such as the classic
Food Chemistry (Belitz and Grosch, 1999), contained almost
nothing about culinary transformations; even as late as 1999,
most of the chapter on meat described either meat compos-
ition and structure, or industrial products (e.g., sausages, meat
extracts), but nothing was said about braisés, sauté, roasts and
all other preparations including meat. Also, less than 0.5% of the
text considered “culinary phenomena” (such as meat shrinkage
during heating because of collagen denaturation). In addition,
nothing was said about the effect of thermal processing on wine,
despite the wide use of this beverage in culinary activities; 48%
of French classical sauces contain wine (This, 2015).
Probably because culinary transformations are complex, and
also because the food industry did not support studies outside
its field, food science and technology had drifted slowly toward
the science of ingredients and toward technological questions,
neglecting phenomena that occur when making home or res-
taurant dishes, such as cassoulet, goulash, hollandaise sauce, etc.
It was even considered a conspicuous eccentricity when an article
on bearnaise was published in a scientific journal in the 1970s
(Perram et al., 1977).
This is why the late Nicholas Kurti (1908–​1998), former pro-
fessor of physics in Oxford (This, 1999), and I decided in March
163
164
1988 that a “new scientific discipline” was to be created under
the name “molecular and physical gastronomy”. The choice of
“gastronomy” in this title was obvious, as it means “knowledge”
(Brillat-​Savarin, 1825). Personally, I had the feeling that the “and
physical” part of the name was useless, but Kurti, being a physi-
cist, insisted on keeping it, because he thought that there would
be too much emphasis on chemistry with only “molecular gas-
tronomy”, and perhaps also because he wanted to escape from a
former proposal (which I did not know about) by the American
cook Elizabeth Thomas to name all culinary activities based on
scientific explorations “molecular gastronomy” (Lersch, 2018).
The objective of molecular and physical gastronomy was and
remains scientifically clear; if one wants to discover new phe-
nomena, which is the goal of the natural sciences, the explor-
ation of a new range of phenomena is probably a better strategy
than looking to already well-​considered objects. Indeed, at the
beginning of molecular gastronomy, Kurti was upset by the tech-
nically old-​fashioned way of cooking, and he insisted on transfer-
ring ideas from physics into the kitchen, which was technology;
he felt, however, that knowledge of the mechanisms was needed.
Personally, from March 1980, I thought that it was necessary to
collect and to test rigorously “old wives’ tales” and analogous
pieces of information. More generally, we both thought that
culinary traditions should be scientifically analysed, and because
our ideas on science and technology were not clear enough,
we introduced the following faulty scientific programme (This,
1995; This, 2002): (1) collect and test old wives’ tales; (2) model
recipes; (3) introduce new tools, products and methods; (4) invent
new dishes using the three previous steps; (5) use the appeal of
food in order to promote science.
It is easy to see, today, that this “scientific programme” was
not clear, and it is surprising that nobody noticed: if aims (1) and
(2) are really scientific, items (3) and (4) are technological
applications of the first two, and (5) is the educational applica-
tion of the first four. As it had been clear for Kurti and me since
the beginning of our discussions that molecular and physical gas-
tronomy was a scientific discipline, and not a technology, a new
programme had to be designed. Considering recipes, it appeared
in 2003 that the technical information in any traditional recipe
includes two parts. The first one is a “definition”: for example, a
soufflé is a foamy product that expands during cooking and goes
down as it is opened (otherwise it is a cake); a mayonnaise sauce
is an emulsion obtained with only egg yolk, salt, pepper, vinegar
and oil, etc. (This, 2005a). Generally, culinary definitions are
given as protocols, and they are mixed with “culinary precisions”,
i.e., added technical information that is not absolutely necessary
for making the dish but conveys ideas of technical improvement.
Such pieces of information were often called old wives’ tales,
lore, ways of doing, tricks, sayings, adages or maxims. It was
recognized that all these terms were too many, and that one term
could encompass all of them: “culinary precisions”. Finally, any
recipe can be considered to have three main parts: (1) one “defin-
ition”, (2) “culinary precisions” and (3) commentaries about the
art or social components of the culinary practice (This, 2003).
Let us examine, for example, a recipe from a culinary
book published in France at the beginning of the 20th century
(Anonymous, 1905):
Hervé This vo Kientza
Take a dozen pears of middle size, remove the skin and put
them immediately in cold water. Then melt 125 g of sugar
with some water in a pan at low heat: as soon as the sugar
is melted, add the pears, add some lemon juice if you want to
keep the pears white; if you prefer them red, do not add lemon
juice and cook them in a pan lined with tin.
In this recipe, the words in bold are enough to give the definition
of the dish (this definition here is less than 10% of the recipe). The
words in italics add “precisions”. Here, there is no “third part”
(art, social). However depending on the recipe and the author, the
precision ratio of recipes can vary considerably; for example, in
some recipes from the French chef Jules Gouffé (Gouffé, 1867),
the percentage of culinary precisions is nil.
Making this difference between culinary definition and culinary
precisions was the basis for an improved scientific strategy for
molecular and physical gastronomy: (1) modelling definitions
and (2) exploring precisions. However, this new programme was
rapidly discovered to be insufficient, because the main point in
culinary practice is to produce “good” dishes; this is art, and not
technique, because “good” means “beautiful to eat”. Moreover,
even technically successful and artistically well-​designed dishes
are not appreciated if the guests are neglected or treated poorly,
so that the “social component” of culinary practice also needs to
be considered. Of course, the natural sciences cannot have the last
word on such topics, but evolutionary biology of psychology, for
example, can explain a lot about human behaviour and, accord-
ingly, about culinary practice. Today, the scientific programme of
molecular and physical gastronomy could be more appropriately
stated as: (1) explore scientifically the technical part of cooking
(definitions and precisions); (2) explore scientifically the art
component of cooking; (3) explore scientifically the “social link”
component of cooking.
Now that this scientific programme is clearer, what is the
most rational way of exploring the field of culinary phenomena?
As culinary transformations are dynamic processes involving
systems with structure (Dickinson, 2006), it is natural to make
complementary descriptions of the physical state, on one hand,
and of the chemical state, on the other. The bioactivity (organo-
leptic, nutritious or toxic) of such systems is considered later as
the result of the two states (This, 2012).
Testing Culinary Precisions
We now see why culinary precisions (in short, “precisions”)
are important for molecular and physical gastronomy. Since the
1980s, a lot of precisions have been tested, and the number of
precisions now collected from French culinary books alone is
more than 25,000. Many have been given online (This, 2019a) or
discussed within the framework of a yearly course on molecular
and physical gastronomy (This, 2019b); some have been analysed
in a book (This, 2010). The open questions are many, and we give
some examples in the following.
Is it true that pears (Pyrus communis L.) stay white when
lemon (Citrus citrus L.) juice is added, in a pear jam? The
answer is yes, and it is well understood that ascorbic acid
Cooking: Culinary Precisions
((R)- ​ 5 - ​ ( (S)- ​ 1 ,2- ​ d ihydroxyethyl)- ​ 3 ,4- ​ d ihydroxyfuran-​ 2 (5H)-​
one) inhibits o-​diphenol O2 reductase enzymes (EC 1. 10. 301,
PPO) (Zawistowski et al., 1991). When pears and other fruits
are cut, these enzymes transform the released phenolics, such
as chlorogenic acid and (−)-​epicatechin, into reactive quinones,
which can in turn polymerize into dark pigments (Grotte et al.,
2000; Goupy et al., 1995).
Is it true that pears turn red when cooked in tin-​covered copper
pans? During public lectures and laboratory experiments with
Valérie Michaud, we performed numerous tests with common
pears (Passe Crassane, Williams, Red Williams, Comice,
Conference …) and never observed the expected red colour.
Model tests using tin ions did not produce the red colour, but
such a colour was obtained for many varieties when the pH was
lower than 2 (Figure 24.1).
This is attributed to anthocyanidins from the fruits, which turn
red in an acidic medium (UV-​visible spectroscopy: λmax = 522 nm)
(Belitz and Grosch, 1999). Colouration is particularly deep near
the pear skin, where anthocyanidins are more concentrated.
Is it true that mayonnaise sauce fails when made by menstru-
ating women? This precision has been tested experimentally and
was proved wrong (of course!).
Is it true that mayonnaise sauce fails when made during a
full moon? Students at Tours University (MST Le goût et son
environnement, promotion 2000–​ 2001) tested this old wives’
tale, and the first mayonnaise that they made failed; i.e., a phase
separation occurred after the addition of oil. However, the same
students were told to repeat the process and got well-​formed
emulsions. Of course, this means only that the students did not
perform well the first time, probably adding oil too rapidly, but
anyway, as only one counter-​example is enough to refute a gen-
eral law, it can be said that the culinary precision about the influ-
ence of the moon is wrong.
Is it true that for making mayonnaise sauce, eggs and oil have
to be at the same temperature? Experiments were performed with
eggs at 4 °C and oil (sunflower) at 24 °C, and with eggs at 24 °C
and oil at 4 °C, and emulsions were obtained. Indeed, there is no
FIGURE 24.1 Pears of the Passe Crassane variety were cooked with water
(same weight as fruits) and sucrose (same weight as fruits). For some, Sn2+
ions were added, and the colour did not change much (centre). On the other
hand, a red colour appeared when the pH was low (left and right).
165
reason why the sauce could fail in all the circumstances described
by old wives’ tales, as it is only an emulsion, i.e., a dispersion of
oil droplets in the water of the yolk and vinegar, with proteins and
phospholipids from the egg stabilizing (it would be more appro-
priate to write “metastabilizing”, as emulsions are not thermo-
dynamically stable) the droplets.
Finally, tests of culinary precisions performed since 1980 show
that all possibilities arise: (1) some precisions seem wrong, and
they are wrong; (2) some seem wrong, and they are true; (3) some
seem true, and they are wrong; and (4) some seem true, and they
are true. We shall now give a brief example of each, adding a fifth
class, of uncertain precisions (5).
(1) As we demonstrated, it does not seem true, and indeed it
is not true, that women’s menstruation prevents them from pre-
paring mayonnaise, as has been proposed in France. Indeed, it
is strange that this culinary precision is widespread in France
yet not in England (on the other hand, in England, menstruating
women should not rub meat with salt) (Kurti, 1995) or in other
countries. This demonstrates how much cooking is rooted in cul-
ture, and why comparative molecular gastronomy across coun-
tries and cultures could be an interesting project (This, 2006).
(2) In 1994, the question of whether the skin of suck-
ling pigs crackles more (i.e., becomes more crispy) when the
head of the pig is cut off immediately after being roasted was
examined (This, 1994). This advice seemed intuitively wrong,
but it was proven to be true. The culinary precision was found
in L’Almanach des gourmands from the French gastronome
Alexandre-​ Balthazar Grimod La Reynière: “suckling pigs
should have the head cut immediately when the pigs are taken
out from the oven, otherwise their skin softens”. The same
advice is found in many other culinary books. For example, the
French chef Marie Antoine Carême indicates that a cut should
be made around the neck (“When you are ready to serve, you
separate immediately, with the tip of the knife, the skin of the
neck, so that the skin says crisp, which makes most of the
interest of roasted sucking pigs”). These remarks are strange,
as in roasted pigs, no fluid seems to exchange between the head
and the skin; it was highly unlikely that the advice was true,
but the experiment was performed (public experiment at Saint-​
Rémy-​l’Honoré, Yvelines, France, July 7, 1993) with four suck-
ling pigs from the same parents, reared together on the same
farm, weight 7.1–​7.3 kilograms, cooked on a large outside fire
from 4.00 PM to 9.00 PM, with one head being cut off for each
pair of pigs. Blind tasting for 143 people showed that the skin
of pigs with head cut was crispier.
The mechanism behind this was easily discovered, as it was
observed during cooking that a stream of steam was escaping one
pig from a hole made during the preparation. This means that heat
is causing water to evaporate from the surface of the meat during
cooking, making the crust, and that vapour formed inside the
meat is not enough to compensate for the loss of surface water.
When the pigs are no longer being heated, the crust softens if
vapour goes through; cutting the head prevents vapour perfusion,
as it escapes through the opening.
(3) It is said that a pan in which green beans are cooked should
not be covered, as this would entrap volatile acids, which would
promote pheophytinization of chlorophylls, but tests show that
166
there is no colour difference. The idea seemed true, but it is
wrong (Valverde, 2008).
(4) It is sometimes said that the soufflés should be made from
very firm whipped egg whites, added to a viscous preparation.
It was demonstrated that this precision holds, as vapour bubbles
formed in the bottom part of soufflés during cooking escape less
easily through a firmer foam (This, 2002; see also the chapter on
Expansion in this book). The advice seemed true, and it is true.
(5) Let us now discuss a fifth class, with culinary precisions
having a non-​clear-​cut status. For example, it was said by the
French chef Pierre Gagnaire (This, 2005b) that wine sauces are
more “brilliant” (in French, this would mean shiny or luminous)
when shaken than when whipped. Such a declaration, even by
a famous chef, should be considered with caution, because in
many circumstances experiments have shown that cooks were
influenced by tradition rather than experiments and facts.
In such cases, one has to carefully define the question, because
it would be useless to make tests in conditions different from the
ones in which an observation is made; in particular, the production
of wine sauces depends both on the authors and on the period in
the history of cooking. The “wine sauces” discussed by Gagnaire
are made from a veal “fond”, wine and butter. The “fond” is a
solution obtained by grilling veal bones until they have a brown
(not black) colour. Then water, carrot (Daucus carota L.) roots,
onions (Allium cepa L.) bulbs and possibly other plant tissues are
added. A thermal treatment at a temperature lower than 100 °C
20
15
×106
10
5 0.5
0
0 5 10 15 20 25 30
×103
FIGURE 24.2 Wine sauces with the butter emulsified by whipping with a whisk (left) or by shaking the pan (right). Under each picture, a histogram of the
diameters of fat droplets is given.
Hervé This vo Kientza
for some hours (between 2 and 20, depending on the author,
but also on each particular sauce) is achieved. Then, the fond is
filtered, and its volume is reduced by boiling to about one-​tenth
of its initial volume, after which red wine is added. The sauce is
reduced again, and red wine is again added, before butter is added
while the sauce is heated at a temperature lower than 100 °C,
so that there is no boiling (when details are not given here, it has
to be understood that the cooks themselves can make changes
depending on any particular sauce; e.g., the exact quality of the
“red wine” is not considered, and any red wine can be used).
In our studies, the sauce was first modelled by a system
containing distilled water (instead of stock and wine), gelatine
(because gelatine is extracted in the first steps of fond produc-
tion) and butter (approximate quantities of 100 mL water, 6 g
gelatine and 60 g butter, based on preliminary analyses). The
initial mixture of water and gelatine was divided into two parts,
and the same quantity of butter was added to each. Then, the
model sauce was heated and either shaken (the pan was moved
forward and back over a distance of 5 cm, 23 times in 10 s, for
65 s) or whipped using a whisk (four whipping movements per
second).
Initial (triangular) visual tests with 52 people did not detect
any difference in visual appearance. However the observa-
tion of the model sauces using optical microscopy (microscope
Meiji Techno ML Série 2000, Model ML2300) showed clear
differences (Figure 24.2) that were characterized.
2.0
1.5
1.0
0.0
0 50 100 150 200 250 300
Cooking: Culinary Precisions
These differences can be explained by the fact that the energy
given to the sauce is very different when the sauce is shaken or
whipped. This energy is used to increase the surface energy of
the emulsion and, hence, the size of the melted butter droplets
dispersed in the water phase.
At that point, it should be concluded that, if there is no diffe-
rence in “brilliancy” (or shining), there is, however, a difference
in flavour, as it was demonstrated for these model emulsions that
the composition of odorant molecules above an emulsion differs
depending on the microstructure of the emulsion, where compos-
ition is constant (Relkin et al., 2004).
Reasons for Culinary Precisions: The Robustness
Assumption
Why did precisions arise in the past? Considering the empirical
way culinary technique developed, we assume that failures and
successes generated assumptions concerning the experimental
protocol used. For example, in the old recipe for emulsion given
now (Bernardi et al., 1853), the inverse order of ingredients
should have frequently led to failure:
Green Rémolade. Take a handful of chervil, tarragon, you will
blanch these herbs that are called Ravigote; press and grind,
add salt, pepper, mustard: grind all together, then add half a
glass of oil that you amalgamate with the ravigote and mus-
tard; finally you add two or three egg yolks.
Such a process is strange; the authors are describing an emulsion
but they add the surfactants (proteins and phospholipids from
the yolk) at the end (happily, there are some phospholipids and
proteins, as well as water, in the ground herbs!). The frequent
failure of such a method could have led them to investigate the
causes of the irregularity of the process, and culinary precisions
should have arisen naturally. This observation leads to a predic-
tion: if it is true that precisions arise from failures, then an inverse
quantitative relationship should exist between the “robustness” of
a recipe and the number of precisions written in culinary books.
In order to test experimentally this theoretical prediction,
“robustness” has first to be made quantitative. Let us consider
that a recipe R is a function of many variables: various times
(t1, t2 …), temperatures (T1, T2 …), quantities of ingredients
(m1, m2 …) and very general details of process (p1, p2 …).
For example, in a mayonnaise recipe, the process can be
described by the amount of egg yolk (a parameter including water
content, protein content and phospholipid content), the amount
of vinegar (i.e., primarily water), the rate of oil addition and the
energy of whipping.
A product P obtained through the recipe using particular
conditions is given by the equation:
P = R (t1, t2, …, T1, T2, …, p1, p2, …).
Or, more generally, P = R(xi, yj), xi being parameters describing
the ingredients, yj the parameters describing the process, and i
and j being integers from 1 to n and m, respectively.
167
As long as the parameters vary within certain limits (xi, min
< xi < xi, max, yj, min < yj < yj, max), the recipe is successful: a product
is the result of a successful recipe if it is associated with a point
inside a limited hypervolume in the multidimensional space of
the parameters; when the representing point is outside the recipe
hypervolume, the recipe is considered as failed. For each param-
eter of the recipe, the size of the interval [xi, min, xi, max] can be a
measure of the robustness, but, in order to get a non-​dimensional
value that can be compared with others, we need to divide xi, max
− xi, min by a number having the same units. We proposed to nor-
malize by the uncertainty i(xi) on the considered variable xi:
ρi = Δ xi /​  i(xi)
Of course, orders of magnitudes have to be calculated instead of
exact values, as the uncertainty is only known by estimation.
For example, mayonnaise can be defined by the mass of yolk,
m(y), the mass of vinegar, m(v), the mass of oil, m(o), the mass
of salt, m(s), the mass of pepper, m(p), the mass of oil in each
successive addition, m(d), the whipping power, Pw, and the effi-
ciency of dispersion, Ed.
As the critical parameter is clearly the oil addition, let us focus
on robustness related to oil addition; at the beginning of mayon-
naise preparation, oil should not be added too fast, or a water-​in-​
oil (W/O) emulsion is obtained instead of an oil-​in-​water (O/W)
emulsion. As the quantity of water from one yolk and one tea-
spoon of vinegar is about (15 g + 5 g = 20 g), and considering the
uncertainty on the oil quantity added each time (estimation based
on experiments 7.5 g), robustness related to oil addition is equal
to 20/​7.5 = 2.7.
In more “robust” recipes, such as beef meat roasted in the oven,
the calculated robustness is bigger; for a piece of meat of mass 1 kg,
cooked at 180 °C for a time between 20 and 60 min, assuming that
the precision on time measurement is 5 min, robustness is equal to
(60 − 20)/​5 = 8. If the cooking temperature is lower (e.g., 70 °C),
then the cooking time interval would be still bigger, and robustness
higher; the time interval could be estimated to be between 60 min
and 1 day, so that the robustness is equal to 1440/​5 = 276.
For some recipes, parameters are not independent, and success
is obtained only if more than one condition is simultaneously
verified. Particular robustnesses have to be aggregated. In order
to consider this, let us assume that robustness is inversely related
to the number of precisions:
ρi = 1/​ni.
If the n is the total number of precisions:
n = n1+n2+ … +nm.
Then
ρ = 1/​(n1 + n2+ …) = 1/​(1/​ ρ1 + 1/​ρ2 + …).
Does the inverse relation hold? In the corpus of precisions that we
have collected since 1980, there are 105 paragraphs about may-
onnaise preparation, compared with 12 paragraphs for roasts.
168
FIGURE 24.3 Variation of robustness as a function of the number of
culinary precisions. (a) No regular variation appears, but if the last point is
eliminated, then the relationship can be fitted to a 1/​n curve (b). The last
point corresponds to meat stocks, recipes that are robust but have been very
important in the history of human food, because they are a way to keep all the
nutrients from animal tissues during cooking.
In Figure 24.3, we show how the robustness ρ depends on the
number of paragraphs containing precisions for grated carrots,
stock, soufflé, boiled eggs, gougères, mayonnaise and roast beef.
In Figure 24.3a, stock is included, and the curve does not corres-
pond to an inverse relation: stock generated many precisions only
because of its culinary importance, even if there is almost no risk
of failure. In Figure 24.3b, stock has been excluded, and the rela-
tionship is more as expected.
More work now needs to be done to test our assumption, using
the aggregation of partial robustness, for example, and also the
hyperspace needs to be explored more thoroughly.
Hervé This vo Kientza
Comparative Molecular Gastronomy
As we said earlier, molecular and physical gastronomy developed
initially through international workshops, but, after 1999, these
meetings were complemented with monthly seminars and also
courses, numerous lectures, articles and books. Today, more
and more associations of molecular and physical gastronomy
are in existence in various countries, making an international
network of scientists in universities and research centres. Each
country can focus its study on its own particular definitions and
precisions, so that we can look forward to a time when “com-
parative molecular and physical gastronomy” will be possible.
Of course, some assumptions on the origin of such precisions can
be made on the basis of “robustness of recipes”, but a more com-
prehensive collection of culinary precisions associated with time
periods would help to investigate such cases.
As we said, dozens of thousands of culinary precisions were
collected for French culinary books alone, but it would be scien-
tifically interesting for other countries than France to do the same
work, as it would allow some comparative analysis. The disperse
system formalism (DSF, see the chapter on this in this book) also
could be useful in this regard; in the same way as it was used for
studying the evolution of the number of physical categories of
sauces, it could be applied to comparing sauces between coun-
tries and thereby understanding cultural influences and transfers.
This would also have educational interest, perhaps helpful in
view of the current pandemic of obesity. Even in Crete, where the
famous Cretan diet originated, up to one-​third of children aged
12 are now overweight or obese (WHO, 2013). Other important
factors include the increasing concern for the environment, the
increasing proportion of the population living in cities, and
energy issues. Finally, there is a growing gap between the world
of science and laypeople, along with a disaffection for scientific
studies, which could easily be bridged by considering food, the
most complex scientific process most people engage in on a daily
basis, as a scientific phenomenon.
All these data lead toward the idea that children should
receive more information about food and food preparation. In
particular, health programmes promoting a healthy diet cannot
be successful if people cannot rationally choose the food they
eat. In order to adapt food to particular cases, citizens need clear
information. However, tradition is no guarantee of healthy food
or rational preparation of food. This is why educational activ-
ities such as “Ateliers expérimentaux du gout” (This, 2001) and
other programmes were introduced in France, linking cooking
and science at school. Some of these programmes include field-
work by children that contributes to the collection of culinary
precisions in order to create a national database of such precisions.
No doubt, all countries could do the same.
REFERENCES
Anonymous. 1905. Le livre de cuisine de Tante Colette, François
Tedesco, Paris, France.
Belitz HD, Grosch W. 1999. Food Chemistry, Springer, Heidelberg,
Germany.
Cooking: Culinary Precisions
Bernardi M, Viart M, Fouret M, Délan M. 1853. Le cuisinier national
de la ville et de la campagne (ex Cuisinier royal), augmenté de
200 articles nouveaux, Gustave Barbu, Paris, France, 49.
Brillat-​Savarin JA. 1825. La physiologie du goût, Brillat-​Savarin,
Paris, France.
Cadet de Vaux AA. 1818. De la gélatine et de son bouillon, L. Colas
fils, Paris, France.
Dickinson E. 2006. Colloid science of mixed ingredients, Soft
Matter, 2, 642–​652.
Geoffroy le Cadet M. 1733. Mémoires de l’Académie royale, histoire
de l’Académie royale des sciences, année MDCCXXX, Pierre
Mortier, Paris, France, 312.
Gouffé J. 1867. Le livre de cuisine, Henri Veyrier (new edition 1988),
Paris, France.
Goupy P, Amiot MJ, Aubert S, Nicolas J. 1995. Apple phenolic
compounds and enzymatic oxidation in model solutions by
apple polyphénoloxydases, in Scalbert A., Ed., Polyphenols
94, Inra editions, 183–​184.
Grotte M, Duprat F, Loonis D, Pietri E. 2000. Aspects des
meurtrissures des pommes, Science des aliments, 20, 575–​590.
La Varenne PF. 1651. Le cuisinier françois, Pierre David, Paris.
Lavoisier AL de. 1783. Œuvres complètes (1862–​1893), Imprimerie
Nationale, 3, 563–​578.
Lersch M. 2018. https://​blog.khymos.org/​molecular-​gastronomy/​his-
tory/​, last access 27 November 2018.
Perram CM, Nicolau C, Perram JW. 1977. Interparticle forces in
multiphase colloid systems: the resurrection of coagulated
sauce béarnaise, Nature, 270, 572.
Relkin P, Fabre M, Guichard E. 2004. Effect of fat nature and aroma
compound hydrophobicity on flavour release from complex
food emulsions, Journal of Agricultural and Food Chemistry,
52(20), 6257–​6263.
Thiebaut E. 2017. Ceci est mon sang, La Découverte, Paris,
France.
This H. 1994. La cuisson: usages, tradition et science, in La cuisson
des aliments, 7 e rencontres scientifiques et technologiques des
industries alimentaires, Agoral, 94, 13–​21.
This H. 1995. La gastronomie moléculaire, L’Actualité chimique,
42–​46.
This H. 1999. Froid, magnétisme et cuisine: Nicholas Kurti (1908–​
1998, membre d’honneur de la SFP), Bulletin de la Société
française de physique, 119, 24–​25.
This H. 2001. Les Ateliers expérimentaux du gout, Presse de la
Sorbonne, Paris, France.
169
This H. 2002. Molecular gastronomy, Angewandte Chemie,
International Edition in English, 41(1), 83–​88.
This H. 2003. La gastronomie moléculaire, Sciences des aliments,
23(2), 187–​198.
This. 2005a. Modelling dishes and exploring culinary “precisions”:
the two issues of Molecular Gastronomy, British Journal of
Nutrition, 93(4), S139–​S146.
This H. 2005b. Séminaire INRA de gastronomie moléculaire
N°49: le vannage des sauces les rend-​elles plus brillantes?, 15
September 2005.
This H. 2006. When shall we have comparative Molecular
Gastronomy? Japanese Journal of Cookery Science, 4(2006).
This H. 2010. Les précisions culinaires, Cours de gastronomie
moléculaire N°2, Editions Quae/​Belin, Paris.
This H. 2012. Solutions are solutions, and gels are almost solutions,
Pure and Applied Chemistry, 1–​20.
This H. 2015. Gastronomie moléculaire. Applications. Techniques
de l’ingénieur, September 2015. www.techniques-​ingenieur.fr/​
base-​documentaire/​procedes-​chimie-​bio-​agro-​th2/​biochimie-​
alimentaire-​ a nalyses-​ e t-​ a limentation-​ h umaine-​ 4 2470210/​
gastronomie-​moleculaire-​f1016/​, last access 22 August 2018.
This H. 2019a. http://​blogs.inra.fr/​herve_​this_​cuisine/​, last access 16
November 2019.
This H. 2019b. www2.agroparistech.fr/​podcast/​-​Gastronomie-​
Moleculaire-​.html, last access 16 November 2019.
This H, Kurti N. 1994. Physics and chemistry in the kitchen, Scientific
American, 270(4), 44–​50.
This H, Méric R, Cazor A. 2006. Lavoisier and meat stock. Comptes
rendus de l’Académie des Sciences Chimie, 9, 1510–​1515.
TLFI. 2018. Emulsion, http://​stella.atilf.fr/​Dendien/​scripts/​tlfiv5/​
search.exe?23;s=2841039660;cat=0;m=%82mulsion, last
access 22 August 2018.
Valverde J. 2008. Study of the modifications induced by various
culinary and industrial treatments of pigment systems from
immature pods of green beans (Phaseolus vulgaris L.): intro-
duction of new analytical methods for the study of these
systems, PhD dissertation of the University Paris 6, www.
theses.fr/​2008PA066677, last access 5 December 2020.
WHO. 2013. Data and statistics, www.euro.who.int/​en/​health-​topics/​
noncommunicable-​diseases/​obesity/​data-​and-​statistics, last
access 6 December 2020.
Zawistowski J, Bibarderis CG, Eshin NAM. 1991. Polyphenol
oxidase. In Robinson DS and Eshin NAM (Eds.) Oxidative
Enzymes in Food, Elsevier Applied Science, 217–​273.
Cryogenics in the Kitchen
Peter Barham
H H Wills Physics Laboratory, University of Bristol, Bristol, BS8 1TL, United Kingdom
Probably, most foodies’ introduction to the use of cryogens
(substances that change state at low temperatures, typically
well below ca. −60 °C) was to see an ice cream or sorbet being
frozen at the tableside using liquid nitrogen. As the cold liquid
is added to the ice cream mix, the bowl overflows with white
clouds of fog, creating the appearance of an alchemist’s caul-
dron and setting up expectations of a wonderful dish to come.
But although many restaurants do indeed use liquid nitrogen in
this way, there are far more uses that occur behind the scenes that
customers never see.
A Short History of How Cryogens Came into
Restaurants
The use of liquefied gases in food preparation has a long and
interesting history, much of which involves the regular ‘redis-
covery’ of particular uses over time. The first recorded descrip-
tion of the use of a cryogen came very soon after gases were first
liquefied.
However, it was only recently that Myhrvold and Young (2001)
managed to unearth the full history of making ice cream using
nitrogen. As recently as 1994, in their Scientific American article,
Kurti and This noted that, despite many people who said they had
heard of someone making ice cream in this way, the first time they
could be certain it had happened was when I demonstrated the
process at one of the Erice Workshops on Molecular and Physical
Gastronomy (Kurti and This-Benckhard, 1994). However, the
story is much more interesting and is full of strange twists and
turns as the wheel kept on being invented and re-​invented over
a century.
In her magazine The Table, A.B. Marshall wrote: “Persons
scientifically inclined may perhaps like to amuse and instruct
their friends as well as feed them when they invite them to the
house. By the aid of liquid oxygen [sic], for example each guest
at a dinner party may make his or her own ice cream at the table
by simply stirring with a spoon the ingredients of ice cream to
which a few drops of liquid air has been added by the servant”
(Marshall, 1901).
Air was first liquefied in 1877 (Dumas, 1877) by Louis Paul
Cailletet in France and Raoul Pictet in Geneva (but in very
small quantities), but it was only in 1884 that James Dewar
managed to prepare enough liquid air that he could show it to the
public. By 1894, Dewar was ready to make a number of public
demonstrations of cryogenic gases including whole air, nitrogen
and oxygen. On 19 January 1894, he gave a particularly well-​
attended lecture, which is when it has been surmised, but not
proven, that Marshall may have seen liquid air added to water
to make it freeze quickly and got the idea for the article she later
published in her magazine (Marshall et al., 1988). One of the
reasons why the achievements of Marshall are not as well known
as they probably should be is that after her death in 1905, her
estate was sold to Isabella Beeton’s publisher Ward Lock, who
had little interest in keeping her work in print.
In 1957, William Harrison in the USA patented a process to
make nitrogen ice cream using pretty much the same method
described by Marshall, but no records have been found as yet
showing anyone actually using this patent, and it seems to
have been forgotten for many years. Then, in 1974, William
Chamberlain, working in the USA, took some liquid nitrogen
home and played with it to try to make ice cream; after a number
of failures, he settled on a technique, adding nitrogen to an ice
cream mix that was being stirred in a food mixer on a stand. This
proved successful (and is the technique used in most nitrogen ice
cream parlours today), but although he gave a number of parties
where he made ice cream for his friends and colleagues, again, it
appears to have been forgotten.
In 1976, after seeing a demonstration of using nitrogen to
freeze bovine sperm, André Daguin, the chef of the restaurant
Jardin des Saveurs, started using nitrogen to make ice cream at
the tables of his guests; this spectacular dish gained a lot of publi-
city, which continued into the early 1980s, but no other chefs took
it up, and once again it was largely forgotten (see the chapter by
Daguin in this book).
During the early 1980s, as a result of a drive to encourage
school pupils to take up studying science, scientists started giving
‘inspirational’ talks in schools. I had used a demonstration of
making ice cream with liquid nitrogen to teach undergraduates
about entropy and adapted this in 1983 to give a series of school
talks. These soon expanded rapidly as more young physicists, and
later chemists and biologists, were trained to give similar talks.
Brian Coppola and colleagues published details of how to make
nitrogen ice cream in 1994 (Coppola et al., 1994). Today, pretty
much any science department in any university can provide an
171
172
inspirational school speaker to talk about their research and com-
bine that with a demonstration of ice cream making.
As the use of nitrogen to make ‘instant’ ice cream became
widely known through these school talks, chefs were introduced
to liquid nitrogen not just to provide spectacle and make
instant ice cream but also, for example, as a means to prepare
finely ground herbs and spices. I well recall taking a dewar of
nitrogen to the Fat Duck restaurant as I was driving close by on
my way back from a school talk one afternoon in 2000 and then
spending an afternoon in the garden with Heston Blumenthal
and his chefs, simply seeing what could be done. Not long after
that, Heston Blumenthal introduced the green tea amuse bouche
(Nitro Poached Green Tea and Lime Mousse) with the accom-
panying spectacle of dragon’s breath coming from the nostrils
of the diners, that captivated his customers at the start of their
meals. This was quickly adopted by Ferran Adria, and similar
dishes were served at El Bulli. As these two restaurants were
at the time frequently voted the best in the world, it is not sur-
prising that other chefs quickly started investigating how to use
liquid nitrogen for themselves, and so, within less than a decade,
it became almost compulsory for any Michelin restaurant to make
some use of liquid nitrogen in the restaurant, usually at the table,
to provide a combination of spectacle and deliciousness.
How Do Cryogens Work?
When a substance undergoes a change of state (or first-​order phase
transition), its internal energy changes, often by a large amount,
so it either absorbs or releases this energy in the form of heat,
which is often termed the latent heat. When a solid is heated (for
example. an ice cube in a cocktail), it starts to melt when the tem-
perature reaches its melting point (0 °C in this case). But, to be
able to melt, it has to absorb a large amount of heat (334 kJ/​kg),
so it can take some time for all the ice to melt as it absorbs this
heat from its surroundings. As we know from experience, it takes
typically 10 to 20 minutes for the ice in our drinks to melt. As the
ice melts, it extracts the heat required for the change of state from
a solid to a liquid from the surrounding liquid, and so cools that
liquid. In the case of a typical ice cube in a gin and tonic, the tem-
perature is reduced by about 5 °C as the ice cube melts.
Similarly, if we heat a pan of water to its boiling point (100 °C),
as it boils, the temperature remains at 100 °C, and the heat
applied to the pan goes into the latent heat required to change
its state from a liquid (water) to a gas (steam). The absorption of
heat as a liquid boils is the basis for the operation of nearly all
air conditioners, refrigerators and freezers we use at home and
work. In these devices, the coolant is kept in a closed system and
is made to boil by reducing the pressure in the place where the
temperature needs to be reduced and then compressed back into a
liquid outside, where the excess heat released as the gas turns into
a liquid can be vented into the atmosphere.
One disadvantage of using conventional freezers in cooking
is that the cooling can take a long time, as the heat is extracted
from the foodstuff by transfer through the air inside the freezer. It
would be much more efficient if the food were placed in a liquid
rather than a gaseous environment, as the heat transfer through
Peter Barham
the denser liquid would be much greater. So, if we have a liquid
that boils at a very low temperature, we can use that to cool things
down to that temperature; as the liquid boils, it will extract heat
from whatever we want to cool, and, provided there is enough of
the liquid available, it can cool our object right down to its own
boiling point.
Cryogens change state at low temperatures (typically well
below ca. −60 °C) and as a result can be used to cool systems
down to these temperatures. Provided they are also inert and safe
to eat, drink or breathe, then they can be used in direct contact
with foods. Two commonly available cryogens that are com-
pletely non-​toxic are liquid nitrogen and solid carbon dioxide.
Nitrogen boils at −196 °C, and carbon dioxide sublimes (turns
directly from a solid to a gas) at −78.5 °C. Thus, if solid carbon
dioxide (often referred to as dry ice) is dropped into some liquid
that you want to freeze, or cool rapidly, it will very quickly sub-
lime, causing the liquid to bubble up as if it were boiling as the
carbon dioxide gas emerges; the sublimation process extracts a
lot of heat, and, depending on how much carbon dioxide was used
and the amount and type of liquid being cooled, it will cool to its
own freezing point and then, as further heat is extracted, it will
start to turn solid. Dry ice is often dropped into buckets of water
in this way, not to cool the water but to generate white clouds of
carbon dioxide vapour and small ice crystals to provide the spe-
cial effects of rolling clouds of fog we see at many events. Indeed,
I recall back in the 1960s buying dry ice from a local supplier as
the stage manager for my school play, Macbeth, so we could have
a bubbling cauldron on stage for the witches’ scenes.
Similarly, solid foods can be dropped into liquid nitrogen,
which, as it boils, will rapidly cool them down (again generating
the fog-​like effects). The limit to how fast the food cools depends
largely on the size of the pieces; larger pieces take longer to cool,
as the heat has to get from the inside to the outside by thermal
conduction, which can take some time. If there is sufficient liquid
nitrogen that it does not all boil off, then the food will eventually
cool down to its boiling point (−196 °C), which is of course far
too cold to eat!
If cryogens are used in the kitchen, great care must be taken
to ensure safety. First, the proper equipment is needed to store
the cryogens safely and to prevent them from simply evaporating
before they can be used. Secondly, as any contact of cryogens with
naked skin can lead to severe frostbite (or cryo-​burns), personal
protective equipment such as gloves, aprons, and face masks or
goggles are necessary. You will also need to take care that, if the
gas evaporates quickly, there is sufficient oxygen in the room to
avoid the risk of suffocation. For example, when liquid nitrogen
boils, the resulting gas takes up about 700 times the volume of
the liquid, so 1 litre of liquid nitrogen dropped on the floor will
rapidly become 700 litres of nitrogen gas, which will displace
700 litres of air from the room. If you work in a small kitchen, a
few litres of liquid nitrogen boiling could easily displace enough
oxygen that you could die of suffocation. Some details of the
necessary equipment and safety procedures and how to perform a
proper risk assessment for the use of cryogens in the kitchen will
be addressed later in this chapter.
You might reasonably ask whether, with all these difficulties,
it is worth the trouble of using cryogens in the kitchen. However,
Cryogenics in the Kitchen
there are plenty of things that can only be achieved with the
use of cryogens. For example, some disadvantages of normal
freezing are easily overcome by freezing at low cryogenic
temperatures. The very low temperatures that can be quickly
achieved greatly affect the formation of ice crystals; very many
crystals nucleate in a short time and grow quickly, so the overall
crystallite size in ice formed using cryogenic cooling is very
much smaller than that of ice formed by freezing in a standard
freezer. Thus, instead of large needle-​or tree-​like crystals that
grow relatively slowly during relatively slow cooling and can
puncture the cell walls inside many foodstuffs, giving coarse
and grainy textures as well as leading to both structural damage
and negative flavour changes, cryogenically frozen foodstuffs
contain very small (sub-​micron) crystals that have smooth and
creamy textures and do not normally damage structure or fla-
vour. The food industry uses significant quantities of both dry
ice and nitrogen in the preparation of many different foods,
not just in frozen produce. As we shall see later, many of these
techniques have in recent years found their way into professional
(and even a few domestic) kitchens.
Why Are Cryogens Useful in Food Preparation?
Many foods have soft textures that make them difficult to cut into
very precise thin slices, or to grate or grind to make garnishes,
etc., as they will tear, rather than cut, and will, if ground, simply
turn into a paste rather than a fine powder. However, if they can
be frozen completely solid, then they can readily be ground or
sliced as required. The problem is then how to freeze them so that
they are truly solid throughout and the texture is not damaged.
When you freeze food, the liquid water turns to ice, but the water
is not pure, as it has dissolved in it a number of molecules that
provide taste and flavour to the food, e.g., salt, sugar and amino
acids. We all know that salty water has a lower freezing point
than fresh water, which is why putting salt on the roads helps
prevent ice formation in winter. So it is when we freeze our food;
as some of the water turns to ice, it increases the concentration
of these solutes in the remaining water, which in turn reduces
the freezing point of the water, and thus a very low temperature
is required before all the water becomes frozen and the product
is fully solid. This temperature is known in the sciences as the
eutectic point, i.e., the temperature below which all the phases
present will, at equilibrium, be solid. For a solution of common
salt, the eutectic temperature is −21.1 °C, for sucrose it is −9.5 °C,
and for most amino acids it is around −12 °C. Most oils, such
as olive oil, only freeze at temperatures around −40 °C, and any
alcohol in the food generally will not freeze until the tempera-
ture is below −110 °C. So, to be sure your food is truly a solid so
that it will not form a paste when ground, for example, it needs
to be at a very low temperature (lower than −30 °C and pref-
erably much colder). Although some professional freezers can
reach such low temperatures, it is much simpler (and faster) to
use either dry ice or liquid nitrogen to freeze the food you want
to grind or slice.
Further, if the texture is to be retained (for example, if you
want to make sub-​millimetre slices of a cream cake), then it is
173
most important that the ice crystals do not damage the texture
when they form. There are two ways in which ice crystals damage
the texture of foods: firstly, from the expansion of ice when it
freezes, and secondly, by growing large needle-​shaped crystals
that can puncture cell walls.
We all know that ice floats on water, because, when ice forms
at 0 °C, its density is lower than that of water at the same tem-
perature; another way of looking at the phenomenon is to note
that water expands when it freezes (anyone who has lived through
a really cold winter will probably have had experiences of water
mains that burst because the water inside froze and the expansion
fractured the pipe). So, if the water inside a cell freezes at around
0 °C, it will expand and is likely to burst the cell wall, which can
massively affect the texture of some frozen foods. However, if the
ice forms at a much lower temperature (lower than −20 °C) then
it is much denser and the effect is much smaller, resulting in little
or no damage to cells.
The size and shape of ice crystals that grow when food is
frozen depend not only on the molecules that are present in the
water but also on the temperature at which the crystals actu-
ally grow. While ice can form at any temperature below 0 °C
(in pure water), in practice creating a crystal is very difficult, so
crystals do not usually start to grow at all until the temperature
is well below zero, typically from about −4 °C and lower. The
reason is that, as a crystal forms, there is an interface between
the growing crystal and the surrounding liquid. While converting
a liquid to a solid (at a temperature below the freezing point)
liberates a lot of energy in the form of heat (the latent heat noted
earlier), creating the interface actually requires the use of signifi-
cant amounts of energy, so there is an energy balance between the
interfacial energy (which is proportional to the surface area) and
the energy released due to crystallization, which is proportional
to the volume of the crystal. Thus, small crystals, which have a
relatively high surface energy, are unstable, while large crystals,
where the surface to volume ratio is low, are stable. The upshot is
that, if a liquid is cooled quickly, it is possible to start the crystal-
lization process at temperatures well below the melting tempera-
ture, where small crystals are quite stable and can grow quickly,
leading to an overall small crystal size. If foods are frozen at a
low temperature in this way, the ice crystals can be small enough
that they cause little or no damage to the surrounding tissues and
hence leave the texture of the frozen food, once defrosted, just as
it was before it was frozen.
Industrial Use of Cryogens in Food
Processing
Before looking in detail at the various uses of cryogens in the
professional and domestic kitchen, it is worth quickly reviewing
industrial uses, as many techniques that have found their way into
modernist kitchens and much of the so-​called molecular cuisine
lexicon have their origin in industrial food science. However, as
we shall see, many of the real advantages that are offered by cryo-
genic freezing are much better suited to the small professional
kitchen than to industrial-​scale processing, where long transport
and storage times are involved.
174
Freezing Fresh Foods
When food is frozen by simply placing it in a freezer and waiting
for it to cool down, there is a major risk of damage to the food
caused by the formation of large ice crystals. These can not only
break the cell membranes and distort the surrounding tissues but
can also lead to denaturation of some proteins (Sanz et al., 1999;
Gomez and Calvelo, 1982).
Liquid nitrogen has been used in the industrial-​ scale
freezing of foods since the early 1960s (Davidge, 1981). In
particular, it was used in tunnel freezers, where the food to be
frozen is passed on a conveyor belt through a tunnel where the
temperature is steadily decreased until the food is completely
frozen. In the entrance region, food is cooled using cold air (or
gaseous nitrogen or carbon dioxide); then, as it passes along,
liquid nitrogen is sprayed from nozzles in the top of the tunnel
directly onto the food; this reduces the temperature below ca.
−30 °C so that the food is ready for packaging, storage and
distribution.
In tunnel freezers, the rate of cooling can be quite high. High
freezing rates lead to small crystal sizes, due to the high nucle-
ation rate of ice at the lower temperatures achieved during rapid
cooling. This can have a number of benefits; the food quality
is perceived as being better in terms of texture and appearance
as well as flavour (Awonorin, 1997). Furthermore, not only can
using cryogenic freezing lead to reduced losses from dehy-
dration, prolonging the overall shelf life (Awonorin, 1997;
Ramakrishnan et al., 2004), but it is also more cost effective
(Miller and Roberts, 2001). It can be shown through a number
of simulations, modelling and experimental studies that the
faster the freezing, the smaller the ice crystal sizes and the better
the quality of the products. For example, Martino and Zaritzky
(1988) demonstrated that frozen meat benefits from the fastest
possible freezing using the greatest possible thermal gradients.
Although they note that direct immersion in liquid nitrogen
gives the best results, they also speculated on how much better
it could be if greater thermal gradients were to be achievable
(e.g., using pressurized systems). Other authors (e.g., Sanz
et al., 1999) have demonstrated the superior quality of a range
of produce, including dates, peas, pineapples and even sausages,
when frozen at the highest possible rates (Alhamdan et al., 2001;
Biglia et al., 2016).
Despite the multiple advantages imparted by the very small
crystals (<< 1 micrometre) that form when using fast cryogenic
freezing, these all have limited lifetimes, as the overall average
crystal size will slowly and steadily increase as the product is
stored (Carrington et al., 1994; Carrington et al., 1996). This pro-
cess, known as Ostwald ripening, is driven by the fact that, the
smaller a crystal is, the less stable it is, so larger crystals grow
larger at the expense of smaller ones in a continuous process. As
a result, some of the advantages of freezing in liquid nitrogen
can be short lived and hence not suitable for industrial processes
where long storage or transport times are required. This is most
notable with nitrogen-​frozen ice cream, where the initially extra-
ordinarily smooth and cream texture can change noticeably
within an hour after preparation.
Peter Barham
Producing Powders and Ground Spices
When spices are ground to make fine powders that can be used to
season or flavour dishes, it is important to try to avoid the loss of
important odorant compounds during the process. These volatile
compounds can be lost if the spices are heated by friction from
the action of the grinding machine itself. Also, as the surface area
of the ground spices continually increases as they are ground to
ever finer sizes, so the rate of loss of volatile compounds will
increase. However, of course, to impart the maximum possible
flavouring effect in food, it is usually preferred to have spices
ground to as fine a size as possible.
If spices are ground at very low temperatures, this will avoid all
these issues, and so in specialist and high-​quality systems grinding
machines are often cooled with cryogens (Balasubramanian
et al., 2012). Cryogenic grinding of spices leads not only to
improved flavour retention but also to better perceived physical
qualities such as colour and texture in products that use ground
ingredients, such as sausages.
The production of powders by spray freeze drying, where a
liquid is sprayed into nitrogen, is a relatively expensive tech-
nique used in industry mainly for pharmaceuticals. However, it is
becoming more widely used in the food industry, as it can provide
much lower loss of volatiles and better final products. Examples
of foodstuffs for which the use of spray freeze drying has
proved beneficial include apple juice and egg albumin (Malecki
et al., 1970).
Flavour Extraction in Wine and Oil Production
(Cryomaceration)
The fact that the solubility of different flavour compounds in
fruits varies quite widely with temperature means that the rela-
tive amounts of different compounds present in extracted juices
will depend on the temperature at which they are extracted. In
winemaking, the solubility also depends on the alcohol con-
tent at the time of extraction. To maximize the extraction of
some desirable compounds, it has been found that it is best to
carry out the initial extraction (or maceration) at a low tem-
perature when fermentation is very slow or absent, thus keeping
the alcohol content at that time as low as possible. Generally,
many winemakers seem to prefer a cold maceration for 2 to
10 days at a temperature below 15 °C (in some cases as low as
ca. 0 °C), depending on the grape variety and desired final wine
characteristics.
Some winemakers use a cryomaceration process whereby the
grapes are cooled directly with nitrogen or dry ice. This leads to
freezing of the interior of the grapes and, if the cooling is not too
fast, to fracture of internal structures in the grapes, leading to an
internal process of flavour development prior to the grapes being
mechanically broken to extract the juice further (Allen, 2007).
Wines prepared in this way (especially those for which the grapes
were cooled using dry ice) have been shown to have more aroma,
intensity and stability of taste properties than those prepared by
traditional processes (Parenti et al., 2004; Salinas et al., 2005;
Gordillo et al., 2010).
Cryogenics in the Kitchen
In a similar fashion, the processing of olives to make high-​
quality olive oils can be markedly improved when the olives are
exposed to dry ice before the initial milling process. The expan-
sion of the water in the olives, as it freezes to form ice crystals
during the relatively gentle cooling rates applied, breaks up the
cells and promotes the diffusion of these flavour compounds that
are eventually dissolved in the oil, providing both increased fla-
vour and nutritional value. In particular, the use of dry ice has been
shown to significantly increase the yield of tocopherols without
increasing the phenolic fraction of the oil. Furthermore, the use
of carbon dioxide creates a gaseous layer that prevents oxida-
tion and thus further improves the final quality of the product (Di
Giovacchino et al., 2002).
Ice Cream
Liquid nitrogen has found several uses in the manufacture of
some luxury ice creams; for example, dipping an ice cream in
liquid nitrogen or spraying it with liquid nitrogen gives the outer
layers a quick hard freeze so that, when it is put into a bath of
warm chocolate, the chocolate will stick well without melting the
ice cream at all and then set to give a nice hard outer coating.
Similarly, layers of different flavours can be built up by successive
dipping in liquid mixes that are then quickly set with a short dip
in liquid nitrogen.
There is one commercial ice cream product that is entirely
frozen by immersion in liquid nitrogen. ‘Dippin dots’™, which
are exclusively available in the USA, are produced by letting ice
cream mixes drip into a liquid nitrogen bath where they freeze
into small (ca. 5 mm diameter) spheres. To prevent them from
sticking together, they have to be stored, transported and kept
before being sold at quite low temperatures (<−40 C), and so
they are only available at specialist outlets, usually shops or
stands in shopping malls and at major events that serve only
‘Dippin dots’.
In the last decade or so, a number of nitrogen ice cream
parlours have sprung up around the world, and these stalls and
shops all make ice cream to order in front of the customers. Using
nitrogen as the coolant means that they can make single servings
of any desired flavour, so the wastage can be very low.
Use of Cryogens in Restaurants
A whole host of techniques that involve the use of cryogens to help
prepare dishes in the professional kitchen have been developed in
the past 20 years or so. Many of these techniques have been given
names that refer to the ways in which they appear similar to more
conventional techniques. In this section, I will address a range of
these techniques and try to show how they can be used to advan-
tage in any kitchen.
Cryo-​Grinding (Cryo-​Powders)
As noted earlier, if a food is fully frozen so that no liquid phase
remains, then it can be ground down to a powder. The key is
to ensure both that there are no liquid phases present, which
175
normally means the temperature needs to reach below ca. −50 °C,
unless there is any alcohol in the product, in which case it may
need to be as low as ca. −120 °C, and that the temperature is
not raised by the frictional heat that will be generated during the
grinding process, to avoid any melting of the solid phases.
The simplest way to achieve this is to immerse the food to
be ground in liquid nitrogen and then grind it. However, the
bowls of many electric grinders may not withstand being cooled
to liquid nitrogen temperatures, and the seals around the shaft
may fail at such low temperatures, so it is usually best to use a
manual grinder. Most coffee grinders are quite robust enough
to withstand having liquid nitrogen in the grinding bowl, and
any ceramic mortars are also usually more than strong enough
to withstand the low temperatures. Of course, using manual
grinders means that care must be taken not to touch any cold
surfaces with bare hands, so proper protective gloves should be
worn. In kitchens with Pacojet machines, these can be used as
grinders, provided the temperature of the frozen Pacojet beaker
is not lower than about −40 °C, which is low enough to freeze
most foods completely solid. The food can be simply frozen ini-
tially in liquid nitrogen and then the beaker allowed to warm up
in a freezer set to about −30 °C.
Once ground, powdered foods usually store well, provided
they are kept in air-​tight containers in the freezer so as to prevent
any melting and coagulation of the fine powder particles. Herbs
and spices really work very well, as the low temperatures prevent
the loss of most, if not all, of the volatile compounds that impart
their aromas. The ground herbs and spices are ideal for adding at
the last minute, just before service, as garnishes or as seasonings
to provide truly fresh flavours.
Finely ground powders used as a garnish can also create a
remarkable aromatic effect, as the powders can simply blow up
into the nose of the diner if they get close; a particular favourite of
mine, which arose during some experimentation at the University
of Copenhagen when I was working with one of the chefs from
Noma in the experimental kitchen at the University, is a very fine
pea powder prepared from a deeply frozen pea puree at about
−40 °C using a Pacojet machine. If this is sprinkled on a plate, a
diner can get a small dose of pea powder in the nose as they start
to eat, and the result is a spectacular and clean hit of pure pea
flavour.
A slightly different use for cryogens with grinders or mincers
is simply to use them to prevent the heat generated during normal
grinding from damaging the product being ground. This can be
particularly useful when grinding or mincing meats to make
sausages, when it is desirable to retain the fat in individual
pieces rather than allowing it to melt from the heat generated
as it passes through the mincing blades. In this case, it is prob-
ably not necessarily desirable to freeze the meat, but the grinding
blades should be kept as cool as possible. Thus, simply grinding
in small batches using either dry ice or liquid nitrogen to cool the
blades and hopper of the grinder between batches will signifi-
cantly reduce the effect of frictional heating –​not only will the
cold blades and walls of the grinder keep the meat cool, but also,
since the colder the blades are the sharper they become, it will
also reduce the amount of heat generated in the process.
176
Cryo-​Grating
This is very similar to cryo-​grinding in that it involves freezing
a soft piece of food (such as a mushroom) so that it is com-
pletely solid throughout before grating off small pieces using, for
example, a micro-​plane. It differs from cryo-​grinding in that it is
a technique used at the pass to add a final garnish to a dish. Here,
the purpose is to enable a chef to create small flakes of odorant
and flavoursome foods that will release sudden short bursts of
flavour on a dish. By grating at very low temperatures, the all-​
important volatile molecules that impart the required aromas
and flavour cannot escape during the grating and remain in the
tiny flakes on the food until they are warmed as they enter the
diner’s mouth.
It is, of course, necessary for any chefs using this technique to
ensure they use appropriate safety equipment (most importantly,
they must never hold the food being grated with a bare hand, and
either a thick insulating glove or a pair of tongs should be used to
avoid the risk of frostbite).
Cryo-​Slicing
It can be difficult to slice some vegetables. meats, breads and
cakes very thinly. But if they are pre-​frozen using a cryogen so
as not to damage their texture and to ensure there are no liquid
phases present, it is possible to produce extremely thin slices
using a standard bacon slicing machine. If even thinner slices are
required, then a carpenter’s plane can be used to produce slices of
thickness less than 1/​10 mm. Such thin slices can be very delicate
and prone to fall apart when handled, especially after they have
warmed up and defrosted when the ice that was helping to hold
them together melts. Thus, it is advisable to keep the frozen slices
in the freezer until required and put them on the plate as required
while still frozen –​note that such thin slices will heat through
very quickly, so they should reach room temperature in the time
it takes to complete the dish and deliver it to the table.
An example of the use of cryo-​slicing is to make a salad
that looks as though it is still in plastic packaging, but is in fact
simply covered with extremely thin slices of cucumber. Use a
plane (or mandolin set to cut slices of about 0.1 mm or less) to
shave off thin slices along the length from a peeled cucumber
that has been frozen in liquid nitrogen. The squared-​off cucumber
can be held in a cloth in a bench vice and the plane pre-​cooled
by pouring liquid nitrogen over it. As the resulting transparent
slices of cucumber are shaved off, they are allowed to fall dir-
ectly into liquid nitrogen. The thin slices can then be taken dir-
ectly from the dewar (using long stainless steel forceps) and
carefully placed on top of a previously prepared salad –​as the
slices heat up and defrost, they quickly deform to provide a shiny
transparent covering over the salad, giving the appearance that
someone forgot to unwrap it! Thinly cryo-​sliced lemons or limes
make interesting garnishes floating on the surface of cocktails.
Cryo-​Shattering
This is a technique whereby thin sheets of liquids or gels are
cast in trays that are frozen by the application of cryogens before
being shattered using a small hammer or spoon –​the remnants
Peter Barham
are then stored either under liquid nitrogen, in a cool box lined
with a layer of dry ice or in a standard freezer, depending on the
temperatures at which the now shattered fragments will melt. The
thin sharp fragments make excellent, if short-​lived, garnishes for
a number of dishes.
Many liquids, such as honey, olive oil and wines, lend them-
selves to being made into fractured frozen shards in this way.
However, the ways in which they can be used and the time it takes
between service and their melting can vary widely, depending not
only on the actual liquid used but also on the thickness and tem-
perature of the frozen sheets. So, for example, 1 mm thick shards
of frozen honey that have been stored at −20 °C can be put into
an ice cream cone and will remain frozen for up to five minutes,
while a similarly sized shard of frozen port will need to be kept
on dry ice until service if it is to remain solid for up to two or
three minutes after being served on top of a slice of stilton. This
is an area where careful experimentation, varying the thickness of
the sheets to be fractured and their storage temperatures, is essen-
tial if a useful product is to be made.
Cryo-​Searing
Another process where very careful experimentation is required to
achieve a consistent outcome is cryo-​searing. Cryo-​searing –​the
dipping of a foodstuff that is already partially cooked into a bath
of liquid nitrogen for a short time –​can allow deep frying to crisp
up the outside without overheating the interior. Because of the
large thermal gradient created between the outside and the centre
by the short dip in liquid nitrogen, the outside can be heated up
to ca. 200 °C, while the interior is heated much less and remains
near room temperature (or at the original cooking temperature).
The technique has been used to create well-​browned crusts on
steaks that have been cooked sous vide at low temperatures (ca.
40 °C) and as a means of creating a wonderfully crisp skin on
almost rare duck breasts.
Cryo-​Shaping
Many foods are so soft that even if they can be extruded or
moulded into any required shape, they will quickly deform
and lose that shape. But if the shaped object is quickly frozen
by either spraying with liquid nitrogen or simply dropping into
liquid nitrogen, it will become hard and stiff enough to retain the
shape as long as it is kept cold –​so, provided the shaped objects
are kept cold in the fridge or freezer until service, it becomes
relatively simply to make some very delicate and complex shapes
from the most unlikely foods –​maybe a rose shaped from frozen
honey, or a basket made from a salmon mousse. In another twist,
the Indian chef Gaggan Anand makes a wonderfully light, almost
spherical mango mousse –​by preparing the mousse mix in a
cream whipper, then injecting it into a balloon, which is dipped
into liquid nitrogen to set. Then the balloon is peeled off and the
mousse kept refrigerated until service.
Another form of cryo-​shaping is to drizzle molten chocolate
into liquid nitrogen. If chocolate is piped into liquid nitrogen, it
freezes in shapes that resemble twigs on a tree –​these can be used
to great effect in some creative desserts. However, if the choc-
olate is allowed to pour slowly so that it forms a very thin stream,
Cryogenics in the Kitchen
then when it comes into contact with a very cold surface, it will
solidify into a fine hair-​like strand –​with practice, it is possible to
create the same sort of delicate structures that can be made with
spun sugar simply by careful pouring of molten chocolates onto a
mould that has been cooled using liquid nitrogen.
Cryo-​Carbonation
The usual method to make fizzy drinks is simply to inject carbon
dioxide at high pressure into the carbonated liquid. However, it
is also possible and (if you have dry ice pellets at hand) easier
to make fizzy drinks of all kinds simply by adding a couple of
dry ice pellets to the liquid to be carbonated. Carbon dioxide is
much more soluble in water at low temperatures than it is at high
temperatures, and it is also much more soluble in alcohol than in
water (Cargill and Macphee, 1981) –​so if you add solid carbon
dioxide to cold water, then as it sublimes, much of the resulting
gaseous carbon dioxide simply dissolves in the water. Now, if the
liquid with dissolved carbon dioxide is warmed up –​e.g., by put-
ting it in a glass, or in the mouth during drinking – ​the tempera-
ture rises and the solubility of carbon dioxide decreases, releasing
gaseous carbon dioxide in the form of bubbles, thus making the
liquid appear fizzy.
So, you can make any drink become fizzy simply by dropping
in a couple of pellets of dry ice and waiting until all the solid
carbon dioxide has sublimed. The maximum solubility of carbon
dioxide in water at 0 °C is 3.4 g carbon dioxide per litre of water
(and up to twice that amount in alcoholic drinks), so as a typical
dry ice pellet weighs around 2 g, just adding one pellet to a glass
is sufficient to make the drink fizzy. But always make sure you
check the weight of your own pellets, as they do vary!
One interesting example is the slightly fizzy milk shakes
invented by Juan Mari Arzark. You simply make a thick milk shake
base (using pureed fruits, skimmed milk powder and a thickener,
e.g., gum Arabic, with sugar and cream to taste) and once it is in a
glass (make sure it is no more than half full), add about 3–​5% by
weight dry ice and allow it to foam up as the dry ice sublimes –​
remember to only serve once all dry ice has sublimed!
You can also use cryo-​carbonation to make fizzy solids, for
example carbonated whole fruits such as oranges or peaches.
Simply use a cool box with a layer of dry ice on the bottom, put
the fruit or other food to be carbonated on a grill above it, and
leave for 30 minutes to 4 hours, depending on the size of the fruit
to be carbonated. The dry ice will sublime and cool the fruit, and
the carbon dioxide gas released will slowly permeate into it and
dissolve in the water inside –​remember to keep the carbonated
fruits in a fridge so they retain their fizz until they are warmed up.
As the fruit is put into the mouth it heats up and the solubility of
carbon dioxide is reduced, making the fruit taste fizzy. This will
work for any fruit or foodstuff that contains a reasonable amount
of water, but some fruits will taste ‘off’ as people associate the
petillance with the start of fermentation.
Cryo-​Peeling
Partially freezing some foods can make it much simpler to peel or
shell them without damage, which can otherwise be taxing. For
example, it is notoriously difficult to remove the shell from soft
177
boiled eggs leaving a perfect undamaged egg –​but if you simply
dip the egg into liquid nitrogen for up to one minute, the outer
layers of the white will freeze, leaving a much less delicate struc-
ture. Then, if you dip the egg into water at room temperature so
you can hold it without getting frostbite, you will be able to peel
off the shell easily without damage to the egg. Just remember to
let the whole peeled egg warm up before serving …
Cryo-​Freeze Drying
Many foods, such as leafy vegetables, bananas and cucumber,
do not freeze well –​the large crystals that can grow during con-
ventional (slow) freezing can damage the cell walls so that on
defrosting they become soggy and textureless. Fast freezing in
nitrogen can ensure very small crystals and reduce or remove
the cell damage so that on defrosting the structure is maintained.
Freezing a bunch of flowers is a standard demonstration used in
many talks about cryogenics in schools and colleges around the
world. If you want to serve freeze dried vegetables (which really
pack a punch of flavour as they rehydrate in the mouth), you
need to retain their structural integrity during the initial freezing
process –​so for example you can prepare freeze dried Brussel
sprouts only by first freezing them in nitrogen before transfer-
ring them into the freeze drying chamber –​at the completion of
the freeze drying process they come out looking exactly as they
did when freshly picked but without any of the water. Olives do
not freeze dry well as they tend to break during freezing but if
quenched in liquid nitrogen the damage is limited so they can be
freeze dried and then re​hydrated in, for example in vodka to make
the perfect garnish for a Martini.
On Show –​in the Dining Room
Customers generally love the spectacle associated with any use
of cryogens in the dining room. The very cold nature of cryogens
means that they cool the atmosphere around them enough that
the water vapour present in the air condenses and creates a fog.
If the cryogen is mixed with water, then the fog simply pours over
the top of the vessel and rises a little before rolling over the sides
and gently flowing to the floor before dispersing. If the water to
which the cryogen is added is perfumed, then the resulting fog
also carries its odour direct to the diner, providing a hint of what
is to come from the food being served at the same time. This is
great theatre and can set up expectations that the restaurant is
an exciting place to be, so it is hardly surprising that increasing
numbers of chefs are finding ways to use cryogens front of house
with a range of frozen mousses and ice creams and as an odour
enhancer.
However, a word of caution is needed here. While dry ice and
liquid nitrogen are not toxic or poisonous, they are extremely
cold, and anything that contains them or has been on contact
with them for any period of time also becomes extremely cold;
so cold that if touched it can cause serious cold burns (or frost-
bite). Therefore, just as no-​one would ever provide cutlery that is
red hot or serve food on a plate straight out of a hot oven, or at a
temperature that will instantly burn the inside of the mouth, it is
178
necessary to make sure that any food or utensils that have been
in contact with cryogens are not so cold that they can harm any
diners –​remember that while chefs in the kitchen and waiters
in the restaurant can wear insulating gloves and will have been
trained in the safe use of cryogens, customers will have bare
hands and most likely will not have had any training in the pos-
sible dangers from the extreme cold of cryogenic materials.
Probably the most important message here is that it is imperative
never to allow anyone to ingest liquid nitrogen or dry ice. Serving
staff need to make sure all cryogens have evaporated from any
food before presenting it to the customer.
Cryo-​Poaching
The Nitro Poached Green Tea and Lime Mousse served as an
amuse bouche at the Fat Duck from 2001 can be seen as the
origin of modern use of cryogens in restaurants. As Heston
Blumenthal’s Fat Duck became more and more famous and other
leading chefs started visiting, so the dish was copied and modi-
fied around the world. This is a relatively simple dish to prepare.
Basically a mousse made from green tea, lime juice and vodka
is put in a cream whipper. At the table, the waiter squirts out a
little of the mousse onto the surface of a bowl of liquid nitrogen,
where it floats –​the waiter then carefully spoons more liquid
nitrogen over the mousse until it forms a hard external shell.
This is picked up with a spoon, any liquid nitrogen remaining is
allowed to drain off, and then it is placed on a plate in front of the
diner, who is told to wait a few moments and then pick it up and
eat it whole. The result is a melt-​in-​the-​mouth immediate hit of
flavour as the whole mousse initially crunches and then dissolves
in the mouth. At the same time, the cooling effect from the still
very cold mousse makes some of the moisture in the mouth con-
dense, so that when the customer breathes out through their nose
some foggy water vapour is emitted, giving the appearance of a
dragon’s breath, much to the amusement of their companions.
Liquid Nitrogen Ice Creams
Preparing ice cream at the table is always a spectacle, but also,
and more importantly, by making the ice cream as quickly as pos-
sible you achieve an incredibly smooth and creamy texture. The
smaller the size of the ice crystals in ice cream, the smoother it
tastes. Our tongues can detect solid particles as small as 1/​500th
of a millimetre, so keeping the ice crystals smaller than that will
ensure the ice cream is truly smooth. As anyone who has tasted
freshly prepared liquid nitrogen ice cream will attest, it can have
a truly wondrous texture. However, it is not quite as simple as it
may at first sight appear to achieve this end. First, speed matters –​
as soon as any ice crystals form, they also start to grow larger. In
the first few minutes after making some ice cream, the overall
average crystal size will rapidly increase as the smallest crystals
melt and larger ones just grow larger. So, it is important both
to ensure the actual crystallization takes place at as low a tem-
perature as possible (by adding and mixing the liquid nitrogen as
quickly as possible, getting the maximum possible cooling rate)
and then to serve it as soon as possible after it has frozen (but
always remembering that it cannot be served until all the liquid
Peter Barham
nitrogen has evaporated and there are no super-​cooled lumps in
the ice cream).
To achieve the best results, you need first to know how much
liquid nitrogen is required to freeze the amount of ice cream
being made. There is no easy answer to this, as it depends not
only on the water content of the ice cream mix but also, and
more importantly, on how much of the nitrogen boils off into the
surrounding atmosphere rather than just cooling the ice cream
mix. So, you will need to experiment to find out just what is
required in any particular situation. But as a very rough guide,
and a starting point for any experimentation, I find I use about 0.8
litres of liquid nitrogen to make 1 litre of ice cream.
My favourite use of liquid nitrogen to make ice cream is the
one used to make the bacon and egg ice cream that formed part of
the ‘breakfast’ dessert at the Fat Duck some time ago. The waiter
bought out a plate with the main ingredients of a full English
breakfast: ‘fried bread’ represented by a sweet pain perdu, some
tomato jam and some caramelized bacon. Next the waiter, stating
that they were going to prepare some scrambled eggs, cracked a
couple of eggs into a frying pan, added liquid nitrogen and stirred
until an ice cream the consistency of scrambled eggs was ready,
and then served that onto the plate alongside the other elements.
The ice cream mix, which had been injected into previously
blown egg shells, was a slightly over-​cooked custard infused with
bacon that had a remarkable bacon and egg flavour. The overall
multisensory experience and the juxtaposition of breakfast and
dessert makes this still one of my personal favourite dishes of
all time.
Fogging Odours
A further use for cryogens in the restaurant is to provide localized
odours at a table to correspond with the food being served. If water
(preferably warm water) to which the required aroma has been
added is poured over some cryogenic material (dry ice generally
gives the best effect), then the resulting clouds of foggy vapour
will carry along the volatile odorants –​the effect is somewhat
localized, as the cool vapour droops over the vessel containing
the dry ice and then flows over the table before dropping to the
floor and dispersing –​thus, it is possible to limit the main effect
of the odour to just the particular table where it is wanted.
Liquid Nitrogen Cocktails
Over the past decade or so, many cocktail bars have started
incorporating liquid nitrogen into their repertoire. A few drops of
liquid nitrogen added to a cocktail give it that smoky effect that
can excite customers. Holding empty glasses by their stems and
dipping them in liquid nitrogen before adding the cocktail ensures
a very cold glass to keep the drink cold as well as allowing a trail
of smoky fog to trail from the glass when held.
Anyone using liquid nitrogen in this way needs to be very
aware of the inherent risks and must ensure that all traces of liquid
nitrogen have evaporated before the cocktails are consumed.
There have been a number of serious incidents where staff were
not properly trained or procedures not followed. For example,
in 2015, Oscar’s Wine Bar and Bistro in Lancaster was fined
Cryogenics in the Kitchen
£100,000 as a result of an incident where a customer was taken
to hospital and had to have a large part of her stomach removed.
It was found that no proper risk assessment had been carried
out and bar staff had not received proper training or adequate
warnings of the importance of not drinking the cocktail until all
the nitrogen had boiled off.
Safely Storing, Handling and Using Cryogens
Training
Cryogens are not commonplace materials, so it is necessary to
ensure that anyone dealing with them is properly trained so they
understand the risks involved and know the correct procedures
to follow. While much of this may be thought of as really little
more than common sense, for example, just as most people do
not need to be told not to grab hold of the metal handle of a pan
that has just come out of a hot oven with their bare hands, not
everyone will automatically appreciate that it is probably even
more dangerous to grasp the metal handle of an uninsulated pan
that is full of a cryogen. Both situations can cause severe burns,
but when you hold a very hot or cold pan, your immediate reac-
tion is to let it go as quickly as possible. This is fine for the
hot pan –​you can let it go and the burning will stop; but with
the cold pan, as soon as you grab it, the moisture on your hand
will turn to ice and your hand will become stuck to the pan so
you cannot let go; the damage continues for some time and you
can get a very severe cold burn, or frostbite. Hence the need for
proper training.
Training can take many forms: on the job training from a very
experienced person who has had some formal training and is able
to pass on the necessary knowledge; online courses, external
training courses run by professionals; or, probably the best, an
external expert trainer giving workplace training to all the staff.
All universities and colleges that use cryogens have to provide
training for their staff and students and so run regular courses.
Most universities are willing to allow external people to attend
their courses; for example, many local chefs have been trained in
the use of cryogens in my own Physics department by our cryo-
genics manager. All cryogen suppliers will be able to point their
customers to appropriate training resources.
One very important aspect of providing staff training in cryo-
genics is that in the event of any accidents occurring, there will
be a clear paper trail demonstrating that proper training has taken
place. Without evidence of proper training, in many countries,
there may be considerable legal liability for negligence.
Equipment
To store cryogens, properly insulated vessels are required –​most
suppliers of cryogens will only deliver to you if you have an
appropriate container. For liquid nitrogen, a proper dewar (named
after James Dewar, who invented the idea of using a vacuum flask
to hold cryogenic liquids) is required. The dewar may be com-
pletely open so the liquid is constantly simmering, or it may be
closed and held under a small pressure with a safety vent valve
179
to prevent build up of pressure inside as the nitrogen boils off.
All liquid nitrogen-​filled dewars will give off a slow stream of
cold nitrogen gas as it slowly boils off. The larger the dewar and
the better its vacuum, the longer the liquid will last before it all
evaporates. Most restaurants find that medium-​sized dewars (typ-
ically around 40 litres), which they have re-​filled weekly are
sufficient. To store dry ice, an open dewar is best, but a well-​
insulated Styrofoam box, such as a cool box, can suffice.
Note that liquefied gases must never be stored in uninsulated
or sealed containers –​if an uninsulated container is used, the out-
side will cool to cryogenic temperatures and pose a major risk to
anyone accidentally touching it. If a cryogen is stored in a closed
container, then as it evaporates the internal pressure will build up,
most likely resulting in the eventual fracture of the container in a
powerful explosion.
When transporting cryogens, even within the premises for
short distances, it is essential to follow proper safety guidelines –​
one of the most important being never to enter a closed space
(such as a lift) with a container of a cryogen, as if you are trapped
or the cryogen spills, the resulting gas will displace the air in the
room or lift, leading to asphyxiation.
In the restaurant, the best way to carry and use liquid nitrogen
is to obtain a small stainless steel dewar –​these are available from
many scientific suppliers –​which can be used as required and
will fit in with any modern restaurant décor.
When working with cryogens you need to wear appropriate
protective clothing –​thermal gloves and eye protection are usu-
ally the most important items. But you also need to be aware of
risks of spills and splashes, so make sure that all clothing (espe-
cially footwear) worn is non-​absorbent and will not trap any
cryogens in the weave.
Risk Assessment
As with any process that carries the risk of injury, you should
always perform a risk assessment. For cryogens, you can find
many examples of standard risk assessment questions for storing,
handling and use on the internet, which you can adapt for your
own particular circumstances. Briefly, you will need to assess and
quantify the various risks associated with the use of cryogens:
small spillages that can cause burns, larger spillages that may
lead to oxygen depletion, the presence of very cold surfaces and
the attendant risk of frostbite or cryo-​burns if people come into
contact with them, as well as the risk to customers from any
accidental ingestion of cryogens. For each identified risk, you
will need to develop procedures to reduce the likelihood of any
potential injury to an acceptable minimum. This can be through
the use of personal protective equipment, staff training detailed
working practices, etc. Finally, you will need to ensure there
are procedures in place to make sure that all the instructions are
followed by all staff. For example, once you have prepared risk
assessments and procedures for storing, transporting and using
cryogens, you should make sure that all staff who may come into
contact with cryogens have read and are fully aware of these. It
is also advisable to make sure they are checked and revised as
necessary, at least annually if not more often.
180
As cryogens can be dangerous if not handled correctly, you
should display appropriate signage in the areas where they are
used alongside copies of the risk assessments.
REFERENCES
Alhamdan A, Hassan B, Alkahtani H, Abdelkarim D, Younis M.
2001. Cryogenic freezing of fresh date fruits for quality pres-
ervation during frozen storage. Journal of the Saudi Society of
Agricultural Sciences, 17(1), 9–​16.
Allen D. 2007. Pre fermentative cryo maceration. Australian & New
Zealand Grapegrower & Winemaker, August, 59–​64.
Awonorin SO. 1997. An appraisal of the freezing capabilities of
tunnel and spiral belt freezers using liquid nitrogen sprays.
Journal of Food Engineering, 34(2), 179–​192.
Balasubramanian S, Gupta MK, Singh KK. 2012. Cryogenics and its
application with reference to spice grinding: A review. Critical
Reviews in Food Science and Nutrition, 52(9), 781–​794.
Biglia A, Comba L, Fabrizio E, Gay P, Aimonino DR. 2016. Case
studies in food freezing at very low temperature. In Masoero
MC, Arsie I (Eds.), 71st Conference of the Italian Thermal
Machines Engineering Association, 101, 305–​312.
Cargill RW, Macphee DE.1981. The solubility of carbon dioxide in
mixtures of water and ethanol. Journal of Chemical Research,
S(8), 232.
Carrington AK, Goff HD, Stanley DW. 1996. Structure and stability
of the glassy state in rapidly and slowly cooled carbohydrate
solutions. Food Research International, 29(2), 207–​213.
Carrington AK, Sahagian ME, Goff HD, Stanley DW. 1994. Ice crys-
tallization temperatures of sugar/​polysaccharide solutions and
their relationship to thermal events during warming. Cryo-​
Letters, 15(4), 235–​244.
Coppola BP, Hovick JW, Daniels DS. 1994. I Scream, You Scream …
: A new twist on the liquid-​nitrogen demonstrations. Journal of
Chemical Education, 71(12), 1080.
Davidge H. 1981. Cryogenics in the food-​ industry. Cryogenics,
21(5), 287–​290.
Di Giovacchino L, Sestili S, Di Vincenzo D. 2002. Influence of olive
processing on virgin olive oil quality. European Journal of
Lipid Science and Technology, 104(9–​10), 587–​601.
Dumas JB. 1877. Liquéfaction de l’oxygène. Comptes rendus de
l’Académie des sciences, 85, 1212.
Peter Barham
Gomez TP, Calvelo A. 1982. On the ice growth mechanisms during
beef freezing. Proceedings of the Meeting of Commissions C2,
D1, D2 and D3 of the I.I.F., Hamilton, New Zealand.
Gordillo B, Lopez-​Infante MI, Ramirez-​Perez P, Gonzalez-​Miret
ML, Heredia FJ. 2010. Influence of prefermentative cold
maceration on the colour and anthocyanic copigmentation
of organic Tempranillo wines elaborated in a warm cli-
mate. Journal of Agricultural and Food Chemistry, 58(11),
6797–​6803.
Kurti N, This-​Benckhard H. 1994. Chemistry and physics in the kit-
chen. Scientific American, 270(4), 66–​71.
Malecki GJ, Shinde P, Morgan AI, Farkas DF. 1970. Atmospheric
fluidized bed freeze drying. Food Technology, 24(5), 601–​603.
Marshall AB. 1901. The Table. 24 August 1901.
Marshall AB, Weir R, Brears P, Deith J, Barham P. 1988. Mrs
Marshall: The Greatest Victorian Ice Cream Maker, Including
a Facsimile of Her 1885 Book of Ices. London: Smith Settle
Ltd for Syon House.
Martino MN, Zaritzky NE. 1988. Ice crystal size modifications
during frozen beef storage. Journal of Food Science, 53(6),
1631–​1637.
Miller JP, Roberts W. 2001. How to minimize startup costs. Process
Cooling and Equipment. Retrieved from www.process–​
cooling.com.
Myhrvold N, Young C. 2011. Modernist Cuisine. Washington:
Bellevue.
Parenti A, Spugnoli P, Calamai L, Ferrari S, Gori C. 2004. Effects of
cold maceration on red wine quality from Tuscan Sangiovese
grape. European Food Research and Technology, 218(4),
360–​366.
Ramakrishnan S, Wysk RA, Prabhu VV. 2004. Prediction of pro-
cess parameters for intelligent control of tunnel freezers using
simulation. Journal of Food Engineering, 65(1), 23–​31.
Salinas MR, Garijo J, Pardo F, Zalacain A, Alonso GL. 2005.
Influence of prefermentative maceration temperature on the
colour and the phenolic and volatile composition of rose
wines. Journal of the Science of Food and Agriculture, 85(9),
1527–​1536.
Sanz PD, de Elvira C, Martino M, Zaritzky N, Otero L, Carrasco JA.
1999. Freezing rate simulation as an aid to reducing crystal-
lization damage in foods. Meat Science, 52(3), 275–​278.
Dairy: Milk Gels –​a Gastrophysics View
Judith Hege, Marta Ghebremedhin, Bhagyashri L. Joshi, Christine Schreiber, H.-​C. Gill and Thomas A. Vilgis
Max-​Planck-​Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany
This chapter explores the gastrophysical potential of bovine
milk. This specific colloidal structure, defined by casein micelles
embedded in water and proteins (whey), can be the basis for
unusual applications that are derived from the two physical
ways of milk destabilization, i.e., adding rennet or lowering the
pH. Both methods yield casein-​based particle gels with roughly
similar structures but different taste and mouthfeel. Both fun-
damental principles can be extended to new gastronomic ideas
with controlled textures and taste. Finally, we discuss gels made
from colostrum, which are quite common in different Nordic and
Asian (Indian) cultures.
Introduction
Milk is a very common food in most cultures (Belitz et al.,
2008). The consumption of non-​human milk by humans started
at different places independently, with indications dating back
to as early as about 9000–​7000 BC in Mesopotamia (Bellwood,
2005). It is a white, opaque liquid and is defined as the liquid
secreted by the mammary glands of female mammals. It is the
only food for the new-​born and contains all important nutrients.
This chapter will focus on bovine milk and the physics of
derived dairy products such as yogurts and cheese; we focus on
the basic physical process, rather than food technology. Indeed,
these aspects provide deep insight into the gastrophysics of nat-
ural colloids, and their molecular view yields new ideas in the
molecular gastrophysics of food derived from milk and thus,
molecular cooking.
Raw milks (before pasteurization, heat treatment and hom-
ogenization) do not always have exactly the same composition.
The precise composition of the fat and protein fraction depends
on the breed, feed, season, stage of lactation and many other
factors. Thus, it is necessary to consider the “universal” aspects
of the milk only, which are the properties of the whey proteins
and the caseins, and how they react to pH changes, tempera-
ture and time scales. Therefore, the main emphasis will be on
native raw milk and its gelling to yogurt by changing the phys-
ical parameters accordingly. To form gels, native milk needs pH
changes or enzymes (e.g., chymosin from rennet) (Vilgis, 2015).
In contrast, early cow’s milk (colostrum) gels by simply raising
the temperature, due to a high concentration of immunoglobulins
and other key proteins that are denatured during the heat treatment
and subsequently rearrange themselves.
The chapter is organized as follows. To understand the different
physical treatments of milk, a general introduction to the physical
nature of the milk structure is given. From this, we will discuss
well-​known procedures to form soft particle gels, which allow
us to determine which part of the milk is affected by tempera-
ture changes and pH changes. In the last part, we provide new
insights into the early milk of cows (colostrum), rich in protein
molecules such as, for example, IGF-​1 (insulin-​like growth factor
1), which will form, together with other proteins, including the
“classical” whey proteins α-​lactalbumin or β-​lactoglobulin, very
soft elastic gels.
Native Milk, a Simplified Picture
The main constituents of raw milks are water (about 87%), lipids
(3.8–​4%) and proteins (about 3.6%) (Vilgis, 2014), which can
be as high as 7% lipids and 6% proteins depending on the breed,
for example in Jersey cows. Water is the continuous phase for
the colloidal dispersion of fat droplets and colloidal particles of
casein micelles made from proteins and mineral salts; it is also
the solvent for mineral nutrients. In milk, lipids are dispersed in
droplets with a broad size distribution from 0.2 μm to 20 μm and
an average size of about 3–​6 μm (Töpel, 2016). Their average
density is 0.9 g/​mL, which is lower than that of water (1.0 g/​mL).
In native milk that has not been homogenized, lipid particles
aggregate and move to the surface of the milk due to their lower
density (a phenomenon referred to as creaming). Since the lipid
droplets are in an aqueous environment, they are stabilized by
emulsifying agents that prevent coalescence and formation of
larger droplets (Vilgis, 2014). Like all nutritional fats, milk fat
mainly consists of triacylglycerols (TAGs), with three fatty acid
(FA) residues esterified by a glycerol residue. In bovine milk,
there are many different FA residues (about 430 different kinds,
but 15 make up the majority), with 4 to 24 carbon atoms (Töpel,
2016; Haug et al., 2007). The chain length and the saturation
degree of the carbon–​carbon bonds in the FA residues deter-
mine the melting point of the TAGs. At room temperature, some
TAGs are liquid (TAGs having short or unsaturated FA residues),
whereas others (consisting of long and saturated FA residues)
181
182
are already crystalline. In a fat droplet, the liquid fats will be
found in the middle of a fat droplet spherically surrounded by the
crystallized fats (Gallier, 2010; Vilgis, 2014). When the milk is
heated, the crystalline phase melts.
Stability under heating is ensured by a complex emulsifying
layer around each fat droplet. This layer consists of a monolayer
of phospholipids and a membrane (Gallier, 2010; Vilgis, 2014).
A phospholipid is mainly constructed from a glycerol residue
esterified with a phosphate group and two FA residues. The phos-
phate group is hydrophilic, and the hydrocarbon chains of the
FA residues are hydrophobic. The membrane is built by a double
layer of phospholipids. The hydrophilic heads (phosphate) of the
phospholipids are arranged at the outside of the membrane, and the
hydrophobic tails (alkyl chain) towards the inside (Alberts et al.,
2008). In a biomembrane, proteins and enzymes are found as well
as cholesterol, which makes the membrane flexible (Vilgis, 2014).
Two types of proteins are present in milk: caseins, which
are one of the main components of cheese, and whey proteins,
which are present in the whey (mainly water) that arises as a by-​
product of the cheese making process and is used today by many
bodybuilders as a source of protein. Milk contains about 3.3%
protein, about 2.7% of which is caseins and about 0.6% whey
proteins (Belitz et al., 2008). Whey protein contains residues of
all nine essential amino acids, which cannot be produced by the
human body and therefore have to be consumed via the food. The
whey proteins include the globular proteins α-​lactalbumin, β-​
lactoglobulin, immunoglobulins and serum albumins.
There are four types of caseins: αS1-​, αS2-​, β-​ and κ-​caseins
(Ternes, 2008). Except for the κ-​casein, they are insoluble or only
FIGURE 26.1 Simplified model of casein micelles. The submicelles (red) in the centre contain the hydrophobic α-​ and β-​caseins, whereas the perimeter
contains mainly hydrophilic κ-​caseins.
Judith Hege et al.
scarcely soluble in water, as only κ-​casein contains a hydrophilic
part. This is the reason why the caseins in milk form micelles
(Vilgis, 2014), i.e., spherical structures stabilized by amphiphilic
molecules (molecules that include hydrophilic as well as hydro-
phobic parts). In a watery environment, the hydrophilic parts of
these molecules will arrange themselves at the outside, whereas the
hydrophobic parts are in the interior of the sphere. In Figure 26.1,
a simplified model of a casein micelle is shown. The αS1-​, αS2-​
and β-​caseins in the inner part of the micelles form hydrophobic
submicelles (in red) that are connected via calcium phosphate
bridging bonds (black dots). The κ-​casein molecules are situated
on the outside of submicelles to form the overall micelle surface.
Thus, the hydrophilic parts of the κ-​caseins (blue arms) are facing
the watery environment, ensuring solubility in the whey (Tuinier
and de Kruif, 2002). The hydrophilic part of the κ-​casein shows a
weak negative net charge in raw milk, where the pH is about 6.8.
Therefore, two mechanisms contribute to the casein stability: steric
hindrance by the κ-​casein arms and the overall negative net charge
of the micelles at their surface. Casein micelles are thus highly
stable even when the milk is heated. Only whey proteins denature
(at around 70 °C) and form the basis of the skin during cooking.
The volume-​average radius of a casein micelle is about 61 nm
(de Kruif et al., 2012), much smaller than the fat particles present
in milk. The density of the casein micelles is 1.078 g/​mL (de
Kruif et al., 2012; Walstra et al., 1999), slightly higher than that
of water. Casein micelles still do not sediment but are uniformly
distributed in the whey due to different interactions.
The hydrophilic arms of the κ-​casein that extend into the
whey stabilize the system sterically. For entropic reasons, it
Dairy: Milk Gels – a Gastrophysics View 183

is unfavourable for the arms of two different casein micelles

that come close to each other to interpenetrate (Vilgis, 2015;
McMahon and Oommen, 2013). For this reason, the radius of
the micelle is basically extended by the length of the arms. Thus,
steric repulsion has a very short range, since the spheres have to
come as close as r = R + 2a, where R is the radius of the casein
micelle without the arms and a is the length of one arm.
Besides the steric effect, there is also an electrical repulsion.
Like all proteins, κ-​casein consists of amino acid residues that
can be electrically charged depending on their lateral groups. The
charge depends on the pH of the environment. At the surface of a
casein micelle, charges add up to a net charge which is the same for
all casein micelles, which leads to a repulsion of the micelles. At
the native pH of milk (6.7), casein micelles are negatively charged.
The charge repulsion is weakened by free ions (salts and mineral
nutrients) present in the whey. Consequently, the electrical inter-
action is strongly dependent on the amount of free ions present
in the milk. If an acid is added to the system, protons (positive
charges) are added, and the charge of the κ-​casein arms is screened.
In this case, casein micelles form large agglomerates, because their
distribution in the whey is no longer stabilized by electrical repul-
sion. The electrical repulsion is a longer-​range interaction than the FIGURE 26.2 The fundamental colloid interactions define the stability of
steric repulsion. An attractive interaction (due to van der Waals the casein micelles in milk: steric repulsion (by the stretched arms of the κ-​
attraction) can only be observed after elimination of all repulsive casein), charge repulsion (negative net charge of the κ-​casein arms) and van
der Waals attraction (hydrophobic core). These basic interactions define the
effects, e.g., by lowering the pH as in yogurt production (Tuinier
different methods of milk treatment for cheese and yogurt: either manipulate
and de Kruif, 2002). Figure 26.2 illustrates the impact of the three
the charge and the steric interaction by changing the pH-​value, or cleave the
interactions: steric repulsion, electrical repulsion and van der
κ-​casein arms by enzymes like rennet.
Waals attraction. It stands out from the diagram that steric inter-
action and electrical interaction are repulsive effects, whereas the (Tuinier and de Kruif, 2002; redrawn Vilgis, 2015)
van der Waals interaction is attractive. Also, it can be seen from
the diagram that charge repulsion is a longer-​range interaction than huge but loose clusters are formed. At some point, the casein
steric repulsion (Tuinier and de Kruif, 2002). micelles are caught in a particle gel. As they can no longer move
freely but only locally, a soft gel is formed (Haque et al., 2001;
Vilgis, 2015). Figure 26.3 illustrates this process of gel for-
mation. The diagram on the left shows how repulsion between
Yogurt and Its Production
micelles decreases with decreasing pH. For pH 6.6 (close to the
Yogurt formation occurs when bacteria called lactobacilli are pH of native milk), a strong repulsion can be observed. At pH
responsible for milk fermentation. Classically, Lactobacillus 5.0, the interaction already has a shorter range. At pH 4.8, an
delbruekii subsp. bulgaricus and Streptococcus thermophilus are attraction can be observed for particles that touch each other, but
most commonly used (Ternes et al., 2005). The lactobacilli bac- for particles at a small distance, a potential wall has to be passed
teria convert the sugar lactose that is naturally present in the milk in order to attract another micelle, whereas at pH 4.6 no repulsion
into lactic acid. can be observed. The schematic representation on the right shows
To produce yogurt, a small amount of natural yogurt or lacto- how by lowering the pH of a stable casein dispersion, a gel is
bacilli is added to (raw) milk and stirred well. Depending on the formed over time.
desired consistency and the particular species of bacteria being At higher fermentation temperatures of about 45 °C, the con-
used, the milk is fermented at temperatures between 25 °C and sistency of the yogurt will be firmer. Whey and fat particles are
46 °C. At these temperatures, the lactobacilli accrete and produce caught in the particle gel. Thus, the gel strength (firmness of the
enzymes, which metabolize the milk sugar lactose into lactic yogurt) and also the mouthfeel are dependent on the structure of
acid. With time, the pH decreases: due to this acidification, casein the particle gel (Vilgis, 2015).
micelles are destabilized and agglomerate. With increasing
time, larger agglomerates are formed, and consequently the vis-
cosity increases. This process starts at random positions, leading
to agglomerates of different size and shape. With the rising Yogurt from Raw and Boiled Milk
number of aggregated casein micelles, the mass of the agglom- In the experiment that we describe now, yogurt was produced
erate increases, leading to an increase in viscosity. With further from fresh raw milk. The milk came from the farm of H.-​C.
decrease of the pH, more and more agglomerates form. With a Gill in Bodenheim, close to Mainz, Germany. Two types of
rising number of agglomerates and their simultaneous growth, yogurt were produced. For one type, the raw milk was directly
184 Judith Hege et al.

10
pH=6.6
lowering pH-value

5 pH=5.0
Energy (kBT)

pH=4.8
0

pH=4.6

–5

–10 milk stable casein forming aggregates particle gels

0 2 4 6 8 dispersion (collapsed κ-casein arms)
Distance (nm) time

FIGURE 26.3 Sketch of the formation of a particle gel (Tuinier and de Kruif, 2002; Vilgis, 2015) by lowering the pH-​value, which acts systematically on the
charge and conformation of the κ-​casein arms.

fermented with lactobacilli from natural yogurt. The natural

yogurt contained lactobacilli from the strains Lactobacillus
acidophilus, Bifidobacterium bifidum and Lactobacillus casei.
For the other yogurt, the raw milk was first heated to boiling
and then cooled down in a water bath. All other procedures were
the same for both types of yogurt, and the fermentation took
place at 35 °C for 15 h. After fermentation, the two types of
yogurts showed a very different consistency. The yogurt made
from boiled milk was much firmer than the one made of non-​
heated raw milk. Also, the odour of the yogurt of the two types
was different. The one made from unheated milk smelled more
“acidic” than the other one. Additionally, both samples had a
cream layer at the surface, which is normal for milk that was
not homogenized. The viscosity of both yogurts was measured
using a rheometer.
FIGURE 26.4 Newtonian, shear-​
thinning and shear-​
thickening flow
behaviour.

Rheology Basics
Rheology is the science of deformation (solids) and flow behav-
iour (liquids) of materials. In a rheometer, different types of tests
can be done to investigate the behaviour of a material under
shear. There are three main ways a material can behave under
shear: Newtonian, shear-​thinning, and shear-​thickening behav-
iour. The ideal or Newtonian flow behaviour is shown, e.g., by
water, where the shear stress τ is proportional to the shear rate
γ̇. The Newtonian law τ = η × γ holds, where η is the dynamic
viscosity. Shear-​thinning as well as shear-​thickening materials
show a non-​linear relation between shear stress τ and shear rate γ̇ .
For shear-​thinning materials, the viscosity decreases under shear
strain, whereas for shear-​thickening materials, it increases. In a
shear stress –​shear rate diagram (Figure 26.4) – the slope of the
graph decreases with increasing shear rate for shear-​thinning flow
behaviour. For shear-​thickening materials, the slope increases
with increasing shear rate in a τ–​γ̇ diagram.
To investigate the viscoelastic and flow behaviour of the two
different yogurts, an amplitude sweep and a flow sweep were
performed. An amplitude sweep is a type of oscillatory test in
which the amplitude of the deformation γ or the shear stress τ is
varied, while their frequency and the temperature of the sample
are kept constant (Mezger, 2010). The results of an amplitude
sweep are in terms of the sample’s loss modulus and storage
modulus in dependence on either oscillation strain or shear stress.
The storage modulus, G′, is a measure of the deformation
energy stored in the material during the shear process. This
energy is fully available again after relaxation from the shear
and drives the back deformation. The deformation of the material
caused by the shear is compensated fully or partly after relaxation
of the system. Materials that completely save the deformation
energy are the same after relaxation from the shear as they were
before shearing. The elastic behaviour of a sample is described
Dairy: Milk Gels – a Gastrophysics View
by the storage modulus G′, which refers to the solid fraction of
the material.
The loss modulus, G″, describes the viscous behaviour of a
material and is a measure of the energy that is used in the shearing
process to change the structure of the material or is emitted to the
environment as heat. Materials that flow have a high loss modulus
because molecules or particles of the material move relative to
each other and rub each other. The friction produces heat, which
is emitted to the environment or heats up the sample. Substances
with a loss modulus are different after the shearing process than
before it. G″ represents the liquid fraction of a material.
Rheological Measurements of Yogurts
For the yogurts, the diagram of the amplitude sweep is shown in
Figure 26.5. The loss modulus and storage modulus are displayed
in dependence on the oscillation strain. A constant part of the
graph can be seen in all four graphs, and then at some point the
slope becomes negative. The point at which this change happens
is called the yield point. At this point, the gel starts to break. At
even higher oscillation strain, there is a point where the curves of
the storage modulus and the loss modulus for one sample inter-
sect. This point is called the flow point, because at this point the
material starts to flow. It is evident that both loss modulus and
storage modulus have much higher values for the yogurt made
from boiled milk than the yogurt made from unboiled milk. This
confirms the visual observation that the yogurt produced from
boiled milk is firmer than the other one.
Furthermore, the viscosity of both yogurts was measured using
the flow sweep method. Figure 26.6 shows the diagram of the
flow sweep. Here, the viscosity η was measured and displayed in
dependence on the shear rate γ̇. Each measurement was performed
three times. Shown in the diagrams are the mean values. All
measurements were performed at 20 °C.
FIGURE 26.5 Amplitude sweep of the two yogurt samples. The yogurt
made from boiled raw milk shows higher values for the moduli, indicating
a firmer gel.
185
FIGURE 26.6 Flow sweep of the two yogurt samples. The yogurt made
from boiled milk shows higher viscosity.
One can see from the diagram that both yogurts show very
similar behaviour. First, a negative slope is shown. With rising
shear rate, the viscosity decreases. This behaviour is typical
of shear-​thinning materials. Therefore, we can learn from the
measurement that both yogurt types are showing shear-​thinning
behaviour. At shear rates of about 10 s−1, there is a narrow almost
Newtonian sector observed before the viscosity decreases again
with further increase of the shear rate. The parallel behaviour
of both yogurt types indicates that both yogurts have a very
similar structure. However, the viscosity of the yogurt made from
boiled milk is higher than the viscosity of the one produced from
unboiled milk. This is congruent with the observation from the
amplitude sweep and visual experience.
Such differences in the rheological behaviour of boiled
and unboiled yogurt could be explained by the following two
reasons:
(a) Native, raw milk contains many different
microorganisms from the cow. With time, these bac-
teria change certain properties of the milk. On boiling
the milk, all these bacteria are killed, so they no longer
have any influence on the milk. During the fermenta-
tion of the yogurt, it is kept at a temperature at which
bacteria are very active, because the activity of the
lactobacilli is desired. However, in the unboiled milk,
the other bacteria from the cow are also still active.
This difference could be a reason for the difference in
the viscosity of the two types of yogurt.
(b) During the boiling of the milk, whey proteins are
denatured. When denatured, the arrangement of the
protein structure is changed. The initial links between
different amino acid residues of the whey protein are
broken and new links can be formed. If there are free
positive charges, like, for example, bivalent posi-
tively charged calcium ions, they can bind to nega-
tively charged amino acid residues of whey proteins.
A network of ionic links is formed. This leads to
aggregation of the whey proteins. More clusters can
186
be formed than before, leading to enhanced viscosity.
Caseins are very stable against heat treatment. They
only start to show structural changes at about 120 °C
(Ternes, 2008). This does not necessarily mean that
the micelles are also stable at these temperatures. If
the casein micelles were disrupted, calcium would be
set free because the calcium phosphate bridging bonds
would be broken. The release of more calcium ions
would enhance the described effect. However, even if
the micelles are not disrupted, there are enough free
ions in the milk, and the whey proteins are denatured
during the heat treatment of the milk. Thus, it is highly
probable that the denaturation of the whey proteins
enhances the viscosity of the yogurt gel. For further
literature see, e.g., Lucey and Singh, 1998; Vasbinder
et al., 2003; Koutina et al., 2016.
Colostrum, a Brief Insight
The first milk produced by a mammal after giving birth is called
colostrum. It is given by the mother shortly before and during the
first four to five days after the birth (Ternes et al., 2005; Eisenbrand
and Schreier, 2006). Colostrum looks yellowish-​brownish, and
its consistency is thicker than that of the milk produced later on
(Eisenbrand and Schreier, 2006). Since the newborn mammal is
unable to catch, collect, eat and digest any solid food, it has to
rely fully on its mother’s milk. Colostrum not only provides a
complete diet to the neonate, providing all essential nutrients,
but also contains a high amount of immunoglobulins, which are
essential for the immunological protection of the newborn, espe-
cially for ruminants. For ruminants, like cows, for example, no
immune factors are exchanged between mother and fetus in the
uterus. Without the passive immunization of the neonate via the
colostrum, the ruminant young animal would not have a chance
of survival (Larson et al., 1980; Stelwagen et al., 2009). The dom-
inant immunoglobulin in bovine milk is IgG1. However, others
like IgG2, IgA and IgM are present as well, even though in much
smaller concentrations. All these immunoglobulins show a higher
concentration in colostrum than in normal milk. The concentra-
tion of IgG1 in bovine milk is about 80 times higher in colostrum
than in normal milk (Stelwagen et al., 2009).
The protein content in colostrum is about four to six times
higher than in normal milk. It contains much more whey protein
(about 20 times more) and also immunoglobulins, as mentioned
before. There is also a higher amount of mineral nutrients
and vitamins in colostrum compared with milk (Töpel, 2016;
Eisenbrand and Schreier, 2006). Furthermore, it has been found
that immunoglobulins present in bovine colostrum are specific
to human pathogens and thus can be used for immunization of
people (McConnell et al., 2001).
In different countries, colostrum is used to make cheese or
pudding. For example, the Finnish leipäjuusto, which means
“bread cheese”, is a type of squeaky cheese (Neher, undated;
Worldnews Inc., undated). In India, people make junnu or
kharvas out of colostral milk. Both terms describe the same
food, but different regions and languages have different names
Judith Hege et al.
for this dessert. It is a pudding that is served either warm or cold
(Swasthi, 2019).
Creaming of Colostrum
To characterize the rheological properties of colostrum and its
different behaviour under heating compared with normal milk,
several measurements were taken. The colostrum for these
measurements came from the same farm as the raw milk for the
yogurt experiments (H.-​C. Gill, Bodenheim). A phase separation
naturally occurred in the colostrum; the upper phase looked more
yellow and the lower phase whiter. For the characterization of the
colostrum, three phases were probed: the upper phase, the lower
phase and a sample of the colostrum after stirring and thus mixing
upper and lower phase, which is referred to as mixed phase.
To learn more about the different phases, differential scanning
calorimetry (DSC) was used to measure the thermal properties
for each of the colostrum samples. The temperature was varied
from 20 °C to 70 °C at a rate of 1 K/​min and then kept at 70 °C for
30 min. The DSC results of all phases are shown in Figure 26.7.
A large endothermic peak for the upper phase can be seen at
44 °C and a smaller one for the mixed phase, whereas the lower
phase did not show any peak at this temperature. The integra-
tion over the peak gives the enthalpy change ΔH: it is a measure
of the absorbed or supplied heat in an enclosed system. So, the
enthalpy change was higher for the upper phase than for the mixed
phase. This means that the upper phase needs a higher enthalpy
for melting. From the temperature ramp (green line), one can see
at what temperature the endotherm peak occurs and thus conclude
which material is melting. In this case, it is the milk fat, which
melts at about 44 °C. This means that the upper phase must contain
most fat. So, the measurement shows that, like normal raw (not
homogenized) milk, colostrum also shows creaming behaviour.
To gain more insight into the three different phases, light
microscopy was used. Figure 26.8 shows pictures taken with
a light microscope using polarized light, under which the fat
droplets can be easily distinguished. To observe the behaviour of
FIGURE 26.7 Differential scanning calorimetry (DSC) of the different col-
ostrum phases. A peak occurred at 44 °C, which can be assigned to milk fat.
Dairy: Milk Gels – a Gastrophysics View 187

FIGURE 26.8 Microscopy pictures of colostrum. (a) Microscopy with polarized light of the upper phase at 20 °C (left) and 70 °C (right). (b) Microscopy
with polarized light of the mixed phase at 20 °C (left) and 70 °C (right). (c) Microscopy with polarized light of the lower phase at 20 °C (left) and 70 °C (right).

the fat droplets under heating, the same procedure as in the DSC
measurement was used. The pictures on the left were taken before
heating the samples, and the ones on the right after heating.
Similarly to the DSC results, these pictures indicate that there is
less fat in the lower phase than in the upper phase. It also can be
seen that after the milk was heated to 70 °C, the fat crystals were
melted.
Gel Formation from Colostrum
To investigate the gel formation of colostrum, rheology was
used again. A temperature sweep combined with a time sweep
was performed. In a temperature sweep, storage modulus and
loss modulus are measured under variation of the temperature
while the frequency and amplitude are kept constant. In a time
sweep, all conditions are kept constant but the sample’s behav-
iour is observed for a defined time. The temperature was varied
from 20 °C to 70 °C at a rate of 1 K/​min. Afterwards, the sample
was kept at 70 °C for 30 min. The oscillation strain was 0.01%
for both temperature sweep and time sweep. The measurement
results are shown in Figure 26.9. The graph of temperature vs.
time (temperature ramp) is indicated by a green line with linear
slope first (temperature sweep) and constant value (time sweep)
later in the diagram. For all colostrum samples, the moduli rose
at first and later showed a plateau. At the point where storage
modulus and loss modulus cross, the sample undergoes a sol–​gel
transition. This point is called the gelation point, because that
188 Judith Hege et al.

FIGURE 26.9 Temperature and time sweep for upper phase, lower phase and mixed phase of colostrum. The temperatures mentioned are where each phase
started to gel. For comparison, raw milk is shown as well. It does not gel. The green line shows the temperature ramp of the measurement.

FIGURE 26.10 Amplitude sweep of upper phase, lower phase and mixed phase of colostrum at 70 °C. For comparison, raw milk is also shown.

is when the material starts to gel. From the diagram, it can be
seen that the upper phase starts to gel first: both in the sense of
time and also temperature-​wise, meaning that it begins to gel at a
lower temperature. The gelation point for the upper phase was at
45 °C. Next, at 59 °C, the mixed phase started to gel, and at 70 °C
the lower phase followed.
For normal milk (orange squares), completely different
behaviour is experienced. It does not gel when heating, and so
no rise of the moduli’s values can be observed. This is due to
the different composition of normal milk. As mentioned before,
colostrum contains much more protein, especially whey protein
and immunoglobulins, which denature under heating. There is
also a higher calcium content in colostrum than in normal milk
(McGrath et al., 2016). Free calcium ions are bivalently positively
charged and bind to negatively charged lateral groups of amino
acid residues of the whey protein. Ionic links are formed, leading
to aggregation. Thus, the viscosity is enhanced. Colostrum also
contains more casein than milk, and the casein micelles are larger
(McGrath et al., 2016). This is probably the reason why colos-
trum forms a gel when heated while normal milk does not.
Last but not least, an amplitude sweep was performed at 70 °C
to understand the gel network properties (Figure 26.10). From
this diagram, it can be observed that the curves belonging to the
colostrum show the typical behaviour of a gel: first an area where
the moduli stay constant even with increasing oscillation strain,
and then, after the breaking of the gel, a negative slope. So, this
Dairy: Milk Gels – a Gastrophysics View
measurement confirms that colostrum forms a gel due to heating.
Regarding the different phases of the colostrum, it stands out that
the moduli of the upper phase have the highest values, which
means that this phase has the highest viscosity. Because of the
higher fat content, the gel became thicker than for the other two
phases. With decreasing fat content, the moduli showed smaller
values. For the normal milk, the values of storage modulus and
loss modulus are a lot smaller (about five orders of magnitude!).
This is because normal milk stays liquid under heating, and thus
no gel formation can be observed.
Conclusions
As has been shown in this chapter, milk is a complex and
interesting system that mainly consists of four components:
water, fat, caseins and whey proteins. Some of its properties
were discussed in this chapter. The focus here was on gel for-
mation, especially the production of yogurt and the gelation of
colostrum. It was shown that the mechanisms leading to the for-
mation of a yogurt gel are different from those leading to the for-
mation of a colostrum gel. The mechanism that leads to a yogurt
gel is lowering the pH. This is done by adding lactobacilli to
milk. The lactobacilli convert lactose into lactic acid, thus acid-
ifying the system. The addition of acid causes a destabilization
of the micelle distribution in the whey because the charge of
the micelle arms is screened. Thus, the electrical repulsion is
weakened. Also, the steric repulsion is weakened, because the
arms may lie at the micelle surface now. The casein micelles are
no longer uniformly distributed but start to coagulate. With time,
and thus the addition of more acid, a particle gel is formed. Fat
globules and whey proteins are caught in the particle gel. With
longer fermentation time, the yogurt will become firmer. In the
experiment described, it was shown that the pretreatment of the
milk used to prepare yogurt also has a crucial impact on the con-
sistency of the yogurt. In this specific case, raw milk was used
to prepare yogurt, but one of the two samples was heat-​treated
by boiling it and cooling it down again before starting to prepare
the yogurt. It was observed that the yogurt made of boiled milk
became much firmer than the one made out of unboiled milk.
The reasons for this could be the presence of other bacteria in
fresh raw milk and also denaturation of the whey proteins in
heated milk.
Gel formation in colostrum was investigated as well. However,
the responsible mechanism, which was the focus of our study,
was completely different from the one used to prepare yogurt. In
contrast to normal milk, colostrum also gels by simple heating.
The reason for this lies in the denaturation of the proteins, espe-
cially the whey proteins. There is a much higher amount of
proteins in colostrum than in normal milk. During denaturation,
the arrangement of the protein structure is changed. The initial
links between the amino acids of the whey protein are broken,
and new ones can be formed. If there are free ions present, e.g.,
the bivalent positively charged calcium ion, they can link to
negatively charged amino acids of the protein. In this way, a net-
work of ionic links is formed. The growth of such clusters leads
to gel formation and thus increased viscosity. Colostrum has a
189
higher content of calcium than normal milk, which enhances the
described effect.
These are only a few of milk’s interesting properties. However,
there are many more astonishing features to be discovered, motiv-
ating further research in this field.
REFERENCES
Alberts B, Jonson A, Lewis J, Raff M, Roberts K, Walter P. 2008.
Molecular Biology of the Cell. 5th edition, Garland Science,
New York.
Belitz HD, Grosch W, Schieberle P. 2008. Lehrbuch der
Lebensmittelchemie. Springer Verlag, Heidelberg.
Bellwood P. 2005. First Farmers: The Origins of Agricultural
Societies. Blackwell, Malden.
De Kruif C, Huppertz T, Urban SV, Petukhov A. 2012. Casein
micelles and their internal structure. Advances in Colloid and
Interface Science, 36, 171–​172.
Eisenbrand G, Schreier P. 2006. RÖMPP Lexikon Lebensmittelchemie.
2. Auflage, Thieme, Stuttgart.
Gallier SYFC. 2010. Understanding the Structure of the Bovine
Milk Fat Globule and its Membrane by Means of Microscopic
Techniques and Model Systems. PhD Thesis, University of
Otago, Dunedin, New Zealand.
Haque A, Richardson RK, Morris ER. 2001. Effect of temperature
on the rheology of set and stirred yogurt. Food Hydrocolloids,
15, 593–​602.
Haug A, Høstmark AT, Harstad OM. 2007. Bovine milk in human
nutrition –​a review. Lipids in Health and Disease, 6, 25.
Koutina G, Christensen M, Bakman M, Andersen U, Skibsted LH.
2016. Calcium induced skim-​milk gelation during heating as
affected by pH. Dairy Science & Technology, 96, 79–​93.
Larson BL, Heary HL, Devery JE. 1980. Immunoglobulin produc-
tion and transport by the mammary gland. Journal of Dairy
Science, 63, 665–​671.
Lucey JA, Singh H. 1998. Formation and physical properties of acid
milk gels: a review. Food Research International, 30(7), 529–​542.
McConnell MA, Bucha GAA, Borissenko MV, Brooks HJL. 2001.
A comparison of IgG and IgG1 activity in an early milk
concentrate from non-​ immunised cows and a milk from
hyperimmunised animals. Food Research International, 34,
255–​261.
McGrath BA, Fox PF, McSweeny PLH, Kelly AL. 2016. Composition
and properties of bovine colostrum: a review. Dairy Science
and Technology, 96, 133–​158.
McMahon DJ, Oommen BS. 2013. Casein Micelle Structure,
Functions, and Interactions. Chapter 6. In McSweeney PLS,
Fox PF (eds.), Advanced Dairy Chemistry: Volume 1A:
Proteins: Basic Aspects, 185, 4th edition, Springer.
Mezger TG. 2010. Das Rheologie Handbuch. 3. Auflage, Vincentz
Network, Hannover.
Neher J. undated. www.kaesewelten.info/​ kasesorten/​kuhmilch/​
leipajuusto/​
Stelwagen K, Carpenter E, Haigh B, Hodgkinson A, Wheeler TT.
2009. Immune components of bovine colostrum and milk.
American Society of Animal Science, 87(Suppl. 1), 3–​9.
Swasthi’s Recipes. 2012–​ 2019. www.indianhealthyrecipes.com/​
junnu-​kharvas-​recipe/​
Ternes W, Täufel A, Tunger L, Zobel M. 2005. Lebensmittel-​Lexikon.
4. Auflage, Behr’s Verlag, Hamburg.
Ternes W. 2008. Naturwissenschaftliche Grundlagen der Lebens­
mittelzubereitung, Behr’s Verlag, Hamburg.
Töpel A. 2016. Chemie und Physik der Milch. Naturstoff. Rohstoff.
Lebensmittel. 4. Auflage, Behr’s Verlag, Hamburg.
190
Tuinier R, de Kruif CG. 2002. Stability of casein micelles in milk.
The Journal of Chemical Physics, 117, 1290–​1295.
Vasbinder AJ, Alting AC, de Kruif KG. 2003. Quantification of
heat-​
induced casein-​ whey protein interactions in milk and
its relation to gelation kinetics. Colloids and Surfaces B:
Biointerfaces, 31, 115–​123.
Vilgis T. 2014. Frische Milch –​kolloidale Struktur, Physik und
Verdauung. Journal Culinaire, 18, 55–​59.
Judith Hege et al.
Vilgis T. 2015. Wenn Milch sauer wird –​Änderung ihrer Struktur
und praktische Anwendungen. Journal Culinaire, 20 , 20–​36.
Walstra P, Geurts TJ, Noomen A, Jellema A, van Boekel MAJS.
1999. Dairy Technology –​Principles of Milk Properties and
Processes. Marcel Dekker, Inc., New York.
Worldnews, Inc. undated. www.cheese.com/​juustoleipa/​
Dairy: Culinary Uses of Milk, Butter and Ice Cream
Alan L. Kelly and David S. Waldron
School of Food and Nutritional Sciences
University College Cork, Ireland
The Complexity of Milk
Milk is a nutritionally rich food material, which has formed a
significant part of the human diet for millennia. Milk is a com-
plex physical system, with the fat content dispersed as globules
in a serum within a natural emulsifier system (the protein-​and
phospholipid-​ rich milk fat globule membrane, MFGM). The
serum contains two families of proteins, the majority of which
is casein (80%), assembled into macromolecular structures
called casein micelles. The serum dissolves other proteins and
a fermentable saccharide called lactose, among other chemical
species, including mineral ions, such as phosphate and calcium,
that exchange with the micelles. Milk contains protein (3–​4%),
fat (3–​5%), lactose (4.4–​4.8%), and minerals, vitamins and trace
nutrients (0.7–​0.8%).
Due to its nutritional content and physico-​chemical makeup,
milk is an unstable raw material, with the large fat globules rising
to give a cream layer on standing: traditionally, raw milk would
have been left in a separating dish for the rich cream to rise over
the course of a day, and the surplus skimmed milk would be
poured off.
In addition to physical separation, the multiple sources of
microbial contamination in the production environment (farms)
lead to a host of safety and spoilage challenges. Hence, milk for
consumption in liquid form is processed to physically stabilize it
by homogenization (which reduces the size of the fat globules)
or treated to make it microbiologically more stable using heat
such as pasteurization (which kills vegetative bacteria, yielding a
product that must be kept refrigerated) or more severe ultra-​high
temperature (UHT) treatment, which kills spores and results in a
product that can be stored at ambient temperatures.
For 6,000 to 8,000 years, cows have been milked, and the milk
has been processed into products to preserve it through different
principles, such as separation, concentration or fermentation.
A well-​known process is the concentration of fat and proteins
into cheese through coagulation with rennet. Preservation is
provided by a combination of acidity (generated by bacterial
cultures producing lactic acid from lactose), salt and low water
activity. The diversity of cheeses is the result of differences in
making conditions, cultures used, and the conditions under which
the cheese is stored after manufacture to allow reactions catalysed
by enzymes and micro-​organisms to develop flavour and texture
(i.e., ripening).
Milk has been a traditional feature of culinary uses to extents that
differ greatly from country to country. Countries such as Ireland
have a long tradition of consumption of ‘fresh’ pasteurized milk
as a beverage with a meal, whereas in countries such as France
this is not common; 95% of milk sales in France are in the form of
UHT milk, which has a longer shelf life and lower price (Belgium,
Portugal and Spain similarly consume >90% UHT milk). On the
other hand, these countries have a culture of offering cheese after
the main course, sometimes before the sweet desserts.
Milk has also been deconstructed in recent decades by a range
of increasingly powerful techniques to extract its components in
purified or enriched form, and a modern kitchen might include
several components of milk (such as lactose, or whey proteins
extracted from the fluid that arises as a by-​product of cheese
manufacture). These can also find applications in molecular
cuisine due to the unusual techno-​ functional properties of
components like whey proteins, which, depending on concen-
tration, conditions like pH, and treatment, could be effective
emulsifiers, gelling agents or foaming agents.
Two common uses of the recovered cream are butter and ice
cream, which will be the focus of this chapter. Butter and ice
cream are two traditional dairy products, which, in common
with many products and dishes discussed in this handbook, have
a deep scientific basis underpinning the art with which they are
produced and enjoyed. In this chapter, the principles of produc-
tion of both will be considered, alongside consideration of their
main characteristics and culinary applications.
The Production of Butter
Butter is very simple to produce by allowing raw milk to sep-
arate under gravity, recovering the floating cream layer and then
agitating it vigorously, incorporating air. This leads first to a
stiffened whipped cream structure, which under continued agita-
tion (for example, through manual rotation of a cylindrical churn)
‘breaks’, leading to the release of butter milk and butter grains of
released milk fat, which are consolidated into butter, often with
salt added for flavouring and preservative reasons.
191
192
(a) (b)
Casein micelle Milk fat globule membrane
FIGURE 27.1 Schematic illustration of the transition from milk (a) to cream (b), then whipped cream (c), and finally butter (d).
This simple process has been applied in countries around
the world for millennia. For example, in Ireland, samples of
‘bog butter’ have been found buried in peat bogs spanning at
least 3,500 years, from the Early Bronze Age (c. 1700 BCE) to
the 17th century CE (Smyth et al., 2019). Across North Africa
and through the Middle East into Asia, the keeping properties
of butter were more often improved by cooking-​off the water,
leaving a more stable ghee (or samna).
The scientific principle underpinning the production of butter
is the inversion of the natural oil-​in-​water emulsified system
found in milk and cream to produce a pseudo-​emulsion, which
includes water droplets dispersed in a semi-​solid fat matrix, or
even a system that behaves as a gel, depending on temperature.
As said before, the natural dispersion of milk fat globules in
milk is stabilized by the presence of the MFGM, which keeps oil
and the aqueous solution separate in the aqueous milk environ-
ment. However, the agitation applied during churning mechanic-
ally ruptures this membrane, exposing hydrophobic fat; in order
to minimize the instability and retained energy caused by being
exposed to the aqueous environment, damaged fat globules, on
collision, coalesce.
This process is usually undertaken at temperatures (10–​12 °C)
where the milk fat is partially liquid but still largely crystalline,
and so the globules can flow, deform and coalesce, but still retain
a certain solid-​like character. Air is incorporated into the cream
by a shear action, forming and breaking bubbles, which are ini-
tially stabilized by the adsorption of native soluble proteins to
the air/​water interface (Brooker, 1986). Fat globules collide,
inducing aggregation and partial coalescence at the interface,
to create a continuous enveloping matrix to stabilize the foam.
Without sufficient solid fat content (SFC), the whipping process
would cause release of the liquid fat and premature collapse of
the foam. Further agitation increases the air volume, making the
fat lamellae between the bubbles progressively thinner. The fat
globules concentrated in the serum phase form larger aggregates,
leading to distortion and eventual penetration of the MFGM by
fat crystals, causing a leakage of internal liquid oils and leaving
Alan L. Kelly, David S. Waldron
(c) (d)
Crystallised fat globule Water Air bubble
masses of coalesced fat globules that form the basis of the butter
grains (Schmidt and van Hooydonk, 1980; Brooker, 1993).
This collapse is accompanied by the release of ‘butter milk’,
which has a composition quite similar to that of skim milk but
with a higher fat content and a high content of released MFGM
material, particularly phospholipids, which give butter milk
important functional properties for use in a wide array of foods
(Sodoni et al., 2006) and a range of health benefits (Conway
et al., 2014).
The steps in the production of butter are schematically
illustrated in Figure 27.1.
The Texture of Butter
The desirability of butter is strongly linked to its melt-​in-​the-​
mouth appeal and its performance in cooking and baking.
The principal characteristic of butter that influences its use is
hardness. This is primarily determined by the proportions of
the fat that are liquid and solid at a given temperature, which
is determined by the types of fatty acid residues that comprise
the triglycerides in the fat. A higher proportion of long-​chain
and saturated fatty acid residues (such as palmitic acid) will give
fat a higher melting point and make it harder at typical food use
temperatures, whereas higher levels of short-​chain or unsaturated
fatty acid residues (such as oleic acid) will give lower melting
points and softer butter. The greatest influence on the proportions
of different fatty acid residues present is the diet of the cows, and
hence the hardness of butter may differ from region to region,
from country to country and even with different seasons, due to
changes in availability of grass as feed (O’Callaghan et al., 2016).
The use of grass as a main component of cows’ diet also greatly
influences butter colour; butter from countries such as Ireland
that practise a seasonal grass-​based milk production system tends
to have a notably stronger golden-​yellow colour (due to high
levels of β-​carotene) than that from countries or systems where
different feeding practices dominate, e.g., 5 µg/​100 ml in the
Dairy: Milk, Butter and Ice Cream
USA (indoors) compared with 21 µg/​100 ml in the UK (summer
grazing) (Schönfeldt et al., 2012)
So, much of the eventual texture, especially the hardness, of
butter is already set by the time milk or cream is received in a kit-
chen or factory, as the particular mix of fatty acid residues present
will determine what percentage of the fat is solid at a particular
temperature. There is still the possibility, however, of manipu-
lating not the amount of solid fat but rather, the form of that fat.
Fat, when it passes its melting temperature, will form crystals,
and the ultimate self-​supporting structure of butter is due to an
interlocking crystal network. The exact texture then depends
on the microstructure of this network, and the size and shape of
crystals can influence texture. Samples of butter on a dish with
exactly the same proportion of fat solid, but where in one case
there are a lot of small crystals and in another a smaller amount
of larger crystals, will have very different spreadability.
In industrial butter production, this is controlled by the tem-
perature history of the cream. The cream, after separation from
the milk, is pasteurized at temperatures higher than those used for
milk due to the higher fat content and viscosity (min. 80 °C for
15 s) (Juffs and Deeth, 2007; IDFA., 2016). Pasteurization has the
added benefit of inactivating the native milk enzyme lipoprotein
lipase (LPL; EC 3.1.1.34), which otherwise could hydrolyse the
fat released when the MFGM (which normally keeps enzyme and
substrate apart) is broken and lead to rancidity. At pasteurization
temperatures, the milk fat is completely liquid, and on cooling to
churning temperatures a proportion of the fat (determined by the
diet and tempering conditions) will crystallize. The key is then
to control the rate of cooling, and complex profiles, which can
differ at different times of the year, will be applied over several
hours to tailor the exact crystalline structure in the fat and hence.
the eventual butter texture (Lee and Martini, 2018; Vilgis, 2015).
After manufacture, butter also hardens during storage due to
continued crystallization during cold storage over a period of
weeks or months following production (Rønholt et al., 2014),
and butter may often be reworked or physically manipulated
after a period of storage to soften it; this may often be combined
with the point of division of a bulk of butter into smaller con-
sumer packs. The deformation induced by the rotating propellers
and perforated plates can break up crystalline structures and
reduce SFC due to the applied shear (Rønholt et al., 2013). Upon
cooling, the SFC increases to the original levels, but some of the
Newtonian viscosity may be lost (Mazzanti et al., 2011).
Butter possesses a mild creamy flavour with a delicate odour,
making it well suited as a culinary ingredient. The profile of
butter includes a fine balance of short-chain free fatty acids, i.e.,
odorant volatile compounds associated with the hydrolysis or oxi-
dation of triglycerides in the milk. The distribution, concentration
and potency of butanoic acid (buttery, cheesy, sweaty), γ-​ and
δ-​lactones (sweet, peach, coconut), indole and skatole (mothball/​
faecal), phenols and hydrocarbons yield this subtle odour (Mallia
et al., 2008).
Controlled fermentation with bacteria can be used to intro-
duce a lactic acid taste with odorant diacetyl, butanoic and δ-​
decalactones to create the distinctive lactic or cultured butter. So
subtle, in fact, is the flavour of butter that it is quite common to
193
add herbs, spices, capers and allium for use in the kitchen. Butter
is very suitable for addition of parsley, lemon and pepper (beurre
maître d’hôtel), garlic cloves (beurre à la bourguignonne –​
Burgundy or garlic butter) or Roquefort (to make a piped
topping), or it can be prepared sautéed with onions or simmered
with brown sugar to make butterscotch.
The perception of the flavour of foods involves complex and
multidimensional layers of gustatory sensations. Before con-
sumption, the volatile odorant compounds associated with the
food enter the nose with inhaled air and are perceived by way of
the olfactory neurons (antenasal odour). In the mouth, the tongue
manipulates the food to assist breakdown by mastication and
forms a bolus before swallowing, while odorant molecules travel
through the nasal pharynx, connecting the back of the mouth with
the nose, to the olfactory neurons (retronasal odour). During this
direct contact with the food, other compounds within the food
activate the taste buds on the tongue, sending signals that are
perceived as taste by the brain.
These sensations can guide our eating habits; bitter and
sour tastes warn of potentially spoiled, toxic or strongly acidic
foods, while salty and sweet desires direct us to foods rich in
saccharides, amino acids or salt. Beyond these core four tastes,
other chemoreceptors on the tongue and in the mouth have been
identified that can enhance our perception of the food: the savoury
taste of monosodium glutamate (MSG) is detected as umami (as
additional taste not elicited by any combination of the others);
heat and spiciness of food are perceived as irritation or pun-
gency; and mechanical stimulation allows touch sensors to inter-
pret texture, whether creamy and smooth or crunchy and granular
(Faurion, 1988; Finger and Kinnamon, 2011). From a dairy and
nutrition perspective, recent research has identified receptors that
may inform the brain of sources of mineral-​rich, e.g., calcium
(Tordoff et al., 2012) or energy-​rich fatty foods (Besnard et al.,
2016; Hartley et al., 2019). It is postulated that this may influ-
ence human eating behaviour and aid the body in finding foods
containing lipid-​soluble vitamins (A, D, E, K). Conversely, this
association may be linked to the over-​consumption of some foods
by obese individuals (Gilbertson and Khan, 2014).
The combination with heat opens a further exploration of the
transitions within butter odour and flavour. On heating butter
with flour to 120 °C to form a white roux, a roasted odour of
aldehydes, carboxylic acids and lactones is released (Kato, 2003).
At higher temperatures, butter transforms into beurre noisette
and beurre noir, as Amadori rearrangement associated with
Maillard browning causes changes in the colour and aroma of
fats, sugars and proteins, forming furans, cyclic ketoenols (sweet,
fruity, floral) and nitrogen-​containing pyrazines (nutty, roasted)
(Shigematsu et al., 1977; Maga and Katz, 2009).
The desirable nature of butter goes beyond flavour and is
linked to its behaviour as it warms in the mouth –​a gradual but
quick melting in an almost uniquely pleasurable manner. The
small solid α and β′ fat crystals in butter melt over a broad range
of temperatures, softening to a liquid easily as they reach 35 °C
(Devi and Kahatkar, 2017).
In baking and confectionery, the performance of butter is of
utmost importance; it must provide aeration, dispersibility and
194
plasticity properties on mixing, allow even expansion during
baking and a desirable aroma from the oven, and play a role, with
native starches, in retarding staling during storage (Gray and
BeMiller, 2003; Mamat and Hill, 2014).
In application, the use of butter depends on its spreadability,
which depends on its temperature history and also its struc-
ture, as the presence of some globular fat will actually help with
spreadability, with the spherical globules thought to act like ball
bearings to facilitate the lateral movement of the material under
the shearing force of the knife.
The concept of butter as a moisture barrier in dishes often
escapes our attention. In sandwiches, butter spread evenly can
prevent moisture in fillings making the bread soggy. In croissants
and puff pastry, butter is used as a laminate between layers of
dough, which during baking traps the moisture as it turns to
steam and yields the mille feuille of these Viennoiserie delights
(Mattice and Marangoni, 2017). The performance of butter as a
bakery fat requires a certain consistency of fat crystal size, type
and dispersion with defined melting characteristics. Traditionally,
for optimum baking performance, bakery fats required a min-
imum of 20% SFC at 25 °C and a minimum of 5% SFC at 40 °C
(Podmore, 2002).
In its many diverse ways, butter has complemented and
enhanced all forms of snacks and meals, from savoury to sweet.
However, the most popular accompaniment for desserts must be
ice cream.
The Science of Ice Cream
It is believed that ice cream has been produced for several cen-
turies, with its origins ‘hotly’ debated between Italy and China
and tales of Marco Polo bringing recipes with him in the 13th
century. Iced desserts developed from blends of fruits and
powders with crushed ice to the more complex process involving
milk being cooled in vessels cooled by chilled brine while being
whipped. Today, ice cream is a product that is produced and
enjoyed around the world, the properties of which are dependent
on a surprising amount of physics and physical chemistry.
One of the best ways to understand the science of ice cream is
to visualize it down a microscope. Frozen ice cream is a complex
multi-​phase system, with elements as follows:
• Ice crystals, which give the product its name and char-
acter, but which must be below a certain size to prevent
their being detected on the tongue while eating;
• Air bubbles, for ice cream is an aerated product (10–​
50% of the total volume), which give the light airy
structure of ice cream, and mean the product is typically
sold on the basis of volume rather than weight (CFR,
2019). These bubbles are entrapped by;
• A network of partially coalesced fat globules. The ice
cream mix is homogenized, which coats fat globules
in protein (the greatly increased interfacial surface
area being too much for existing MFGM material to
cover), but such globules do not whip and entrap air
well. Thus, the mix usually includes emulsifiers (such
as lecithin, provided through the inclusion of egg in the
Alan L. Kelly, David S. Waldron
recipe), which compete successfully for access to the
globule surfaces, eventually displacing the protein but
giving a weaker coating, which can be damaged during
whipping, giving the desired effect;
• A thick syrupy surrounding liquid, which contains milk
minerals and, critically, sugars, natural (lactose) and
added (sucrose, fructose), which depress the freezing
point so much that the liquid, once a high proportion
of the water has been effectively removed through the
formation of ice crystals, is unable to freeze even at the
temperatures (−18 °C) used in typical freezers.
So, ice cream is a sugar-​rich liquid within which air bubbles
are trapped by a partially coalesced fat network and ice crystals
of defined size float; part liquid, part emulsion, part foam, part
frozen. It may, of course, also include flavourings and inclusions
(cookie pieces, chocolate or fruit), but these are the core elements
of every ice cream.
Ice cream is produced by making a mix of milk, cream (if
desired to enhance the fat content), sugar, emulsifiers (perhaps
egg), stabilizers (to be discussed later) and flavours, pasteur-
izing and homogenizing, to ensure microbial stability and create
the emulsion already described. It is then held at refrigeration
temperatures for several hours (referred to as ageing) to allow
the emulsion to ‘sort’ in terms of displacement at the globule
surfaces, fat to crystallize and stabilizer to hydrate fully.
It is then frozen in either a batch or a continuous process (the
latter using a scraped-​surface heat-​exchanger) to remove heat,
form ice crystals while mechanically disrupting them to prevent
their growing too large, and whip in air to be entrapped by fat
within the freezing structure. At the end of this initial freezing,
the mix may be at −5 °C; it could be consumed directly (in a tub
or on a cone) and is still liquid enough for inclusions to be added
and mixed in. If not consumed directly, it is, however, now deep-​
frozen in suitable packaging to −18 °C, at which temperature the
final structure emerges, and ideally stored at that temperature
until consumption.
The steps in the production of ice cream are schematically
illustrated in Figure 27.2.
However, ideal conditions are rarely met, and frozen storage of
ice cream is often punctuated by points at which its temperature
increases, even slightly, perhaps due to opening of a freezer door
allowing in a small amount of warm air. Even a tiny increase in
temperature allows some ice to melt, and upon the temperature
decreasing again the water tends to freeze onto existing crystals,
which grow as a result; crystal growth by accretion or Ostwald
ripening can also occur.
The net effect of such crystal growth is a progressive movement
towards the ultimate fate of all ice cream, which is a deterioration
towards a state where consumers will reject the product due to
perceived iciness or grittiness, as ice crystals exceed the sensory
threshold. The key function of stabilizers in the ice cream mix
(typically hydrocolloids such as carrageenan, guar or locust bean
gums) is to delay this point of end of acceptable life as effectively
as possible, by essentially building molecular cages around ice
crystals or otherwise entrapping and immobilizing water, such
that any water that thaws does not have the mobility to enlarge
Dairy: Milk, Butter and Ice Cream
(a) (b)
Casein micelle Milk fat globule membrane Crystallised fat globule
FIGURE 27.2 Schematic illustration of the transition from ice cream mix (a) to homogenised mix (b), and after aeration and freezing, to a semi-frozen foam
(c) and final hardened ice cream (d).
existing ice crystals, and acceptable structure is retained for
longer.
Ice cream has been a popular basis for experimentation and
innovation within the field of molecular cooking. For example,
freezing using liquid nitrogen results in extremely fast freezing
(due to a very large temperature difference between the freezing
medium and the mix), leading to very small ice crystals and a
structure dominated by the perception of creaminess due to the
fat present. Fast freezing using liquid nitrogen can also result in
a product with a hard-​frozen outside crust and a creamy semi-​
liquid interior. Freezing with liquid nitrogen also allows the
possibility of inclusion of alcohol in an ice cream mix, which
otherwise would undesirably interfere with the freezing proper-
ties of the product due to the low freezing point of ethanol.
Ice cream flavours can also be manipulated to contain counter-​
intuitive flavour combinations, and savoury ice creams are avail-
able in some restaurants (such as a frozen crab bisque or egg and
bacon ice cream at The Fat Duck); inclusion of spicy ingredients
such as chilli can result in an unusual mix of sensations of heat and
cold on consumption. Ice cream can also be dried into a powder as
an ingredient of desserts (a Pacojet system, which micro-​purées
frozen materials, is used commonly for this purpose) or freeze-​
dried into a solid form which does not require frozen storage (so-​
called astronaut ice cream). Addition of methylcellulose (around
2%) can also lead to a ‘hot ice cream’, which has an ice cream-​
like consistency at hot temperature and actually melts (liquifies)
as it cools due to the formation of a thermo-​reversible gel; the ice
cream can be heated in a microwave oven or on a hob and, when
placed on a plate, gradually melts as the temperature falls.
Structural Considerations for Butter
and Ice Cream
Butter and ice cream are, as explained earlier, complex entities,
and also ancient products, but consideration of their physical
characteristics leads to possibilities for future innovation.
195
(c) (d)
Emulsifier New emulsifier layer Ice crystal Sugar serum phase Air bubble
A key consideration in evaluating their structure is the
properties of milk fat, which is entirely solid at <−10 °C and
entirely liquid at >50 °C. Between these temperatures, part of
it is solid, and part is liquid. For example, using melting curves
(plus nuclear magnetic resonance (NMR) analysis), one can
say that about 70% is liquid at room temperature (Bouteille
et al., 2013).
Then there is the question of how the liquid and the solid are
distributed in systems like ice cream and butter. If the liquid is
enough to make a continuous liquid phase intermixed within a
continuous solid matrix, then we could use the Dispersed System
Formalism term:
D3(O) × D3(S)
However, if the solid is less than 5% (the limit for having a con-
tinuous solid network), then we can have a dispersion:
D0(S)/​D3(O)
Moreover, the issue of the water phase has to be considered,
which adds complexity because, for some temperatures, we
can imagine dispersions of water and liquid fat in the solid,
or other possibilities. As mentioned earlier, butter is often
described as a water-​ in-​oil emulsion, but technically it is
not an emulsion but rather a pseudo-​emulsion, as the water
droplets are not individually covered in an emulsifying barrier
layer but are entrapped in the semi-​solid fat matrix by physical
dispersion.
When the butter is consumed, the first phenomenon that
happens in the mouth will be melting (leading to cooling, as
latent heat of fusion for the fat crystals is absorbed) followed by
rapid coalescence of the water droplets (there being no barrier to
this happening once the fat liquefies) and a perceptible sudden
rise in perception of salty flavour, in the case of salted butter,
as pools of salted water react with the relevant receptors on the
tongue.
196
For ice cream, there are many different possibilities, but if
there are air bubbles (D0(G)), dispersed fat (D0(O)) and dispersed
solids (D0(S)) (for example ice crystals, fat crystals or egg solids),
and a continuous dispersing liquid (concentrated sugar syrup)
D3(W), we can have
[D3(G) + D3(O) + D0(S)]/​D3(W)
However, this will be highly changeable with temperature,
since the relative proportions of frozen and unfrozen phases
will change, such as in the case of heat shock (leading to partial
melting and later refreezing of ice crystals).
As with butter, consumption results in rapid melting in the
mouth, and the absorption of latent heat to fuel the melting of
both ice and fat crystals results in a pleasurable cold sensation,
followed by a rapid release of flavour and sweetness as the system
reverts to a liquid and highly mobile state.
Conclusion
Ice cream and butter are two traditional dairy products, which can
be produced on scales from the large to the small and are staples
in many cuisines and culinary traditions. While the principles
of their production are relatively simple, this is underpinned, as
is the case for so many food products, by complex science. In
both cases, the key principle is destabilization of an emulsion and
its reconstruction in a modified form that gives specific textural
characteristics. In addition, key attributes of both products relate
to melting, in one case of fat and in the other of ice crystals, and
the stability and eating experience of both are dependent on con-
trol of this process. Understanding these principles allows fine
control of their use in culinary settings, as well as innovating
new recipes and experiences based on their integration into other
dishes, sauces and desserts.
REFERENCES
Besnard P, Passilly-​Degrace P, Khan NA. 2016. Taste of fat: a sixth
taste of modality? Physiology Review, 96, 151–​176.
Bouteille R, Gaudet M, Lecanu B, This H. 2013. Monitoring lactic
acid production during milk fermentation by in situ quantita-
tive proton nuclear magnetic resonance spectroscopy, Journal
of Dairy Science, 96, 4, 2071–​2080.
Brooker BE. 1986. The development of structure in whipped cream.
Food Structure, 5, 2, 277–​285.
Brooker BE. 1993. The stabilisation of air in foods containing fat –​a
review. Food Structure, 12, 115–​122.
CFR. 2019. Ice cream and frozen custard. Food for Human
Consumption (21 C.F.R. § 135.110). Code of Federal
Regulations, US Food & Drug Administration.
Conway V, Gauthier SF, Pouliot Y. 2014. Buttermilk: much more
than a source of phospholipids. Animal Frontiers, 4, 2, 44–​51.
Devi A, Kahatkar BS. 2017. Thermo-​ physical properties of fats
and oils. International Journal of Engineering and Technical
Research, 7, 2, 45–​50.
Faurion A. 1988. Birth and obsolescence of the concept of four
tasting qualities. Journal d’Agriculture Traditionnelle et de
Botanique Appliquée, 35, 21–​40.
Alan L. Kelly, David S. Waldron
Finger TE, Kinnamon SC. 2011. Taste isn’t just for taste buds any-
more. F1000 Biology Reports, 3, 20.
Gilbertson TA, Khan NA. 2014. Cell signalling mechanisms of oro-​
gustatory detection of dietary fat: advances and challenges.
Progress in Food Lipid Research, 53, 82–​92.
Gray JA, BeMiller JN. 2003. Bread staling: molecular basis and con-
trol. Comprehensive Reviews in Food Science and Food Safety,
2, 1–​21.
Hartley IE, Liem DG, Keast R. 2019. Umami as an ‘alimentary’
taste: A new perspective on taste classification. Nutrients, 11,
1, 182–​200.
IDFA. 2016. Pasteurisation [Online]. International Dairy Foods
Association. 2016.
Juffs H, Deeth HC. 2007. Scientific Evaluation of Pasteurisation for
Pathogen Reduction in Milk and Milk Products. Food Standards
Australia New Zealand. www.idfa.org/​pasteurization
Kato Y. 2003. Chemical and sensory changes in flavour of roux
prepared from wheat flour and butter by heating to various
temperatures. Food Science and Technology Research, 9, 3,
264–​270.
Lee J, Martini S. 2018. Effect of cream aging temperature and agi-
tation on butter properties. Journal of Dairy Science, 101, 9,
7724–​7735.
Maga JA, Katz I. 2009. Pyrazines in foods: An update. Critical
Reviews in Food Science and Nutrition, 16, 1, 1–​48.
Mallia S, Escher F, Schlichtherle-​ Cerny H. 2008. Aroma-​ active
compounds of butter: A review. European Food Research and
Technology, 226, 315–​325.
Mamat H, Hill SE. 2014. Effect of fat types on the structural and
textural properties of dough and semi-​sweet biscuit. Journal of
Food Science & Technology, 51, 9, 1998–​2005.
Mattice KD, Marangoni AG. 2017. Matrix effects on the crys-
tallization behaviour of butter and roll-​ in shortening in
laminated bakery products. Food Research International,
96, 54–​63.
Mazzanti G, Megyu L, Marangoni AG, Idziak SHJ. 2011.
Effects of shear rate variation on the nanostructure of crys-
tallizing triglycerides. Crystal Growth & Design, 11, 10,
4544–​4550.
O’Callaghan TF, Faulkner H, McAuliffe S, O’Sullivan MG, Hennessy
D, Dillon P, Kilcawley KN, Stanton C, Ross RP. 2016. Quality
characteristics, chemical composition, and sensory properties
of butter from cows on pasture versus indoor feeding systems.
Journal of Dairy Science, 99, 12, 9441–​9460.
Podmore J. 2002. Bakery Fats. In Rajah KK (ed.) Fats in Food
Technology. CRC Press.
Rønholt S, Kirkensgaard JJK, Mortensen K, Knudsen JC. 2014.
Effect of cream cooling rate and water content on butter micro-
structure during four weeks of storage. Food Hydrocolloids,
34, 169–​176.
Rønholt S, Mortensen K, Knudsen JC. 2013. The effective factors
on the structure of butter and other milk fat-​based products.
Comprehensive Reviws in Food Science and Food Safety, 12,
468–​482.
Schmidt DG, van Hooydonk ACM. 1980. A scanning electron
microscopal investigation of the whipping of cream. Scanning
Electron Microscopy, 3, 653–​658.
Schönfeldt HC, Hall NG, Smit LE. 2012. The need for country spe-
cific composition data on milk. Food Research International,
47, 2, 207–​209.
Shigematsu H, Shibata S, Kurata T, Kato H, Fujimaki M. 1977.
Thermal degradation products of several Amadori compounds.
Agricultural and Biological Chemistry, 41, 12, 2377–​2385.
Dairy: Milk, Butter and Ice Cream
Smyth J, Berstan R, Casanova E, McCormick F, Mulhall I, Sikora M,
Synnott C, Evershed RP. 2019. Four millenia of dairy surplus
and deposition reavealed through compound-​specific stable
isotope analysis and radiocarbon dating of Irish bog butters.
Scientific Reports, 9, 4559–​4569.
Sodoni I, Morin P, Olabi A, Jiménez-​Flores R. 2006. Compositional
and functional properties of buttermilk: A comparison between
197
sweet, sour and whey buttermilk. Journal of Dairy Science, 89,
525–​536.
Tordoff MG, Alarcón LK, Valmeki S, Jiang P. 2012. T1R3: A human
calcium taste receptor. Scientific Reports, 2, 496.
Vilgis TA. 2015. Soft matter food physics –​the physics of food and
cooking. Reports on Progress in Physics, 78 (12), 124602.
Dairy: Ginger Milk Curd

Martin Lersch
https://khymos.org

While researching gel formation in foods where no “external” egg whites (claiming the preparation requires “a lot of skills and
hydrocolloid or gelling agent is added (Lersch, 2014), I came a pinch of luck”). Kitchen myths arise when people randomly fail
across ginger milk curd (in Chinese: 薑汁撞奶). With only three and succeed with a recipe. Depending on the outcome, they try
ingredients –​milk, ginger and sugar –​it immediately caught my to attribute the result to something they did differently from last
attention. I found a recipe and my first attempt was successful. time. There is a chance that this was a relevant change, but more
I was amazed! With three seemingly simple ingredients, I was often than not it has nothing to do with the outcome. With the
able to form a tender, fragile gel within minutes. I loved the strong wrong conclusion drawn, the kitchen myth is a reality.
ginger taste with a touch of sweetness. After my first success, Based on what I’ve read so far in the scientific literature, the
I had several failed attempts, so I looked up some more recipes on following is my “best bet” for a ginger milk curd that will not fail
the internet. What puzzled me was that, as I dug up more recipes, (Figure 28.1). The most important piece of equipment is a digital
the different instructions were specific, but also contradictive. kitchen thermometer.
I couldn’t let go at this point, so I continued reading –​also sci-
entific papers (Mazorra-​Manzano 2013 and Su 2009). Now that
I have a fairly good understanding of the science behind ginger Fool-​Proof Ginger Milk Curd
milk curd, it is clear that the many recipes I had found were full
250 mL skimmed milk
of kitchen myths.
18 g fresh* ginger juice**
I found recommendations for using fresh ginger alongside
20 g sugar
warnings that the gel will only set if you use old ginger. One
recipe claimed that the gelling was caused by the white sediment
* See comment later regarding the stability of the enzymes.
in ginger juice and that this “starch” was located right below
Ginger juice should not be made in advance due to a short
the skin, hence requiring careful peeling with a spoon. Advice
half-​life at room temperature.
regarding temperature was very confusing. Some recommended
** 18 g ginger juice = ca. 31 g peeled ginger = ca. 43 g raw ginger.
boiling followed by cooling –​or no cooling at all –​while others
recommended against boiling the milk. Some indicated optimal
Combine milk and sugar in a pan and heat carefully to 65 °C. Peel
temperatures in the range 60 to 95 °C, while others suggested
and microplane ginger, and squeeze out the juice. Place juice in
pouring milk back and forth between containers 10 times,
a bowl and pour milk into the ginger juice from a height to allow
which obviously will help cool the milk. The recommendations
sufficient mixing. Do NOT stir, as this will interfere with the gel
regarding milk included full cream, pasteurized, non-​pasteurized,
formation. Leave to set at room temperature and, after 5–​10 min,
non-​ homogenized, micro-​ filtered, high-​
protein and even soy
a gel has formed. The curd may be served immediately or kept
milk. Indeed, very confusing! For a volume of 250 mL of milk,
in the fridge.
the recommended amount of ginger juice ranged from 10 to
In this recipe, I used a milk:ginger juice ratio of 14:1. You can
30 mL, which actually seemed like a reasonable range from a
certainly use more ginger juice, but the taste of ginger may then
culinary perspective. Regarding the preparation, some sources
become too powerful. You can also reduce the amount of ginger
emphasized the importance of pouring milk in one go into the
juice, but I have not tested this yet.
ginger juice (hence the Chinese name “ginger hits milk”), and not
the other way around. Most recipes agreed not to stir after milk and
ginger juice have been mixed. For firmer gels, recommendations
included use of less sugar and addition of a few drops of vinegar. Mechanism of Gelling
It’s fascinating how a recipe calling for only three ingredients Ginger milk curd belongs to a large group of foods where enzymes
can lead to so many awkward instructions. No wonder, then, that are used to curdle milk (including cheese, of course). Traditionally,
I found reports from people who only succeeded in half of their one would use rennet to prepare a curd known as Junket in the US
attempts, succeeded on their ninth attempt, or gave up and added and Cuajada in Spain. Rennet is found in the stomach of young

199
200 Martin Lersch

FIGURE 28.1 (Left to right): The fragile gel is strong enough to support the weight of a spoon. The ginger is grated and pressed to extract the juice. The gel
easily loses its liquid, indicating syneresis.

mammals and is essential as they digest their mothers’ milk. The

active enzyme in rennet is chymosin, also known as rennin. This is
a proteolytic enzyme, also known as a protease, which is capable
of breaking proteins into smaller fragments.
Ginger (the root of Zingiber officinale) enters the scene
because it also contains proteolytic enzymes. The ginger
proteases (GP, sometimes called zingipain) are very sensitive
to temperature. At temperatures above 70 °C, they are rapidly
denatured (irreversibly inactivated). This explains why so many
people fail when trying to make ginger milk curd. The milk
clotting activity (MCA) of GP peaks around 63 °C and falls off
rapidly above 65 °C and below 60 °C (Figure 28.2). This means
that one is left with a relatively narrow temperature window of
60–​65 °C. GP does show proteolytic activity (PA) outside this
window, but this PA is more of a non-​specific kind. MCA, on the
other hand, is related to a specific hydrolysis of κ-​casein (more
on casein in a second; the greek letter κ is pronounced “kappa”).
By now, you may be wondering if it’s possible to make cheese
with ginger –​and the answer is yes. Scientists have studied sev-
eral plant extracts and found that ginger, kiwi and melon all con- FIGURE 28.2 Milk clotting activity (MCA) of ginger proteases vs. tem-
tain proteases with a relatively high MCA/​PA ratio (albeit not perature (in °C).
as high as that of chymosin found in rennet). The temperatures
for maximum MCA of kiwi and melon proteases are 40 °C and
70 °C, respectively. the micelles can collide, and the calcium present in milk aids in
Enzymes are catalysts, the tireless workers responsible for the formation of aggregates (Figure 28.3). These aggregates of
building the gel. However, it is caseins, a group of proteins “shaved” micelles make up the actual gel. It all happens within
found in milk, that are the actual building block of the gel. The a couple of minutes. The resulting gel is very fragile and easily
casein proteins group together to form large “balls” known as loses water, a process known as syneresis (Figure 28.1).
micelles, which are held together by calcium ions. The outside I mentioned the very narrow temperature window for GP, but
of the micelles is covered with κ-​casein, which is composed of a one remaining question was whether heating milk to a higher tem-
water-​soluble part (an acidic glycopeptide) and another part that perature (before cooling to the same temperature window) would
is much less soluble in water (para-​κ-​casein). The water-​soluble be beneficial. It turns out that if milk is heated above 65 °C, the
part is found at the surface and leaves the micelles covered by strength of the resulting gel is reduced. The reason is that the heat
a “hairy” layer. This layer both keeps the micelles dissolved in causes other proteins in the milk (specifically the whey protein
water and prevents the micelles from coming so close to one β-​lactoglobulin) to precipitate onto the κ-​casein, which interferes
another that they coalesce and aggregate. with the gel formation. The same is true for milk fat, so skimmed
It is thanks to this “hairy” layer that milk is stable (i.e., does milk is the choice for a stronger gel. Since calcium plays a role in
not spontaneously form a gel), and all is well until we add the the aggregation of the “shaved” micelles, a higher calcium con-
above-​mentioned proteases to milk. What chymosin, GP and centration will also result in a stronger gel.
other proteases do is cleave off the water-​soluble part of κ-​casein, The ginger juice also deserves a couple of extra words. In
leaving the less water-​soluble para-​κ-​casein behind. Suddenly, freshly squeezed ginger juice the GP has a half-​life of 20 min
Dairy: Ginger Milk Curd
FIGURE 28.3 Protease and aggregated micelles.
at 30 °C, so in a warm kitchen, half of your enzyme activity is
lost 20 min after you have grated and squeezed your ginger. If
you leave it another 20 min, you’re only left with 25% of the ori-
ginal activity. This means that the ginger juice can’t be prepared
in advance or stored unless you use a little trick. The reason for
the instability is that ginger also contains another enzyme, poly-
phenol oxidase (or PPO for short) –​the same enzyme that is
responsible for the browning of apples (an example of enzym-
atic browning, as opposed to the Maillard reaction, which is an
example of non-​enzymatic browning). Once the ginger has been
grated, PPO attacks phenolic groups, yielding ortho-​quinones.
These, in turn, can react with the GP enzymes to inactivate them.
A well-​known trick to prevent the browning of apples is to use
ascorbic acid (more commonly known as vitamin C). Ascorbic
acid blocks the action of PPO, which in turn prevents the inactiva-
tion of the GP enzymes (Figure 28.4). The same trick also works
for ginger juice. If you need to make ginger juice in advance, just
add a pinch of vitamin C (0.2%, to be precise).
201
Ideas for Further Experimentation
Even though the recipe given here works fine, there are sev-
eral claims that remain to be tested. Feel free to join in with
the experimentation or share it as an idea for a school science
project:
• Is there any difference between old and young ginger?
Which works better?
• What is the minimum amount of ginger juice required
to make the gel set? Does more ginger juice make the
gel harder or softer?
• Does it matter if you pour the milk from a low or a high
height?
• Is it true that mixing within the first couple of seconds
destroys the gel?
• Can milk and ginger juice be mixed cold and then
heated carefully to 65 °C? For instance, this could be
heated in a microwave oven.
202 Martin Lersch

FIGURE 28.4 The enzyme polyphenol oxidase (PPO) present in ginger can convert phenolic compounds into ortho-​quinone capable of blocking the active
ginger protease (GP) enzymes. Addition of ascorbic acid (vitamin C) can block the action of PPO.

• Does the amount of sugar influence the firmness of
the gel?
• Will a few drops of vinegar promote the gelling?
• Instead of adding milk to the ginger juice, try to add
ginger juice to the milk. What happens?
REFERENCES
Lersch M (ed.). 2014. Texture -​A hydrocolloid recipe collection
(v.3.0, 2014). Available for free download from http://​khymos.
org/​recipe-​collection/​.
Mazorra-​Manzano MA, Perea-​Gutiérrez TC, Lugo-​Sánchez ME,
Ramirez-​Suarez JC, Torres-​Llanez M, González-​Córdova
AF, Vallejo-​Cordoba B. 2013. Comparison of the milk-​
clotting properties of three plant extracts, Food Chem., 141,
1902–​1907.
Su HP, Huang MJ, Wang HT. 2009. Characterization of ginger
proteases and their potential as a rennin replacement, J. Sci.
Food Agric., 89, 1178–​1185.
Dehydration
José M. Aguilera
Department of Chemical Engineering and Bioprocesses, Pontificia Universidad Católica de Chile, Avenida Libertador Bernardo
O’Higgins 340, Santiago, Chile
Introduction
Water is an important component of foods, even when they are
“dry” (Table 29.1). Dehydration (or drying) is the application of
heat under controlled conditions to remove by evaporation (or
sublimation in the case of freeze-​drying) most of the water nor-
mally present in a food. Although dehydration is an important
unit operation in food engineering and in the food processing
industry, it is not extensively practised as such in the kitchen
(Guiné, 2018). In fact, there is no such kitchen apparatus as a
“dehydrator”, even though some cooks utilize a sort of tray drier
to remove the moisture from fruits and vegetables, as well as the
kitchen oven (Boyer, 2018). This is unfortunate, since, among
other things, the quality of a dry product or ingredient depends on
a precise programming of the time–​temperature conditions and
the humidity of the air.
Dried products and ingredients are ubiquitous in our kitchens,
e.g., most cereals and legumes, corn and wheat flour, spices,
pasta products, etc. Partial loss of moisture occurs unnoticed in
many culinary processes, such as frying, baking, toasting and
roasting; on other occasions it is deliberately induced (e.g., stock
reductions or oven-​dried vegetables). Oftentimes, chefs collect
fresh materials directly from the field (wild herbs, edible flowers
and leaves, seaweeds, etc.) that need to be dried in their kitchens
for later use.
An important characteristic of dry foods in gastronomy is that
they become brittle, emit sounds on biting, and are sensed as
“crispy” in texture (Tunick et al., 2013). Today, some chefs are
exploiting these properties to produce amazing products such as
a crispy freeze-​dried yogurt and crunchy fish flakes.
Moisture in Foods
Cooks have many problems in dealing with moisture in foods
because recipes are usually expressed by weight of ingredients,
and this includes the water content. Except for ingredients like
salt, table sugar, fats and oil, most food ingredients and food have
water in them. The common term to refer to how much water a
food contains is the “moisture content” (W), which is defined in
at least two ways:
Wetweight − Dryweight
Wwb = ×100
Wetweight
Wetweight − Dryweight
Wdb = ×100
Dryweight
These definitions lead to much confusion, because moisture on
a wet basis (Wwb) is numerically very different from moisture
on a dry basis (Wdb). For example, an apple has a Wwb = 85%
and a Wdb = 567% (for a quick moisture converter, see https://​
booksite.elsevier.com/​9780123985309/​content/​calc/​Chapter_​
1/​Ex_​1_​4.html). The moisture of a food is usually determined
by weighing a sample and exposing it in an oven to a high tem-
perature (e.g., 105 °C) with circulating air or preferably, using
a vacuum oven. The sample is removed after a couple of hours
and cooled in a desiccator (to avoid moisture pickup) before
reweighing. Moisture content is calculated as the difference
between wet and dry weight (see formulae earlier). In food
science, there are “official methods” for determining the mois-
ture content of different foods (AOAC, 2016). In actual practice,
laboratory moisture balances for routine moisture determination
are fast and convenient.
The moisture of foods has major relevance for their chemical,
microbiological and physical stability. In general, dry foods are
quite stable at ambient temperature, while moist foods are not
and need to be refrigerated. A few products, such as honey, jams,
some sausages (e.g., salami) and hard cheeses, look moist but
are stable at room temperature because they contain a combin-
ation of a high concentration of sugars or salt or have a low pH
as well as a reduced moisture content. From the chemical and
microbiological stability viewpoints, they are a separate class of
foods known as “intermediate moisture foods”, or IMF.
Moisture in Air
Air on this planet contains water vapour that we cannot see. The
amount of water vapour in the air is expressed as percentage
relative humidity (% RH). The term “relative” means that it is in
203
204
TABLE 29.1
Approximate Moisture Content of Some Foods and Food Products
(on a Wet Basis)
Food Percent moisture (wb)
Milk 87
Fish 80
Eggs 73
Wheat dough 55
Honey 23
Raisins 18
Wheat flour 14
Rice (and other grains) 14
Dry pasta products 12
Nuts 5 in the field by direct exposure to the sun (e.g., beans,
reference to 100% RH, which corresponds to saturation or fog,
the maximum amount of water that air can take at a given tem-
perature. Interestingly, liquid water is constantly evaporating and
being captured by the unsaturated air even at room temperature;
otherwise, a floor mopped with water would never dry. This is
due to a property called the “vapour pressure” of water in air, or
the tendency of liquid water molecules to become vapour when-
ever they can. The higher the temperature of liquid water, the
larger its vapour pressure, until it equalizes with the atmospheric
pressure and water boils.
So, exposing “dry foods” to air usually results in the uptake
of moisture (see later). This has important consequences when it
comes to crispness. A food may be crisp coming out of the oven
or fryer and become “soggy” if left in a humid environment.
Hygroscopicity of Foods
Another concept that cooks can use is hygroscopicity, or the cap-
acity of a food to exchange water with air. This means that, for
most foods, there is not a “single” moisture content; it depends
on the % RH of the air in which they are allowed to equilibrate.
Moreover, within a food there may be a moisture gradient between
the outside and the interior. Think of fresh bread right out of the
oven and French fries just removed from the fryer. In the latter
case, while the crust is dry and crispy (moisture around 10%),
the mealy interior may have a moisture of 80%. Eventually, the
whole fried potato piece will come to an equilibrium moisture
content and become limp (Miranda et al., 2005).
The Engineering Approach
From an engineering viewpoint, dehydration involves heat transfer
and mass transfer. For water to become a vapour (or steam), heat
has to be supplied to account for the sensible heat (increase in
enthalpy due to temperature) and latent heat (enthalpy needed for
a phase change at a constant temperature). Mass transfer refers to
how moisture migrates from the interior of the food to its surface
and then is captured by air that is hot and dry, and carried away
José M. Aguilera
from the product. Thus, in most dehydration processes, we need
a heat source and a moisture sink. The subject of food dehydra-
tion methods and industrial equipment was reviewed by Fellows
(2017).
Methods for Dehydrating Foods
For a comprehensive review on the subject, the reader is referred
to an article by Maisnam et al. (2017), but key points are as
follows:
• “Natural” drying: sun drying of foods has been
practised since ancient times. Most grains are dried
lentils, corn, wheat, etc.). Dry grains become brittle
and can be ground into flours and meals (e.g., wheat
flour, corn meal, etc.) that are widely used as convenient
ingredients by chefs. The same can be said for some
fruits and vegetables (e.g., sun-​ dried tomatoes and
raisins). Other naturally dried products have found their
way into several local gastronomic dishes. Charqui is
a typical IMF product (aw of 0.70–​0.75) obtained by
salting and sun-​ drying meat from different sources
and used in soups or dishes or as a snack (Torres et al.,
1994). The Incas of the highlands of Peru froze pota-
toes during the cold nights and let them thaw in the
sunny days while exuding the water. The resulting nat-
urally freeze-​dried chuño (black dried potatoes) or tunta
(white dried potatoes) were later ground into a flour
(Peñarrieta et al., 2012).
• Air drying is the process of controlled water removal,
usually by circulating hot and dry air over a piece of
food in a chamber (Figure 29.1a). In high-​moisture
cellular material such as plant and animal tissues, it is
common to observe shrinkage of food pieces during air
drying (e.g., as happens to a grape when it becomes a
raisin). Product shrinkage is mainly due to the exist-
ence of moisture gradients within a product, which
generate structural stresses leading to contraction and
collapse of the cellular structure (Aguilera et al., 2003).
Rehydration of dried fruits and vegetables does not
locate the water where it was in the fresh tissue and gen-
erally leads to poor quality (or different) products (e.g.,
raisins, charqui).
• Spray drying is a method for drying liquid foods (fluid
milk) or extracts (coffee), in which a nozzle distributes
the liquid as droplets (like the spray nozzle of a
garden hose) that fall into a chamber with circulating
hot air (between 120 °C and 180 °C) (Figure 29.1b).
Interestingly, as long as there is liquid water present
in the falling particle, the temperature will remain
much lower than that of the surrounding air (i.e., at
the so-​called wet-​bulb temperature). The droplets con-
tinuously lose water to the warm air and become dry
particles at the bottom of the chamber, from where they
must be quickly removed to avoid excessive heating and
browning of the dry powder. Although this would be an
attractive method for cooks to prepare dry ingredients
Dehydration
from liquid extracts and juices, it has not found its
way into modern kitchens. However, there are several
models of lab bench spray driers available for research
and product development.
• Drum drying takes place when a thin layer of a wet and
viscous paste is dried over a hot surface and continu-
ously removed as a sheet of dry product. This may be
accomplished in the kitchen by contacting a wet food
mix or a slurry with a hot metal surface (Figure 29.1c).
The final products are dried flakes, as is the case with
instant mashed potatoes and gelatin sheets.
• Freeze-​drying, or lyophilization, involves removing
water in the form of vapour from a frozen product (e.g.,
at −40 °C) held under high vacuum (e.g., at less than
5 Pa, or 5 millionths of atmosphere), a process called
sublimation. As water sublimates (passes directly from
ice to vapour), the boundary separating a nearly dry outer
layer from the remaining central frozen portion moves
to the interior of the product (Figure 29.1d). Since there
is never any liquid water present, the product’s matrix
does not collapse, and the original structural appearance
is preserved. Also, micropores are formed in places
occupied by tiny ice crystals, facilitating fast and homo-
geneous rehydration. Moreover, the food never reaches
very high temperatures, so freeze-​dried foods maintain
the shape, texture, colour and flavour of the natural
product much better than with any other method of food
drying. The main applications are freeze-​dried instant
coffee, berries and small fruit pieces, and vegetables for
high-​quality foods.
FIGURE 29.1 Main modes of drying foods.
205
Dried Products
Almost any wet food material can be “dried”, which does not
mean that, in the dry state, it becomes functional for culinary
applications. However, high-​quality dried foods can be obtained,
and a good example is the freeze-​dried menus of the astronauts
developed in the 1960s. Pretreatments are in many cases decisive
for the quality and properties of the dried material. A fast
freezing rate minimizes structural changes in freeze-​dried foods.
Blanching, or briefly exposing the food to boiling water or steam,
is used to prevent enzymatic reactions, shorten the drying time
and eliminate spoilage microorganisms. Some pretreatments are
more elaborated, as in the case of instant mashed potatoes, where
intact potato cells must be present in the dry product. Cell rup-
ture and the liberation of starch results in poor texture (pasting
and sticking) and a gel-​ like consistency of the reconstituted
potato puree.
Overall, dried products have many advantages in the kitchen
and menus. They can be suitably stored at ambient temperature
(usually packaged) and converted into their hydrated state by
immersing them in water or another liquid. Dry foods exhibit
a desirable textural property called crispness, which means that
they fracture during mastication while emitting a loud, high-​
pitched sound (Tunick et al., 2013).
REFERENCES
Aguilera JM, Chiralt A, Fito P. 2003. Food dehydration and structure.
Trends in Food Science & Technology, 14(10), 432–​437.
AOAC. 2016. Official Methods of Analysis of AOAC
INTERNATIONAL, 20th Edition.
Boyer R. 2018. Using dehydration to preserve fruits, vegetables, and
meats. Virginia Cooperative Extension. Virginia Tech, Virginia
State University Publication, 348–​597.
Fellows PJ. 2017. Dehydration. In Food Processing Technology,
Boca Raton: Woodhead Publishing, 661–​716.
Guiné RPF. 2018. The drying of foods and its effect on the physical-​
chemical, sensorial and nutritional properties. International
Journal of Food Engineering, 4(2), 93–​100.
Maisnam D, Rasane P, Dey A, Kaur S, Sarma C. 2017. Recent
advances in conventional drying of foods. Journal of Food
Technology and Preservation, 1, 25–​34.
Miranda M, Aguilera JM, Beristain C. 2005. Limpness of fried potato
slabs during the post-​frying period. Journal of Food Process
Engineering, 28(3), 265–​281.
Peñarrieta JM, Alvarado JA, Bravo JA, Bergenståhl B. 2012. Chuño
and tunta: The traditional Andean sun-​ dried potatoes. In:
Caprara C, Potatoes: Production, Consumption and Health
Benefits, New York: Nova Science Publishers Inc., 1–​12.
Torres EAFS, Shimokomaki M, Franco BDGM, Landgraf M,
CarvalhoJ BC, Santos JC. 1994. Parameters determining the
quality of charqui, an intermediate moisture meat product.
Meat Science, 38, 229–​234.
Tunick MH, Onwulata CI, Thomas AL, Phillips JG, Mukhopadhy
S, Sheen S, Cooke PH. 2013. Critical evaluation of crispy
and crunchy textures: A review. International Journal of Food
Properties, 16(5), 949–​963.
Dispersed System Formalism
Hervé This vo Kientza1,2
1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
In order to describe existing or possible food systems, as well as
physical changes of food ingredients during “culinary” processes,
a preliminary version of the “disperse system formalism” (DSF)
was first introduced under the name “complex disperse system”
(CDS) formalism, later improved with the addition of the “non-
periodical organization of space” (NPOS) formalism. The final,
comprehensive DSF can give a description of the physical state of
dispersion at any scale. It was shown to be important for the study
of food, as well as for food innovation, which will be particularly
useful for the preparation of “note by note” dishes (indeed syn-
thetic food; see the corresponding chapter in this book), in which
pure compounds are used instead of plant and animal tissues as
food ingredients.
Dishes are indeed physico-​chemical systems, with chemical
composition and physical structure. Indeed, “cooking” was once
said to be a “chemical art” because new compounds are produced
during thermal processing (Cuvier, 1810; This, 2009). However,
in this description, it was overlooked that physical transform-
ations occur as well. How should we describe these?
Food ingredients and dishes are frequently made of various
colloidal subsystems (Everett, 1988; Lyklema, 1991; Hiemnez,
1986; De Gennes, 1997; Jones, 2002), colloids being defined by
the International Union for Pure and Applied Chemistry (IUPAC)
as follows:
a state of subdivision, implying that the molecules or
polymolecular particles dispersed in a medium have at least in
one direction a dimension roughly between 1 nm and 1µm, or
that in a system discontinuities are found at distances of that
order. It is not necessary for all three dimensions to be in the
colloidal range: fibers in which only two dimensions are in this
range, and thin films, in which one dimension is in this range,
may also be classified as colloidal. Nor is it necessary for the
units of a colloidal system to be discrete: continuous network
structures, the basic units of which are of colloidal dimensions
also fall in this class (e.g. porous solids, gels and foams).
(IUPAC, 2001)
For instance, emulsions have been known under this name
since 1560, when the French surgeon Ambroise Paré (1509–​1590)
understood that white thick liquids like milk or cream were often
composed of water and fat (Paré, 1560). However, food colloids
are generally more complex than such biphasic systems. Potatoes,
for example, are suspensions of solid starch granules dispersed in
the liquid inside cells, with cells themselves organized into a solid.
Ice cream includes gas bubbles, protein aggregates, fat crystals
and water crystals dispersed in a liquid concentrated solution
(Israelachvili, 1992; Dickinson, 1994). For systems such as butter
at room temperature (20 °C, for example), the physical microstruc-
ture remains poorly known (Lopez and Ollivon, 2009), but at least
three phases coexist: solid fat, liquid water and liquid fat.
In order to describe the microstructure of colloidal systems, and
of food systems in particular, the CDS formalism was introduced
at the European Congress on Interface Science (ECIS) in 2002
(This, 2007). Later, in 2003, another formalism called NPOS was
added for the overall description of dishes and the relative distri-
bution of materials described by the CDS formalism. Finally, it
was recognized that these two formalisms could be mixed into a
more comprehensive description (DSF), which can be applied for
the description of formulated products as well as for innovation
(This, 2009).
Here, in order to present this description, we begin by
describing an egg, before moving to the general formalism that
we shall use for innovation.
A Simple Case for Training: How to Describe a
Raw Egg
The main issue of descriptions is that, like definitions, they can
be a never-​ending enterprise, but, at the same time, the idea of
going into more precision by orders of magnitude can be useful
(Raiman, 1991; Mavrovouniotis and Stephanopoulos, 1988). This
idea has to be used in DSF: a choice of the pertinent descriptions
to make is needed. In the following example of the description of
a chicken (Gallus gallus) egg, of which the physical structure is,
of course, known, we shall see how the various characteristics of
DSF are needed.
Let’s begin by observing simply that an egg is made of a shell
including the egg white, in the inside of which the yolk lies (Belitz
et al., 2009). In order to make a shorter and faster description
207
208 Hervé This vo Kientza

of this structure, we can use the “@” symbol, also used inter-
nationally in chemistry for describing inclusion in interstitial
compounds. Here, using the letters S for shell, W for white, Y for
yolk, the egg could be described as:

Y@W@S

For colloidal systems, and for physical objects made of colloidal

parts, it can be interesting to indicate the physical phases of the
various parts: solid (S), or liquid (L), or gaseous (G). Focusing
now on phases, the egg formula could be:

(L1@L2)@S

However, the shell does not have the same “importance” as the
white and the yolk; it is a thin layer, whereas the two other parts
are bulky. This is why the “physical dimension” (number of
dimensions, not size) is usefully taken into account in the formal
description of DSF. Of course, the number of dimensions of
physical objects is not exactly the number of dimensions of math-
ematical objects, but a formal definition can be given considering FIGURE 30.1 The ultrasonic picture of an egg yolk shows its internal
the diameters in the three directions of space, compared to a ref- organization.
erence size, for example, the size of the plate in which a dish is
served (Mandelbrot, 1982). In this regard, an object whose size
in the three directions of space is smaller than the reference size
by more than one order of magnitude should be considered to
be of zero dimension (e.g., a rice seed on a plate). An object for
which two directions are smaller than the reference size by more
than one order of magnitude is of dimension one (a piece of spa-
ghetti on a plate), and an object for which one direction is smaller
than the reference size by more than one order of magnitude is
of dimension two (a salad leaf on a plate). The other objects are
three-​dimensional.
Using D0, D1, D2 and D3 to describing objects of dimension 0,
1, 2 and 3, respectively, one can improve the previous description
of eggs: if the reference size is the length of the long axis of the
egg, this one could be described as: FIGURE 30.2 On this microscopic picture of an egg yolk, the granules
appear as dispersed in the plasma.
[D3(L1)@D3(L2)]@D2(S)

The example of eggs is interesting, because the yolk is made of con-
centric layers called light and deep yolk (Anton, 1998), deposited
during the day and the night, respectively; the number of these is
about nine, as shown from ultrasound scan pictures (Figure 30.1)
(This, 2003; This, 2006). As each layer is composed of solid
granules (S) dispersed in a plasma (aqueous solution of proteins
and other compounds W), the full yolk could be also described as:
(D0(S)/​D2(W))@9
However, the question is whether this second formula should be
grouped with the first, as in:
different thicknesses around the yolk when the egg is poured on
a flat surface. The shell, as well, could be described in more and
more detail. The answer to this question of the precision of the
formal description is to be based on the goal of the description
using DSF, keeping in mind that it is often not appropriate to mix
two different scales. For sure, the egg yolk used for making an
emulsion is made of granules (Figure 30.2), but they are unim-
portant when only the emulsion is concerned. Indeed, granules
would not appear in a microscopic picture of a mayonnaise at
a magnification that would visualize the oil droplets dispersed
in the liquid. At the end of the next paragraph, we discuss this
question more formally.

D3((D0(S)/​D2(W))@9)@D3(L2)@D2(S)

Is it useful? The same question could be asked about the egg The Objects for the Formal Description
white, because it is not a simple solution of proteins but has a Starting from this example of an egg, one can now present the
colloidal sub-​structure, as demonstrated by the fact that it makes DSF system more generally. For the formal description of the
Dispersed System Formalism
physical structure of systems, the DSF describes three kinds
of aspects: the nature of the material, the organization of the
dispersions, and the dimensions of colloidal parts.
Let us observe that the DSF is a generalization of the old O/​W
description used in colloid science.
Indeed, for phases, the letters G, O, W and S are proposed to
stand for “gas”, “oil” (any hydrophobic liquid phase), “water” (any
aqueous solution) and “solid”. If needed, indexes can be added
(for example, when two different solids are present), and other
symbols could be used when needed (such as E for “ethanol”).
Operators are used, between the letters representing phases, to
make formulas. As said before, the IUPAC symbol “@” describes
an inclusion, and the symbol “/​” has been used in physical chem-
istry to describe a random dispersion, such as in emulsions and
foams. The symbol “+” is used to describe a mixture of phases
that can be dispersed into another one, such as (G+O)/​W for an
aerated emulsion: a water solution is the continuous phase, and
gas bubbles and oil droplets are randomly dispersed into this
solution. The symbol “σ” indicates a superposition.
If needed, geometrical connectors such as σx, σy and σz can
be used instead to represent superposition in the direction x, y
and z, respectively. Any particular direction can be given by the
Cartesian coordinate of the vector, such as in (u, v, w), or other
coordinate systems such as {r, θ, φ} for a spherical organiza-
tion. Finally, the “x” symbol is used to describe the mixing of
FIGURE 30.3 The four distributions for which energy is calculated.
209
two continuous phases, such as in gelatin gels (SxW). Up to now,
these connectors have been enough to describe all food systems
that have been considered, but, again, more operators could be
introduced if needed.
Now, for describing the non-​periodical organization of all the
phases in space, the elementary objects are described according to
their “physical dimension”: D0, D1, D2 and D3. D0 stands for zero-​
dimensional objects (“dots”), i.e., objects whose size in the three
directions of space is more than one order of magnitude lower
than a “reference size” that was chosen initially (the scale where
the full object is considered). D1 stands for lines, i.e., objects with
only one macroscopic dimension. D2 stands for surfaces, i.e.,
objects with one dimension lower than the two others by more
than one order of magnitude. D3 stands for volumes. If necessary,
Dk objects could be considered, “k” being non-integer, and these
objects then being fractals (Mandelbrot, 1982).
Added rules give coherence to this formalism. First, in order
to avoid having more than one formula for a particular physical
system, the components of a sum have to be written in alphabet-
ical order, for example, (G+O+S)/​W. Secondly, simplifications
sometimes have to be made; for instance, G/​G is equal to G. More
importantly, an order also has to be respected for operators. In
order to do this, the determination of a free energy can be used,
assuming, for example, simply an interface energy γAB between
the phases A and B (Figure 30.3).
210
Considering two phases that do not mix at a temperature T, such
as oil and water, the ratio of free energy (or more precisely, free
enthalpy) between superposition and inclusion can be calculated
simply. Let us assume two material substances, A and B (A being
less dense than B), with respective volumes VA and VB. The sur-
face energy between A and air is neglected, and the two phases
have a contact surface of area c² (where c is the length of the
edge of A, assuming it has a cubic shape). The free enthalpy is
then equal to γAB VA1/​3. For the inclusion of a phase into another,
corresponding to the operator @, it would be equal to 6. γAB VA2/​3,
showing that the free enthalpy for inclusion is about one order of
magnitude higher than for superposition.
To compare the free enthalpy for mixtures (intermixing oper-
ator x) and dispersions (operator /​), the calculation was made on
a network.
Let us assume that, for the system A x B, the phase A is around
the edges of a cubic cell with n elementary cells per side of the
total volume. The free enthalpy would then be:
4.√3.γAB . n.VA1/​2.(VA+VB)1/​6
For the dispersion, n3 elementary cubes of A are considered
dispersed in B, and the total free enthalpy is:
6.γAB . n.VA2/​3
Numerical comparison (with γAB chosen to be 80, n = 100,000,
T = 300 K, VA = 1, VB = 1) shows that intermixing needs about
1.3 times more free energy than dispersion.
The formal description can be increased in many ways,
if needed. For example, repetitions can be described by
exponents, as could be seen for the example of the egg. Another
possibility is to give the quantity of each phase by a subscript,
such as O95/​W5 to describe an emulsion of 95 g oil into 5 g
water. The size of structures can be also given in brackets, such
as in O200[10–​6–​10–​5]/​W5, where the powers of 10 indicate the
minimum and maximum radii of dispersed oil droplets (SI
units should be used). In order to be more precise about the
size of structures, the size of the smallest structures considered
can be given in brackets as a “reference size” at the end of the
formula. For example, O[10–​5–​10–​4]/​W [d > 10–​5] indicates that
the structures considered are larger than 10–​5 m, so that smaller
granules are not taken into account. Kinetic parameters such
as time or energy can be added to describe the evolution of the
system. The equation O/​W + G → (G+O)/​W can be replaced
by the following formula:
(Gt=0…50 + O30(100 − t)/​100)/​W70(100 − t)/​100
where the time t is in seconds, the gas would be introduced at a
regular rate and indices give volume instead of mass.
Up to now, no food system has resisted description by this
formalism. However, do all formulae correspond to possible
systems? Many dispersed systems are metastable and not thermo-
dynamically stable. Indeed, they evolve, depending on the size of
their structures or on the nature or quantity of stabilizing elements
Hervé This vo Kientza
like surfactants in emulsions. This is also a question of kinetics,
not of thermodynamics.
Mixing Orders of Magnitude for an Overall
Description
Let us finish this discussion by considering many orders of mag-
nitude together. Using DSF, one can describe any physical struc-
ture at any scale. For example, let us consider a dish P1 made
of many macroscopic parts P1,1, P1,2,…, P1,n(1), each part being
made itself of parts at a smaller scale, and so on. Using the DSF
operators, this could be described as:
P1 = P1,1 op1,1 P1,2 op1,2 P1,3… o1,n(1) − 1 P1,n(1)
where the opi,j are chosen among the DSF operators and where
P1,i is itself made of parts, such as in the equation:
P1,i = P2, i,1 op2, i,1, P2,i,2 op2,i,2 … op2,i,n(2, i) − 1 P2,i,n(2,i)
and so on until molecular scale.
All scales are not always useful. For example, if the first ref-
erence size is between 2 × 10−2 and 2 × 10−1 m, the second scale
could be between 2 × 10−3 and 2 × 10−2 m, the third between
2 × 10−4 and 2 × 10−3 m, and so on until 2 × 10−10 and 2 × 10−9 m
(molecular scale). It is perhaps not useful to describe the system
at all scales (nine orders of magnitude), because some objects
would not appear to be different on more than two orders of mag-
nitude. For food, studies in our group show that it would often be
enough to consider macroscopic sizes, between 10 cm and 1 mm
(scale 1); microscopic, between 0.1 mm and 0.001 mm (scale 2);
nanoscopic, between 1 μm and 0.01 μm; and molecular, between
10 nm and 0.1 nm (Hornyak, 2009).
Use of DSF for Scientific Explorations and for
Innovation
DSF was initially introduced for the analysis of the differences
between different kinds of “gels”, but it is also useful for innov-
ation. The importance of algebraic notation has been recognized
for many centuries, and it was a major success of René Descartes,
Wilhelm Gottfried von Leibniz and Isaac Newton to use it in
mathematics and physics. In a treatise on logic published in 1918,
the French logician Edmond Goblot discussed how notation can
lead to discovery (Goblot, 1918):
For the algebra of logic, its inventors probably never thought
that it was only a notation of concepts, relationships and elem-
entary operations for logicians, and they had never had any
doubt on the difference between discovery of a truth and the
invention of a notation for expressing it when it is discovered.
Notation can lead to discovery, as it occurred frequently in
algebra. To general and abstract concepts, intractable without
formula, cumbersome to use with words and common
Dispersed System Formalism
language, the algebra of logic, as ordinary algebra, substitutes
concrete and regular symbols which can be organized in a
wealth of combinations and reduce heavy mind operations to
very simple written processes.
This idea to simplify operations through an algebraic formalism REFERENCES
was already developed by Antoine Laurent de Lavoisier (1743–​ Anton M. 1998. Structure and functional properties of hen egg yolk
1794), and it was the basis of the introduction of modern chem- constituents. Recent Research Development in Agricultural
ical notation (Lavoisier, 1782): and Food Chemistry, 2, 839–​864.
In order to better show the state of the issue, and to present
synthetically the result of what is going on during metal
dissolutions, I build formulas, that could be confused with
algebra, but do not derive from the same principles; we are
very far from the time when the precision of mathematics can
be introduced in chemistry, and I invite the reader to consider
the formula that I shall give only as simple annotations, whose
aim is to think easier.
Indeed, in 1995, a new dish named “Chocolate Chantilly” (see
the chapter on Chantillys in this book) was based on the equation
O/​W + G → (G + O)/​W (This, 1996). First, a chocolate disper-
sion ((O+S)/​W) is made by heating chocolate in water with the
same final fat/​water ratio as in ordinary cream: here O stands for
the melted cocoa fat and S for the cocoa plant particles, the sugar
crystals having dissolved in the aqueous phase. Then, this disper-
sion (mainly an emulsion) is whipped (+G) at room temperature
while cooling. The very unstable hot (G+O)/​W system is slowly
transformed into a more stable “chocolate mousse” G/​g(O, S,
W), when part of the fat crystallizes and traps the dispersed oil
and gas structures (here we use again a function g to indicate
the unknown continuous structure made of liquid fat, solids –​fat
and plant pieces –​and aqueous solution). This mousse contains
no eggs, in contrast to a traditional chocolate mousse (Larousse
Gastronomique, 1996), and its texture can be the same as that of
whipped cream. Of course, chocolate can be replaced by other
food products, such as as cheese, foie gras or even butter, leading
to “cheese Chantilly”, “foie gras Chantilly” or “butter Chantilly”,
respectively.
The use of DSF can lead to a wealth of other new systems with
both scientific and culinary interests. For example, using four
phases and four connectors, the number of formulae is 114,688,
and more than 106 with six phases; there is plenty of room for
innovation.
Indeed, DSF will be a valuable tool for the creation of “note
by note dishes” (This, 2016a, 2016b), i.e., dishes made from
pure compounds, as the consistency will have to be built, and it
is likely that the gels of class 1 will not be enough to achieve the
target systems.
Conclusions and Perspectives
As we said, DSF can lead to innovation (technology), but it can
also contribute to the scientific development of molecular and
physical gastronomy, as for the study of flavour release. For
211
example, considering the many possible gels of class 2, one can
now ask whether they can be grouped into categories of flavour
releasing systems. This is shown in the chapter about gels in
this book.
Belitz HD, Grosch W, Schieberle P. 2009. Food Chemistry, Springer,
Heidelberg.
Cuvier G. 1810. Rapports à l’Empereur sur les progrès des sciences,
des lettres et des arts depuis 1789. T. II. Chimie et Sciences de
la nature, Belin, Paris, re-​edition 1989.
De Gennes PG. 1997. Soft Interfaces, the 1994 Dirac Memorial
Lecture, Cambridge University Press, Cambridge.
Dickinson E. 1994. Structure, properties and functions. In Nishinari
K, Doi E. (eds.). Food Hydrocolloids, Plenum Press, New York.
Everett DH. 1988. Basic Principles in Colloid Science, Royal Society
of Chemistry, London.
Goblot E. 1918. Traité de logique, Armand Colin, Paris, 18.
Hiemnez PC. 1986. Principles of Colloid and Surface Chemistry.
Marcel Dekker Inc, New York.
Hornyak GL. 2009. Fundamentals of Nanotechnology. Taylor &
Francis, Boca Raton, FL.
Israelachvili J. 1992. Intermolecular & Surface Forces, 2nd Ed.
Academic Press, London.
IUPAC. 2001. Manual of Symbols and Terminology for
Physicochemical Quantities and Units, http://​ old.iupac.org/​
reports/​2001/​colloid_​2001/​manual_​of_​s_​and_​t/​node33.html
Jones RAL. 2002. Soft Condensed Matter. Oxford University Press,
Oxford.
Larousse Gastronomique. 1996. Larousse-​Bordas, Paris.
Lavoisier AL. 1782. Considérations générales sur la dissolution
des métaux dans les acides. Mémoires de l’Académie des
sciences, 492.
Lopez C, Ollivon M. 2009. Triglycerides obtained by dry frac-
tionation of milk fat 2. Thermal properties and polymorphic
evolutions on heating. Chemistry and Physics of Lipids,
159, 1–​12.
Lyklema J. 1991. Fundamentals of Interface and Colloid Science,
Academic Press, London.
Mandelbrot B. 1982. The Fractal Geometry of Nature. W. H. Freeman
and Co., New York.
Mavrovouniotis ML and Stephanopoulos G. 1988. Formal order-​
of-​
magnitude reasoning in process engineering, Computer
Chemical Engineering, 12(2/​10), 867–​880.
Paré A. 1560. Emulsion. Trésor de la langue française informatisé,
CNRS-​ Université de Lorraine. Access: http://​ atilf.atilf.fr,
research “emulsion”.
Raiman O. 1991. Order of magnitude reasoning, Artificial
Intelligence, 51, 11–​38.
This H. 1996. Le chocolat Chantilly, Pour la Science, 230, 12.
This H. 2003. La gastronomie moléculaire, Sciences des aliments,
23(2), 187–​198.
This H. 2006. Let’s have an egg. In Hosking R (Ed.), Eggs in
Cookery, Proceeding of the Oxford Food Symposium on Food
and Cookery, Prospect Books, Oxford.
This H. 2007. Formal descriptions for formulation, International
Journal of Pharmacy, 344(1–​2), 4–​8.
This H. 2009. Molecular gastronomy, a scientific look at cooking.
Accounts of Chemical Research, 575–​583.
212
This H. 2013. Molecular gastronomy is a scientific discipline, and
note by note cuisine is the next culinary trend, Flavour, 2(1).
This H. 2016a. Statgels and dynagels. Notes Académiques de
l’Académie d’agriculture de France, /​Academic Notes from
the French Academy of Agriculture, 12, 1–​12.
Hervé This vo Kientza
This H. 2016b. What can “artificial meat” be? Note by note cooking
offers a variety of answers, Notes Académiques de l’Académie
d’agriculture de France (N3AF), 6, 1–​10.
Distillation: The Behaviour of Volatile Compounds during
Distillation of Hydro-​Alcoholic Solutions and during
Hydro-​Distillation
Martine Esteban-​Decloux
Unité Mixte de Recherche Ingénierie Procédés Aliments, AgroParisTech, INRAE, Université Paris-​Saclay, F-​91300 Massy,
France
When food ingredients are cooked, many volatile compounds are
formed and escape. The goal of this chapter is to give some infor-
mation on the behaviour of odorant compounds during the distil-
lation of hydro-​alcoholic solutions, and to make a link with their
elimination through steam distillation.
Characterization of Volatility
Distillation is a unit operation that makes it possible to separate
compounds all together present in one phase by using volatility
differences. It is generally performed at pressures close to atmos-
pheric pressure (from 0.2 bar to some bars).
For each pure species separately, it is possible to calculate the
saturating vapour pressure Psat as a function of temperature by
using equation (31.1). The coefficients are generally given in
handbooks or databases (Perry et al., 1997; ProSim, 2019).
 b 
P sat = exp  a + + c.ln(T ) + d.T e  (31.1)
 T 
where T (kelvin) is the absolute temperature, Psat (Pascal) is the
saturating vapour pressure, and a, b, c, d and e are particular
constants for the various compounds. For example, Table 31.1
gives the values for water, ethanol and some categories of vola-
tile compounds of the families of alcohols, esters, aldehydes and
terpenes.
Separately, at a particular temperature, water and ethanol do not
have the same partial pressure; the vapour pressure of ethanol is
always higher than that of water. For example, at 100 °C, the vapour
pressure for water is 101.3 kPa, and it is 224.5 kPa for ethanol.
Ethanol, having a vapour pressure much higher than that of water
(by a factor of 2.2), is thus said to be more volatile than water.
Partial Pressures for the Liquid–​Vapour Equilibrium
According to Dalton’s law, the total pressure for a vapour of n
compounds is equal to the sum of the partial pressures for each
compound of the mixture. In equation (31.2), Pi is the partial
pressure of the ith compound in the vapour phase, and P is the
total pressure:
n
∑P = P
i =1
i
(31.2)
At the pressure traditionally used in distillation or hydro-​
distillation, the behaviour of the vapour phase can be described
using the ideal gas law. As a consequence, the partial pressure Pi
of the ith compound in the vapour phase can be expressed as a
function of the molar fraction yi of this compound in the vapour
phase and the total pressure P (31.3):
Pi = yi .P (31.3)
Interactions in the liquid phase can be described using activity
coefficients. Hence, the partial pressure Pi of the ith com-
pound can be expressed using the activity coefficient γi of the
molar fraction xi of the ith compound of the liquid phase and
the saturating vapour pressure at the considered temperature
(31.4):
Pi = γ i (T , xi ) xi .Pisat (T ) (31.4)
Absolute and Relative Volatility
In order to know the behaviour of the volatile compounds in a
solution, one needs to look for the data for the liquid–​vapour
equilibrium (how the compounds distribute between the boiling
liquid and the saturated vapour phases). Each compound, i, is
characterized by an absolute volatility coefficient, Ki, which is
the ratio of molar concentrations yi of the ith compound in the
saturated vapour and the molar concentration xi in the boiling
liquid (31.5):
213
214 Martine Esteban-Decloux

TABLE 31.1
Coefficients for the Riedel Equation (31.1) According to ProSim (2019) for Water and Some Compounds of the Families of Alcohols, Esters,
Aldehydes and Terpenes
Compound Tmin (°C) Tmax (°C) a b c d e
Water 0 374 73.64900 −7258.20000 −7.30370 4.1653 × 10−6 2
Ethanol −114 241 73.30400 −7122.30000 −7.14240 2.8856 × 10−6 2
Methanol −98 239 82.11780 −6905.50000 −8.86220 7.4664 × 10−6 2
1-​propanol −126 263 84.66416 −8307.24422 −8.57673 7.0509 × 10−18 6
3-​methyl-​1-​butanol −117 304 117.0740 −10743.2000 −13.1654 1.1670 × 10−17 6
2-​phenylethanol −27 411 65.07600 −8985.10000 −5.70340 4.6087 × 10−18 6
Ethyl acetate −84 250 66.82400 −6227.60000 −6.41000 1.7914 × 10−17 6
Hexyl acetate −81 345 74.72700 −8352.60000 −7.29240 7.1794 × 10−18 6
Ethyl caproate 5 189 64.83120 −7853.33462 −5.82651 0.000 0
Ethyl lactate −26 334 72.66870 −8249.65000 −6.91582 7.5520 × 10−18 6
Acetaldehyde −124 193 52.91070 −4643.14000 −4.50683 2.7028 × 10−17 6
Linalol 10 198 128.0810 −12163.6320 −14.7440 0.000 1
D-​limonene −74 380 75.57400 −8079.70000 −7.55960 8.3872 × 10−18 6

yi behaves like a pure species, i.e., the vapour has the same compos-
Ki = (31.5) ition as the liquid. Hence it is impossible, even in a highly com-
xi
plex distillation system, to produce pure ethanol.
In the European food industry, the official unit for the char-
Generally, the volatile compounds are present in very low
acterization of the ethanol concentration of a solution is the
concentrations, and the interactions between them can
Alcoholic strength By Volume (ABV in % v/​v) (Official Journal
be neglected. In contrast, the influence of the solvent is
of the European Regulation, 2019, article 4, 23rd point). This
often very important. For mixtures, one can define relative
is the ratio in percentage between the volume of pure ethanol
volatilities αi/​j, for example for the ith compound relative to
at a temperature of 20 °C, in a solution, and the total volume
the jth one (31.6):
of this solution at the same temperature. Using relationships
Ki y x published by the International Organization of Legal Metrology
α i = = i i (31.6) (1975) between ABV and densities, a polynomial relation was
j K j yj xj
established between ethanol molar fraction and ABV (31.7),
where xEt is the ethanol molar fraction and ci the constants of
All these concepts being defined, we shall now examine (1) the Table 31.2.
principle of the distillation of a hydro-​alcoholic distillation,
(2) the behaviour of volatile compounds in a hydro-​alcoholic 5

solution during a simple distillation and (3) the behaviour of ABV = ∑ c .( x

i =1
i Et )
i (31.7)
volatile compounds in water during hydro-​distillation.

Distillation of a Hydro-​Alcoholic Solution
The equilibrium data for the water–​ethanol binary mixture are
well known. They can be represented in a diagram of composition
(Figure 31.1) giving, for a particular pressure, the molar concentra-
tion of the boiling liquid in equilibrium with the saturated vapour.
These data were calculated using the Simulis Thermodynamics
software (ProSim, 2019) using the Non-​ Random Two-​ Liquid
(NTRL) coefficients given by Puentes et al. (2018a). The two
phases in equilibrium are at the same temperature. Using the dia-
gram, one can determine, for a pressure of 1.013 bar, the tempera-
ture at which the liquid reaches its boiling temperature and the
composition of the vapour in equilibrium at this temperature. One
can observe that the water–​ethanol system forms an azeotrope at a
molar fraction equal to 0.8943: for this composition, the mixture
One can trace the composition diagram using the ABV unit
system, more convenient for culinary uses (Figure 31.2).
Such plots of the composition of a hydro-​alcoholic solution at
its boiling temperature are useful for a simple batch distillation.
Let us consider the example of a hydro-​alcoholic solution with
an ABV equal to 10% v/​v. At a pressure of 1.013 bar, it reaches
its boiling temperature at 92.8 °C (at this temperature, the vapour
pressure due to the liquid is equal to 1.013 bar). The vapour from
this solution is at 54.5% v/​v. The production of this vapour, richer
in ethanol than the liquid, generates a drop in the ethanol concen-
tration of the liquid, and in this way, an increase of the boiling
temperature and a decrease in the ethanol concentration of the
vapour produced. This phenomenon goes on until nearly all the
ethanol of the liquid is evaporated. It is important to understand
that, in a solution with many compounds, the vapour is composed
of most volatile compounds, but in different proportions. In the
Distillation: Volatile Compounds 215

FIGURE 31.1 Composition diagram for a water–​ethanol system at a pressure of 1.013 bar in molar fractions.

TABLE 31.2
Coefficients for the Determination of ABV Using Equation (31.7)
c1 = 320.68 c2 = −560.14 c3 = 654.94 c4 = −439.14 c5 = 123.76

FIGURE 31.2 Composition diagram for a water–​ethanol system at a pressure of 1.013 bar in ABV (% v/​v).

case of a water–​ethanol solution with 10% v/​v that is set to boil Distillation of Volatile Compounds in a
with condensation of the vapour being formed, the evolution of Hydro-​Alcoholic Solution
the ethanol content of the liquid and the vapour can be described
The goal of distilling hydro-​alcoholic solutions is to extract from
as in Figure 31.3.
the initial liquid the volatile compounds and to separate them
In some cases, distillations (simple evaporation) are performed
when possible or when needed. For fermented must (grapes,
at a pressure below atmospheric pressure (so-​called “vacuum
apples, pears or grain), the main compound (except water) is, of
pressure”). The equilibrium data are modified, with a strong
course, ethanol, but the primary interest is to fractionate volatile
decrease in boiling temperatures when the pressure is low (as
compounds that influence the quality of the distillate. The key
shown in Figure 31.4 for the water–​ethanol system) and a slight
question is then their volatility in the boiling medium.
increase of the relative volatility of ethanol.
216 Martine Esteban-Decloux

FIGURE 31.3 Evolution of the ABV for the produced vapour (_​ _​ _​) and the liquid remaining in the boiler ( _​_​_​ ) as a function of time.

FIGURE 31.4 Equilibrium data in ABV (% v/​v) for a water–​ethanol system as affected by pressure.

For the distillation of hydro-​alcoholic solutions, the vola-
tility of most compounds depends on the ethanol content of the
solution. Puentes et al. (2018a) reviewed the published equilib-
rium data and determined the coefficient of the NRTL model
(Non-​Random Two Liquid) for using such data in simulation
software. It is then possible to determine the absolute volatility
of these compounds according to the ethanol concentration of
the liquid. As an example, Figure 31.5 shows the shape of the
absolute volatility (Ki in equation (31.5)) of some compounds,
including ethanol. Obviously, some compounds are always more
volatile than ethanol (ethyl acetate and acetaldehyde), whereas
others have a volatility that varies with the ethanol content
in the liquid (3-​methylbutan-​ol and methanol) or a volatility
always lower than that of ethanol (ethyl lactate, acetic acid and
2-​phenylethanol).
Puentes et al. (2018a) proposed a classification for 44
compounds based on their relative volatility (31.6) with respect
to ethanol or water (αi/​Et or αi/​W): (i) light compounds, which are
always more volatile than ethanol, whatever the ABV of the liquid
(αi/​Et > 1); (ii) intermediate compounds, with a relative volatility
higher than that of ethanol at low ABV but a lower volatility at
high ABV; and (iii) heavy compounds, always less volatile than
water (αi/​W < 1) (Table 31.3). Depending on the distillation pro-
cess (simple or with a column), the compounds do not have the
same distribution in the distillate and the residue.
During simple distillation (evaporation), the compounds
are more concentrated in the first vapour fractions when their
volatility with respect to ethanol is higher; this is the case, for
example, for ethyl acetate. Because of their strong extraction in
the vapour phase, the concentration of such compounds in the
Distillation: Volatile Compounds
FIGURE 31.5 Variation of the absolute volatility (Ki) of some compounds as a function of the ethanol content of the liquid: (a) volatilities always higher than
ethanol and (b) volatilities lower than ethanol.
liquid decreases rapidly. As a consequence, their concentration
in the vapour also decreases. Such compounds belong to “heads”
because they are extracted at the beginning of the distillation
process. The compounds with a volatility comparable to that of
ethanol, such as methanol or 3-​methylbutan-​1-​ol, have a distilla-
tion behaviour similar to that of ethanol and are particularly dif-
ficult to separate from ethanol. Among intermediate compounds,
some, such as 2-​phenylethanol or ethyl lactate, have a volatility
that is always lower than that of ethanol, but their volatility with
respect to water increases when the ethanol content decreases.
These compounds are evaporated at the end of distillation, when
the ethanol content is greatly decreased. Finally, the compounds
with a volatility lower than water remain in the liquid. As an
example (Figure 31.6), the behaviour of some compounds during
a distillation of cognac was shown by Douady et al. (2019).
One can consider that, in a column distillation, when the
initial solution is separated into a distillate with high ethanol
content and a residue with low ethanol concentration, all light
compounds are extracted by the distillate and heavy compounds
by the residue. In contrast, the distribution of the intermediate
compounds depends on the ethanol content of the distillate; the
lower this content, the greater the extraction of the compounds in
the distillate. This occurs, for example, during Armagnac produc-
tion, for which the ABV of the liquid in the column is less than
10% v/​v (Puentes et al., 2018b).
Finally, for the production of ethyl alcohol of agricul-
tural origin with a minimum ABV of 96.0% v/​v and limited
concentrations of other volatile compounds (Official Journal of
217
the European Regulation, 2019, article 5), more than seven dis-
tillation columns, each with a high number of plates, are coupled
in order to eliminate all the volatile compounds except ethanol
(Esteban-​Decloux et al., 2014).
It is important to remember that the matrix and the method
of distillation play an essential role in the behaviour of volatile
flavouring compounds during distillation.
Behaviour of Volatile Compounds in Boiling Water
Hydro-​distillation (or steam distillation) is a particular distilla-
tion for which the solvent is water or steam. As we have seen
before, the compounds are more volatile when the ethanol con-
tent is low. This is due to their relative affinity for organic and
aqueous phases, which can be characterized by the partition coef-
ficient kow (octanol/​water). The partition coefficient is a measure-
ment of a compound’s differential solubility in octanol and water,
but, because of very large values, the logarithm of this partition
coefficient is used. The higher the log kow, the more lipophilic the
compounds are (and the less hydrophilic they are).
When the concentration of a compound in a solution is higher
than its solubility limit, its partial pressure in the vapour does
not depend on its concentration but only on the temperature.
It can then be extracted at a temperature lower than its own
boiling temperature when pure. For example, limonene is the
main compound of the essential oil of most citrus fruits and
extracts (Deterre et al., 2011). Its solubility in water at 25 °C is
218 Martine Esteban-Decloux

TABLE 31.3
Classification of Volatile Compounds According to Their Relative Volatilities with Respect to Ethanol (αi/​Et) and Water (αi/​W) over a Wide
Ethanol Composition Range in the Liquid Phase
αi/​Et αi/​W αi/​Et αi/​W
min max min max min max min max
Light compound Intermediary compound
3-​methylbutanal 1.3 19.8 1.2 239.6 3-​methyl-​1-​butanol 0.1 2.7 0.1 32.1
Propan-​2-​yl ethanoate 1.7 14.2 1.7 172.0 2-​methyl-​1-​propanol 0.4 2.5 0.3 30.0
2-​methylpropanal 1.1 12.8 1.0 154.6 2-​methyl-​1-​butanol 0.1 2.5 0.1 30.2
Ethyl acetate 1.8 10.8 1.6 131.2 1-​butanol 0.2 1.9 0.2 23.1
Butanal 1.7 10.5 1.5 125.9 cis-​3-​hexen-​1-​ol 0 1.9 0 23.4
Acetaldehyde 5.1 6.3 5.3 61.6 1-​propanol 0.6 1.4 0.5 16.9
1,1-​diethoxyethane 3.3 6.8 5.6 40.0 Methanol 0.6 1.5 1.4 6.8
Propanal 1.8 6.5 1.7 51.3 2-​phenylethyl ethanoate 0 1.5 0 18.8
2-​methylpropyl methanoate 1.7 6.0 1.6 72.8 Octanoic acid 0 1.4 0 17.2
Prop-​2-​enal (acrolein) 2.4 4.5 2.4 30.1 Allyl alcohol 0.5 0.9 0.5 10.8
Intermediary compound Hexanoic acid 0 0.7 0 8.1
Ethyl decanoate 0 82.9 0 1,005.3 Furan-​2-​carbaldehyde 0.1 0.4 0.1 5.3
Ethyl 2-​methylpropanoate 1.0 33.0 0.9 399.3 Ethyl lactate 0.1 0.3 0.1 3.9
Ethyl octanoate 0 28.0 0 339.4 3-​methylbutanoic acid 0 0.3 0 4.0
3-​methylbutyl ethanoate 0.1 18.5 0.1 224.0 2-​methylpropanoic acid 0 0.2 0 2.1
Ethyl 3-​methylbutanoate 0.5 18.1 0.4 219.5 Butanoic acid 0 0.2 0 2.0
Pentanal 0.9 10.1 0.8 122.9 Diethyl butane-​1,4-​dioate 0.1 0.2 0.1 1.9
Linalool 0 8.2 0 99.5 2-​methylbutanoic acid 0 0.2 0 2.0
Ethyl hexanoate 0.6 6.7 0.5 81.6 2-​phenylethanol 0 0.1 0 1.3
Linalool oxide 0 5.2 0.1 63.4 Propanoic acid 0.1 0.1 0 1.0
2-​methyl-​2-​propanol 0.7 2.7 0.7 33.2 Heavy compound
1-​hexanol 0.1 2.9 0.1 35.3 Acetic acid 0.1 0.1 0.1 0.8
2-​propanol 0.9 1.8 0.8 22.0 Formic acid 0 0.1 0.1 0.5

Source: Puentes et al. (2018a)

13.8 mg.L-​1, its boiling temperature at 1.013 bar is 176 °C and
its molar mass is 136.23 g.mol-​1 (The Good Scents Company,
2019). When a water–​limonene system (with a concentration of
limonene higher than the solubility limit) is put to the boil, the
partial pressure for each compound depends only on temperature
(equation (31.1) and the coefficients in Table 31.1), and the total
pressure is equal to the sum of partial pressures (equation (31.2)).
Because the solution reaches its boiling temperature when the
total vapour pressure is equal to the atmospheric pressure, the
boiling temperature of water with an excess of limonene is
97.49 °C (Table 31.4). This value is lower than the individual
boiling temperatures of limonene and water at the same pressure.
The molar fraction of limonene in the vapour phase can be
calculated from the pressures of each compound at the boiling
temperature and equation (31.3), giving
sat
Plim P sat
ylim = = sat lim sat
P Plim + Pwater
Thus, the molar fraction in the vapour phase at 97.49 °C is
8.66 × 10−2, corresponding to a mass fraction of 0.4178 g.g−1.
This value is much higher than the solubility limit, and so the
condensed vapour separates into two phases: a light organic
phase (the “essential oil”) and a heavy aqueous phase, called the
floral water.
Conclusion
This chapter intended to show the behaviour of volatile compounds
when solutions are put to the boil, during distillation or hydro-​
distillation. The importance of knowing the properties of these
compounds, in particular their saturating pressure vapour with
respect to temperature, and their absolute volatility with respect
to the nature of the solvent, is stressed. During the simple distil-
lation of hydro-​alcoholic solutions, the volatile compounds dis-
tribute between some that are concentrated in the first fractions
of vapour, some that distribute as ethanol, and some that are more
concentrated at the end of the process. With hydro-​distillation or
steam distillation, one can extract poorly water-​soluble compounds
at a temperature much lower than their own boiling temperature;
(31.8)
indeed, their vapour pressure depends only on temperature. Their
concentration in the vapour depends only on the ratio of the sat-
urating vapour pressures, and through the condensation of the
vapour, they can be separated by simple decantation as essential
oil. Finally, by simple distillation or by steam distillation, one can
lower the boiling temperature by reducing the pressure.
Distillation: Volatile Compounds 219

FIGURE 31.6 Evolution of some odorant compounds during the distillation of cognac; the experimental values are compared with simulated values with
BatchColumn software from ProSim®.
(Douady et al., 2019)

Douady A, Puentes C, Awad P, Esteban-​Decloux M. 2019. Batch dis-

TABLE 31.4
tillation of spirits: experimental study and simulation of the
Boiling Temperatures of Limonene and Water at Various Total behaviour of volatile aroma compounds. J. Inst. Brew. 125,
Pressures P 268–​283.
Esteban-​Decloux M, Deterre S, Kadir S, Giampaoli P, Albet J, Joulia
T (°C) Plim (Pa) Pwater (Pa) P (Pa)
X, Baudouin O. 2014. Two industrial examples of coupling
97.47 8768 92,459 101,227 experiments and simulations for increasing quality and yield
97.48 8771 92,493 101,264 of distilled beverages. Food Bioprod. Process. 92, 343–​354.
97.49 8775 92,526 101,300 International Organization of Legal Metrology. 1975. International
97.50 8778 92,560 101,338 Alcoholometric Tables, OIML R 22.
97.51 8782 92,593 101,375 Official Journal of the European Regulation. 2019. Regulation
(EU) 2019/​787 of the European parliament and of the council
of 17 April 2019 on the definition, description, presentation
and labelling of other foodstuffs, the protection of geograph-
REFERENCES ical indications of spirit drinks, the use of ethyl alcohol and
Deterre S, Rega B, Delarue J, Decloux M, Lebrun M, Giampaoli P. distillates of agricultural origin in alcoholic beverages, and
2011 Composition of aroma volatiles of bitter orange essen- repealing Regulation (EC) No 110/​ 2008. https://​
eur-​
lex.
tial oil (Citrus aurantium L.) and its distillate to evaluate the europa.eu/​legal-​content/​EN/​TXT/​?uri=CELEX:32019R0787
impact of the maceration-​distillation process, Flavour Fragr. Perry JH, Green DW, Maloney JO. 1997. Perry’s Chemical Engineers’
J. 27, 77–​88 Handbook. New York, Mc Graw Hill.
220 Martine Esteban-Decloux

ProSim. 2019. Simulis thermodynamics. www.prosim.net/​ en/​ Puentes C, Joulia X, Vidal JP, Esteban-​Decloux M. 2018b. Simulation
software-​simulis-​thermodynamics-​mixture-​properties-​and-​ of spirits distillation for a better understanding of volatile
fluid-​phase-​equilibria-​calculations. aroma compounds behavior: application to Armagnac produc-
Puentes C, Joulia X, Athès V, Esteban-​Decloux M. 2018a. Review tion. Food Bioprod. Process. 112, 31–​62.
and thermodynamic modeling with NRTL model of vapour-​ The Good Scents Company. 2019. D-​ Limonene. www.thegood­
liquid equilibria (VLE) of aroma compounds highly diluted in scentscompany.com/​ data/​
rw1013772.html. Last access 19
ethanol-​water mixtures at 101.3 kPa. Ind. Eng. Chem. Res. 57, November 2019.
3443–​3470.
Eggs: Let Us Have an Egg
Hervé This vo Kientza1,2
1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
Using codes for describing egg transformations can lead to a
wealth of new culinary preparations. Sometimes, the physical
or chemical mechanisms of the transformations are known, but
often, more scientific research (molecular and physical gas-
tronomy) is needed in order to explain the changes observed
during simple processes. Since the beginning of molecular and
physical gastronomy, scientific knowledge has been used for
making technical innovations (This, 2006a). This “culinary tech-
nique” has been called “molecular cooking”, when new tools (in
particular, equipment from chemistry or physics laboratories) or
methods were used. For example, when considering innovative
transformations of eggs, the following systematic system has
been proposed (This, 2006b; This, 2007).
A Code for Innovation
This system is given in Table 32.1, applied to eggs, but one can
easily generalize it to any other food ingredient, and indeed, this
has also been done for plant and animal tissues. But let us begin
with eggs: the full, whole egg, comprising a shell, yolk and egg
white, is given the code number 1, and this makes the first row
of a table.
Then, the various possibilities of dividing the whole egg are
considered as cells of the second row. The full, intact egg is
assigned the label 1.1, the shell alone 1.2, the non-​mixed yolk
and the white out of the shell 1.3, the mixed yolk and white 1.4,
the yolk alone 1.5, and the white alone 1.6 (Figure 32.1).
In the third row of the table examples of transformations of
these various constituent “parts” by various processes are found;
options include the addition of nothing (1), gas (2), “water” (i.e.,
any aqueous solution) (3), “oil” (any fat liquid phase) (4), solids
(5), ethanol (6), acid (7), alkali (8) or heat (9).
Generally, the gas used in the kitchen is air, but many possibil-
ities exist; for example, during an educational dinner organized in
2008 by the Institut des hautes etudes du goût, de la gastronomie
et des arts de la table (“Institute for the advanced studies for fla-
vour, gastronomy and arts de la table”) at the Cordon Bleu School
in Paris, the dessert was made using helium, so that the guests
had a strange duck voice for some seconds after consumption
(as sound velocity changes depending on the nature of the carrier
gas, hence resulting in a modification of frequency when sounds
are emitted in helium instead of air). The “water” in option 3 can
be any aqueous solution, as long as the concentration is low and
solutes do not noticeably change the properties of the solution.
Likewise, “oil” can be any liquid fat, such as ordinary kitchen
oil (olive oil, corn oil, sunflower oil, etc.), but also melted butter,
melted foie gras, melted chocolate, etc.
Of course, ethanol, acids and alkalis should be edible: this
means that they have to be food grade, but also they can be
dissolved, for example, in vodka, vinegar, baking powder, etc.
Heat, finally, can be applied in many different ways, which
means that more slots could be added to the table if necessary
(see chapter “Uncook the Egg and Eggs at 6X °C”). In particular,
it is useful to consider systematically that heat can be supplied by
a hot solid, a hot liquid (water or oil) or a hot gas, or by radiation
(microwaves, infrared and also all kinds of electromagnetic radi-
ation, as they transmit energy when they are absorbed).
Using these ingredients, new “products” or culinary concepts
can be made, and numbers can be used to describe what they
are. For example, in the third row, three numbers describe results
of combinations of transformation. Now we shall consider only
some cases, because, essentially, the table offers infinite numbers
of possible combinations.
Some Examples
As the colour indicates, the table contains old results as well as
new ones. A first interesting result is given by the code 1.1.6: an
egg (1) is stored, whole in the shell (1), in alcohol (6). Because
the brandy can enter the egg through the pores, a result called a
“baumé” is obtained, after more than one month (Figure 32.2),
due to precipitation and possibly coagulation of proteins. The
name “baumé” was introduced for this new product in honour
of the French pharmacist Antoine Baumé (1728–​ 1804), who
devised a density scale.
The next code, 1.1.7, describes eggs that are stored in vin-
egar, for example, and that coagulate after some time (about one
month), making “minus one century eggs” (the “contrary” to “one
221
222 Hervé This vo Kientza

TABLE 32.1
A Coded System for Innovation Using Eggs (Blue Colour Indicates Traditional Applications, While Red Indicates Novel Ones)
1. Whole intact egg
1.1. Whole egg in 1.3. Yolk and white out of 1.4. Mixed yolk
the shell 1.2. Shell alone the shell, unmixed and white 1.5 Yolk 1.6. White
1.1.1. Nothing 1.2.1. Nothing 1.3.1. Nothing 1.4.1. Nothing 1.5.1. Nothing 1.6.1. Nothing
1.1.2. Gas 1.2.2. Gas 1.3.2. Gas 1.4.2. Gas: Genoise 1.5.2. Gas 1.6.2. Gas: Whipped
1.1.3. Water 1.2.3. Water 1.3.3. Water before cooking 1.5.3. Water egg white
1.1.4. Oil 1.2.4. Oil 1.3.4. Oil 1.4.3. Water 1.5.4. Oil: Mayonnaise 1.6.3. Water
1.1.5. Solid 1.2.5. Solid 1.3.5. Solid 1.4.4. Oil: Mayonnaise 1.5.5. Solid 1.6.4. Oil: Geoffroy
1.1.6. Ethanol: Baumé 1.2.6. Ethanol 1.3.6. Ethanol (one kind) 1.5.6. Ethanol: Thenard (emulsion based on
1.1.7. Acid: Minus one 1.2.7. Acid 1.3.7. Acid 1.4.5. Solid of yolk the white)
century eggs 1.2.8. Alkali 1.3.8. Alkali 1.4.6. Ethanol: Thenard 1.5.7. Acid 1.6.5. Solid
1.1.8. Alkali: One 1.2.9. Heat 1.3.9. Heat: Fried eggs, of white and 1.5.8. Alkali 1.6.6. Ethanol: Thenard
century eggs poached eggs, low-​ yolk 1.5.9. Heat: Pommade of white
1.1.9. Heat: hard-​ temperature poached 1.4.7. Acid egg (cooked 1.6.7. Acid
boiled eggs, eggs 1.4.8. Alkali yolk at low 1.6.8. Alkali
eggs at XX °C 1.4.9. Heat: omelette, temperature) 1.6.9. Heat: Cooked white
flan, flan at low
temperature
1.1.1.1. 1.2.1.1. 1.3.1.1. 1.4.3.9. Lavoisier 1.5.3.9. Royale, flans, 1.6.2.2. Chaptal (raw
1.1.1.2. 1.2.1.2. 1.3.1.2. (extreme flans at low preparation for
1.1.1.3. 1.2.1.3. 1.3.1.3. royales), temperature wind crystal)
1.1.1.4. 1.2.1.4. 1.3.1.4. Avogadro 1.6.3.9. Avogadro of white
1.1.1.5. 1.2.1.5. 1.3.1.5. 1.6.4.5. Liebig (physically
1.1.1.6. 1.2.1.6. 1.3.1.6. gelled emulsion)
1.1.1.7. 1.2.1.7. 1.3.1.7. 1.6.4.9. Gibbs (chemically
1.1.1.8. 1.2.1.8. 1.3.1.8. gelled emulsion)
1.1.1.9. 1.2.1.9. 1.3.1.9.
1.1.2.1. 1.2.2.1. 1.3.2.1.
1.1.2.2. 1.2.2.2. 1.3.2.2.
…. … …
1.6.2.2.9. Wind crystal
1.6.3.2.9. Vauquelin
(wind crystal
preparation
cooked in the
microwave
oven)

century eggs” prepared by Asian populations by storing eggs in
mixtures containing alkali such as lime or potash) (Figure 32.3).
Label 1.1.9 describes in particular the hard-​ boiled egg
(Figure 32.4), as the full egg in its shell (1.1) is heated (9). But
considering the various denaturation temperatures of proteins
from the egg, many other possibilities exist, such as “eggs at
6X °C”, i.e., eggs that are heated at 62 °C, 63 °C, 64 °C… up to
100 °C, until the temperature is constant in the egg (This, 1997a),
with many different results being found at different temperatures
between 61 °C and 100 °C (Figure 32.5).
The code 1.3.9 corresponds to fried egg, or “oeuf cocotte”, and
it is not new. The mechanism of this transformation was shown
in This (1997b): for example, in egg white, the denaturation of
proteins can lead to the exposure of thiol groups from cysteine
residues, and the formation of disulfide bridges (oxidation) can
make a three-​dimensional protein network trapping the 90% of
water making up the egg white.
The code 1.5.6 is a yolk coagulated by ethanol (and 1.6.6 is
an ethanol-​coagulated egg white) (Figure 32.6). It was named a
“thenard” (of white, or of yolk), from the name of the French
chemist Louis-​Jacques Thenard (1777–​1857), who made many
discoveries (hydrogen peroxide, Thenard blue, etc.) and in par-
ticular introduced “osmazome”, i.e., the result of extraction of
meat in ethanol.
Label 1.6.4 describes what was considered impossible by
chefs of the past, who published that “the slightest trace of egg
white with the egg yolk prevents making a successful mayon-
naise” (Gencé, 1900). Here, considering that proteins are much
better surfactants than phospholipids, it was proposed to whip oil
in an egg white in order to make an emulsion. The oil used can
be ordinary plant oil (e.g., sunflower, soybean) but also melted
butter (ordinary, clarified or brown) or even melted foie gras or
cheese, as long as its temperature remains lower than 61 °C, i.e.,
the first coagulation point of a protein of egg white. The obtained
emulsion was called a “geoffroy”, from the French chemist
family of scientists, including Claude Joseph Geoffroy (1685–​
1752), a botanist; Claude-​François Geoffroy (1729–​1753), who
discovered bismuth; and Etienne Louis Geoffroy (1725–​1810), an
Eggs: Let Us Have an Egg 223

FIGURE 32.1 An egg can be divided into parts that can be described by a code number.

FIGURE 32.2 This “beaumé” (a) was obtained by storing a whole egg in rum for one month. The name is after the French pharmacist Antoine Beaumé (b).

entomologist (Figure 32.7). Other products, differently labelled,
can also be produced but have not received names up to now.
When the fourth row of the table is considered, products are
now labelled with codes written with four figures. Again, some of
them are traditional (a few), and many new possibilities arise. The
“gibbs” (1.6.4.9), for example, is obtained by heating a geoffroy:
the chemical gelation of the proteins previously used for the
emulsion creates a three-​dimensional network trapping the oil
droplets (stabilities of more than three months were observed)
(Figure 32.8). The name is from the American physical chemist
Josiah Willard Gibbs (1839–​1903).
On the next line of the table, there are more possibilities,
such as 1.6.3.2.9, a product to which the name of the French
chemist Nicolas Vauquelin (1763–​1829) was given. This product
consists of egg white, with added air, to which water is added,
and the result is heated (in this case, using a microwave oven).
Figure 32.9 shows a vauquelin made by the chef Denis Martin
(Restaurant Denis Martin, Vevey, Switzerland).
And, of course, there are many other new possibilities, because,
as said, the table is infinite in length. If one wants to create a new
dish, one can just select a code and turn it into food!
224 Hervé This vo Kientza

FIGURE 32.3 The various steps of transformation of an egg stored in vinegar. First the shell is dissolved, then the egg expands by osmosis, and finally the
proteins coagulate.

FIGURE 32.6 A “thenard” of egg white.

FIGURE 32.4 A hard-​boiled egg. When the temperature of 100 °C is

applied for a long time, sulphur-​containing proteins are degraded, and they
release hydrogen sulfide, as can be shown, for example, by applying a filter
paper previously dipped in a lead acetate solution.

FIGURE 32.7 A “geoffroy” is an emulsion obtained by whipping a liquid

fat in an egg white, as for the making of a mayonnaise.

FIGURE 32.5 An egg cooked at 65 °C. The name initially given, “perfect
egg”, is still used in many restaurants of the world.
Eggs: Let Us Have an Egg 225

Conclusion
This digital system for eggs can be applied to any other food
ingredient, such as meat, vegetables, fruits, etc. Moreover,
the system is indeed a “code”, an example of many that have
been produced in the last decades, such as the “table of double
cooking” (This, 2002), the formalism for dispersed systems (see
chapter “Dispersed System Formalism”) and the “tree of doughs”
(This, 2002). All of them arise from the same idea, as used by
Antoine Laurent de Lavoisier for chemistry in 1791 (Lavoisier,
1791) and later formalized by Goblot (1918).
FIGURE 32.8 A “gibbs” is a chemically jellified emulsion, obtained, for
example, by thermal processing of a “geoffroy”.

REFERENCES
Gencé. 1900. Encyclopédie de la vie pratique, Librairie nationale des
beaux arts, Paris, p. 476.
Goblot E. 1918. Traité de logique, Armand Colin, Paris.
Lavoisier AL. 1791. Considérations générales sur la dissolution
des métaux dans les acides, in Oeuvres, t. 3, Imprimerie
nationale, Paris.
This H. 1997a. L’oeuf parfait, www.pierregagnaire.com/​ pierre_​
gagnaire/​travaux_​detail/​76
This H. 1997b. Can a cooked egg white be uncooked? The Chemical
Intelligencer, 10, 51.
This H. 2002. Molecular Gastronomy, Columbia University Press,
New York.
This H. 2006a. Cooking in schools, cooking in universities,
Comprehensive Reviews in Food Science and Food Safety, 5
(3), 48–​50.
This H. 2006b. Questions d’œuf, Pour la Science, 344, 4.
This H. 2007. Let’s have an egg. Eggs in Cookery, Proceedings of the
Oxford Food Symposium on Food and Cookery 2006, Prospect
Books, Totnes, 250–​258.
FIGURE 32.9 A “vauquelin” produced by Denis Martin in Vevey
(Switzerland).
Emulsions: Emulsified Systems in Food
Markus Ketomäki, Trivikram Nallamilli, Christine Schreiber and Thomas A. Vilgis
Max–​Planck-​Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany
Principles behind Emulsions and Foams
In life, it is often said that opposites attract, but in the kitchen this
is not necessarily the case. For example, try to mix oil (or gener-
ally speaking, lipids; the following text uses the terms “fat” and
“oil” to refer to solid and liquid natural food fats, respectively;
the general term “lipid” encompasses both) and water: no amount
of mixing or shaking will help you to reach a stable system with
oil dispersed in water. In order to understand the fundamental
reasons behind this behaviour, we need to consider two things:
the electrostatic forces between the individual molecules and
the concept of entropy (i.e., the degree of disorder in a system)
(Silverstein, 1998).
One characteristic of water molecules is their polarity.
Although each molecule is electrically neutral, the electron dis-
tribution within the molecule is unbalanced, making the mol-
ecule polar. This unbalance arises from the fact that oxygen has
a higher electronegativity than hydrogen: the shared electron
pairs are more attracted by the oxygen atom than by the hydrogen
atoms (McNaught and McNaught, 1997). The oxygen atom thus
carries a slight negative charge, and the two hydrogen atoms
carry a corresponding positive charge. This is also the reason
why water molecules “stick together” at room temperature; the
“negative” oxygen attracts the “positive” hydrogen atoms from
the neighbouring water molecules.
With most molecules from fat, however, the situation is
different, and this has something to do with the structure of the
molecules. Generally speaking, natural food fats are made of
triacylglycerides (Vilgis, 2011). In oil, such molecules are pri-
marily triacylglycerols, which consist of a glycerine residue
attached (esterified) to three fatty acid residues. These fatty acid
residues can be of various lengths (number of carbon atoms)
and they can either be saturated or unsaturated (single or double
bonds between carbon atoms). These long hydrocarbon chains
are electrically balanced (unlike water molecules) and are called
non-​polar.
Thus, upon adding triglycerides into water, the water molecules
would have to rearrange (due to their mutual attraction) around
the non-​polar molecules (Huque, 1989). This process, in turn,
would limit the ability of water molecules to move freely, thus
resulting in a decrease in entropy in the system (Alger, 1994).
However, this decrease in entropy caused by the organized struc-
ture around the triacylglycerol molecules is not compensated
through any enthalpy gains upon mixing the two substances
(Streitwieser et al., 1992). This leads the non-​polar molecules to
be driven by entropy to “clump” together, which, in turn, again
increases the entropy of water within the system. This entropy-​
driven separation of non-​ polar and water molecules is also
referred to as the hydrophobic effect (“water-​fearing”, as opposed
to hydrophilic or “water-​loving”) (Huque, 1989). Finally, the
differences in density between the solvent (water) and the solute
(triacylglycerol molecules) cause the lighter substance (fat) to
rise to the top.
So, in order to mix water and oil, substances that help in
bringing oil and water together (emulsifiers) are needed. There
are many different mechanisms by which the emulsifiers work,
but this effect is primarily due to the fact that emulsifiers typically
consist of a water-​soluble (hydrophilic) and a hydrophobic (also
referred to as lipophilic) part. When the emulsifier is mixed into
an oil–​water mixture, the emulsifying molecules tend to move
toward the oil–​water interface, where the hydrophilic part tends
to remain in the water phase and the hydrophobic part tends to
stay in the oil phase. When the oil–​water system is given mech-
anical energy (e.g., by whipping), the freshly formed small oil
droplets are stabilized by the layer of the emulsifier at the oil–​
water interface. Emulsions are, however, only metastable and will
eventually separate when given enough time.
Depending on the medium in which the emulsifier is more sol-
uble, it can be used to make either water-​in-​oil emulsions (when
the emulsifier is more soluble in/​attracted to oil) or oil-​in-​water
emulsions (when the emulsifier is more soluble in/​attracted to
water) (Ohshima, 2016; Griffin, 1949; Griffin, 1954). This rule
(also called “Bancroft’s rule”) does not hold in all cases. There
are several factors that can promote or inhibit the efficiency of
an emulsifier (e.g., temperature, dissolved salts or presence of
proteins). However, as a rule of thumb, it provides a good starting
point when preparing emulsions.
Looking at the chemistry and physics behind emulsions,
there are several mechanisms that are involved in the process
of stabilizing and destabilizing emulsions. Among other things,
emulsifiers help in reducing surface tension (and thus the energy)
in the water–​oil interface (McClements, 2015). Each time an
interface is formed in a physical system (e.g., an oil droplet
in water), a certain amount of energy is required to form the
boundary between the substances. Consequently, large surface
227
228
FIGURE 33.1 Schematic representation of different processes that lead to destabilization of emulsions and foams.
areas require more energy and are, because of this, inherently
unstable.
This instability leads to a phenomenon called Ostwald
ripening, which takes place in emulsions (Kabalnov et al.,
1992). In order to minimize the surface energy in the interface,
triacylglycerol molecules migrate from the small oil droplets
toward the larger ones, as the energy requirement for forming
one large droplet is less than that required to form several small
ones. Emulsifiers also limit the fusion of droplets, a phenomenon
known as coalescence.
Furthermore, differences in density can cause creaming in
emulsions (Adams et al., 2007). The emulsified oil droplets remain
separate but start floating upwards in the water, thus forming a
zone of small oil droplets on top of the water. Ingredients (such
as gelatine, starch granules or pectin) and certain preparation
methods (such as reducing the water content by boiling or simply
whipping), which increase the viscosity of the emulsion, oppose
creaming. Also, increasing the number of emulsified droplets in
an emulsion can lead to a phenomenon known as jamming, which
in turn leads to higher viscosity (Siemens et al., 2010). However,
it is also possible to break an emulsified system or even cause it
to undergo a phase inversion with the wrong treatment. Making
Markus Ketomäki et al.
butter by whipping cream is the best example of this. Initially,
the cream is a dispersion of oil droplets in water, but continued
whipping turns the cream into butter, which is essentially an
emulsified system consisting of water droplets in fat. Figure 33.1
shows various types of destabilization mechanisms in emulsions
and foams.
Types of Emulsifiers in Food Emulsions and Foams
For food applications, a wide range of emulsifiers can be used,
depending on the type of food being produced. In food, both nat-
ural and synthetic emulsifiers are used. Natural emulsifiers can be
divided into three main categories: (1) phospholipids, (2) proteins
and polysaccharides and (3) particulate emulsifiers.
Phospholipids have small molecules containing one polar
“head” group (phosphate moiety) and a nonpolar “tail” (one or
two fatty acid residues). One of the most versatile and widely
used phospholipid emulsifiers is lecithin, which can be obtained
from egg yolks or from any plant tissues (it is the main material
of biological membranes), including soya beans. Lecithin is
used in a wide range of culinary applications, such as improving
Emulsions: Emulsified Systems in Food
the stability and texture of coffee cream, salad dressings and
mayonnaise.
Proteins are polymers whose monomers are amino acids.
Many proteins have regions that are alternately hydrophobic
and hydrophilic, giving them good emulsification properties.
For this reason, alongside their nutritional properties, they are
the most widely used emulsifiers in the food industry. Protein-​
based emulsifiers include egg whites (Drakos and Kiosseoglou,
2006), whey proteins (beta-​lactoglobulin) (Hu et al., 2003), milk
proteins (caseins) (Dickinson et al., 1998) and gelatine (Olijve
et al., 2001) from animal sources.
Polysaccharides are polymers whose monomers (over 100 of
them by definition) are saccharides, and they exhibit thickening
properties, which aid in enhancing emulsion stability. Examples
of polysaccharides include xanthan and gellan gum (extracted
from bacterial cultures), carrageenans (extracted from sea
weeds), pectins (extracted from fruit peels), natural and modi-
fied starches (from grain sources), galactomannans (extracted
from seeds of guar beans and konjac) and chitosan (from shells
of crustaceans).
Xanthan consists of monomers of β-​D-​glucopyranose glucan
as backbone, and its side chains consist of D-​mannopyranose and
β-​D-​glucuronic acid. Gellan is a tetrasaccharide, which consists
of two residues of D-​glucose and one residue each of L-​rhamnose
and D-​glucuronic acid. Carrageenan is a polysaccharide made
up of repeating galactose units and 3,6-​ anhydrogalactose
(3,6-​AG), both sulfated and non-​sulfated; the units are joined by
alternating α-​1,3 and β-​1,4 glycosidic linkages. Starch is made of
amylose and amylopectin, of which the monomers are D-​glucose.
Polysaccharides belonging to the galactomannan family, such as
guar gum, etc., consist of monomer units of galactose and man-
nose. Chitosan is a polysaccharide consisting of sub-​units of
D-​glucosamine and N-​acetyl-​D-​glucosamine.
Particulate or Pickering emulsifiers are a recent addition to
the repertoire of food emulsifiers, and include microparticles of
materials like starch, microcrystalline cellulose, chitin, cocoa
fibres, zein particles (from corn), soy protein particles and fla-
vonoid particles. In recent years, these have been recognized for
their roles in stabilizing emulsions and foams in food products.
Unlike traditional emulsifying agents, Pickering emulsifiers are
much larger (in the size range of a few tens of nanometres to
micrometres) and can impart better texture and mouthfeel to food
products (Linke and Drusch, 2018). Figure 33.2 shows the micro-
scopic structure of soy oil droplets stabilized by hydrophobic rice
starch particles.
Synthetic emulsifiers are artificially made from non-​natural
sources and include sugar-​ alcohols like sorbitol, sorbitan
monolaurate (Polysorbate 20), sorbitan monostearate (Polysorbate
80), sodium and calcium salts of fatty acids, sucroglycerides, etc.
For example, in ice creams, Polysorbate 80 is often added up to
0.5% (v/​v) concentration to make the ice cream smoother and
easier to handle, as well as increasing its resistance to melting
(Goff, 1997). Adding this substance prevents milk proteins from
completely coating the fat droplets. This allows them to join
in chains and nets, which hold air in the mixture and provide a
firmer texture that holds its shape as the ice cream melts.
229
FIGURE 33.2 Dark field polarization microscopy image of a Pickering par-
ticle stabilized emulsion made with soybean oil emulsified in water at pH 3
with rice starch particles modified with octenyl-​succinic anhydride. Starch
particles adsorb on to oil droplets, thus stabilizing the emulsion.
Examples of Emulsified Systems in Food
Milk
A classic example of food emulsion is milk. As shown in
Figure 33.3, the milk fat globules (in the size range 0.1 to 10 μm)
are dispersed in water. Fat globules are stabilized by a thin layer
(consisting of a mixture of phospholipids and proteins) called
the milk fat globule membrane (MFGM). The MFGM consists
of about 30% phospholipids (sphingomyelin, phosphatidyl-
choline and phosphatidylethanolamine) (Lopez and Ménard,
2011), and the remaining fraction consists of glycosylated and
non-​glycosalated proteins (Kanno, 1990). Additional stability in
some cases is provided by micelles of caseins, vital milk proteins.
Caseins are found in the form of aggregates called micelles, which
are negatively charged at the normal pH of milk, i.e., 6.6 (Sinaga
et al., 2017). These charged casein particles repel each other, thus
avoiding fat droplet coalescence and stabilizing the milk.
Milk, when left to stand, separates into a visible fat layer over
an aqueous one, in a process called creaming. To avoid creaming,
the fat globules in milk are reduced in size by a process called
homogenization: the fat globule membrane is ruptured and
fat globules are stabilized by casein micelles (Cano-​Ruiz and
Richter, 1997). Thus, homogenized milk may be considered as a
Pickering emulsion.
In contrast to stabilization of milk, in some cases it is necessary
to destabilize milk, for example in cheese making. In the manu-
facture of cheese, acid or rennet is added to milk, reducing the
pH or enzymatically hydrolysing a key stabilizing protein in the
casein micelle, which destabilizes the casein micelles, resulting
in coagulation of fat and protein particles and eventually separ-
ation, leading to the formation of cheese. Fermented products
such as curd, buttermilk and yoghurt (Bongard, 2009) are also
examples of emulsified systems consisting of oil/​fat and water.
Emulsions in which the suspended phase is more than 74% by
230
FIGURE 33.3 Schematic representation of microscopic structure of milk showing fat globules and casein micelles suspended in solution of whey proteins
and milk sugars.
FIGURE 33.4 Picture showing (from left to right) oat-​based emulsion, coconut-​based emulsion, soy-​based emulsion and almond-​based emulsion.
volume are called high internal phase emulsions (HIPEs) (Patel
et al., 2014).
“Plant-​Based Emulsions” (Non-​Dairy)
Plant-​based emulsions are oil-​in-​water emulsions obtained from
nuts, grains, seeds, legumes and cereals. Plants store oils (i.e.,
lipids) in the form of droplets (in the size range of ca. 300–​
350 nm) encapsulated by a layer of phospholipids and proteins
(Waschatko et al., 2012). These droplets are called oil bodies
or “oleosomes”. Plant-based drinks are generally extracted by
blending and then straining soaked grains (barley, rice, oat and
wheat), legumes (soy, lupin, pea and ground nuts), nuts (cashew,
almond, hazel nut, pistachio, walnut, etc.), seeds (chia, flax,
hemp, pumpkin, sesame, sunflower, safflower, etc.), coconut, etc.
Figure 33.4 shows some examples of plant drinks, some of which
have been used across cultures around the world for thousands of
years. Just as with dairy milk, plant drinks can be used as such
or further processed/​fermented to products analogous to yoghurt
and cheese. Plant-​based recipes include horchata (a north African
beverage from soaked and sweetened tiger nuts), Indian/​Thai
Markus Ketomäki et al.
chicken curry with coconut milk, tofu-​based recipes from soy
milk, and wheat milk halwa (a type of dessert from India). Plant
drinks have attracted attention in the vegan community world-
wide. Apart from being used in food products, plant drinks are
also used in a variety of cosmetic and skin care products like skin
creams, moisturizer, etc.
Butter and Margarine
Butter is an example of an emulsified system that consists of
more than 83% fat, 13% water, about 3% ash and 1% milk
proteins (Farmer, 2011). Butter is also a rich source of fat-​
soluble vitamins like vitamins A, D and E (Delgado-​Zamarreño
et al., 1995). Butter is made by churning fresh or fermented
cream to separate butterfat from buttermilk. In order to make
butter, the cream recovered over cow milk is cooled to around
5–​7 °C before churning to allow some of the fat to crystal-
lize (Gunstone, 2006). The fat crystals serve as a basis for
further crystallization and help to break the milk fat globule
membranes during churning (Provost et al., 2016). The milk
fat aggregates first into fat particles and later into solid masses
Emulsions: Emulsified Systems in Food
FIGURE 33.5 Picture of butter (a) at the temperature of 22 °C and schematic representation of the microscopic structure of butter (b).
during churning. The fat mass collected in this process is sub-
sequently kneaded to remove any remaining pockets of butter-
milk and to achieve the desired texture. The final texture of
butter depends on the size and structure of the fat crystalline
network in the butter that is brought about by the kneading pro-
cess (typically around 14–​16 °C) (Fuquay et al., 2011). Butter
is available in varieties such as salted butter, spreadable butter,
clarified butter, ghee and maitre d’hotel butter. Figure 33.5b
also shows a schematic representation of the microscopic
structure of butter.
Margarine is another example of an emulsified system
consisting of water droplets in a continuous phase of fat, which
is used for flavouring, baking and cooking. Margarine was first
created by the French chemist Hippolyte Mège-​Mouriès in 1869
and was originally made as a cheaper substitute for butter using
beef tallow. Today, margarine is made from a variety of oils
such as safflower, corn, soy bean and animal fats (Carpenter and
Slover, 1973). Just as with butter, margarine consists of about
80% oil, 15–​18% water phase and 1–​2% salt and other solids.
The oil is often hydrogenated to make it less susceptible to oxi-
dation and improve shelf life. The hydrogenated oil and milk
(water base) is mixed with lecithins, and the mixture is blended
together in an emulsification chamber at 38 °C. This process
leads to the formation of an emulsified system consisting of tiny
water droplets (0.42–​2.7 μm) suspended in a crystalline oil phase
(Balinov et al., 1994). Then, the contents are cooled to about 7 °C
(in a scraped-​surface heat-​exchanger called a votator) to a semi-​
solid state that is further processed and packed.
Spreads, Dips, Dressings and Sauces
Spreads, dips, dressings and sauces are also emulsified systems
of an oil phase suspended in a water base. In the case of spreads
and dips, the proportion of the oil/​fat, i.e., the suspended phase, is
very high compared with the suspension medium, i.e., the watery
base. Such emulsified systems with a very high percentage
(>74% by volume) of oil phase are called HIPEs (Pulko and
Krajnc, 2012). Hummus is a good example of a spread/​dip and is
made from a mixture of chickpea (Cicer arietinum) puree, tahini
231
FIGURE 33.6 Picture of hummus (left) and hollandaise sauce (right). In
most homemade spreads and sauces, proteins from mustard, sesame and egg
yolk act as emulsifiers.
(paste from ground sesame seeds), lemon juice (Citrus limon),
olive oil (Olea europaea), cumin (Cuminum cyminum) and salt
(Figure 33.6).
In this mixture, chickpea puree acts as the water base, and
proteins present in the tahini (sesame proteins) act as emulsifier
that stabilizes it (Al-​Mahasneh et al., 2017).
Hollandaise is an example of a sauce, which contains egg
yolks, pepper, lemon juice, butter and salt. For the sauce, the
egg yolk is mixed with water (or wine, vinegar or lemon juice),
and mixed while heating. This allows the proteins present in egg
yolk to denature partially and coagulate to allow the formation
of a foam while mixing (Hopia et al., 2013). The melted butter
is added subsequently into the mixture, which is then emulsi-
fied (together with the lipids from the egg yolk) by the proteins
and lecithin from the egg yolk. The resulting sauce is a combin-
ation of an (oil-​in-​water) emulsion, a foam and a suspension, as
it contains melted fat droplets, air bubbles and protein aggregates
from the egg (This, 2016; Rognsa et al., 2014). The final texture
of the sauce, in turn, depends on the amount of air incorporated
into the sauce and the size of the fat droplets (Hopia et al., 2013;
Rognsa et al., 2014).
232
FIGURE 33.7 Liver and blood sausages, and a smoked bratwurst (a); schematic of microstructure of sausage emulsion (b).
Sausages
Sausages, such as German frankfurter, Austrian wiener, hot dogs,
mortadella, bologna, liver sausages and pâté, are also examples of
emulsified dispersions of fat in water together with particulates of
meat. The manufacturing process of sausages involves the prep-
aration of an emulsified mixture of finely chopped or minced
meat in water together with seasonings, salts and, where neces-
sary, binders or extenders (Belitz et al., 2004). The exact com-
position of ingredients depends on the type of sausage being
produced. Initially, the (muscle and fat) tissue is comminuted
(i.e., chopped, flaked, ground and/​ or minced) to the desired
degree (Toldra, 2010; Varnam et al., 1995). The final texture of
the sausages is determined by the duration and the method of
comminution (Belitz et al., 2004). This process generates a lot
of heat, however, due to friction between the sausage batter and
the blades (heating the contact surface up to 80 °C) and can
increase the temperature of the meat mixture by 10–​20 °C during
the first 15 minutes of processing (Toldra, 2010; Pearson and
Gillett, 2012). To minimize cooking losses and achieve optimum
stability for the emulsified system, the comminution temperatures
should be kept below 18 °C (Belitz et al., 2004; Essien, 2003;
Zayas, 2012). This is achieved by using ice or iced water. When
meat is cut into fine particles, the muscle fibres and connective
tissue are broken into pieces and mixed with fat droplets and
particles of fatty tissue. During the processing, the fat partially
melts and leads to the formation of an emulsion system with the
proteins present in the system (Belitz et al., 2004). The different
types of protein contribute to different degrees to the formation of
a monomolecular layer (ca. 130 nm in thickness) around the fat
globules (in decreasing order of importance: myosin, actomyosin,
sarcoplasmic proteins and actin) (Belitz et al., 2004). The fat
droplets (together with water) in the mixture are thus suspended
in a matrix of (salt-​soluble) proteins, leading to a thick paste,
which is an emulsified suspension (Figure 33.7) (Zayas, 2012).
To further improve the texture and mouthfeel, other components
such as water binders and emulsifying agents are added to the
mixture. Finally, a mixture of spices, salt and sodium nitrite is
added, which also helps in curing the meat. Depending on the
Markus Ketomäki et al.
type of sausage being produced, the sausage batter may be loaded
into casings and sold raw (e.g., bratwurst and breakfast sausages)
or smoked (e.g., kielbasa), fermented (e.g., sucuk), cooked (e.g.,
liver sausage, blood sausage) or dried (e.g., salamis and cervelats)
(Belitz et al., 2004; Pearson and Gillett, 2012).
Anise-​Flavoured Liquors and Spirits (Microemulsions)
A special type of emulsion called a microemulsion occurs when
a highly hydrophobic oil (e.g., anethole in anise-​ flavoured
drinks) dissolved in a water–​ethanol solution (as, for example,
in ouzo) is diluted with water. In this case, a spontaneous oil-​
in-​water emulsion is formed, the mixture turning from being
clear to opalescent (cloudy) (Carteau et al., 2008). This is com-
monly called the “ouzo effect” (also “pastis effect”), or “louche”,
or “spontaneous emulsification” (Lopez-​Montilla et al., 2002).
Microemulsions form spontaneously and are highly stable even
without any surfactant.
The underlying physical principles are depicted in Figure 33.8
as a simplified system of three components, with ethanol being
the emulsifier (with a hydrophobic hydrocarbon tail and a hydro-
philic OH group). The (yellow) oily odorant molecules in ouzo
surround themselves with ethanol molecules (red dots) and
are thus dispersed in the continuous water phase (blue dots).
However, ethanol is not a particularly strong emulsifier (with
a partition coefficient of around −1.0 to −1.5) due to its short
hydrophobic tail of only two carbon atoms (Leo et al., 1971;
Vilgis, 2011). Thus, the ethanol molecules are only loosely bound
to the oil–​water interface and retain much of their freedom to
diffuse and move around in the solution. A thermodynamic equi-
librium is thus formed between the states of being bound to the
oil–​water interface and being diffused in the water. This equi-
librium can then, in turn, be disturbed by adding water into the
drink. The ethanol molecules diffuse within a larger volume of
water, thus reducing the number of available alcohol molecules
near the anethole molecules. This causes the anethole molecules
to come together to “share” the available ethanol molecules with
each other. This process continues with the addition of water until
the anethole droplets become large enough to scatter light, which
Emulsions: Emulsified Systems in Food
FIGURE 33.8 A schematic representation of the ouzo effect/​pastis effect. At higher concentrations of water, molecules of essential oil (anethole, or 1-​
methoxy-​4-​[(1E)-​prop-​1-​en-​1-​yl]benzene) come together to maximize their exposure to alcohol, leading to the formation of nanoscopic oil droplets, which
scatter light, resulting in the development of the characteristic cloudiness of ouzo or pastis.
causes the solution to become turbid. The droplet size of such
emulsions ranges from 300 nm to about 2.4 μm and can change
with composition and time (Grillo, 2003).
Pastis (France), aquavit (Scandinavia), arak (Persia and the
Middle East), sambuca (Italy), aguardiente (Columbia), anis
(Spain), ouzo (Greece), absinthe (Europe), mastik (Balkans), raki
(Turkey) and xtabentun (Mexico) are a few examples of liquors
that form microemulsions when served with water. These liquors
generally contain essential oils extracted into alcohol from a
range of spices (like wormwood, anise and fennel), herbs (like
peppermint, coriander, hyssop, etc.) and botanicals (like roots,
barks, flowers and seeds). This effect is also demonstrated in
Figure 33.9: it is apparent that a small amount of water is enough
to disturb the delicate balance in the drink and to cause it to turn
cloudy.
Beverages
Beverages such as fruit/​vegetable juices and milk-​based beverages
like chocolate milk, milkshakes and several types of coffee-​
based beverages fall into the category of emulsified suspensions
(a suspension and also an emulsion). A simple example such as
fruit juice contains crushed fruit pulp suspended in water. To
this base, several different flavouring oils such as citrus peel
(Citrus sinensis), lemon (Citrus limon), lavender (Lavandula
angustifolia), peppermint (Mentha piperita), thyme (Thymus
vulgaris), cinnamon (Cinnamomum verum), tea tree (Melaleuca
233
alternifolia), rose wood (Dalbergia nigra), etc. are added to
improve the flavour of the juice. Further, polysaccharides such as
xanthan (Mirhosseini et al., 2008), carrageenan or pectin (Ibarz
et al., 1996) may be added to improve the mouthfeel and flow
properties, stability and shelf life of the final product.
Another class of beverages are milk based, such as fruit
milkshakes, chocolate milks and coffee-​milk-​based beverages. In
all these beverages, milk forms the water base of the beverage,
and most of the flavouring compounds, such as carotenoids,
polyphenols and essential oils, are emulsified into fine droplets
suspended in this water phase (Voilley and Etiévant, 1995). The
fat component of the milk is also suspended in this complex mix-
ture, thus forming an oil-​in-​water emulsion. Many of the final
properties of these beverages, like texture, mouthfeel, flavour
release, stability and shelf life, depend on droplet size distribu-
tion, viscosity and interactions between different components in
the mixture.
Foams in Food Systems
Foams in food systems are typically dispersions of gas bubbles
in a liquid, which at first sight resemble emulsions. Instead of fat
molecules (or triacylglycerols, to be exact), the dispersed phase
in foams consists of a gas (e.g., air or carbon dioxide), and the
dispersion medium is typically a liquid. However, incorporating
air into a liquid without any additional stabilizers typically results
234
FIGURE 33.9 Picture showing cloudiness (turbidity) developed in mixtures of ouzo and water at different mixing proportions. In all glasses, 80 mL ouzo
was mixed with (from left to right) 0, 10, 15, 17.5, 20, 22.5, 25, 30 and 40 mL of water.
FIGURE 33.10 Examples of emulsified systems (beverages): (from left to
right) latte macchiato, strawberry yoghurt (emulsified gel), mango lassi and
chocolate milk.
FIGURE 33.11 Dark field polarization microscopy image of foam bubbles
stabilized by a mixture of wheat gluten proteins. Microscopic foam bubbles
form a dense network, giving a rich texture to the foam.
Markus Ketomäki et al.
only in short-​lived bubbles. More stable culinary foams are thus
created with proteins, fats or surfactants (abbreviation from “sur-
face active agent”) (Myhrvold et al., 2011). Different types of
foams are stabilized by different underlying mechanisms. For
example, a foam made with egg white is a typical foam stabilized
by proteins. In this case, the protein molecules are partially
denatured by whipping and subsequently coat the surface of the
air bubbles. In this process, the denatured proteins rearrange
at the water–​gas interface in such a way that the hydrophobic
segments of the protein stick to the gas phase while the hydro-
philic parts remain in the water phase (Kijowski, 1995; Brown,
2018). In this way, the proteins form a thin film around the air
bubbles, thus stabilizing the foam.
Whipping cream, on the other hand, is stabilized with
semi-​solid fat crystals (Hasenhuettl and Hartel, 2008). Dairy
triacylglycerides have melting points between −40 °C and
+40 °C, so that at room temperature, whipping cream contains
both solid fat crystals and melted fat (Chandan, 2011; Marshall
et al., 2012). Through whipping, the triacylglycerol molecules
(in crystals and in liquid droplets) move to the edges of the gas
bubbles, where they form a thin membrane around them, stabil-
izing the foam. Thus, as the triacylglycerol molecules have to
be present both as solid crystals and in melted form in order to
make whipped cream, the temperature of the cream should not be
too high. According to the current consensus, the best whipped
cream is achieved by using chilled/​refrigerated whipping cream
at around 5–​10 °C (Hasenhuettl and Hartel, 2008; Ihara et al.,
2010; Myhrvold et al., 2011; Provost et al., 2016; Brown, 2018).
Finally, foams can be formed using emulsifiers, such as lecithin
from egg yolk. The underlying physics in this case are similar to
those described earlier in the context of emulsions: the emulsifier
molecules arrange themselves at the gas–​liquid interface in such
a way that the hydrophobic part of the emulsifier remains in the
gas and the hydrophilic in the water phase (Vega et al., 2012).
Emulsions: Emulsified Systems in Food
Culinary foams, just like emulsions, are inherently not stable
and collapse over time (with the exception of solid foams such
as bread). The mechanisms of foam decay are similar to those
of emulsion instability and phase separation (Hasenhuettl and
Hartel, 2008; Bamforth et al., 2011). To name some, the walls
between bubbles can burst (causing the bubbles to merge), leading
to foam coalescence. As gasses are (to varying degree depending
on the gas and pressure) soluble in water, the gas molecules
can dissolve into the water phase and escape the bubbles. This
process takes place between bubbles with different Laplace
pressures, which leads the bubbles to diffuse from areas of high
Laplace pressure (small bubbles) into areas of lower Laplace
pressure (large bubbles). For example, the Laplace pressure of
an air bubble (at 25 °C) in water with a diameter of 1000 μm is
around 300 Pa (Weaire and Hutzler, 2001). Halving the diam-
eter (500 μm) leads the pressure to increase to almost 600 Pa,
and a bubble of only 100 μm has a Laplace pressure of almost
3 kPa. This behaviour of gas molecules diffusing from smaller to
larger bubbles is called disproportionation (the foam “version” of
Ostwald ripening). Finally, as a foam consists of a dispersed gas
phase separated by a thin membrane from the continuous water
phase, the water slowly drains away from between the bubbles, a
phenomenon called film drainage.
Beer Foams
The beer head is one of the first characteristics we tend to notice
when we receive our drink. Too much of it, and we feel cheated
out of our beer; too little of it, and the beer appears flat and stale.
Not surprisingly, scientific research has also established that
the quality of the beer foam is perceived as one of the most (if
not even the most) important signs of overall quality in a served
beer (Bamforth, 2000). Apart from the perceived quality, how-
ever, beer foam also fulfils several other important functions. To
name some, the foam dampens the oscillations of the liquid when
the glass is carried (about 10% in comparison to a headless beer)
(Narziß et al., 2017). It also serves as a barrier, slowing down
the escape of gas out of the beer; the bitter flavour of hops tends
to concentrate in it, and many aroma compounds also tend to
concentrate in the foam (Bamforth, 2016). Lastly, the foam also
contributes to the mouthfeel of the beer, and in part to the char-
acteristic CO2 tingle (Langstaff and Lewis, 1993a; Langstaff and
Lewis, 1993b).
A lot of research effort has been put into understanding the
life of beer foam. There are several substances that contribute
to (or reduce) foam quantity and quality (such as proteins,
polyphenols, alpha acids, oligo-​/​polysaccharides and metal ions,
to name some) (Bamforth et al., 2011). Essentially, however, beer
foam is a type of foam stabilized by proteins, originating mainly
from the malts (Asano and Hashimoto, 1976, 1980; Bamforth,
1985). A further important factor increasing the stability of the
beer foam is the so called iso-​α-​acids contributed by the hops.
The iso-​α-​acid molecules are surface-​active and contribute to the
head stability by cross-​linking foam proteins. An example of an
alpha acid is humulone, also called α-​lupulic acid ((6S)-​3,5,6-​
trihydroxy-​2-​(3-​methylbutanoyl)-​4,6-​bis(3-​methylbut-​2-​en-​1-​
yl)cyclohexa-​2,4-​dien-​1-​one). Together, these two components
235
lay the foundations for a stable beer foam. Further components
such as polyphenols and melanoidins similarly support the
foam stability (Bamforth, 1998; Bamforth et al., 2011; Lewis
and Lewis, 2003). Finally, substances that increase the beer
viscosity (polysaccharides such as β-​glucans, consisting of six-​
sided D-​glucose monomers) and pentosans ([(2R,3R,4S,5R)-​2-​
hydroxy-​5-​[(2S,3R,4S,5R)-​5-​hydroxy-​3,4-​disulfooxyoxan-​2-​yl]
oxy-​3-​sulfooxyoxan-​4-​yl] hydrogen sulphate) originating from
the malts) slow down the rise of the gas bubbles in the beer, giving
them more time to collect foam-​promoting substances on their
way up (Bamforth et al., 2011). However, research until now has
not been able to show unambiguously that increased viscosity
increases the foam stability by reducing foam drainage. Fats and
detergents, on the other hand, inhibit foam formation and desta-
bilize existing foams, which is why dirty drinking glasses tend to
make themselves noticeable (Langstaff and Lewis, 1993a; Dickie
et al., 2001; Bamforth et al., 2011). Apart from the main players
in beer foam stability (proteins and iso-​α-​acids), factors such
as alcohol content, pH, polyphenols and gas type influence the
foam quality in beer.. Both ethanol and polyphenols have been
shown to be foam-​promoting or foam-​inhibiting depending on
the exact circumstances (Sarker et al., 1995; Brierley et al., 1996;
Bamforth, 1998; Evans et al., 1999, 2002; Fisher et al., 1999;
Lewis and Lewis, 2003). Ethanol generally lowers the surface
tension in beer foam, and thus, it could be expected that high
ethanol content would promote foam stability (weak correl-
ation between ethanol content and improved foam stability has
been established in some studies) (Bamforth, 1998). However,
it has also been observed that a high ethanol content possibly
disturbs the interactions between proteins and iso-​α-​acids and
thus destabilizes the foam (akin to lipids).
Polyphenols, on the other hand, are assumed to slightly pro-
mote foam stability in the same way as iso-​α-​acids do (cross-​
linking proteins in the foam) (Lewis and Lewis, 2003; Bamforth,
1998; Sarker et al., 1995). However, polyphenols are also gener-
ally associated with the precipitation of foam-​promoting proteins
in different stages of the brewing process (and leading to haze
formation later in the finished product) (Lewis and Serbia, 1984;
Barth, 2013; Bamforth, 2016).
The pH of beer has been observed to correlate with foam sta-
bility (Melm et al., 1995; Bamforth and Kanauchi, 2003). The
reason for this is assumed to be the amphiphilic nature of proteins
and iso-​α-​acids, which means that a variation in pH results in a
change in charge in these substances. Thus, a lower pH could lead
to a greater dissociation of the proteins and iso-​α-​acids, which
would in turn promote their migration to the beer–​gas interface.
The change in charge is also expected to aid their interaction.
Lastly, beer is typically carbonated either using carbon dioxide
or nitrogen, as demonstrated in Figure 33.12. Guinness is typic-
ally nitrogenized, whereas in other types of beer (such as wheat
beer), carbon dioxide is used. The advantage of using nitrogen
in beer when compared with CO2 arises mainly from the sub-
stantially lower solubility of nitrogen in beer (about 100 times
less than CO2) and from the fact that it is pH neutral (Bamforth
et al., 2011). Due to the lower solubility of nitrogen, higher gas
pressures can be used to dispense beer through narrower openings
to produce a great number of small bubbles (Carroll, 1979; Barth,
236
FIGURE 33.12 (a) Comparison of beer foam on wheat beer (left) and Guinness (right). (b) Fine and dense foam on Guinness beer. (c) Coarse foam on
wheat beer.
2013). The foam stability is, in turn, enhanced by the lower par-
tial pressures between the inner bubble and the atmosphere when
using nitrogen. Thus, by the time the foam on a wheat beer is long
gone, the foam on a Guinness remains practically unchanged.
Ice Creams
Ice creams, when considering their physics and chemistry, are
surprisingly complex systems. A wide range of frozen desserts
fall under the category of “ice cream”. Dairy ice creams (dairy
ingredients, sugar and flavours), gelato (Italian custard-​based ice
cream) and sorbet (fruit-​based ice cream that contains no fat and
no milk) are a few popular varieties (Clarke, 2015).They are com-
plex emulsified systems consisting of air (about 30–​50% of the
volume), ice crystals (up to 40%), fat crystals (5–​10%) and an
aqueous phase consisting of sugar, milk proteins and flavouring
components (Merkus and Meesters, 2013).
The proteins play a role in stabilizing fat globules and air bubbles
in the ice cream. At the same time, some emulsifiers are added to
ice cream mixtures to displace some of the proteins adsorbed to
the fat interfaces (Marshall et al., 2003). This not only allows the
fat droplets to partially coalesce and entrap air bubbles (similar to
the mechanism in whipped cream) but is also very important for
the right texture and to slow the meltdown (Cropper et al., 2013).
Any remaining proteins that are not adsorbed to the interfaces
remain dissolved in the unfrozen water phase and increase its
viscosity (El-​Zeini, 2016). Finally, yet importantly, the proteins
also influence the flavours of the ice cream. As the milk pro-
tein content of the ice cream increases, there is a corresponding
decrease in flavour intensity due to binding interactions of flavour
molecules with proteins (Hansen, 1996).
The term “sugar” in ice cream manufacturing covers sev-
eral different kinds of sugar compounds, including such sugars
as D-​glucose, D-​fructose, sucrose and sugar alcohols (Ozdemir
et al., 2008). Apart from sugar giving the ice cream its sweetness,
the softness and the viscosity of the ice cream can be modified
Markus Ketomäki et al.
by varying the sugar content and the composition of the added
sugar mix (Clarke, 2015). Generally speaking, a higher sugar
content leads to fewer ice crystals in ice cream, thus making it
softer (Vega et al., 2012). Addition of sugar molecules causes a
depression in the freezing point of water in ice cream. This slows
down the crystallization process and thus leads to the formation
of fewer ice crystals. The same effect can also be attained without
increasing the sweetness of the ice cream by using different kinds
of sugars.
The fat in ice cream also plays several roles. As already
mentioned, the fat helps to stabilize the foam part of the ice
cream, but it is also responsible for the creamy texture, and any
fat-​soluble flavour molecules are dissolved in it. The fat also
influences the mouldability (Goff, 1997; Goff, 2008) of the ice
cream and helps it to retain its shape when it is melting. Butterfat,
cream and vegetable oils are commonly used in both industrial
and homemade ice creams as fat sources.
Water makes up about 60–​72% of ice cream’s weight (Clarke,
2015). Not all of this is frozen into crystals; 15–​27% remains
liquid and contains dissolved sugars and proteins. Smaller ice
crystals are preferred over large ones, as they give a softer tex-
ture, unlike larger crystals, which lead to a coarser, gritty tex-
ture. The problem, however, is the propensity of the ice crystals
for recrystallization, which causes the small ice crystals to merge
into larger ones over time. This is also one of the major factors
limiting ice cream’s shelf life. In order to slow this down, it is
necessary to use stabilizers (usually polysaccharides such as
locust bean gum or carrageenans). Apart from limiting ice crystal
growth, stabilizers also help to achieve a smooth texture, reduce
or slow down lactose crystal growth, and increase the ice cream’s
resistance against meltdown.
Making a short leap from the fundamental physics to manufac-
turing ice cream, there are different possibilities for how to make
the ice cream after mixing the ingredients. Factory-​produced ice
cream is typically homogenized, pasteurized and aged before
further processing (Batt and Tortorello, 2014), but for home use,
these steps are skipped. In the food industry, the freezing is quite
Emulsions: Emulsified Systems in Food 237

often done with devices called scraped surface heat exchangers,

which in their simplest form consist of a cylindrical container
with a double wall and a rotating dasher in the middle. The
working principle is quite similar to that of a refrigerator; a lique-
fied gas (typically liquid ammonia) flows through the double wall
of the cylinder and cools the container (down to around −30 °C)
as it evaporates. The dasher in the middle scrapes the frozen ice
cream off the walls of the vessel while beating and aerating it
simultaneously. After initial freezing, the ice cream is hardened at
around −30 to −45 °C in a freezer with cold air. Alternatively, the
ice cream can also be processed via low-​temperature extrusion,
whereby the ice cream is sent through a cylindrical container with
refrigerated walls and a screw extruder. This method offers sev-
eral advantages in comparison to the first method, such as higher
FIGURE 33.13 Mango-​mint-​olive oil ice cream.
shear stress with lower shear rate, lower processing temperatures,
and the formation of smaller ice crystals and air bubbles.
The basic principles of ice cream manufacturing stay the same Earl Grey ice cream
for homemade ice cream. The easiest way to do this is by using
300 g Full-​fat milk
an ice cream maker, which works in the same way as the scraped
200 g Whipping cream
surface heat exchangers used in professional manufacturing. One
3 bags Earl Grey tea
of the major differences in this case is the freezing temperature 100 g Sugar
(around −20 °C), which is reached by using salted ice water (or 1 Egg yolk
an eutectic mixture of ice and salt, to be precise). Alternatively,
the ice cream mixture can be whisked at regular intervals while Blueberry-​oat ice cream
freezing (e.g., once every hour for two or three times) to break the
ice crystals. In this way, however, the consistency of the ice cream 400 g Oat milk
won’t necessarily be as good as by using an ice cream maker. 60 g Melted butter
300 g Blueberries (frozen)
60 g Oats
Pacojet Ice Cream 100 g Sugar
One possible way of making ice cream in professional gas- 1 Egg yolk
tronomy is by using Pacojet. In this equipment, the ice cream
mixture is frozen overnight and the frozen mixture is “blended” For the recipe with Earl Grey, the milk–​ cream mixture was
(or “pacotized”) the following day. We conducted our own warmed up and the tea bags soaked in it for about 10 minutes.
experiments on different flavour combinations in this way. The Otherwise, the ingredients were simply mixed together/​blended
basic recipe used in our simple experiments consisted essen- and frozen overnight. The frozen mixture was pacotized the
tially of the main flavour component (e.g., mango juice, red wine, following day and served immediately. For better consistency,
tea-​flavoured milk or plant-based drink), an emulsifier (e.g., egg the pacotized ice cream can be stored in a freezer for about an
yolk or pure lecithin), fat (e.g., olive oil or butter), and sugar. The hour before serving. The ingredients and amounts can easily be
following are four examples for recipes we used (Figure 33.13). varied for different flavour combinations and consistencies. As
quite often, the only limit is your imagination!
Mango-​mint-​olive oil ice cream
440 g Mango juice
60 g Olive oil Conclusion
100 g Caster sugar This chapter is intended to give a rough overview on the topic
1 Egg yolk of emulsified systems and foams in the context of food systems.
Handful of mint Emulsified systems play a big role in the food industry, but they
can also be found in the cosmetic, pharmaceutical and chemical
Glühwein ice cream industries, to name a few. Among foods, there is a great variety of
220 g Glühwein /​mulled wine emulsified systems ranging from beverages to solid food. As both
220 g Berry juice (mixed berries, e.g., black currants, emulsified systems and foams are fundamentally metastable, it
raspberries) is important to understand the physical principles underlying
60 g Melted butter this instability. This allows one to improve the shelf life, flavour,
100 g Caster sugar aroma retention, texture and many other aspects of food products.
1 Egg yolk Understanding these principles will also help with home cooking,
238 Markus Ketomäki et al.

whether saving a failing mayonnaise or appreciating why chilled Carteau D, Bassani D, Pianet I. 2008. The “Ouzo effect”: Following
whipping cream is better suited for whipping than warm. the spontaneous emulsification of trans-​anethole in water by
NMR. Comptes Rendus Chimie, 11 (4–​5), 493–​498.
Chandan RC. 2011. Dairy ingredients for food processing: An over-
REFERENCES view. In Dairy Ingredients for Food Processing (eds Chandan
Adams F, Walstra P, Brooks B, Richmond H, Zerfa M, Bibette J, RC and Kilara A). Blackwell Publishing Ltd, 3–​33.
Hibberd D, Robins M, Weers J, Kabalnov A. 2007. Modern Clarke C. 2015. The science of ice cream. Royal Society of Chemistry.
aspects of emulsion science. Royal Society of Chemistry. Cropper S, Kocaoglu-​Vurma N, Tharp B, Harper W. 2013. Effects of
Al-​Mahasneh M, Rababah T, Alu’Datt M. 2017. Effect of palm oil locust bean gum and mono-​and diglyceride concentrations on
(PO) and distilled mono-​ glycerid (dmg) on oil separation particle size and melting rates of ice cream. Journal of Food
and rheological properties of sesame paste. Journal of Food Science, 78 (6), C811–​C816.
Processing and Preservation, 41 (3), e12896. Delgado-​ Zamarreño M, Sanchez-​ Perez A, Gomez-​ Perez M,
Alger DB. 1994. Erroneous explanations for the limited water solu- Hernández-​ Méndez J. 1995. Automatic determination of
bility of organic liquids. Journal of Chemical Education, 71 liposoluble vitamins in butter and margarine using Triton X-​
(4), 281. 100 aqueous micellar solution by liquid chromatography with
Asano K, Hashimoto N. 1976. Contribution of hop bitter substances electrochemical detection. Analytica Chimica Acta, 315 (1–​2),
to head formation of beer. Report of the Research Laboratories 201–​208.
of Kirin Brewery Co. Ltd, 19, 9–​16. Dickie KH, Cann C, Norman EC, Bamforth CW, Muller RE. 2001.
Asano K, Hashimoto N. 1980. Isolation and characterization of Foam-​negative materials. Journal of the American Society of
foaming proteins of beer. Journal of the American Society of Brewing Chemists, 59 (1), 17–​23.
Brewing Chemists, 38 (4), 129–​137. Dickinson E, Semenova MG, Antipova AS. 1998. Salt stability of
Balinov B, Söderman O, Wärnheim T. 1994. Determination of water casein emulsions. Food Hydrocolloids, 12 (2), 227–​235.
droplet size in margarines and low-​calorie spreads by nuclear Drakos A, Kiosseoglou V. 2006. Stability of acidic egg white protein
magnetic resonance self-​diffusion. Journal of the American emulsions containing xanthan gum. Journal of Agricultural
Oil Chemists’ Society, 71 (5), 513–​518. and Food Chemistry, 54 (26), 10164–​10169.
Bamforth C. 1985. The foaming properties of beer. Journal of the El-​Zeini H, El-​Abd M, Mostafa A, El-​Ghany F. 2016. Effect of
Institute of Brewing, 91 (6), 370–​383. incorporating whey protein concentrate on chemical, rheo-
Bamforth C. 2000. Perceptions of beer foam. Journal of the Institute logical and textural properties of ice cream. Journal of Food
of Brewing, 106 (4), 229–​238. Processing and Technology, 7 (2), 1000546.
Bamforth C. 2016. Brewing materials and processes: A practical Essien E. 2003. Sausage manufacture: Principles and practice.
approach to beer excellence. Academic Press. Elsevier Science.
Bamforth C, Kanauchi M. 2003. Interactions between polypeptides Evans DE, Sheehan M, Stewart D. 1999.The impact of malt derived
derived from barley and other beer components in model foam proteins on beer foam quality. Part II: The influence of malt
systems. Journal of the Science of Food and Agriculture, 83 foam-​positive proteins and non-​ starch polysaccharides on
(10), 1045–​1050. beer foam quality. Journal of the Institute of Brewing, 105 (3),
Bamforth C, Russell I, Stewart G. 2011. Beer: A quality perspective. 171–​178.
Academic Press. Evans DE, Sheehan MC. 2002. Don’t be fobbed off: The substance
Bamforth CW. 1998. Bringing matters to a head: the status of research of beer foam –​A review. Journal of the American Society of
on beer foam. In Monograph European Brewery Convention. Brewing Chemists, 60 (2), 47–​57.
Verlag Hans Carl Getraenke-​Facherverlag, 10–​23. Farmer FM. 2011. The Boston Cooking School cook book. Read
Barth R. 2013. The chemistry of beer: The science in the suds. Wiley. Books Ltd.
Batt CA, Tortorello ML. 2014. Encyclopedia of food microbiology. Fischer S, Sommer K. 1999. Cling of beer foam to different
Academic Press, Vol. 10. surfaces. In Proceedings of the Congress-​European Brewery
Belitz HD, Grosch W, Schieberle P, Burghagen MM. 2004. Food Concention, 183–​190.
chemistry. Springer. Fuquay JW, McSweeney PL, Fox PF. 2011. Encyclopedia of dairy
Bongard S, Meynier A, Riaublanc A, Genot C. 2009. Protein-​ sciences. Academic Press.
stabilized oil-​in-​water emulsions and low-​fat dairy stirred gels. Goff HD. 1997. Colloidal aspects of ice cream—​ a review.
Journal of Food & Nutrition Research, 48 (1). International Dairy Journal, 7 (6–​7), 363–​373.
Brierley ER, Wilde PJ, Onishi A, Hughes PS, Simpson WJ, Clark Goff HD. 2008. 65 Years of ice cream science. International Dairy
DC. 1996. The influence of ethanol on the foaming proper- Journal, 18 (7), 754–​758.
ties of beer protein fractions: A comparison of Rudin and Griffin WC. 1949. Classification of surface-​active agents by “HLB”.
microconductivity methods of foam assessment. Journal of the Journal of the Society of Cosmetic Chemists, 1, 311–​326.
Science of Food and Agriculture, 70 (4), 531–​537. Griffin WC. 1954. Calculation of HLB values of non-​ionic surfactants.
Brown AC. 2018. Understanding food: Principles and preparation. Journal of the Society of Cosmetic Chemists, 5, 249–​256.
Cengage Learning. Grillo I. 2003. Small-​angle neutron scattering study of a world-​
Cano-​Ruiz M, Richter R. 1997. Effect of homogenization pressure wide known emulsion: Le Pastis. Colloids and Surfaces
on the milk fat globule membrane proteins. Journal of Dairy A: Physicochemical and Engineering Aspects, 225 (1–​ 3),
Science, 80 (11), 2732–​2739. 153–​160.
Carpenter D, Slover H. 1973. Lipid composition of selected Gunstone FD. 2006. Modifying lipids for use in food. Woodhead
margarines. Journal of the American Oil Chemists Society, 50 Publishing.
(9), 372–​376. Hansen A, Booker D. 1996. Flavor interaction with casein and whey
Carroll T. 1979. The effect of dissolved nitrogen gas on beer foam protein. ACS Publications.
and palate. Technical Quarterly Master Brewers Association of Hasenhuettl GL, Hartel RW. 2008. Food emulsifiers and their
America, 16 (3), 116–​119. applications. Springer, Vol. 19.
Emulsions: Emulsified Systems in Food
Hopia A, Sillanpää M, Tuomisto M. 2013. Hollandaise sauce and the
chemistry behind old and new preparation techniques. LUMAT
(2013–​2015 Issues), 1 (2), 189–​196.
Hu M, McClements DJ, Decker EA. 2003. Lipid oxidation in corn
oil-​in-​water emulsions stabilized by casein, whey protein iso-
late, and soy protein isolate. Journal of Agricultural and Food
Chemistry, 51 (6), 1696–​1700.
Huque EM. 1989. The hydrophobic effect. Journal of Chemical
Education, 66 (7), 581.
Ibarz A, Garvin A, Costa J. 1996. Rheological behaviour of sloe
(Prunus spinosa) fruit juices. Journal of Food Engineering, 27
(4), 423–​430.
Ihara K, Habara K, Ozaki Y, Nakamura K, Ochi H, Saito H, Asaoka
H, Uozumi M, Ichihashi N, Iwatsuki K. 2010. Influence of
whipping temperature on the whipping properties and rheo-
logical characteristics of whipped cream. Journal of Dairy
Science, 93 (7), 2887–​2895.
Kabalnov AS, Shchukin ED. 1992. Ostwald ripening theory:
Applications to fluorocarbon emulsion stability. Advances in
Colloid and Interface Science, 38, 69–​97.
Kanno, C. 1990. Secretory membranes of the lactating mammary
gland. Protoplasma, 159 (2–​3), 184–​208.
Kijowski J. 1995. Muscle proteins. In Chemical and functional
properties of food proteins (ed. Sikorski ZE). CRC Press, pp.
233–​239.
Langstaff S, Lewis M. 1993a. Foam and the perception of beer
flavor and mouthfeel. Technical quarterly –​Master Brewers
Association of the Americas (USA), 30 (1), 16–​17.
Langstaff SA, Lewis M. 1993b. The mouthfeel of beer –​a review.
Journal of the Institute of Brewing, 99 (1), 31–​37.
Leo A, Hansch C, Elkins D. 1971. Partition coefficients and their
uses. Chemical Reviews, 71 (6), 525–​616.
Lewis M, Serbia J. 1984. Aggregation of protein and precipitation
by polyphenol in mashing. Journal of the American Society of
Brewing Chemists, 42 (1), 40–​43.
Lewis MJ, Lewis AS. 2003. Correlation of beer foam with other beer
properties. Technical Quarterly & the MBAA Communicator,
40 (2), 114–​124.
Linke C, Drusch S. 2018. Pickering emulsions in foods –​opportun-
ities and limitations. Critical Reviews in Food Science and
Nutrition, 58 (12), 1971–​1985.
Lopez C, Ménard O. 2011. Human milk fat globules: Polar lipid
composition and in situ structural investigations revealing the
heterogeneous distribution of proteins and the lateral segrega-
tion of sphingomyelin in the biological membrane. Colloids
and Surfaces B: Biointerfaces, 83 (1), 29–​41.
López-​Montilla JC, Herrera-​Morales PE, Pandey S, Shah DO. 2002.
Spontaneous emulsification: Mechanisms, physicochemical
aspects, modeling, and applications. Journal of Dispersion
Science and Technology, 23 (1–​3), 219–​268.
Marshall RT, Goff HD, Hartel RW. 2003. Ice cream ingredients. In
Ice cream (eds Goff HD and Hartel RW), Springer, 55–​87.
Marshall, RT, Goff HD, Hartel RW. 2012. Ice cream. Springer.
McClements DJ. 2015. Food emulsions: Principles, practices, and
techniques. CRC Press.
McNaught AD, McNaught AD. 1997. Compendium of chemical ter-
minology. Blackwell Science Oxford, Vol. 1669.
Melm G, Tung P, Pringle A. 1995. Mathematical modeling of beer
foam. Technical quarterly (Master Brewers Association of the
Americas) (USA), 32 (1), 6–​10.
Merkus HG, Meesters GM. 2013. Particulate products: Tailoring
properties for optimal performance. Springer, Vol. 19.
239
Mirhosseini H, Tan CP, Hamid NS, Yusof S. 2008. Effect of Arabic
gum, xanthan gum and orange oil on flavor release from
diluted orange beverage emulsion. Food Chemistry, 107 (3),
1161–​1172.
Myhrvold, N, Young C, Bilet M. 2011. Modernist cuisine: The art
and science of cooking. The Cooking Lab.
Narziß L, Back W, Gastl M, Zarnkow M. 2017. Abriss der
Bierbrauerei. John Wiley & Sons.
Ohshima H. 2016. Encyclopedia of biocolloid and biointerface
science, 2 volume set. John Wiley & Sons, Vol. 2.
Olijve J, Mori F, Toda Y. 2001. Influence of the molecular-​weight
distribution of gelatin on emulsion stability. Journal of Colloid
and Interface Science, 243 (2), 476–​482.
Ozdemir C, Dagdemir E, Ozdemir S, Sagdic O. 2008. The effects of
using alternative sweeteners to sucrose on ice cream quality.
Journal of Food Quality, 31 (4), 415–​428.
Patel AR, Rodriguez Y, Lesaffer A, Dewettinck K. 2014. High
internal phase emulsion gels (HIPE-​gels) prepared using food-​
grade components. RSC Advances, 4 (35), 18136–​18140.
Pearson AM, Gillett TA. 2012. Processed meats. Springer US.
Provost JJ, Colabroy KL, Kelly BS, Wallert MA. 2016. The science
of cooking: Understanding the biology and chemistry behind
food and cooking. John Wiley & Sons.
Pulko I, Krajnc P. 2012. High internal phase emulsion templating –​
a path to hierarchically porous functional polymers.
Macromolecular Rapid Communications, 33 (20), 1731–​1746.
Rognså GH, Rathe M, Paulsen MT, Petersen MA, Brüggemann DA,
Sivertsvik M, Risbo J. 2014. Preparation methods influence
gastronomical outcome of hollandaise sauce. International
Journal of Gastronomy and Food Science, 2 (1), 32–​45.
Sarker DK, Wilde PJ, Clark DC. 1995. Control of surfactant-​induced
destabilization of foams through polyphenol-​ mediated
protein-​protein interactions. Journal of Agricultural and Food
Chemistry, 43 (2), 295–​300.
Siemens AO, Van Hecke M. 2010. Jamming: A simple introduction.
Physica A: Statistical Mechanics and Its Applications, 389
(20), 4255–​4264.
Silverstein TP. 1998. The real reason why oil and water don’t mix.
Journal of Chemical Education, 75 (1), 116.
Sinaga H, Bansal N, Bhandari B. 2017. Effects of milk pH alter-
ation on casein micelle size and gelation properties of milk.
International Journal of Food Properties, 20 (1), 179–​197.
Streitwieser A, Heathcock CH, Kosower EM, Corfield PJ. 1992.
Introduction to organic chemistry. Macmillan.
This H. 2016. Solution to Hollandaise challenge. Analytical and
Bioanalytical Chemistry, 408 (27), 7543.
Toldrá F. 2010. Handbook of meat processing. Wiley.
Varnam A, Sutherland JM, Sutherland JP. 1995. Meat and
meat products: Technology, chemistry and microbiology.
Springer US.
Vega C, Ubbink J, Van der Linden E. 2012. The kitchen as labora-
tory: Reflections on the science of food and cooking. Columbia
University Press.
Vilgis TA. 2011. Das Molekül-​Menü S. Hirzel Verlag.
Voilley A, Etiévant P. 2006. Flavour in food. Elsevier Science.
Waschatko G, Junghans A, Vilgis TA. 2012. Soy milk oleosome
behaviour at the air–​ water interface. Faraday Discussions,
158, 157–​169.
Weaire DL, Hutzler S. 2001. The physics of foams. Clarendon
Press.
Zayas JF. 2012. Functionality of proteins in food. Springer.
Emulsions and Foams: Ostwald Ripening and Disproportionation in
Practice
Hervé This vo Kientza1,2
1 INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​
AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
The physical phenomena called “Ostwald ripening” (particu-
larly applied to emulsions) and “disproportionation” (for foams)
are often discussed in molecular gastronomy circles, because
dispersed systems such as foams or emulsions are frequently part
of dishes. Much has been done since the discovery of these two
destabilizing mechanisms (among others) at the very beginning
of the 20th century (Ostwald, 1901), but in a handbook such as
this one, an important question is to know their real, quantitative
importance in culinary practice. Here, we give only elementary
data about what was made about the theory of these mechanisms,
and we add some comments about how the theory can be applied.
Emulsions and Foams
Dispersed systems are ubiquitous in food, because plant and
animal tissues used as culinary ingredients are formally gels (see
the Gels chapter in this book) and also because energy is added
through culinary processes, making emulsions from fat and water,
and foams from air and liquid. Even some raw plant tissues are
aerated systems; for example, apples contain up to 25% air
(Verboven et al., 2010). For dishes composed by assembling trad-
itional food ingredients (as opposed to ingredients used in note by
note cooking; see Note by Note Cooking and Note by Note Cuisine
chapter, by Hervé This vo Kientza and Róisín Burke), most systems
are dispersed, such as breads, cakes, desserts, etc. In particular,
sauces have complex structures, comprising emulsions, foams,
suspensions and more complex structures based on the elementary
(two phases) dispersed systems (This, 2012). If a minimum of 25%
of dispersed fat or air is arbitrarily used for deciding the physical
state of sauces, 56% of the traditional French savoury sauces from
the Guide culinaire (Escoffier et al., 1903) are emulsions, and 2%
of them are foamy (also called “aerated”) systems.
Regarding food emulsions, one the simplest (in terms of
dispersed system formalism (DSF) complexity) is mayonnaise
(Depree and Savage, 2001; This, 2016a). It is often said to be “an O/​
W emulsion made from egg, vinegar, and oil, containing typically
70 to 80% oil” (Ishibashi et al., 2016), but food science and tech-
nology has been too reliant on the data from the food industry for
these proportions instead of looking at culinary books, which con-
tain different and often more balanced information. In particular, for
mayonnaise, culinary books give different oil and water proportions.
For example, in the recipes given by Jules Gouffé (1867) or by a
“Group of Professionals” (Réunion de professionnels, 1946), the
mass proportions are: water 18%, oil (including both triglycerides
and phospholipids) 66% and proteins 4%. For Carême (1828), who
was closer to the time when mayonnaise appeared (he called it
“magnonnaise”), the oil proportion is even lower, with 51% and 49
% of water and oil, respectively. However, according to the Guide
culinaire, which has been used by many chefs since its publication,
the proportion of oil can reach 90%.
Of the hot savoury sauces that are said to be “emulsified”,
bearnaise, hollandaise, bechamel and white sauce are so popular
that they appear in the Codex Alimentarius (FAO, 2018), again
with definitions that do not correspond to culinary practices (and
they are more legitimate than the Codex Alimentarius, which
was designed by the industry, from traditional cooking). For
these sauces, the oil/​water proportions (w/​w) are respectively
65/​30, 65/​30, 18/​64 and 37/​65. However, it should be observed
that these sauces are not simple O/​W emulsions; bechamel and
white sauce owe much of their rheological behaviour to swollen
starch granules rather than packing of oil droplets; hollandaise,
bearnaise and custard (Figure 34.1) (This, 2012) owe their char-
acter to the coagulation of egg yolk proteins; they are sometimes
“suspensions” first, before being emulsified, contrary to what has
sometimes been published (Perram, 1977).
Indeed, even for the 17% of French savoury sauces that are
said to have the biphasic O/​W organization because of the use of
butter or oil a correct description should go along with an indi-
cation of temperature, as butter is often used for savoury sauces;
since the melting of milk fat begins at −10 °C and ends at 60 °C
(Lopez and Ollivon, 2009; Bouteille et al., 2012), an important
proportion of fat is in the solid state at room temperature, and
systems containing butter have a more complex DSF formula
than simply O/​W. Also, for many systems, the Ramsden (also
mistakenly called Pickering) description should replace the sim-
plistic idea of stabilization by molecular surfactants adsorbed at
the oil–​water interface (Binks, 2002; Rayner et al., 2014). For
241
242
FIGURE 34.1 Microscopic picture of bearnaise before (a) and after
(b) cooking. The proteins aggregate in solids dispersed in water, making a
(S+W)/​O system rather than a pure O/​W emulsion.
sauces in which the oil/​water ratio is reduced (e.g., less than
10%), other stabilization mechanisms can be observed, such as
an increase of the viscosity of the aqueous phase through the con-
centration of various compounds such as gelatin or saccharides
(mono-​, oligo-​or polysaccharides) or of aggregates obtained
through the thermal/​mechanical modifications of animal or plant
tissues (see the chapter on applesauce in this book).
For culinary foams as well, the issue of stability is more dif-
ficult than for the simple dispersion of air bubbles in a solution
of surfactants, because very few aerated food systems are limited
to only one liquid and one gaseous phase, as in whipped egg
white. Solid foams (continuous solid networks with dispersed
gas bubbles) are considered as outside the realm of this chapter,
even if many solid foams in foods, e.g., cakes, pastries, bread,
meringues, soufflés, etc., begin life as liquid foams, and during
this period instability may occur before their stabilization via for-
mation of a continuous solid network.
Let Us Remember That Metastable Systems
Are Not Thermodynamically Stable
One important question about culinary foams and emulsions
concerns their stability, because, if it is recognized that energy
is needed to make them, this means that there is a possibility of
Hervé This vo Kientza
destabilization, lowering their energy. The Gibbs free energy of
formation of an emulsion can be written simply:
∆G = γ∆A − T ∆S f (34.1)
where γ is the surface tension, ΔA the area increase, T the abso-
lute temperature and ΔSf the variation of entropy.
During the making of a dispersed system, the total interface
area of the dispersed phase increases (the ratio is equal to n
for one object divided into n smaller ones), which means that
energy is needed in the process. Of course, the entropy increases,
because there are more microscopic configurations having the
same energy after division than before; however, the energy
term outweighs the entropy ΔSf associated with the formation
of the dispersed structures from the bulk constituents. All this
results in ΔG ≥ 0, leading to the thermodynamic instability of the
emulsions.
The instability of dispersed systems is not due only to the
immiscibility of the hydrophobic phase and water; the difference
in density of the various phases is a main factor driving the phys-
ical evolution.
However, the physical description of dispersed systems is
not enough to understand the evolution of such systems, as the
particular molecular constitution of the phases (with solutes)
and of the surfactants (phospholipids do not behave in the
same way as proteins, for example, in terms of auto-​diffusion,
steric hindrance, interactions, etc.) determine their metasta-
bility. Moreover, it has to be observed that, for emulsified or
foamed culinary systems, which generally have more than two
phases, there are reasons to predict more complex changes
than for simple O/​W or G/​W systems. The volume fraction of
the dispersed phase (very concentrated emulsions can behave
rheologically as gels) and the size of structures have to be
considered (Robins et al., 2002).
Destabilization Mechanisms
Various destabilization mechanisms are known in which small,
dispersed structures (oil droplets or air bubbles) disappear while
large corresponding structures grow, reducing the stability of the
whole system (Wang et al., 2016). As is shown schematically in
Figure 34.2, these include:
1. creaming with or without aggregation and increase in
the droplet diameter;
2. aggregation with or without creaming;
3. increases in the droplet or bubble diameter through
Ostwald ripening;
4. droplet or bubble coalescence, leading to the production
of a separated oil phase;
5. gas diffusion from air bubbles.
As a result of their thermodynamic instability, emulsions and
foams will generally tend to reduce their total free energy through
an increase in droplet or bubble diameter, thereby reducing their
total interfacial area. Young (1805) and Laplace (1880) recognized
Emulsions: Ostwald Ripening
FIGURE 34.2 Various destabilization mechanisms for dispersed systems.
that the pressure in small, dispersed structures increases when the
radius decreases:
γ droplets are not in permanent contact (Schmitt et al., 2004). In
Pinside = Pmedium + 2 (34.2)
R
 ∂F 
where γ is the surface tension  and R the radius of the
bubble.  ∂A  T,V,n
The consequence is easily drawn for foams using Henry’s law,
which states that the concentration in a compound making up the
gas of bubbles, in the liquid continuous phase, can be described
as being proportional to the partial pressure of this compound
in the bubble (McQuarrie and Simon, 1997). This creates a con-
centration difference between small and large bubbles, triggering
the net diffusion of the compound toward the large bubbles. As
a whole, this all results in an overall increase in average bubble
size of the foam; this is the process of disproportionation, which
contributes to coarsening of foams. In the literature, this has also
been referred to as isothermal distillation or molecular diffusion
(Kabalnov et al., 1987; Varescon et al., 1990).
The same analysis can be done for emulsions, with the chem-
ical potential including a variation with concentration in the
liquid phase instead of pressure in gas. However, one should
observe immediately that, whereas Ostwald ripening and dis-
proportionation have often been compared, the actual effects are
different due to huge solubility differences of air (8 mg/​L for
oxygen at 20 °C) and triglycerides in water (less than 10–​9 mol/​L
for tristearin) (Funasaki et al., 1976).
Changes in Emulsions
In simple terms, Ostwald ripening is the growth of the lar-
gest droplets at the expense of the smallest ones as a result of
243
the difference in chemical potential of the material within
the droplets (which is an obvious way of saying that energy is
driving the phenomena of the world). Because of its importance
in food quality, Ostwald ripening in emulsions has been much
studied (Lifshitz and Slyozov, 1961; Wagner, 1961; Enomoto
et al., 1987; Kabalnov and Shuchkin, 1992; De Smet et al., 1997;
Yarranton and Masliyah, 1997; Dalgliesh, 1997). A comprehen-
sive review of Ostwald ripening in emulsions was published by
Taylor (1998), and a recent article studied how to distinguish
Ostwald ripening from coalescence (Santos et al., 2017).
Theoretically, a calculation of the rate of Ostwald ripening
(i.e., the rate at which the radius of the droplets in an emulsion
increases with time) has been proposed. However, this calcula-
tion holds only with important simplifications. The process is
assumed to be diffusion-​controlled, the rate-​limiting step of the
growth rate being the diffusion of the dissolved droplet phase
(solute) through the bulk phase (Meinders et al., 2002; Novales
et al., 2003). This excludes any other effects at the interface that
would dominate the rate. In particular, it is assumed that there
is no barrier to the passage of the solute across the interface: in
emulsions stabilized by small surfactants, this is a reasonable
assumption, but it may not be true in systems stabilized by thick
polymeric species such as proteins (present in most culinary
systems), as in the simple culinary emulsion that is mayonnaise.
In order to test such theoretical analysis, many experimental
studies have been performed at the limit of highly dilute emulsions
(volume fraction lower than 1%), i.e., in conditions where the
the kitchen, this can occur, for example, when an emulsion is
diluted, such as in soups from Gascony that are made by adding
a vegetable stock to aiolli or mayonnaise (Daguin, 1981). But it
does not hold for mayonnaise and other cousins.
The models used to describe Ostwald ripening are based on
and similar to the models used to describe the dissolution of an
air bubble and disproportionation of a foam. For a single droplet
with radius R in an infinite medium, the transport of molecules
over the surface per unit time can be written using Fick’s diffusion
law, in the steady state solution, as:
d
n = 4 π RD (Cm − Cs ) (34.3)
dt R
with D the diffusion coefficient, Cm the concentration of the
dispersed phase in the medium, and Cs the solubility of the
dispersed droplet, which is given by Kelvin’s equation:
 α
C s = C∞  1 +  (34.4)
 R
with C∞ the bulk solubility and α the so-​called capillary length,
given by:
Vm
α=2 σ (34.5)
RgT
where Vm is the molar volume of the dispersed phase, Rg the gas
constant, T the temperature and σ the interfacial tension.
244
Now, for a entire emulsion, with many droplets, the most com-
plete theory was proposed independently by Lifshitz and Slyozov
(1961) and by Wagner (1961). This so-​called “LSW theory”
assumes that the system has been ripening for a sufficiently long
time for quasi-​steady state conditions to apply (asymptotic solu-
tion). For a full discussion of the LSW theory and its derivation,
see Ramsey (1947); Dunning (1973); Kahlweit (1975). One result
is that the coarsening rate, defined as the rate of increase of the
cube of the number-​averaged particle size, is equal to:
R3 dRC 3 4α DC∞
ω= = = (34.6)
dt dt 9ρ
where ρ is the density of the dispersed phase.
Because of the importance of interfacial elasticity, Ostwald
ripening has been studied by means of numerical simulations in
which time-​dependent (elastic) interfacial behaviour is taken into
account (Meinders and van Vliet, 2004). Calculations on the dis-
solution of a single droplet in an infinite medium under saturated
conditions show that the dissolution process can be stopped
only when the interfacial tension goes to zero; when interfacial
stress relaxation is included, which prevents a continuous zero
interfacial tension, no stabilization of the dissolution process is
observed, and the droplet dissolves completely. In the case of a
group of droplets, using the assumptions of LSW theory, numer-
ical calculations on the coarsening of emulsion droplets with
finite interfacial elasticity show that a stable situation occurs at
finite interfacial tensions of the droplets. The coarsening behav-
iour strongly depends on the saturation of the dispersed phase
in the continuous phase. Stabilization can only be accomplished
by adding insoluble species to the dispersed phase, by using
particles as stabilizers, such as in Ramsden systems (Rayner
et al., 2014), or by micro-​encapsulation of the emulsion droplets
by thick insoluble interfacial layers, which have a thickness that
is in the order of the radius of the droplet.
Vincent (1984) has suggested a criterion to determine whether
Ostwald ripening will occur in an emulsion, and suggested that
the presence of surfactant molecules may form interfacial tension
gradients, which may oppose Ostwald ripening. A droplet may be
considered to be in equilibrium if dP/​dr > 0, where P is the Laplace
pressure of the droplet. Consequently, taking into consideration the
area of the droplet and the Laplace equation, 2 dγ/​d ln(A) > γ. The
left hand side of this inequality is just the dilational modulus, ϵ,
of the interface, and so, for mechanical equilibrium:
2ϵ>γ (34.7)
If this inequality is satisfied, then the droplet should be in equilib-
rium and show no Ostwald ripening.
The control of Ostwald ripening has also been the subject of
many studies (Webster and Cates, 1998; Webster and Cates, 2001).
It may be slowed down or halted by the addition of components
that are poorly soluble or insoluble in the continuous phase.
However, phenomena can change completely depending on
particular systems. Taisne and Cabane (1998) examined the
coarsening of concentrated alkane-​in-​water droplets (ϕ ≈ 20–​40%)
Hervé This vo Kientza
stabilized by a non-​ionic poly(ethoxylated) surfactant, following
a temperature quench. Since the rate of ripening does not depend
on the alkane chain length, they concluded that the transfer of
oil from the smaller drops to the larger ones does not occur by
diffusion across the continuous phase but rather, through the
direct contact of the droplets (permeation).
Finally, the traditional view of emulsions is not enough, and
the analysis of culinary emulsions should consider systems such
as nano-​ emulsions, micro-​ emulsions and Ramsden emulsions
(Berton-​Carabin and Schroën, 2015; Solans et al., 2005;
McClements, 2012). For nano-​emulsions, with droplet sizes ran-
ging from 50 to 200 nm, Ostwald ripening is the main destabil-
izing mechanism (Izquierdo et al., 2002). For micro-​emulsions,
coalescence is considered to be the major mechanism of destabil-
ization, but stabilizers can often prevent it.
Changes in Foams
In food systems, there are very few pure G/​L (where G is a gas,
and L a liquid) foams. With molecular cooking, more G/​W (W
stands for an aqueous solution) foams were introduced, using
surfactants such as milk or egg proteins, or various additives such
as lecithins (This, 2016b). In liquid foams, drainage of the con-
tinuous phase between bubbles, followed by the close approach of
bubble surfaces and bubble coalescence, can occur, leading to foam
collapse, loss of gas, and loss of desired foam structure and consist-
ency. In addition, even in the absence of drainage and coalescence,
diffusion of gas can occur, as said earlier, between bubbles that
possess different internal pressures (and therefore concentrations
of gas). The net diffusion flux is usually greatest from small to large
bubbles, due to the greater Laplace pressure of the former, although
the pressure within bubbles at any given time also depends on the
interfacial elasticity of the adsorbed film around the bubbles and
the extent of their shrinkage. More rapid shrinkage against a high
interfacial dilatational elasticity reduces the net internal pressure
more quickly (Dickinson et al., 2002; Ettelaie et al., 2003). The
migration of gas between bubbles eventually leads to coarsening of
the foam. As said before, the effect is called disproportionation and
is analogous to Ostwald ripening in emulsions; it can accelerate
drainage and coalescence (Murray and Ettelaie, 2004).
De Vries (1958a; 1958b), Princen and Mason (1965) and
Princen et al. (1967) provided the basic framework for studying
gas diffusion to an atmosphere from a single bubble. Prins (1987)
showed that the viscoelasticity of the bubble surface would
reduce the rate of gas diffusion.
Lucassen (1981) and Wijnen and Prins (1995) confirmed that
the dynamic surface properties, such as interfacial elasticity, have
a significant effect on the rate of Ostwald ripening: disproportion-
ation can be stopped if the bubble surface is purely elastic and
the surface dilatational elasticity is larger than half the surface
tension.
How Does This Apply?
For sure, the discovery of new phenomena and mechanisms is
a goal of the natural sciences in general, and of molecular gas-
tronomy in particular, but as said in the Introduction, knowing
Emulsions: Ostwald Ripening
their real effect is an important step for improving theories, as
new mechanisms can appear when the known ones do not cor-
rectly explain the phenomena.
Dispersed systems are thermodynamically unstable, but they
are metastable due to the presence of the adsorbed layers at the
O/​W or the G/​W interface; barriers may be electrostatic in nature
(adsorption of an ionic surface active agent), or steric (adsorp-
tion of a non-​ionic surface active agent or polymer), and the two
effects can come into play, for proteins at certain pH values, for
example. These barriers not only prevent dispersed particles
from coming into direct contact but also serve to stabilize the
thin film of liquid between two adjacent droplets or bubbles.
Coalescence occurs when this thin film thins further and
ruptures. The presence of adsorbed surfactants can reduce
the likelihood of rupture through the Gibbs–​Marangoni effect
(Firouzi and Nguyen, 2017). Reduction in interfacial tension may
also play a significant role in stability. In the case of polymers,
other factors may also come into play, such as the viscoelastic
nature of the adsorbed layer. There have been many attempts
to correlate the viscoelastic properties of adsorbed layers with
emulsion stability, but with only limited success.
Application to Emulsions
In the kitchen, emulsions are never of the kinds that were the-
oretically described, because either they are far from the dilute
regime, or the high viscosity of the continuous liquid phase is
too complex for simple diffusion, or particles of different kinds
adsorb at the interfaces. This is sometimes recognized, as when
Tcholakova observed that “the Ostwald ripening is usually neg-
ligible for protein-​stabilized triglyceride-​ in-​
water emulsions”
(Tcholakova et al., 2006).
Moreover, culinary systems usually contain a lot of different
surfactants that protect emulsions against aging factors such
as coagulation and coalescence. If added in sufficiently large
amounts and after complete coverage of the oil–​ water inter-
face, surfactants spontaneously form micelles in the continuous
aqueous phase: the presence of micelles drastically increases
the “solubility” of the oil. Therefore, an effect of micelles on
the Ostwald ripening may be anticipated. This effect can even be
enhanced dramatically by adding micellar surfactant solutions to
ripening emulsions. If enough surfactant is added, a competition
can occur between droplet size increase by Ostwald ripening, on
the one hand, and droplet size decrease by solubilization of the oil
in the added micelles, on the other hand.
Also, theoretical analysis of the distribution of radiuses should
be considered with care, because whereas food science considers
typically highly polydisperse emulsions, with droplet radii easily
spanning two orders of magnitude (Robins et al., 2002), it has
been shown that, for finished mayonnaise, only one order of
magnitude is considered for droplet radiuses, and their droplet
size is so big (up to 0.1 mm) that the droplets are not subject to
Brownian motion, which would resist gravity-​induced creaming.
For emulsions in which egg yolk is used, the emulsifying prop-
erties of egg yolk are regulated by phospholipids, lipoproteins,
hydrophobic and hydrophilic proteins (livetin and phosvitin),
and cholesterol (This, 2009). However, it is not clear whether the
245
interface is covered by low-​density lipoprotein (LDL) (Mizutani
and Nakamura,1985) or high-​density lipoprotein (HDL).
Recently, Ariizumi et al. (2017) reported that mayonnaise
prepared with lower egg yolk concentrations was destabilized
faster, and they showed that long-​term stability (for five months)
was affected by the protein load and protein composition in
the adsorbed layer at the interface. Usually, proteins stabilize
emulsions through steric repulsion, which is highly dependent on
pH and ionic strength (Guilmineau and Kulozik, 2006a; 2006b).
Clearly, this observation is not relevant for culinary practice,
because it would be microbiologically dangerous to keep a sauce
for such a long time in the usual home conditions, and also, cooks
do not try to reduce the yolk concentration (the yolk is used in
kitchens for flavour).
In terms of salt, laboratory conditions also do not match
culinary ones. Salt can certainly neutralize any charge found in
proteins (McClements, 1999), which aids protein adsorption onto
oil droplets, and it can be calculated that there are about 100 neu-
tralizing ions for each charge on a protein, in culinary conditions,
so that this effect could hold. On the other hand, it has been
written (Prost and Rondelez, 1991) that salt could destabilize
emulsions, but this is not true for mayonnaise; in our experiments,
the emulsified structure remained after the addition of 200 g of
sodium chloride to a mayonnaise made from 1 egg yolk, 20 g of
an 8% acetic acid solution and 200 g of sunflower oil. Clearly,
culinary emulsions are different from poly(dimethylsiloxane)
O/​W emulsions, which were destabilized through coalescence at
1 M NaCl concentration (Koh et al., 2000). Of course, it has to be
observed that such a huge quantity of salt is never used.
More generally, the food science literature contains strange
information as compared with culinary practice. For example,
it has been reported that high mayonnaise stability is generally
associated with a fine and uniform oil droplet size (Harrison
and Cunningham, 1985), whereas mayonnaise/​spoonable salad
dressing droplets are smaller, with sizes of about 2–​3 μm (Tung
and Jones, 1981). However, in our experiments, we measured oil
droplet diameters between 1 and 100 μm.
In the last few years, micro-​and nano-​particles of biological
origins have been extensively explored as colloid stabilizers (i.e.,
stabilization of liquid/​liquid and/​or liquid/​vapour interfaces) for
application in bio-​related fields, such as foods and nutraceuticals.
Interestingly, this concept has been applied for centuries in food
products without being scientifically researched (Tavernier et al.,
2016). For instance, in remoulade (a precursor of mayonnaise,
known at least since 1419) (Tirel, 1392), the fine mustard particles
contribute to physical and colloidal stability of the emulsion by
preventing the coalescence of oil droplets (Binks, 2002). Indeed,
the flourishing field of Ramsden emulsions (also called Pickering
emulsions) is discovering outstanding stabilizations against
coalescence and Ostwald ripening compared with emulsions
stabilized with low-​molecular-​weight emulsifiers (Timgren et al.,
2011; Kargar et al., 2012).
Application to Foams
Surface-​active macromolecules, such as most proteins, tend to
form adsorbed layers resembling thin three-​dimensional gels,
246
which essentially have little effect on the mean free path of gas
molecules through them. The interfacial rheology of adsorbed
films can modify the rate and extent of shrinkage of bubbles due
to disproportionation. In order to shrink, bubbles have to do work
against the interfacial elasticity and viscosity, which provide an
energy barrier that could inhibit bubble shrinkage. In a key article,
Kloek et al. (2001) reviewed the potential effects of both bulk
rheology and interfacial dilatational rheology on the shrinkage
kinetics of bubbles. Some estimates of the range of effects in real
food systems are discussed.
For the most part, the behaviour of adsorbed protein films
during disproportionation seems to fit very well to a simple the-
oretical model of gas diffusion and bubble shrinkage, where a
zero or finite value of the dilatational elasticity is included
(McClements, 1999; Robins, 2000). This also indicates that, at the
typical rates of interfacial deformation due to bubble shrinkage
that are observable, dilatational viscosities are not sufficiently
high to exert a significant effect. Loss moduli are expected to fall De Vries AJ. 1958b. Foam stability Part II: Gas diffusion in foams,
with decreasing deformation rate (or frequency) of deformation.
So far, only one notable exception to this good fit of experiment
to theory has been noted (Al-​Malah et al., 2000).
This anomaly occurred with the egg white protein ovalbumin.
Results with ovalbumin were often not very reproducible, and
occasionally very much longer shrinkage times than usual were
observed, which did not fit the simple theory. Ovalbumin is one
of those proteins that microscopically appear to form protein
particles around a bubble as it shrinks, but in fact, it does this
far more readily than other proteins. Aggregates are often visible
before significant shrinkage has taken place, and the aggregates
sometimes appear to be part of some sort of semi-​soluble inter-
facial protein network before shrinkage has even begun.
However, here again, food science departs strangely from what
is really observed in the kitchen. For example, Dutta et al. (2004)
wrote that “some products that benefit from air incorporation are
ice cream, confectionery, salad dressings and mayonnaise” … but
in fact, microscopic analysis of mayonnaise does not show any air
bubbles; mayonnaise is not an aerated system.
Conclusions and Perspectives
After decades of work on very simple models that do not corres-
pond to culinary practices, recent results on Ramsden emulsions Funasaki N, Hada S, Suzuki K. 1976. The dissolution state of a tri-
are now getting closer to the description of culinary systems of
the O/​W and G/​W kinds, but a lot remains to be done to investi-
gate all possible microstructures that can be met in sauces. And
new mechanisms could be discovered during this analysis!
REFERENCES
Al-​Malah KI, Azzam MOJ, Omari RM. 2000. Emulsifying proper-
ties of BSA in different vegetable oil emulsions using conduct- Guilmineau F, Kulozik U. 2006b. Impact of a thermal treament on the
ivity technique, Food Hydrocolloids, 14, 485.
Ariizumi M, Kubo M, Handa A, Hayakawa T, Matsumiya K,
Matsumura Y. 2017. Influence of processing factors on the Ishibashi C, Hondoh H, Ueno S. 2016. Influence of morphology and
stabilityof model mayonnaise with whole egg during long-​
term storage, Bioscience, Biotechnology, and Biochemistry,
81(4), 803–​811.
Hervé This vo Kientza
Berton-​Carabin CC, Schröen K. 2015. Pickering emulsions for food
applications: background, trends, and challenges, Annual
Review in Food Science and Technology, 6, 263–​297.
Binks BP. 2002. Particles as surfactants –​similarities and differences,
Current Opinion in Colloid and Interface Science, 7, 21.
Bouteille R, Perez J, Khifer F, Jouan-​ Rimbaud-​ Bouveresse D,
Lecanu B, This H. 2013. In situ quantitative proton nuclear
magnetic resonance spectroscopy analysis of milk fat fusion,
Journal of Dairy Science, 96, 2071–​2080.
Carême MA. 1828. L’Art de la cuisine française au XIXe siècle,
Chez l’auteur, Paris.
Daguin A. 1981. Le nouveau cuisinier gascon, Stock, Paris.
Dalgliesh DG. 1997. Adsorption of protein and the stability of
emulsions, Trends in Food Science and Technology, 8, 1–​6.
De Smet Y, Deriemaeker L, Finsy R. 1997. A simple computer simu-
lation of Ostwald ripening, Langmuir, 13, 6884.
De Vries AJ. 1958a. Foam stability Part I: Structure and stability
of foams, Recueil des travaux chimiques des Pays-​ Bas,
77(1), 81–​91.
Recueil des travaux chimiques des Pays-​Bas, 77(3), 209–​223.
Depree JA, Savage GP. 2001. Physical and flavour stability of may-
onnaise, Trends in Food Science and Technology, 12(5–​6),
157–​163.
Dickinson E, Ettelaie R, Murray BS, Du Z. 2002. Kinetics of dispro-
portionation of air bubbles beneath a planar air–​water interface
stabilised by food proteins, Journal of Colloid and Interface
Science, 252, 202–​214.
Dunning WJ. 1973. In Smith AL (Ed.), Particle Growth in Suspensions,
SCI Monograph No. 38, Academic Press, London.
Dutta A, Chengara A, Nikolov AD, Wasan DT, Chen K, Campbell
B. 2004. Texture and stability of aerated food emulsions –​
effects of buoyancy and Ostwald ripening, Journal of Food
Engineering, 62, 169–​175.
Enomoto Y, Kawasaki K, Tokuyama M. 1987. Computer modeling of
Ostwald ripening, Acta Metallurgica, 35, 907–​913.
Escoffier A, Gilbert P, Fetu E. 1903. Guide culinaire, Emile
Colin, Paris.
Ettelaie E, Dickinson E, Du Z, Murray BS. 2003. Disproportionation
of clustered protein-​ stabilized bubbles at planar air-​ water
interfaces, Journal of Colloid and Interface Science, 263,
47–​58.
FAO. 2018. Codex Alimentarius, www.fao.org/​fao-​who-​codexalimen
tarius/​fr/​
Firouzi M, Nguyen AV. 2017. The Gibbs–​ Marangoni stress and
non DLVO forces are equally important for modeling bubble
coalescence in salt solutions, Colloids and Surfaces A:
Physicochemal and Engineering Aspects, 515, 62–​68.
glyceride molecule in water and its orientation state at the air–​
water interface, Chemical and Pharmaceutical Bulletin, 24(4),
731–​735.
Gouffé J. 1867. Le livre de cuisine : comprenant la cuisine de ménage
et la grande cuisine. Hachette, Paris.
Guilmineau F, Kulozik U. 2006a. Impact of a thermal treament on the
emulsifying properties of egg yolk. Part 1: Effect of the heating
time, Food Hydrocolloids, 20(8), 1105–​1113.
emulsifying properties of egg yolk. Part 2: effect of the envir-
onmental conditions, Food Hydrocolloids, 20(8), 1114–​1123.
polymorphic transformation of fat crystals on the freeze-​thaw
stability of mayonnaise-​ type oil-​
in-​
water emulsions. Food
Research International, 89, 604–​613.
Emulsions: Ostwald Ripening
Izquierdo, P, Esquena J, Tadros TF, Dederen C, Garcia MJ, Azemar
N. 2002. Formation and stability of nano-​emulsions prepared
using the phase inversion temperature method, Langmuir,
18, 26–​30.
Kabalnov AS, Pertzov AV, Shchukin ED. 1987. Ostwald ripening
in emulsions: I. Direct observations of Ostwald ripening in
emulsions, Colloids Surfaces, 118(2), 590–​597.
Kabalnov AS, Shchukin ED. 1992. Ostwald ripening theory:
Applications to fluorocarbon emulsion stability, Advances in
Colloid and Interface Science, 389, 69–​97.
Kabalnov AS. 2001. Ostwald ripening and related phenomena,
Journal of Dispersion Science and Technology, 22, 1–​12.
Kahlweit M. 1975. Ostwald ripening of precipitates, Advances in
Colloid and Interface Science, 5, 1–​35.
Kargar M, Fayazmanesh K, Alavi M, Spyropoulos F, Norton IT. 2012.
Investigation into the potential ability of Pickering emulsions
(food-​grade particles) to enhance the oxidative stability of oil-​
in-​water emulsions, Journal of Colloid and Interface Science,
366(1), 209–​215.
Laplace PS. 1880. Oeuvres complètes de Laplace, Gauthier Villars,
Paris, ­chapter 4, 385.
Lifshitz IM, Slyozov VV. 1961. The kinetics of precipitation
from supersaturated solid solutions, Journal of Physics and
Chemistry of Solids, 19, 35–​50.
Lifshitz IM, Slezov VV. 1959. Kinetics of diffusive decompos-
ition of supersaturated solid solutions, Soviet Physics Journal
of Experimental and Theoretical Physics, 35(8), 331–​ 339
(English translation).
Lopez C, Ollivon M. 2009. Triglycerides obtained by dry fractionation
of milk fat 2. Thermal properties and polymorphic evolutions
on heating, Chemistry and Physics of Lipids, 159, 1–​12.
Lucassen J. 1981. Dynamic properties of free liquid films and
foams. In Lucassen-​Reynders EH (Ed.), Anionic Surfactants:
Physical Chemistry of Surfactant Action, Dekker, New
York, 217.
Lucassen-​Reynders EH. 1981. Surface elasticity and viscosity in
compression dilatation. In Lucassen-​ Reynders EH (Ed.),
Anionic Surfactants: Physical Chemistry of Surfactant Action,
Dekker, New York, 173.
Martin AH, Grolle K, Bos MA, Cohen Stuart MA, van Vliet T. 2002.
Network forming properties of various proteins adsorbed at
the air/​water interface in relation to foam stability, Journal of
Colloid and Interface Science, 254, 175–​183.
McClements DJ. 1999. Food Emulsions: Principles, Practice and
Techniques, CRC Press, Boca Raton, FL.
McClements DJ. 2012. Nanoemulsions versus microemulsions:
Terminology, differences, and similarities, Soft Matter, 8,
1719–​1729.
McQuarrie DA, Simon JD. 1997. Physical Chemistry, University
Science Books, Sausalito, California.
Meinders MBJ, Kloek W, van Vliet T. 2001. Effect of surface elasticity
on ostwald ripening in emulsions, Langmuir, 17, 3923–​3929.
Meinders MBJ, van Vliet T. 2004. The role of interfacial rheological
properties on Ostwald ripening in emulsions, Advances in
Colloid and Interface Science, 108–​109, 119–​126.
Meinders MBJ, Bos MA, Lichtendonk WJ, van Vliet T. 2002. In
Dickinson E, van Vliet T (Eds.), Food Colloids: Biopolymers
and Materials, Royal Society of Chemistry, Cambridge.
Mizutani R, Nakamura R. 1985. Emulsifying properties of egg yolk
low density lipoprotein (LDL): comparison with bovine serum
albumin and egg lecithin, Lebensmittel Wissenschaft und
Technology, 17, 213–​216.
247
Murray BS, Ettelaie R. 2004. Foam stability: proteins and
nanoparticles, Current Opinion in Colloid & Interface Science,
9(5), 314–​320.
Nguyen Hoang TK, Deriemaeker L, La V, Finsy R. 2004. Monitoring
the simultaneous Ostwald ripening and solubilization of
emulsions, Langmuir, 20(21), 8966–​8969.
Novales B, Papineau P, Sire A, Axelos MAV. 2003. Characterization
of emulsions and suspensions by video image analysis,
Colloids and Surfaces A : Physicochemical and Engineering
Aspects, 221, 81–​89.
Ostwald W. 1901. Analytische Chemie, 3rd ed., Engelmann,
Leipzig, 23.
Perram CM, Nicolau C, Perram JC. 1977. Interparticle forces in
multiphase colloid systems: the resurrection of coagulated
sauce béarnaise. Nature, 270, 572.
Princen HM, Mason SG. 1965. The permeability of soap films to
gases, Journal of Colloid Science, 20, 353–​375.
Princen HM, Oberbeek JT, Mason SG, 1967. The permeability of
soap films to gases: II. A simple mechanism of monolayer
permeability, Journal of Colloid and Interface Science, 24(1),
125–​130.
Prins A. 1987. Theory and practice of formation and stability of
food foams. In Dickinson E (Ed.) Food Emulsions and Foams,
Spec public, The Royal Society of Chemistry, Cambridge,
UK, 30–​39.
Prost J, Rondelez F. 1991. Structures in colloidal physical chemistry,
Nature Supplement, 350, 11–​23.
Ramsey AS. 1947. A Treatise on Hydrodynamics II, G. Bell, London.
Rayner R, Marku D, Eriksson M, Sjöö M, Dejmek P, Wahlgren M.
2014. Biomass-​based particles for the formulation of Pickering
type emulsions in food and topical applications, Colloids and
Surfaces A: Physicochemical and Engineering Aspects, 458
(2014), 48–​62.
Réunion de professionnels. 1946. La cuisine moderne, Aristide
Quillet, Paris.
Robins MM. 2000. Emulsions –​creaming phenomena, Current
Opinion in Colloid and Interface Science, 5, 265.
Robins MM, Watson AD, Wilde PJ. 2002. Emulsions –​creaming and
rheology, Current Opinion in Colloid and Interface Science,
7, 419–​425.
Santos J, Calero N, Trujillo-​Cayado LA, Garcia MC, Munoz J. 2017.
Assessing differences between Ostwald ripening and coales-
cence by rheology, laser diffraction and mutiple light scattering,
Colloids and Surfaces B: Biointerfaces, 159, 405–​441.
Schmitt V, Arditty S, Leal-​Calderon F. 2004. Stability of concentrated
emulsions. I. In Emulsions: Structure Stability and Interactions
(Petsev DN Ed.), Academic Press, Cambridge (USA),
­chapter 4, 607–​639.
Skinner LM, Sambles JR. 1972. The Kelvin equation, a review,
Journal of Aerosol Science, 3, 199.
Solans C, Izquierdo P, Nolla J, Azemar N, Garcia-​Celma MJ. 2005.
Nano-​emulsions, Current Opinion in Colloid and Interface
Science, 10 (2005), 102–​110.
Tavernier I, Wijaya W, Van der Meeren P, Dewettinck K, Patel A.
2016. Food-​grade particles for emulsion stabilization, Food
Science & Technology, 50, 159–​174.
Taylor P. 1998. Ostwald ripening in emulsions, Advances in Colloid
and Interface Science, 75, 107.
Tcholakova S, Denkov ND, Ivanov IB, Campbell B. 2006.
Coalescence stability of emulsions containing globular milk
proteins, Advances in Colloid and Interface Science, 123,
259–​293.
248
Tcholakova S, Denkov ND, Lips A. 2008. Comparison of solid
particles, globular proteins and surfactants as emulsifiers,
Physical Chemistry Chemical Physics, 10, 1608–​1627.
This H. 2009. Molecular Gastronomy, a chemical look to cooking,
Accounts of Chemical Research, 42(5), 575–​583.
This H. 2012. Solutions are solutions, and gels are almost solutions,
Pure and Applied Chemistry, 2012, 1–​20.
This H. 2016a. Statgels and dynagels, Notes Académiques de
l’Académie d’agriculture de France /​Academic Notes from the
French Academy of Agriculture, 12, 1–​12.
This H. 2016b. Mon histoire de cuisine, Belin, Paris.
Tirel G. 1392. Viandier of Taillevent: An Edition of All Extant
Manuscripts, University of Ottawa Press, Ottawa, Canada,
1988.
Tung MA, Jones IJ. 1981. Microstructure of mayonnaise and salad
dressing, In Holcomb DN, Kalab M (eds.), Studies of Food
Microstructure, Scanning Electron Microscopy, AMF, O’Hare,
Illinois, 231.
Varescon C, Manfredi A, Le Blanc M, Reiss JG. 1990. Deviation from
molecular diffusion aging model in fluorocarbon emulsions
stabilized by perfluoroalkylated surfactants, Journal of Colloid
and Interface Science, 137, 373–​379.
Verboven P, Kerckhofs G, Mebatsion HK, Ho QT, Temst K, Wevers
M, Cloetens P, Nicolai BM. 2018. Three-​ dimensional gas
exchange pathways in pome fruit characterized by synchrotron
x-​ray computed tomography, Plant Physiology, 147, 518–​527.
Hervé This vo Kientza
Vincent B. 1984. Introduction to colloidal dispersions, In Tadros ThF
(Ed.), Surfactants, Academic Press, London.
Voorhees PW. 1985. The theory of Ostwald ripening, Journal of
Statistical Physics, 38(1/​2).
Wagner C. 1961. Theorie der Alterung von Niederschlägen durch
Umlösen (Ostwald-​Reifung), Zeitschrift für Electrochemie,
65, 581.
Wang J, Nguyen AV, Farrokhpay S. 2016. A critical review of the
growth, drainage and collapse of foams, Advances in Colloid
and Interface Science, 228, 55–​70.
Webster AJ, Cates ME. 1998. Stabilization of emulsions by trapped
species, Langmuir, 14, 2068–​2079.
Webster AJ, Cates MZ. 2001. Osmotic stabilization of concentrated
emulsions and foams, Langmuir, 17, 595–​608.
Wijnen ME, Prins A. 1995. Disproportionation in aerosol whipped
cream, In Food. Macromolecules and Colloids, Dickinson,
E. and Lorient, D. (eds.), The Royal Society of Chemistry,
Cambridge, UK, 309–​311.
Yarranton HW, Masliyah JH. 1977. Numerical simulations of
Ostwald ripening in emulsions, Journal of Colloid and
Interface Science, 196(2), 157–​169.
Young, T. 1805. An essay on the cohesion of fluids, Philosophical
Transactions of the Royal Society of London, 95, 65–​87.
Emulsions: Lecithin
Elzbieta Kozakiewicz1 and Daniel Cossuta2
1 Quality & Food Safety Coordinator, Bunge
2 Head of Specialty Ingredients – Global Innovation at Bunge Loders Croklaan
Lecithin has been described with several different definitions
in the scientific literature, and there is a simple reason for these
differences: the studies of phospholipids started in the early
1700s, but the sales of lecithin products only started in the 1930s.
The research and development of these products took more than
200 years to reach commercialization, and throughout this time
different definitions appeared, based on the current knowledge
and objectives. For example, in his book, Szuhaj (1989) mentions
three different definitions. The first definition is for commer-
cially used lecithin, which is interpreted as a natural mixture
that contains polar lipids (glycolipids, phospholipids) and neu-
tral lipids (triglycerides) and is obtained from plant or animal
sources. Lecithin has also been defined as a group of phosphorus-​
containing lipids extracted from eggs or brain tissues. The third
definition, called “scientific” by Szuhaj, is that the lecithin is vir-
tually phosphatidylcholine (PC) itself.
According to the International Lecithin & Phospholipids
Society (ILPS, 2020), lecithin is defined as: “A complex mix-
ture of glycerophospholipids of animal, plant, or microbio-
logical origin, containing varying amounts of triglycerides, fatty
acids, glycolipids, sterols, and sphingophospholipids.” Leonard
(2017) defines lecithin relatively simply. Lecithin is a group of
different fatty substances found in plant and animal tissues that
are essential for the proper functioning of cells. Besides books
and journals, lecithin is defined in many encyclopedias avail-
able on the Internet as well. One of these is the Encyclopedia
Britannica (2020), where two definitions can be found. The first
one is the same as the scientific description by Szuhaj, with the
addition of the importance of lecithin in the cell structure and
cellular metabolism. The other is very similar to the ILPS defin-
ition, where they describe lecithin as a natural mixture containing
significant amounts of the following components: phosphat-
idylcholine (PC), cephalin (phosphatidylethanolamine, PE) and
phosphatidylinositol (PI). The various definitions all state that
lecithin is a natural mixture of neutral and polar lipids, where
the main compounds are PC, PE, phosphatidylserine (PS),
PI, phosphatidic acid (PA) and triglycerides. It also contains a
smaller proportion of glycolipids and different sugars. Lecithin
can be found in all living cells, where it plays an important bio-
logical role in the structure of cell membranes.
Lecithin
Lecithin was discovered in 1845 by the French chemist Theodore
Nicolas Gobley (1811–​1876), who was able to extract it from egg
yolk. It was named after the ancient Greek word lekithos, which
means egg yolk. This naming convention has been used since
1850. The full chemical formula of lecithin was unknown until
1874 (Shurtleff and Aoyagi, 2016). The first lecithin products
were obtained solely from egg yolk, but the process was expen-
sive. This driving force and the limited availability opened the
door for plant-​sourced lecithin products. Nowadays, the most
commonly used lecithin is soy lecithin (genetically modified
(GM) and non-​GM), followed by sunflower and rapeseed lecithin.
Lecithin has many forms, such as liquid lecithin, de-​ oiled
powdered lecithin, granules, compounded lecithin (lecithin on
a carrier), and various modified lecithins (enzyme modified –​
hydrolysed, acetylated, hydroxylated). Liquid lecithins from
vegetable sources are brownish, dense, oily products that are gen-
erally sold in a standardized form. In this way, it can be guaranteed
that the customer receives consistent product quality (Table 35.1).
Table 35.2 shows the typical chemical and physical parameters
of sunflower lecithin tested and set during standardization. In
addition, organoleptic and microbiological tests are usually
carried out before the product is released to the market.
Composition of Lecithin
The composition of lecithin is given in Figure 35.1. In Figure 35.1a,
the boxes are, from bottop to top: moisture, triacylglycerol (TAG),
complexed sugars, glycolypids, minor phospholipids, PI, PA,
PC and PE in absolute value (g/100 g lecithin). In Figure 35.1b,
they are C16:0, C18:0, C18:1, C18:2, C18:3 and other fatty acid
residues in relative value (g/100g fatty acid).
Phospholipids
As we can see from Table 35.3, lecithin contains several different
phospholipids. What is common in their molecular structure is
that they have a glycerol backbone, which has two fatty acid
residues attached in the first two positions, while a phosphoric
249
250 Elzbieta Kozakiewicz, Daniel Cossuta

TABLE 35.1
Limits for Food-​Grade Liquid Lecithin
Parameter Unit FAO/​WHO Codex Alimentarius European Union E322
Acetone insoluble % >60 >60
Hexane insoluble % -​ -​
Toluene insoluble % <0.3 <0.3
Moisture % -​ -​
Loss on drying % <2.0 <2.0
Acid value mg KOH/​kg <36 <35
Peroxide value meq/​kg <10 <10
Arsenic ppm -​ <3
Lead ppm <2 <2
Mercury ppm -​ <1

TABLE 35.2
Typical Parameters of Sunflower Lecithin after Standardization
Parameter Unit Specification
Acetone insoluble % min. 60
Hexane insoluble % max. 0.3
Acid values mg KOH/​g max. 35
Peroxide value meq O2/​kg max. 10
Moisture % max. 1
Gardner colour max. 12
Viscosity mPas max. 12,000

FIGURE 35.1 Composition of various lecithins, in terms of compounds (g/100 g lecithin) (a) and fatty acid residues (g/100 g fatty acid) (b).
Emulsions: Lecithin 251

TABLE 35.3
Average Composition of Plant and Animal Oil-​Free Phospholipid Extracts
% of Total Phospholipids
Phospholipids Soya Rapeseed Sunflower Corn Egg yolk Milk
Phosphatidylcholine 24 25 25 30 74 27
Phosphatidylethanolamine 22 22 11 3 19 36
Phosphatidylinositol 15 15 19 16 1
Phosphatidic acid 7 3 9
Sphyngomyelin 2 29
Lyso-​phospholipids 3 5 5 3
Other phospholipids 5 19 1 8

Source: Whitehurst (2004)

structure, such as temperature, presence of different ions and dis-

persion mechanisms.
The phospholipid composition is variable. The ratio of
different phospholipids depends on the source, so a phospholipid
composition is typical of the source from which the lecithin was
extracted.
Usually, the following phospholipids can be found in lecithin
in various ratios (Figure 35.4):

• phosphatidylcholine (PC)
• phosphatidylethanolamine (PE)
• phosphatidylserine (PS)
• phosphatidylinositol (PI)
• phosphatidic acid (PA)
• phosphatidylglycerol (PG)
• lyso-​phosphatidylcholine (LPC)
• lyso-​phosphatidylethanolamine (LPE)

FIGURE 35.2 Amphipathic molecule.
acid residue is bound to the third hydroxyl group of the glycerol
moiety. Phosphoric acid is furthermore linked to a polar nitrogen-​
containing organic molecule (such as ethanolamine, choline or
inositol), except in the case of PA (Wunderlich and Szarka, 2014).
Because of this molecular structure, phospholipids can dissolve
in both polar and non-​polar solvents. In fact, the fatty acid part of
the molecule is hydrophobic (lipophilic), whereas the rest of the
molecule has hydrophilic properties (Figure 35.2). These types
of molecules are called amphipathic molecules (IUPAC, 1972).
Lecithin is present in every living organism. It is the main
constituent of cell membranes. The function of phospholipids in
the formation of various cell membranes is compartmentaliza-
tion. Due to the amphipathic property, hydrophilic parts can form
hydrogen bonds with water molecules. However, due to their
properties, their non-​polar moiety does not participate in bond
formation and above the critical micelle concentration, can create
micelles. Phospholipids can create many types of structures.
They can create spherical micelles, reverse micelles or bilayers
(Figure 35.3). There are plenty of factors that can influence the
Glycolipids
Glycolipids are lipid molecules linked to a short saccharide
chain by glycosidic bonds. In the lipid part, two fatty acid
residues are attached to a glycerol moiety, whereas the sac-
charide moiety is usually D-​glucose, D-​galactose or inositol.
Like phospholipids, they are located in cell membranes. Due to
their size, their saccharide units protrude from the bilayer formed
by the phospholipids. The role of glycolipids is to stabilize cell
membranes. To do this, they use a portion of the sugar mol-
ecule diagonally across the bilayer, while still being able to form
hydrogen bonds with water molecules. Their other main task
is cell recognition, thus helping to induce an immune response
(Biológiai kislexikon, 2007). Glycolipids can be found in both
plant and animal organisms but are more abundant in photosyn-
thetic algae and plants.
The following glycolipids are present in lecithin:
monogalactosyldiacylglycerol (MGDG), digalactosyldiacylgly­
cerol (DGDG), cerebroside, steryl glycosides, esterified steryl
glycosides and sphingomyelin (SM)
Triglycerides
Triglycerides are compounds whose molecules include a glycerol
backbone esterified with three fatty acids. The fatty acids can be
252
FIGURE 35.3 Phospholipid bilayer and micelle.
FIGURE 35.4 1,2-​Diacyl-​sn-​glycero-​3-​phospholipids, nomenclature according to substituent ‘Z’.
(Whitehurst, 2004)
saturated or unsaturated (mono-​or poly-​unsaturated), and they
can differ in their chain length as well. The fatty acid residues
in lecithin are usually in the range of 6 to 26 carbon atoms (C6–​
C26) and typical of the different sources. Animal lecithins are
characterized by longer fatty acid residues, while plant-​derived
lecithins do not contain C20 or longer fatty acid residues. The
unsaturation of fatty acid residues in plant lecithins depends on
the botanical source and the climate of cultivation (Szuhaj, 1989)
(Table 35.4).
Other Components
The ingredients listed so far make up ~94% of the lecithin com-
position. In addition, lecithin contains water (up to 1%) and
sugars. Sugars constitute about 5% of lecithin. Scholfield et al.
(1952) investigated saccharides in soy lecithin and concluded
that saccharides in lecithin can be divided into two groups. One
group is the free sugars, which can be extracted with alcohol,
such as raffinose, sucrose and stachyose. The other group are
Elzbieta Kozakiewicz, Daniel Cossuta
the bound sugars. This group includes mannose, galactose and
arabinose.
As can be seen, phospholipids and glycolipids are the main
active agents in the lecithin, which are responsible for the func-
tionality and many applications of this ingredient in foods.
Applications of Lecithin
Emulsions –​  Types
An emulsion is a fluid colloidal system in which liquid droplets
and/​or liquid crystals are dispersed in a liquid (IUPAC, 1972).
The dispersed units often exceed the usual limits for colloids
in size. When droplets of oil are dispersed in water, an oil-​in-​
water emulsion is obtained. For example, milk and mayonnaise
are emulsions, and cream, ice cream and yoghurt are emulsified
systems. With droplets of water dispersed in oil, water-​in-​oil
emulsified systems are formed; butter and margarine are common
Emulsions: Lecithin 253

TABLE 35.4
Average Fatty Acid Profiles in Plant and Animal Lecithin Extracts
% within Total Fatty Acids
Fatty acid Soya Rapeseed Sunflower Corn Egg yolk
16:0 Palmitic acid 21 18 15 23 30
18:0 Stearic acid 4 1 3 1 16
18:1 Oleic acid 12 21 13 26 29
18:2 Linoleic acid 57 48 69 48 14
18:3 Linolenic acid 6 7 1 1
20:4 Arachidonic acid 5
20:6 Docosahexaenoic acid 3

Source: Whitehurst (2004)

FIGURE 35.5 An emulsion is a fluid colloidal system in which liquid droplets and/​or liquid crystals are dispersed in a liquid. The droplets often exceed the
usual limits for colloids in size (IUPAC, 1972). When the system is more complex, such as when oil droplets are dispersed in a gel, this is called an emulsified
system.

examples (Figure 35.5). Because of the difference in density of
oil and water, food emulsions are natively not stable and will sep-
arate into aqueous and oily layers.
All the following mechanisms (Figure 35.6) may lead to the
physical instability of an emulsion: sedimentation/​ creaming,
aggregation, coalescence and Ostwald ripening (Maphosa and
Jideani, 2018). Sedimentation is reversible, and the initial dis-
tribution of state can be restored by shaking or gentle stirring.
Coalescence results in a change of the drop size distribution of
the dispersed phase; in extreme cases, this leads to complete
phase separation, a so-​called ‘broken’ emulsion. These processes
can occur separately or together.
Emulsifiers are surface-​active substances with a hydrophilic
and a lipophilic moiety. The amphipathic properties (one part
of the phospholipid is lipophilic, another hydrophilic) of leci-
thin (Figure 35.7) make it extremely common in different food
applications.
Lecithin, placed at the phase boundary, reduces the surface
tension between the phases and stabilizes the system. Secondly,
by forming the emulsifier film on the surface of the droplets, it
stabilizes against coalescence.
To influence the stability of emulsions, additionally to the
emulsifiers, stabilizers are used. Vianna-​ Filho et al. (2013)
explained a few of the mechanisms for controlling the consist-
ency of an emulsion. Reducing the interfacial tension is one of
them, which can be performed using pectins or arabic gum. Other
rheology modifiers used in food are: starch, gelatin, xanthan gum,
guar gum and locust bean gum.
Barham et al. (2010) underline the need for creating the surface
when preparing an emulsion, foam or other colloidal dispersion.
The smaller the droplets, the more stable the emulsion. Shearing
power applied in the kitchen environment could be by shaking,
whisking, blending, agitating or stirring, whereas high-​pressure
homogenizers, ultrasonic probes, membranes or pin mixers are
used in industry to decrease the droplet size (Figure 35.8).
Other types of dispersed systems (Figure 35.9) consist of gas
bubbles (foams) or solid particles (colloid suspension) in liquid
phase (foams).
Proposed in the 1980s, and named in 1999, molecular cooking
(see Part III of this book) was intended to introduce an innova-
tive approach for preparing dishes or food products using tools
from laboratories, whereas ‘Note-​by-​Note cooking’ is another
application of molecular and physical gastronomy, in which the
ingredients have to be compounds rather than plant and animal
tissues (Burke et al., 2016). Molecular cuisine was popularized
by chefs such as Raymond Blanc, Pierre Gagnaire, Christian
254 Elzbieta Kozakiewicz, Daniel Cossuta

FIGURE 35.6 Different mechanisms that may lead to the physical instability of an emulsion.

FIGURE 35.7 Amphipathic properties of lecithin (phosphate group binds to the water; fatty acid chains align with the oil).

FIGURE 35.8 Energy needs to be supplied to break the fluid into smaller droplets. It is usually provided by a whisk, immersion blender, blender, ultrasonic
emulsifier or high-​pressure homogenizer.
Emulsions: Lecithin 255

FIGURE 35.9 Examples of other dispersed systems: liquid with trapped air bubbles (foams, aerated systems), and colloidal suspensions, with solid particles
dispersed in a liquid.

TABLE 35.5
Some Recipes by Adrià (2008)
Lecithin Dosing Mixing Type
Base of rose-​air: Mix the 3 ingredients in a bowl with a depth of 25 cm
500 g milk, 4 drops of rose essence, 2.5 g lecithin
Base of milk-​air with spices: Heat up the milk, mix the spices with a hand blender, leave for 24 h, sieve
500 g milk, 3.5 g of spices from Cordoba (Fatema Hal), 1.5 g of lecithin the mixture through a Superbag into a bowl which is 25 cm deep, add
the lecithin and mix with hand blender
Tuna-​air with olive oil: Mix the ingredients in the bowl which is 25 cm deep
400 g olive oil, 400 g water from the tuna can, 2.5 g of lecithin
Base of ponzu: 200 g of ponzu sauce, 0.4 g lecithin Mix the ingredients in a bowl which is 25 cm deep
Base for praline-​air: Chop the chocolate, put into a bowl, add praline & lecithin, heat the water
400 g chocolate, 50 g Piemonte hazelnut praline, 5 g lecithin, 1 kg of water to 90 °C add the chocolate & praline & lecithin, leave for 1 min, mix
with hand blender, keep temp. at 50 °C during air incorporation
Air: foam that is being created on the surface of the liquid during the mixing
or created due to the blending

Conticini, Emile Jung, Ferran Adrià and Heston Blumenthal, to

cite only the first. For example, after 1994, Adrià (Wareing, 2005) TABLE 35.6
was using innovative techniques like siphons to produce light, Nutritional Information of Example of Sunflower Lecithin
soft, foamy textures. In the early days of molecular cooking,
Nutritional information (per 100 g)
foams were generated by adding gelatine to a puree of white
Energy 790/​3302 kcal/​kJ
beans, and mixing it through using a siphon, and serving with sea
Total fat as triglycerides 70 g
urchins. Later, other savoury foams were made from beetroot,
Of which: Saturated 6g
coriander, or almond and confectionery products.
Monounsaturated 20 g
Performing a search on YouTube with the keywords ‘foam & Polyunsaturated 42 g
molecular cuisine’ or the combination ‘foam & molecular gas- Carbohydrate 3g
tronomy’ results in ca. 5460 videos with recipes and explanations Of which sugar 2g
from the amateurs of the technique as well as professional chefs. Fibre 0g
The amount of lecithin added varies between 0.3% and 0.8%. Protein <0.01 g
Foams made with whisk or hand immersion blenders (it is advised Salt <0.01 g
to handle these in the horizontal position to try to create as many
bubbles as possible) may result in different stabilities as well as in
other ‘foams’, ‘espumas’ or ‘airs’ made with a siphon (using N2O Labelling of Lecithin
cartridges to create a foam that is coarser in texture). The larger Maximum levels of lecithins (E 322) have been defined in Annex
number of small air bubbles increases the viscosity and provides II to Regulation (EC) No 1333/​2008 on food additives (European
creaminess in the mouthfeel (see the recipe by Toutain in Part III Parliament, 1995). In this document, these levels are named max-
of this book). imum permitted levels (MPLs) (JECFA, 1993). Currently, lecithin
Some of the recipes mentioned in the book A Day at elBulli (E 322) is an authorized food additive in the EU at quantum satis
(Adrià, 2008) are mentioned in Table 35.5. (QS) in most foods apart from fats and oils essentially free from
256 Elzbieta Kozakiewicz, Daniel Cossuta

water, infant and follow-​on formulae, processed cereal-​based ILPS. 2020. http://​ilps.org/​
foods and baby foods for infants and young children, and other IUPAC. 1972. Manual of symbols and terminology for physico-
foods for young children (EFSA, 2017) (Table 35.6). Lecithin (E chemical quantities and units, Appendix II: Definitions, ter-
minology and symbols in colloid and surface chemistry, Pure
322) is included in the Group I of food additives authorized at QS
and Applied Chemistry, 31, 577–​612.
for such food stuffs as cocoa and chocolate products, bread and JECFA. 1993. Lecithin. www.fao.org/​fileadmin/​user_​upload/​jecfa_​
fresh pasta. additives/​docs/​monograph4/​additive-​250-​m4.pdf
Lecithin is also on the FDA GRAS list (generally recognized Leonard J. 2017. MedicalNewsToday. www.medicalnewstoday.com/​
as safe). articles/​319260
Maphosa Y, Jideani VA. 2018. Factors affecting the stability of
emulsions stabilised by biopolymers, science and technology
REFERENCES behind nanoemulsions, Selcan Karakuş, IntechOpen, DOI:
Adrià F. 2008. Een dag bij elBulli, Bewonder de ideen, methodes en 10.5772/​intechopen.75308. Available from: www.intechopen.
creativiteit van Ferran Adria, Utrecht, The Netherlands. com/​books/​science-​and-​technology-​behind-​nanoemulsions/​
Barham P, Skibsted LH, Bredie WLP, Bom Frøst M, Møller M, factors-​affecting-​the-​stability-​of-​emulsions-​stabilised-​by-​
Risbo J, Snitkjær P, Mørch Mortensen L. 2010. Molecular biopolymers
gastronomy: A new emerging scientific discipline, Chemical Scholfield CR, Dutton HJ, Dimler RJ. 1952. Carbohydrate
Reviews, 110(4), 2313–​2365. constituents of soybean ‘lecithin’, Journal of the American Oil
Biológiai kislexikon. 2007. Typotex Kiadó. Chemists’ Society, 29(7), 293–​298.
Burke R, This H, Kelly AL. 2016. Molecular Gastronomy, Reference Shurtleff W, Aoyagi A. 2016. History of Lecithin and Phospholipids
Module in Food Science, Elsevier. (1850 to 2016): Extensively Annotated Bibliography and
EFSA ANS Panel (EFSA Panel on Food Additives and Nutrient Source Book, Soyinfo Center.
Sources added to Food : Mortensen A, Aguilar F, Crebelli Szuhaj BF (ed.). 1989. Lecithins: Sources, Manufacture & Uses,
R, Di Domenico A, Frutos MJ, Galtier P, Gott D, Gundert American Chemists Society.
M, Remy U, Lambré C, Leblanc J-​C, Lindtner O, Moldeus Vianna-​Filho RP, Oliveira Petkowicz CL, Meira Silveira JL. 2013.
P, Mosesso P, Oskarsson A, Parent-​Massin D, Stankovic I, Rheological characterization of O/​W emulsions incorporated
Waalkens-​Berendsen I, Woutersen RA, Wright M, Younes M, with neutral and charged polysaccharides, Carbohydrate
Brimer L, Altieri A, Christodoulidou A, Lodi F, Dusemund Polymers, 93, 1.
B). 2017. Scientific opinion on the re-​evaluation of lecithins Wareing M. 2005. The Cook’s Book: Recipes and Step-​ by-​
Step
(E 322) as a food additive, EFSA Journal, 15(4), 4742. Techniques from Top Chefs, DK Publishing.
Encyclopedia Britannica. 2020. Lecithin. www.britannica.com/​ Whitehurst RJ. 2004. Emulsifiers in Food Technology, Blackwell
science/​lecithin Publishing Ltd.
European Parliament. 1995. Directive. https://​ eur-​
lex.europa.eu/​ Wunderlich L, Szarka A. 2014. A biokémia alapjai, Typotex Kiadó.
legal-​ c ontent/​ E N/​ T XT/​ P DF/​ ? uri=CELEX:31995L0002
&from=EN
Emulsions: Emulsions and Surfactants in the Kitchen

Hervé This vo Kientza1,2

1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France

In the kitchen, emulsified systems and emulsions are frequently
encountered, because most traditional food ingredients (as
opposed to pure compounds used in note by note cooking; see Part
III of this book), such as animal and plant tissues, contain both
water and fat (Ciqual, 2018). When cooks are preparing dishes,
these two non-​miscible materials are often processed together,
and, because energy is given (mechanical, thermal or chem-
ical), they are frequently arranged in the form of fat “particles”
dispersed in an aqueous solution; when the fat is liquid, or even
when it is present as crystals, these particles are indeed droplets,
and the systems are called “emulsions”, from a name created in
1560 by Ambroise Paré, surgeon of many French kings, who was
considering drugs white and viscous as milk, in which he knew
that there was fat and water; the word is based on the Latine
emulgere, which means “to draw milk” (Paré, 1560).
More recently, the definition of an emulsion was standardized
by the International Union for Pure and Applied Chemistry
(IUPAC):
A fluid colloidal system in which liquid droplets and/​or liquid
crystals are dispersed in a liquid. The droplets often exceed
the usual limits for colloids in size. An emulsion is denoted by
the symbol O/​W if the continuous phase is an aqueous solu-
tion and by W/​O if the continuous phase is an organic liquid
(an “oil”). More complicated emulsions such as O/​W/​O (i.e.
oil droplets contained within aqueous droplets dispersed in a
continuous oil phase) are also possible.
(IUPAC, 1972)
Oil Does Not Mix with Water; Surface Tension
Let us start the discussion of emulsions and surfactants by the
simple experiment of adding water and oil in a vessel (Figure 36.1).
Oil poured over water from some height first sinks into the
water before coming up to the surface at a velocity that can be
calculated using the first principles of dynamics (see the chapter
on decanting, in the Application part of this book). Of course,
this phenomenon of “creaming” is due to the fact that the density
of food oil is lower than that of water (Lide, 2005). But why do
oil drops coalesce? This can be calculated as will be shown now,
using free energy or free enthalpy, for transformations at constant
pressure or at constant volume, respectively.
Let’s assume that the length of the cubic vessel is equal to 1
(arbitrary units, a.u.). Then, the surface where oil and water are
in contact has an area A equal to 1 × 1 = 1. In our world, matter,
energy, electric charge and some other characteristics are always
conserved; i.e., they are constant during the changes of the world.
For our question, let us consider energy: any spontaneous evolu-
tion should take place as:
ΔG = Gfinal − Ginitial < 0  (36.1)

In this definition, we observe that, contrary to what is frequently

said or written, emulsions do not call for immiscible liquids, and
this is fair, because physical chemistry is today discovering W/​W
emulsions, with only aqueous solutions (Bach et al., 2013).
Here, we want to discuss the question of emulsions from both
physical and chemical points of view. What is given is not the
result of personal research but rather, a collection of useful infor- FIGURE 36.1 A system made of two layers of oil and water has less energy
mation about some common culinary systems. when oil is over water.

257
258 Hervé This vo Kientza

where the Gibbs function G is equal to:

G = H –​ T. S. (36.2)

where H is the enthalpy, T the absolute temperature and S the

entropy (McQuarrie, 2011). For example, let us imagine that oil
made a layer at the bottom of the vessel, being covered with water.
If to is the thickness of the layer of oil and tw the thickness of the
layer of water, then the potential energy would be (we decide that
the bottom of the vessel is at zero potential energy and we con- FIGURE 36.2 Hydrogen bonds keeping water from flowing even when its
sider the mass of each material at its centre of gravity): level is over the edge of the vessel, because the result of all hydrogen bonds
on any water molecule at the surface is toward the bulk of the liquid.

tO  t 
Wi = WO + WW = mO g + mW g  tO + W 
2  2 This can be shown experimentally for air and water, as when
water is added to a glass up to more than the maximum level of
 t  t 
= g  mO O + mW  tO + W   (36.3) the glass (Figure 36.2).
 2  2  If water can remain without flowing over the level of the
upper edge of the glass, it means that forces are keeping it
where WO and WW stand for the potential energy of oil and of there in spite of the potential energy that would make it
water, respectively, and mO and mW are the mass of oil and the flow. In this particular case, hydrogen bonds between water
mass of water, respectively. Let us compare this energy with the molecules in the bulk of the liquid keep the water molecules
one with oil on water: from moving down. Let us analyse further. If forces “group”
water molecules, it means that water molecules do not “like”
  t   t  (in terms of energy) to be at a surface. In other words, free
W f = g  mO  tW + O  + mW  W   (36.4)
 2   2  enthalpy depends on the area of the interface. The simplest

possibility would be:
The entropy being the same in both cases, the free enthalpy
G ∞ A. (36.9)
variation is:
Why such a proportionality? Because it is by the interface that
  t  t 
∆G = G f − Gi = g  mO  tW + O  + mw  w  hydrogen bonds can act; the bigger the area, the more bonds there
  2   2 are. However, instead of writing:
t  t 
− m0 O − mW  tO + W   (36.5) ΔG = γ ΔA
2  2 

we write more correctly:

Or:
∂G (36.10)
γ=
∆G = g ( mO tW − mwtO ) (36.6) ∂A

But of course, there is a relationship between the thickness and
the mass:
mi = ρi ti A (36.7)
where i stands for W or O. Thus:
∆G = g ( AtW tOρO − AtO tW ρW )
(36.8)
Here, the sign of ΔG is given by ρO − ρW , and it is negative. Thus,
the evolution from a layer of oil under of a layer of water is less
stable than for a layer of oil over a layer of water.
This result explains why oil comes up, but it does not explain
why droplets may coalesce. Here, we need to remember that
interfaces separating two media are places where molecular
forces are different than in the inside of the two separated media.
The parameter γ is called surface tension (De Gennes et al.,
2004).
Here, we started from the example of the air/​water interface,
but there are other values depending on the two phases that are in
contact. Let us come back to oil floating over water, and imagine
that the oil would be divided into two parts, which are in contact
with water (interfacial energy γ) (Figure 36.3).
The area of the oil increases when it is divided, because of the
surface created by a cut. How much?
Let us assume an initial sample of oil of volume V, whose side
would be n c:
V = (n c)³ (36.11)
The area of its surface is:
Ai = 6 (n c)² (36.12)
Emulsions: Emulsions and Surfactants
Let us divide this sample into small cubes of side c. The volume
of one elementary cube is c³, and their number is n³. The area of
their individual surface is:
a = 6 c²
So that the total area of all small cubes is:
A2 = n³ 6 c²
The area has changed by:
A2 6 n3c 2
= =n
A1 6 n2c 2
This means that one has to give energy to the system in order to
disperse oil in water.
Surfactants
Having observed that emulsions need energy to be produced, let
us now see how the required mechanical energy can be reduced.
As a start, let us examine another experiment, often used in popu-
larization environments: starting from the previous state, with
water over the level of the glass, let us dip the tip of a clean needle
FIGURE 36.3 The surface of oil is increased when oil is divided.
FIGURE 36.4 A triacylglycerol (tristearin) with a glycerol residue esterified by three residues of fatty acid.
259
into the water: the needle pierces the surface without any par-
ticular effect.
Now, let us first dip the needle in liquid soap and repeat the
experiment of piercing the surface: the water above the upper level
(36.13) of the glass flows. We shall reach the conclusion that the “surface
tension” of water was reduced, as this new surface energy is not
enough to overcome the potential energy of water. Such a modifi-
cation is helpful when one is dispersing oil in water; if the surface
(36.14) tension is reduced, this means that the amount of energy neces-
sary to disperse oil in water is reduced.
Why is soap “surface active”? Soap has been produced for
more than two millennia by heating fat and wood ashes (Konkol
and Rasmussen, 2015). The French chemist Michel Eugène
(36.15) Chevreul explained the process chemically, showing in 1823
that potash (KOH, from ashes) could produce “saponification”,
described by the chemical equation (Chevreul, 1823):
R-​COO-​R′ + OH–​ → RCOO–​ + ROH
Oil, for example, is mainly (often more than 98%) a mixture of
triglycerides (Pioch et al., 1998) (Figure 36.4).
A glycerol residue is linked through ester links to residues
R1, R2 and R3, which are alkyl chains that may be saturated or
unsaturated. In other words, saponification generates glycerol, on
the one hand, and carboxylate ions, on the other hand. With such
compounds, there is a negatively charged end (the “head”) and a
hydrophobic end (“tail”). If one puts such molecules in a glass
with water and oil, the head and tail molecules lower their energy
by being more abundant at the interface, so that the hydrophilic
heads are in water and the hydrophobic tails are in contact with
the oil (Figure 36.5).
This reduces the oil–​water surface tension because the inter-
face is changed, and this is why soaps are called “surface active
agents” or “surfactants”. When more surfactants are added, some
can be dissolved, in oil or in water (Figure 36.6). How much?
One way to characterize the distribution of a compound in
a biphasic system is to use the partition coefficient, defined as
follows (Leo et al., 1971). A compound distributes so that there
260 Hervé This vo Kientza

TABLE 36.1
Values of log Kow for Various Organic Compounds
Compound log Kow
CH3-​OH −6.38
CH3-​CH2-​OH −5.54
CH3-​CH2-​CH2-​OH −4.7
CH3-​CH2-​CH2-​CH2-​OH −3.86
CH3-​CH2-​CH2-​CH2-​CH2-​OH −3.0
CH3-​CH2-​CH2-​CH3 4.58
Phi-​OH −3.52
FIGURE 36.5 With a small quantity of surfactants such as alkylcarboxylate
added to an oil-​over-​water biphasic system, the energy is lower when these
molecules come to the interface.
surfactants are ranked between 0 and 14 (De Gennes et al., 2004).
We consider now the way this is derived.
Let us move a surfactant molecule from water toward oil. If the
work needed is W, the proportion of surfactant molecules can be
expressed using the Maxwell–​Boltzmann distribution:

 W 
k = exp  − (36.17)
 kBT 

Let us calculate W, assuming an energy EH for transferring a

hydrophilic group and the energy EL for transferring a lipophilic
group. Let nH be the number of hydrophilic groups, and nL the
FIGURE 36.6 The distribution of a compound between oil and water. The number of lipophilic groups. Then, in this description:
ratio of the concentrations is the partition coefficient.
W = nH EH − nL EL (36.18)
is a concentration cO in oil and a concentration cW in water. The
And then:
partition coefficient k is equal to:
 n E − nL E L 
c k = exp  − H H  (36.19)
k= O (36.16)  kBT
cW

Taking the logarithm:
Because “oils” are not well defined (for milk fat, the number
of possible fatty acid residues on each of the three positions of
log10 ( k ) = − pH E H + pL EL
the glycerol residue is estimated to be 400), the partition coef-
ficient is often expressed for the distribution between octanol or
n-​octane in water (Bouteille et al., 2013). The coefficients pH and pL can be measured.
However, the partition coefficient is difficult to use, because In 1946, Griffin proposed defining HLB from a relationship of
some compounds are very hydrophilic (ions, for example, and this kind, so that the scale would be between 0 and 14:
also saccharides, and more generally polyols, being able to estab-
lish a lot of hydrogen bonds with water molecules) and, on the HLB = 7 + a ( − pH E H + p L EL ) = 7 + ( −qH E H + q L EL ) (36.20)
other hand, other compounds (compounds with long hydrophobic
chains) are very hydrophobic. This is why it is preferable to use the Here, the constant a is chosen so that the scale is similar to that for
decimal logarithm of the partition coefficient, i.e., the “log P” for pH, i.e., between 0 and 14. Table 36.2 gives the q parameters for
the distribution coefficient of compounds between n-​octane and some chemical groups.
water, and the “log KOW” for an octanol/​water system (Table 36.1). Why do we use such a scale? This is because surfactants are
now more easily chosen for particular applications (Table 36.3).

HLB Values
In the case of surfactants, such scales as log P are not used in the
industry. Instead, engineers have another scale, more like the pH
scale: the HLB, or “Hydrophilic Lipophilic Balance”, in which
Making Colloidal Systems
Concerning surfactants, the simple distribution that was
considered between oil and water is simplistic, as the surfactant
Emulsions: Emulsions and Surfactants 261

TABLE 36.2 Why Are Micelles Depicted as Spherical?

Some Parameters That Can Be Used for the Calculation of HLB In the previous paragraph, it was observed that the heads of sur-
(from De Gennes et al., 2004)
factant are sometimes much smaller than the tails. Why are they
Hydrophilic groups Hydrophobic groups depicted as being so big, and why are micelles spherical? The
CO2Na 19.1 CH2-​ 0.47 simple representation with bonds is not enough; instead, the
SO3Na 11.0 “effect” on other molecules has to be taken into account.
N(CH3)3Cl 9.4 The heads are “big” because they are often electrically charged
​O-​ 1.3 or have non-​binding doublets (on oxygen atoms of phosphate
OH 1.9 groups, for example), leading to electric repulsion, whereas tails
avoid contact only through steric effects, even being attracted by
weak van der Waals forces.
One can calculate the structure of micelles, assuming that the
TABLE 36.3 “heads” of surfactant molecules have a projected area (“cross
How to Choose a Particular Surfactant for a Particular Application section”) A and the “tails” a volume V = nq.vq (nq stands for the
number of groups in the hydrophobic chain, and vq is the volume
HLB Use of one group); we assume micelles of radius R.
1.5–​3 antifoaming agents Let N be the average number of molecules of surfactant per
3–​6 emulsions W/​O micelle. The total volume of the micelle is N.V. This volume can
7–​9 foaming agents, wetting agents be put in relation with R:
8–​18 emulsions O/​W
13–​15 soaps 4 3
15–​20 solubilization of organic compounds π R = NV ≈ N .nq .vq (36.21)
3

The surface, on the other hand, has an area:

4 π R 2 = NA (36.22)

From these two equations, the radius can be calculated:

3nq .vq
R= (36.23)
An

And the average number of surfactant molecules per micelle:

FIGURE 36.7 Surfactants grouped in a micelle.

molecules can self-​associate when their concentration is high
enough.
In particular, they can make “micelles” (from Latine mica,
part), with the polar heads being in water and the hydrophobic
tails being far from water. This is frequently depicted as in
Figure 36.7.
There are other possibilities, such as liposomes or bilayers,
but one should remember that all such pictures have to be
interpreted, because representing surfactant molecules with
such heads and tails is simplistic; in general, the “size” of
the head is only a few chemical bonds long, whereas the tails
have a length of 10 to 20 covalent bonds. Moreover, at room
temperature, molecules of liquid systems move; in particular,
surfactant molecules exchange between the liquid environ-
ment and the structure that they form. At any time, a surfac-
tant molecule can leave the structure in which it participates
(e.g., a micelle), being replaced by a water molecule or another
sufactant molecule, or by nothing. The picture is based on the
fact that the residence time of water molecules is very low
(Yamamoto et al., 2014).
36πnq2 vq2
N= (36.24)
A3
Experimentally, the order of magnitude of N is measured to be 100.
Micellar Concentration
Micelles form only when their concentration is higher than a
particular concentration, which is called the “critical micellar
concentration” (cmc). Can we calculate it? There are two
principles involved here (the first and the second principles of
thermodynamics), and we have to use the Gibbs function, or free
enthalpy, in order to take these two principles into account. As
concentrations are involved, we have to make a link with energy.
We start from the concentration c, on one hand, and energy, G,
on the other. The definition of chemical potential μ expresses the
relationship between the two:
∂G
µi = (36.25)
∂ni 
262 Hervé This vo Kientza

where i is a particular chemical species, and ni is the number

of moles of this species. Let us consider a surfactant molecule,
which can either be isolated (μs) or associated with others (μa).
For a charged surfactant:

µ s = µ 0s + 2 kBT ln (c.V ) (36.26)

The expression is given for a molecule (kB) and not for a mole, as
is more common. The factor 2 is because the surfactant is ionic,
made from one ion and one counterion.
For the same molecule, in a micelle:

kBT  V 
µ a = µ 0a + ln  c.  (36.27)
N  N 

Here the entropic term is divided by N, because micelles behave

as unique structures with three degrees of freedom. As N is large,
this term can be dropped, so that the equilibrium between isolated
surfactants and associated surfactants can be written:

2 kBT ln (cmc V ) = µ 0a − µ 0s (36.28)



The second term can be written −nq Uq if we use the results

obtained before for HLB values. Then:

1  nq.Uq 
cmc = exp  −  (36.29)
V  2 kBT 


FIGURE 36.8 When oil is whipped in an aqueous solution of a surfactant,

Surfactants in Emulsions
If surfactant molecules can associate in micelles, they can also
migrate toward a hydrophobic–​water interface, lower the surface
tension, and stabilize oil in water emulsions (O/​W). This is how
clothes are washed in water, using soap, when organic matter
such as fat is adhering to the fibres: soap molecules diffuse to
the fat/​water interface and cause it to curve, forming fat droplets
entirely covered by soap molecules.
The same occurs when making a mayonnaise or a “geoffroy”
(an O/​W emulsion obtained by whipping oil in an aqueous solu-
tion of proteins) (This, 2009), or a “kientzheim sauce” (the same
as a mayonnaise, but with melted –​possibly brown –​butter instead
of oil), for example. When the oil fraction remains below the close
packing coefficient, the oil droplets are spherical, but, when this
ratio increases, the droplets deform (Figure 36.8a and b).
Of course, this description is classic, and nowadays new
kinds of emulsions are being studied. In the kitchen, systems
such as mayonnaise (made from egg yolk, vinegar, oil, salt
and pepper) are called emulsions (IUPAC, 2018), but for some
systems, such as when water is added to Pastis or Ouzo, i.e.,
solutions of anethole in water–​ethanol systems (Scholten et al.,
2008), very different properties can be produced. Other culinary
emulsions will be recognized as Ramsden emulsions, sometimes
called Pickering emulsions, i.e., emulsions for which the oil/​
water interface is covered by solid particles (Rayner et al., 2014;
an emulsion is formed. When the oil fraction is less than the close packing
coefficient, the oil droplets can remain spherical (a) (the largest oil droplets
have a diameter of about 0.1 mm). But when the oil fraction increases, the
oil droplets deform (b). In this second picture, the largest structures have a
diameter of 0.01 mm.
Ramsden, 1904). This is also discussed in the chapter by Fameau
in this book.
Nowadays, one important kind of emulsions is
“microemulsions”, because they are much more stable than the
usual type. There is some confusion about them because they are
poorly named.
In culinary emulsions, such as mayonnaise made with a
fork or a kitchen whisk, the diameter of oil droplets is between
1 and 100 μm, in the micrometre range, so that the term
“microemulsions” could have been given to them. However, this
name is used, rather, to refer to thermodynamically stable iso-
tropic liquids formed by mixing oil, water and surfactants together
(Jonsson et al., 1998; Fanun, 2008). Such mixtures can form
many different systems depending on the composition and on the
physical conditions, such as temperature. In particular, they may
contain a certain number of phases (1, 2, 3 or more) that are in
equilibrium with each other. These phases may or may not be con-
tinuous, and, as shown for micelles, the structures can be diverse:
one-​dimensional (spheroids or cylinders), two-​ dimensional
Emulsions: Emulsions and Surfactants
(lamellar structures) or even intricate three-​dimensional (sponge-​
like) forms (Roux, 1995). On the other hand, “nanoemulsions”
can be considered as conventional emulsions made of very small
particles (Tadros et al., 2004; Mason et al., 2006).
One reason for the confusion between these two kinds of
systems is that “micro” refers to sizes of 10–​6 m, whereas
“nano” applies to sizes of about 10–​9 m; according to this
rule, nanoemulsions would be made of particles that are
smaller than those in microemulsions. However, this termino-
logical rule does not apply: the first article appearing with the
word “microemulsion” was published in 1961 (Schulman and
Montagne, 1961), whereas the word “nanoemulsion” appeared
in 1971 in relation to the development of nanotechnology. Of
course, there are other debates about the upper limit for “nano”
particles: 500 nm (Anton et al., 2008), 200 nm (Huang et al.,
2010) or even 100 nm (Rao and McClements, 2012). Indeed,
the main change in physicochemical properties when the size of
the dispersed structures is reduced occurs when their potential
energy (m g h) is of the same order of magnitude as the thermal
energy of the dispersing medium (kB T).
Concerning Ramsden emulsions, they are stabilized by very
small particles. They have attracted much research in the past
decade due to their properties, such as high stability with respect
to coalescence and Ostwald ripening (see the chapter on this sub-
ject in this book). In particular, they can be made with biomass-​
based particles in the context of foods and topical creams (Rayner
et al., 2014), such as starch granules or egg yolk granules. The
particle stabilization of oil droplets is possible due to partial
dual wettability of particles at the oil–​water interface. Native
starch is not intrinsically hydrophobic; however, its hydropho-
bicity can be increased by chemical modification with octenyl
succinic anhydride. Heat can be applied to induce a partial gel-
atinization of the starch granules, forming a cohesive layer at
the oil–​water interface and increasing barrier properties. Egg
granules, on the other hand, are hydrophobic and, at low ionic
strength (<0.3 M NaCl), are insoluble with a compact structure;
size decrease with increasing concentration can be controlled by
the granule-​to-​oil ratio.
Conclusions
It is interesting to have a historical perspective on the theory of
food emulsions. As late as the 1950s, it was said, even in sci-
entific circles, that emulsions such as mayonnaise had to be
prepared with an iron whisk and a copper bowl because of a
“battery effect”. As late as the 1980s, mayonnaise sauces were
said to be stabilized by phospholipids (rather than proteins). Until
recently, Ramsden emulsions were absent from courses on the
physical chemistry of food (there is nothing about them in the
well-​known Food Chemistry by Belitz et al. (2009)). What will
be the next steps?
REFERENCES
Anton N, Benoit JP, Saulnier P. 2008. Design and production of
nanoparticles formulated from nano-​emulsion templates –​a
review, Journal of Controlled Release, 128, 185–​199.
263
Bach TN, Taco N, Lazhar B. 2013. Stabilization of water-​in-​water
emulsions by addition of protein particles, Langmuir, 29, 34,
10658–​10664.
Belitz HD, Grosch W, Schieberle P. 2009. Food Chemistry, Springer
Verlag, Berlin Heidelberg, 4th ed.
Bouteille R, Perez J, Khifer F, Jouan-​ Rimbaud-​ Bouveresse D,
Lecanu B, This H. 2013. Influence of the colloidal structure
of dairy gels on milk fat fusion behavior: quantification of the
liquid fat content by in situ quantitative proton nuclear mag-
netic resonance spectroscopy (isq 1H NMR), Journal of Food
Science, 78, 4, E535–​E541.
Chevreul ME. 1823. Recherches chimiques sur les corps gras
d’origine animale, FG Levrault, Paris, In-​ 8°, XVI-​ 484 p.
et pl.
Ciqual. 2018. https://​ciqual.anses.fr/​
De Gennes PG, Brochard-​Wyart F, Quere D. 2004. Capillarity and
Wetting Phenomena, Springer Verlag, New York.
Fanun M. 2008. Microemulsions: Properties and Applications, CRC
Press, Boca Raton, FL.
Huang QR, Yu HL, Ru QM. 2010. Bioavailability and delivery of
nutraceuticals using nanotechnology, Journal of Food Science,
75, R50–​R57.
IUPAC. 1972. Manual of Symbols and Terminology for
Physicochemical Quantities and Units, Appendix II:
Definitions, Terminology and Symbols in Colloid and Surface
Chemistry, 31, 577, 606.
IUPAC. 2018. Gold Book, https://​goldbook.iupac.org/​html/​E/​
E02065.html
Jonsson B, Lindman B, Holmberg K, Kronberg B. 1998. Surfactants
& Polymers in Aqueous Solutions, John Wiley & Sons,
Chichester, UK.
Konkol KL, Rasmussen SC.2015. An Ancient Cleanser: Soap
Production and Use in Antiquity. In Chemical Technology in
Antiquity, ed. Rasmussen SC, American Chemical Society, 9,
245–​266.
Leo A, Hansch C, Elkins D. 1971. Partition coefficients and their
uses. Chemical Reviews, 71, 6, 525.
Lide DE (ed). 2005. CRC Handbook of Chemistry and Physics,
Internet Version 2005, CRC Press, Boca Raton, FL.
McQuarrie D. 2011. Physical Chemistry: A Molecular Approach,
Viva Books.
Mason TG, Wilking JN, Meleson K, Chang CB, Graves SM. 2006.
Nanoemulsions: Formation, structure, and physical properties.
Journal of Physics: Condensed Matter, 18, R635–​R666.
Paré A. 1560. https://​archive.org/​stream/​b22008020_​0003/​b2200
8020_​0003_​djvu.txt
Pioch D, Larguèze C, Graille J, Ajana H, Rouviere J. 1998. Towards
an efficient membrane based vegetable oils refining, Industrial
Crops and Products, 7, 2–​3, 83–​89.
Ramsden W. 1904. Separation of solids in the surface-​ layers
of solutions and “suspensions” (observations on surface-​
membranes, bubbles, emulsions, and mechanical coagula-
tion). Preliminary account, Proceedings of the Royal Society
of London, 72, 156–​164.
Rao JJ, McClements DJ. 2012. Lemon oil solubilization in mixed
surfactant solutions: Rationalizing microemulsion &
nanoemulsion formation, Food Hydrocolloids, 26, 1, 268–​276.
Rayner R, Marku D, Eriksson M, Sjöö M, Dejmek P, Wahlgren
M.2014. Biomass-​ based particles for the formulation of
Pickering type emulsions in food and topical applications,
Colloids and Surfaces A: Physicochemical and Engineering
Aspects, 458, 48–​62.
Roux D. 1995. Sponge phases: An example of random surfaces,
Physica A, 213, 168–​172.
264 Hervé This vo Kientza

Scholten E, van der Linden E, This H. 2008. The life of an anise-​ Tadros T, Izquierdo R, Esquena J, Solans S. 2004. Formation and
flavored alcoholic beverage: Does its stability cloud or confirm stability of nano-​emulsions, Advances in Colloid Interface
theory? Langmuir, 24, 5, 1701–​1706. Science, 108–​109, 303–​318.
Schulman JH, Montagne JB. 1961. Formation of microemulsions This H. 2009. The Science of the Oven, Columbia University Press.
by amino alkyl alcohols, Annals of the New York Academy of Yamamoto E, Akimoto T, Yasui M, Yasuoka K. 2014. Origin of
Sciences, 92, 366. subdiffusion of water molecules on cell membrane surfaces,
Scientific Reports, 4, 4720.
Essential Oils
Eric Angelini1 and Laure Dziuba2
1 V. MANE FILS 620, route de Grasse, 06620, Le Bar-​sur-​Loup, France
2 Consultant chromatographic analysis of fragrances and related compounds, 1955 chemin des vergers,
06620, Le Bar-​sur-​Loup, France
Essential oils (EO) are sometimes used in kitchens. They are most
often produced by techniques that appeared longer ago than two
millennia BC but have been improved continuously. Information
about the production of EO can be useful at a time when chefs are
producing their own extracts (for safety questions about the use
of these products, see the chapter “How to Safely Use Essential
Oils”).
Let us begin with a question: why do compounds have a smell?
Here is a first fact: some natural extracts are fragrant. If we per-
ceive a smell, it is because “something” has:
• reached our nose: that “something” is in the air we
inhale and it has the property of volatility (Proust, 2006;
Ohloff et al., 2012);
• come through our nasal mucous membrane, which
contains at least 98% water: that “something” has an
affinity, even tiny, for water; it has a “coefficient of
solubility”.
A second observation is important: there is no link between the
quantity of “something” and the perceived intensity. A very small
amount of a substance may be perceived with high intensity,
while 10 times, 100 times, 106 times or more of another substance
would be necessary to obtain a similar intensity impact. This par-
ameter relates to neurobiology and is described by the perception
and recognition thresholds for different odours.
Of course, the “something” is a mixture of various compounds:
for odours, the compounds are organic, i.e., made of carbon,
hydrogen, oxygen, nitrogen and sulphur atoms. And molecules
being very small objects, it is necessary to remember the
huge value of the Avogadro number, which defines a mole as
containing 6.02 × 1023 entities (either molecules or atoms or any-
thing else). For example, one mole of limonene weighs 136 g and
contains 0.6 million million million million limonene molecules.
However, for other compounds, the olfactory effect is so strong
that the quantity of molecules needed is of the order of only
1–​10 ng/​L (for geosmine, for example) (Drouillard et al., 2005;
Ohloff et al., 2012; ASTM, 1973).
Volatility and solubility are important parameters for the
description of the interactions of different molecules facing
various phases/​ media. While talking about the chemistry of
odorant compounds, two phases only are used to describe the
physical phase of a molecule: the vapour (gas) phase and the
condensed phase, which can be either liquid or solid. As a general
rule, a molecule is always confronted with at least one different/​
foreign phase; for example, the surrounding air. As for the phases
that describe the different media/​solvents/​supports, they may be
gaseous, liquid or solid. The mass, the geometry and the chemical
functions of a molecule influence its volatility and its solubility.
Volatility, first, describes the capacity for a given molecule to
extract itself from the condensed phase and go into the vapour
phase, taking energy from the initial environment. The equilib-
rium that is reached can be described by the partition coefficient K,
which is a function of the temperature and the pressure. Volatility
may be described by vapour pressure at a given temperature.
Vapour pressure of a pure substance is:
• the pressure exerted by that pure substance against the
external pressure (in general, the external pressure is the
atmospheric pressure); it is the pressure of a vapour in
equilibrium with its condensed phase.
• the measure of the ability to evaporate: the higher the
vapour pressure, the higher this trend and the more
volatile the molecule. A substance having a vapour
pressure equal to zero would not evaporate. When the
vapour pressure (of a substance) reaches the external
pressure, boiling is observed.
Solubility, secondly, describes the capacity for a given molecule
or a chemical structure, called the solute, to form links with
molecules of a liquid medium, called the solvent, resulting in
a homogeneous phase. This is also characterized by a partition
coefficient K.
When a fraction only of the overall quantity of molecules
interacts, one speaks of solubility. For example, when salt is
added to water, saturation occurs for 357 g/​L at 0 °C and 391 g/​L
at 100 °C (Lide, 2005). When the phase obtained is homoge-
neous whatever the quantities of solute and solvent, one speaks
of miscibility: for example, ethanol and water. If the solvent is
in huge quantities and/​or of greater dimensions relative to the
265
266
solute, one speaks of absorption. Examples include the extrac-
tion of raw materials with a volatile solvent, the chromatographic
phase in gas chromatography to separate volatile constituents of
a mixture, etc.
The Odour of Plants Is a Result of Their
Metabolism
Many of the odorant compounds used for flavourings, and all
those from EOs, come from plants (Table 37.1). Indeed, most
plants capture solar energy via photosynthesis; the mechanisms
are coded in a modus operandi, i.e. the genetic code, and they are
implemented by a workforce (enzymes and related compounds)
and some tools (RNA, DNA and proteins). This ensures the
functioning and the survival of the plant, and the pursuit of life.
To do this, some plants are able to produce seeds and/​or to
proliferate by layering or by stolons. In this case, there is no
(or little) production of nice-​smelling volatile products. On the
other hand, other plants develop secondary metabolisms that
produce a very large number of organic compounds, among
which some are volatile (volatile organic compounds, VOCs)
and nice-​smelling.
Photosynthesis enables water and carbon dioxide (both min-
eral compounds) to be transformed into organic compounds
that store the (solar and of other origin) energy captured during
this reaction. The first molecules produced are hydrosoluble
saccharides, which are transportable by the sap; these saccharides
are also the basic materials from which the plant will build all
that is needed through biochemical reactions (anabolism) and/​
or through chemical/​biochemical reactions for recombinant and
destructive phases (catabolism).
Catabolism and anabolism are both components of what
is called metabolism (equivalent term: pathways). All the
mechanisms serve to transport and use energy and tend to restore
water and carbon dioxide at the end of the cycle (if in contact
with air).
How to Capture Smell? A Chronological Point
of View
Since the beginnings of humanity, at first in the rites of crema-
tion and to neutralize bad odours, smells have been used as a
link between past, present and future. They quickly became a gift
to the gods through fumigation, whence (later) the word “per-
fume” originated: per fumum means “through the smoke”. The
use of chewing gum (mastic in those times) developed in parallel
because the priests had to be cleansed in order to perform their
office (~3200 BC in Egypt). The smell was used for the purifica-
tion of the spirit and the body (Haarmann & Reimer, 1993; Ohloff
et al., 2012).
From before 4000 BC until the 7th century, odorant materials
were burnt in dedication to gods (a gradual replacement for
human or animal sacrifices). They have been used in holy rit-
uals, remedies, magic, body care, food flavouring and wine
macerations.
Eric Angelini, Laure Dziuba
The techniques used were combustion, the use of smoke, and
the preparation of lipophilic extracts (cold macerations (soaking),
hot macerations, decoctions and infusions) and hydrophilic
extracts (water macerations and wine macerations).
The solvents/​supports were almond, hazelnut and olive oils,
sheep/​lamb fat, beef/​ox grease, water and wine. And the results
were unguents/​ ointments, “pomades”/“cerats”, perfumed oils,
perfumed waters and perfumed wines (Roman flower wines).
The raw materials included incense, spices (such as cin-
namon), rose, myrrh, opoponax, galbanum (resins and balsams,
which exude from wounded plants and are liposoluble), woods
(sandalwood), Mediterranean flowers and aromatic plants such as
saffron and violet flower.
The Greek civilization developed expertise in olive oil extracts
and also fat extracts (this technique led to the “enfleurage”
introduced in Grasse, France, between the 17th and the 19th cen-
tury), tinctures and lotions (the birth of cosmetics). The Roman
civilization increased the therapeutic uses.
From the Middle Ages (~7th century) through the Renaissance
to the 18th century (~1700–​1750), the Arab civilization embarked
on two quests, which are of great interest to us:
• the quest for the philosopher’s stone, which led to “Al
Kimja”, alchemy. This led to progress in knowledge
of the constituents of minerals (mainly), though it was
quite nebulous, which allowed chemistry to come to
light later (next period);
• the quest for the elixir of eternal life, which led to dis-
tillation techniques. Note the terms “Al Ambiq” (1265)
and “Al Kohl” (1586) (the latter was the name for
pulverized antimony).
At that time, the techniques were the same as before, but distil-
lation (from Latin distillare, falling drop by drop), known since
the Chaldean civilization, got its name in 1372; distillation as
used nowadays (with an alembic, or still, and a cooling coil) was
improved during the 8th century.
The solvents and supports were the same as before, but now
water and wine were used in distillation techniques.
Since about 1200, perfumed waters have been obtained
through distillation for pharmaceutical uses, body care and
confectionery. Wine distillation afforded (quite pure) alcohol,
alcoholates and brandies; wine and alcohol macerations afforded
liqueurs, spirits and bitters. All these preparations were made in
abbeys and monasteries by religious brotherhoods, which began
cultivating botanical gardens after the 15th century.
The raw materials were the same as before, but also with
lemon, orange, jasmine, mints, rose and, of course, spices (which
were real treasures).
The process of distillation was needed to allow the develop-
ment of perfumes as we know them today; the first still (for rose
distillation?) was invented in Alexandria by alchemists, but it was
probably improved by Avicenna (980–​1037).
Cooling via a “serpentine” was invented by Italian distillers
in 1320: it became possible to distil wines and to obtain some
(quite pure) alcohol. Then in 1370, the first alcoholic fragrance,
the water of Hungary, was developed in France for the queen of
Essential Oils
TABLE 37.1
Odorant Compounds from Plants
Part of the plant Description/​function
Grains, seeds If the seed can be carried, the
insurance for survival is better;
Samara of a it may be physical (plane tree,
hornbeam tree dandelion, …) or chemical
conveyance (the “carrier” will
move and/​or eat a fruit).
Contains the modus operandi, the
tools and a ration of survival
Ambrette seeds (generally as lipids).
The totality may be either physically
protected by a rigid shell or
chemically with a repulsive (even
poisonous) molecule(s).
© 123RF
As the seed is in contact with the ground, a root is sent to check the water supply; at the same time, tools start a stalk to supply water through the sap and
produce leaves to start photosynthesis.
Roots, rhizomes, Roots draw up water and minerals;
tubers they can be fitted out by the plant Glucoalcaloids: association of a
to store reserves (hence their
Rhizome from swelling).
Iris pallida Tubers serve generally to store the
starch. It is necessary to protect
against predators (hence production
of repulsives, poisons) and rot.
Stalk, trunk The trunk and the stalks ensure
the plant’s rigidity and protect
Commiphora the cellulose, which allows the
molmol (Myrrh) passage of the sap.
Leaves, needles Leaves perform photosynthesis
and must be protected either
Mentha arvensis mechanically (urticant hairs,
thorns …), or chemically
(repulsives, poisons, etc.); in
addition, they must be protected
© Fotolia from sunstroke and the loss of
water by evaporation.
Buds It is again a question of protection
(but also of hormone production).
Blackcurrant buds
Flowers In the case of flowers, it is necessary
to attract the attention of
Rose de mai (Rosa pollinating insects by the colour
centifolia L.) and the smell (pheromones).
267
Example of resulting products Examples of applications
No odorant Fruit juices
(Attractive) products of metabolism Walnut oil: food, supply
Oils (sunflower, palm, etc.) Almond oil: food,
cosmetic medium, etc.
Olive oil: food, soap
factory
Tannins, quinones, etc. (Leather) tanneries,
Tannins woodworking
Benzaldehyde Bitterness and poisoning
Essential oils Flavours, pastry
Fragrances, flavours
Roots used as such: Fragrances: Vetiver oil
(sesquiterpenoids)
sugar (palatable) with an alkaloid Fungicides, pesticides:
(poison) solanine (potatoes)
Triterpenoids Fragrances: irones (Iris
pallida)
Sap Concentrated sap: maple
Forest-based industries: cellulose syrup
Manufacturing waste of the paper Paper pulp
pulp: lignins Vanillin
Hydrodistillation of joiner’s Atlas cedarwood EO
workshop waste Cedarwoods, sandalwood,
Wood processings cinnamon bark
Product recycling: resins Myrrh, incense, pine gum
Pathological resins (produced by the Paint industries: lacquers,
plant in reaction to an aggression/ varnishes, solvents
wound) Elemi
Essential oils (when products are Mentha arvensis,
stable and with a good yield) isolate = menthol
Lavender, sage, patchouli,
pine, etc.
Organic volatile solvent extracts Violet leaves absolute
Infusions, tinctures Lemon balm
Essential oils (when products are Clove buds
stable and with a good yield)
Organic volatile solvent extracts Blackcurrant buds
(when products are delicate and/​or
with a low yield)
Essential oils (when products are Rose
stable and with a good yield)
Organic volatile solvent extracts Jasmine, tuberose, rose
(when products are delicate and/​or
with a low yield)
(continued)
268 Eric Angelini, Laure Dziuba

TABLE 37.1
Odorant Compounds from Plants (Cont.)

Part of the plant Description/​function Example of resulting products Examples of applications

Stamens The only example used in our Tincture, infusion Flavours, spices
profession …
Saffron

Fruits Most of the fruits serve to propagate Fruit juices Concentrated juices for
the seed and are not used in flavours
perfumes.
Hydroalcoholic extraction Vanilla oleoresin
Vanilla Vanilla is a specific case, a fruit Organic volatile solvent extraction, Vanilla resinoid
constituted mainly by heterosides, supercritical CO2 extraction
which release the nice-​smelling Tincture, infusion
part after fermentation; the
released products have a rather
hydrophilic character.

Citrus is another special case. Fruit juices for flavours

Essential oil by cold expression Bergamot, lemon,
Bergamot The EO is recovered by (mechanical) orange, lime (Citrus
explosion of the vesicles of the aurantifolia Swingle or
skin, called pericarp, zest or bark. Citrus latifolia Tanaka – ​
the green acid type),
limette (sweet lime)
Essential oil by hydrodistillation or Distilled limette
steam distillation

Grains, seeds … to close the loop … With, for pleasure, the nutmeg, protected by a kind of bark and its
aril, called the mace (EO)

Source: All images © iStock unless otherwise stated.

Hungary. Several other uses were found for floral waters (medi-
cinal or for food), and interest in essential oils (considered at first
as by-​products?) only appeared later (Le Guerer, 2016).
From the 18th century to 1900–​1930, chemistry separated
from alchemical research (including the search for “philosopher’s
stone”); the first elements were identified, and chemical research
led to the creation of the periodic table of elements and the dis-
covery of atoms and subatomic particles.
The search for the “elixir of life” allowed expansion and
improvements in distillation techniques and the discovery of new
solvents/​supports: diethyl ether (Baumé, 1769; Robiquet, 1835;
Büchner, 1837); carbon disulfide, an excellent solvent but a deadly
poison (Millon, 1856); petroleum ether/​ ligroin (Massignon,
1879); methylene chloride (dichloromethane) (Naudin, 1882);
and benzene (Faraday; structure by Kékulé, 1866).
The first concrete oils, obtained through benzene extraction,
were presented at the International Industrial Fair in Vienna in 1873
by the Roure Company (the company was bought by Givaudan
in the 1980s). Benzene was banned for extraction purposes in
1980–​1990 for safety reasons. The techniques and results were the
same as in the previous periods, but hydrodistillation was added:
while decanting the water of perfumed waters, the oily part of the
distillate is recovered, which contains the main odorant part, called
“essential oil”: “oil” for its fatty aspect, viscosity and solubility in
fats (lipophilic aspects) and “essential” because of its volatility
and its odour, which is (nearly always) close to that of the raw
material. Notice the use of vapour to heat alembics from ~1860.
Cold expression of citrus oils had been performed since the
beginning of the 20th century. But now, alcoholic extractions,
producing oleoresins and absolute oils, were added, as well as
volatile solvent extractions leading to concrete oils, resinoids and
oleoresins, while fat and grease macerations produce pomades
(ointments), concretes, infusions and tinctures.
The raw materials were the same as before, but after 1830, a
specific agriculture was created for perfumery purposes. More
recently, improvements in extraction techniques have been
developed: how to better extract, how to increase the yields, and
how to isolate compounds of interest with split/​fractional dis-
tillation and/​or crystallization. Also, chemists began isolating
compounds of interest, understanding and proving structures, and
performing the synthesis and reproduction of natural compounds
and the synthesis of new compounds of olfactory interest.
Chemists such as Otto Wallach (1847–​ 1931) produced huge
works on some odorant categories, such as terpenes.
Essential Oils 269

TABLE 37.2
Partition Coefficient Applications
So-​called “intra”- and Physical state: condensed phase (solid, liquid) or
Molecule (as such) “inter”-​molecular strengths vapour phase
Kind of new Parameter in Description Applications (non-​exhaustive)
phase(s) relation to K
Air Vapour pressure A part of the molecules receives enough energy to Olfaction (partial description)
exit the condensed phase and go into the vapour Hydrodistillation (production of essential oils;
phase; the vapour pressure is a constant at a given concentration of volatiles)
temperature. Vapour phase chromatography (partial description)
Environmental impacts (pollution, dispersion)
Boiling point Each molecule receives enough energy to exit the Distillation (way to purify).
condensed phase and go to the vapour phase; the Split distillation: one-​way to isolate a specific
boiling point is a constant at a given pressure. compound among others; two-​way to obtain
concentrated EO.
Note that routinely, neighbouring compounds must
have a difference of at least 20 °C from the
boiling point of the expected compound to achieve
separation.
Vapour pressure (against air or any other vapour) is always applicable as long as there is an interface
Gas + Liquid Solubility coefficient Solubility in water or hydrophilic media Retro-​olfaction, taste
in aqueous media Volatile solvent extractions (production of absolutes,
washed resinoids)
Vapour-​liquid chromatography on polar phase
Environmental impacts (pollution, biodegradation)
Solubility coefficient Solubility in oils, fats, organic solvents (lipophilic Volatile solvent extractions (production of concretes,
in organic media media) resinoids)
Vapour-​liquid chromatography on non-​polar
(= apolar) phase
Environmental impacts (pollution, biodegradation)
Gas + Liquid + Different solubility Separation based on affinity differences (lipophilic vs. Volatile solvent extractions (e.g., extractions of
Liquid coefficients hydrophilic) commercial products for analytical purposes)
Gas + Solid Adsorption Techniques of capture on appropriate solid phases, gas Vapour-​solid chromatography, chiral analysis
coefficient driven (other than hydrodistillation) Selective extractions (solid phase extraction, solid
phase micro-​extraction, sparging, …)
Environmental impacts (sparging, clean-​up)
Liquid + Solid Solubility + Separation based on affinity differences All kinds of liquid chromatography: high-​performance
adsorption liquid chromatography, thin layer chromatography,
coefficients chiral analysis

From the Point of View of the Parameters Linked to/​ From the Point of View of the Manufacturing
Depending upon the Nature of Molecules: Partition Processes
Coefficients The processes used in the perfume and flavourings professions
The aim is to observe the behaviour of molecules in different are at first the result of observations and more or less sanctioned
media and better understand our “volatility/​solubility” leitmotiv. by the use of trial and error; the theorization only came later
Every time a compound in the pure state and in a condensed to explain a process and/​or improve it. The extraction/​capture/​
phase (liquid or solid) encounters a different medium (a different isolation of nice-​ smelling compounds depends on partition
phase), part of the compound enters the new medium. The quan- coefficients, which are themselves dependent on the tempera-
tity that is able to migrate into the new phase may be evaluated ture, the pressure and the contact areas. Furthermore, within the
through the partition coefficient K (also called a “constant”), material to be extracted, several types of partition coefficients
which is the ratio of the concentration of the molecule in one of may coexist. During the manufacturing processes, there may be
the phases to the concentration of the molecule in the other phase arrangements related to these parameters:
(Table 37.2). K is temperature and pressure dependent.
Any kind of molecule in touch with its congeners is subject to • temperature: work either at room temperature or under
intra-​and inter-​molecular bonds that determine its physical state: cold or hot conditions, using heating pads for extraction
gas/​vapour, liquid or solid. The contact with (at least) one phase battery, pipes for steam circulation, condenser coils, etc.;
of different nature will modify the balance. • pressure: work with atmospheric or vacuum pressure or
under pressure;
270
• contact areas: the (raw) material is put in baskets and/​
or cut, crushed or pulverized prior to loading (static
extraction), or shaken during the extraction (dynamic
extraction). And there are systems for filtration, puri-
fication and recycling of solvents, cleaning of devices
and deodorization (blowing and/​or other methods).
Primary Extracts: Volatile Solvent Extraction and
Essential Oils
The “hydro” production plants (hydro, from hydrocarbons) use
extraction processes that are based on the solubility of the nice-​
smelling compounds in organic volatile solvents (hydrocarbons
such as hexane, cyclohexane, etc., and sometimes mixtures of
hydrocarbons with a hydrophilic solvent as needed, depending
upon the expected products) used in large amounts (compared
with the amount of the extract of interest) and, as said earlier, at
ambient/​cold/​hot temperature and in static or dynamic processes.
The solvents used are lipophilic (not miscible with water) and/​
or small in molecular size and/​or with a density close to that of
the compounds to be recovered. Moreover, they must be easily
removed or separated from the extract (unless they need to be
used as support in a further specific application); thus, their
boiling point must be rather low in order not to damage the
expected product or allow it to escape.
The raw material to be extracted is introduced into the top of
vessels (whose volume is generally from 600 to 3000 L); many
tanks may eventually work together, and this is called an extrac-
tion battery.
Once the vessel is closed, the solvent is fed in, and the material
is submitted to several successive extractions to ensure its exhaus-
tion. The solvent containing the desired compounds (but not only
those) is then recovered, filtered and (almost totally) eliminated
by distillation/​concentration.
Depending on the kind of equipment in use, the solvent may be
evacuated by gravity or driven out by gas pressure (e.g., nitrogen)
or suction or under vacuum (this situation requires a floating filter
to hold the material). The distilled solvent is recovered in a storing
tank dedicated to a specific raw material, so that the solvent in
that tank will be reused for the next extraction of the same kind of
raw material (of course, due to the loss of some solvent during the
process, fresh solvent must be added from time to time).
The exhausted vegetable material (whose trivial name is
“drèche” in the Grasse area, by analogy to the distiller’s grain in
the wines and spirit industry) is desolventized and generally used
as compost.
If the plant material is fresh (or not yet totally faded), the
product obtained is a concrete. Concretes from jasmine, rose,
mimosa and violet leaves, for example, are produced in this way.
If the plant material is dry or resinous, the product obtained is a
resinoid (or raw resinoid), such as labdanum extracted from the
tree Cistus ladanifer or Cistus creticus. These products are pasty,
even hard, and most often highly coloured.
Because of its nature and its quantity relative to the raw material,
the solvent also extracts heavier compounds without olfactory
interest (fats, pigments, etc.) that may prove troublesome for
Eric Angelini, Laure Dziuba
further applications and dilute the odoriferous ingredients; on the
other hand, the concrete and resinoid, having almost non-​existent
moisture content, allow effective preservation of the fragrant
material.
Because of its versatility, it is a very good process to extract
fragile materials such as jasmine.
Essential Oil Production Plants
The production of EO is based on the fact that VOC have a sig-
nificant vapour pressure and may be extracted by an updraft of
steam (in the same way as a glider is carried by an updraft of air);
they are then condensed back to the liquid phase (rose, or mints
containing menthol) or through a condenser.
Let us observe that none of the extracted molecules reaches
its boiling point during this process (their boiling points are in
principle much higher than that of steam, ~100 °C in context),
except for some very volatile molecules, which in fact distil at the
beginning of the process.
Steam may be produced into the still or fed within the material,
overheated, under pressure.
The raw material may be:
• immersed in water if some hydrolysis processes are
necessary: this is the case when cells containing the
fragrant material are difficult to burst (e.g., the rose,
whose petals are protected by a waxy material); the
immersion may be long (soaking) for woods, for
example. In both cases, numerous degradations occur in
the aqueous medium (off-​odours, degradation of sensi-
tive compounds, artefacts, and loss of some molecules
because of their significant solubility in water);
• put on some kind of grid above the water if there is no
risk of a preferential pathway of vapours (when the
raw material is not too hydrophobic, the water vapours
pass through it in a homogeneous way and the volatile
compounds “take the lift” so that you obtain an extract/​
hydrolate; when the raw material is too hydrophobic,
the vapours find a way through the mass without
touching it or taking anything with them, so that the
yield is almost null).
The hydrodistillation may be performed under pressure in
the case of scented compounds that are not easily accessible.
For very low yields, the VOC remain in the hydrolate, and the
product obtained is an aromatic/​ perfumed/​ floral water (e.g.,
witch hazel water). When VOC are in sufficient quantities, they
gather via coalescence and split up (less often down, but this
may happen) from the water in the Florentine flask. The obtained
product is an EO.
It is interesting to note that the EO contains almost all of the vola-
tile compounds and only those (whereas a concrete or a resinoid
contains both the volatile and some non-​volatile compounds), but
due to the different solubilities/​partition coefficients, a fraction
of each compound will remain in the water; to limit these losses
(and not to waste too much water), water is usually continuously
returned to the still (so-​called “cohobage”). Some of the volatile
Essential Oils
extracted compounds are sufficiently hydrophilic to remain in
large amounts in water (e.g., 2-​phenylethanol from rose EO), and
certain difficult settings occur when the EO density is near that of
water (e.g., ambrette seed: d ~ 0.9); in this case, either mechan-
ical means or volatile organic solvent is used (which, of course, is
removed in the next step).
Tinctures, Soluble Oils, Alcoholates
The production plants for tinctures, soluble oils and alcoholates
are based on the solvation properties of ethanol (see also
“absolutes”). They use equipment similar to those described pre-
viously, but in a different context, and here again at ambient, cold
or hot temperature, and under different pressures. The compounds
of interest are extracted by maceration of the raw materials for
what may be a very long time, and the soaked raw materials must
be dry: the fresh plants contain a lot of water, and the goal is not
to obtain a herb tea.
The alcohol used may be pure or diluted with water (different
dilutions with water are referred to as different tincture degrees):
• simple soakings to obtain tinctures and/​or infusions,
used as such or concentrated;
• soaking, then alcohol distillation (the alcohol is the
recovered part) to produce alcoholates;
• percolation at high temperature of a hydro-​alcoholic
mixture, e.g., vanilla (Soxhlet equipment), followed by
filtration and concentration. The compound obtained in
this way is a vanilla oleoresin (sometimes called “res-
inoid”, although that term strictly refers to a volatile
organic solvent extraction).
• cold deterpenation of EO (especially Citrus); terpenic
compounds in these EO are not that soluble in alcohol,
their olfactory impact is quite small and they may be
inconvenient (in terms of the desired application); the
use of a hydro-​alcoholic mixture (about 65° alcohol)
forces their release, and they are recovered under
the name of “terpenes” (from orange, lemon, etc.). The
interesting odorous part remains solubilized in the mix-
ture and is called “soluble oil”. The organic part may be
recovered through new changes in solubility power, but
for cost reasons it is most often used as is.
In dry distillation, the raw material is distilled without water or
vapour addition (pyrogenation/​cracking); e.g., cade oil or birch
barks. This is strictly restricted because the process produces a
number of benzenic compounds.
In cold expression, a mechanical process is used to obtain oils
from Citrus peels. It is named after the first process used, which
consisted of pressing zests and recovering the oil with sponges; the
“Calabrese” machine was invented in 1900 (fruits are crushed in
special bowls with tips), then the “Pelatrice Avena” machine in 1924
(with a spike-​covered mobile drum), and the “Special Machine” in
1928 (four rollers with spikes in a trapeze forming a corridor into
which fruits are introduced via an Archimedes’ screw).
The oil is decanted through the filtered aqueous phase. This
aqueous phase may be hydrodistilled to increase the yield. More
271
modern machines allow juices and EOs to be obtained (almost)
simultaneously (but separately). Note that “distilled” oils of
Citrus fruits can be obtained by treatment of the fruit juice
industry waste.
Secondary (Re-​Processed) Extracts
For making absolutes, the process is based on the solubility
of concretes and resinoids in ethanol, a small-​molecule hydro-
philic solvent in which the undesirable compounds (see earlier
under concretes/​resinoids) are not soluble. The compounds
are mixed with ethanol in specific mixing machines called
“batteuses” at high temperature (but below the boiling point
of ethanol) to ensure that all the interesting components are
solubilized; the mixture is then filtered when still hot in order
to eliminate the first insoluble compounds; the filtrate is kept
at rest in a freezer (temperature between −12 °C and +5 °C) in
order to precipitate fats; then it is filtered, and the solvent is
evaporated/​distilled.
From concrete oil, an absolute, also called absolute oil, is
produced. From resinoid, a washed resinoid (quite often incor-
rectly also called absolute) is made. These products may be quite
liquid (jasmine) or viscous or still very pasty; they may also be
coloured, depending on the nature of the pigments in the primary
extract. What remains is called wax, most often containing fats
and pigments that are not soluble in alcohol, and it may be used
in non-​alcoholic applications.
Absolute waters are produced using a process based on solu-
bility in an organic volatile solvent of molecules of interest
contained in aromatic/​floral perfumes or waters; e.g., the water
obtained through petitgrain (bitter orange) hydrodistillation gives
the “Eau de Brouts” absolute.
Oils obtained from concretes/​resinoids are based on the vapour
pressure of components in concretes and resinoids, which may be
recovered through hydrodistillation (e.g., rose). The compound
obtained is an essential oil, but this term is not allowed because
it was not obtained in a direct/​primary way. Each manufacturer
gives this kind of extract a special name, such as “Superessence”
(for Mane Inc).
Molecular distillation is a re-​processing of any extract injected
as such or with an additive called a co-​solvent into a specific
device under molecular vacuum (in a so-​ called “molecular
vacuum”, the vacuum is from 10–​1 to 10–​5 Pa, which is sufficient
for this kind of application, hence the name). The system may be
single-​staged, or more. In this case, the different parts (meaning
different vacuums) are separated by a siphon; the second part is
under a higher vacuum/​lower pressure than the first one. The raw
material (concrete, resinoid, etc.) is introduced by a mist nozzle
into the system under vacuum and at a high temperature. The
system has different exits to recover:
• the more volatile molecules at the top via a cold trap
(liquid nitrogen) on the vacuum pipe,
• the volatile molecules (generally what is wanted) at the
bottom (of the first stage) or before the siphon (when
there is more than one part),
272 Eric Angelini, Laure Dziuba

TABLE 37.3
Orange EO Deterpenation
Terpene Remaining
Quantity (g) quantity (g) Concentration x Distilled quantity Distilled terpenes Remaining terpenes terpenes (%)
Before distillation: 1000 950 0 0 0 950 95
1000 950 2 500 500 450 90
500 450 4 250 250 200 80
250 200 8 125 125 75 50

TABLE 37.4 TABLE 37.5

Constituents of Orange EO Examples of VOC Concentration and Number
Boiling point at atmospheric Raw material Type VOCC (mg/​kg = ppm) VOCN
Constituents % pressure (°C) Raspberry Raw fruit 2–​5 150
Monoterpenes from: Tomato Raw fruit 3–​5 250–​300
Alpha-​pinene to 0.5 155–​156 Coffee Roasted fruit 100 800–​1000
Limonene to 90 to 96 175–​177
Cocoa Fermented and 100 500–​700
Terpinolene <0.2 183–​185
roasted fruit
Molecules with
Rose Flowers 200–300 >300
olfactory impact
Octanal 0.2 171–​173 Note: VOCC, VOC concentration; VOCN, number known currently.
Nonanal 0.05 191–​193
Decanal 0.3 207–​209
Citral < 0.2 ~228
Furocoumarins, e.g.
terpenes, the deterpenated EO, and a mixture of menthol and its
Aurapten 455
Meranzin 415 isomers (iso-​and neo-​menthols mix).
Rectification is a way to improve the first distillation pro-
cess. Indeed, with all distillation processes, based on boiling
point differences, the extracts obtained contain a great number of
• the less volatile molecules at the bottom of any system,
either as such or in a specific solvent (e.g., a glycol compounds with similar boiling points, so that it is not easy to get
polymer) or in the co-​solvent, as needed. a 100% pure isolate in this way. Rectification is often performed
under reduced pressure in order to avoid damaging the materials.
This process is very respectful of fragile compounds and makes it It is used for:
possible (among other things):
• purification, discarding of uninteresting molecules,
• splitting, fractional distillation,
• to recover essential oils or VOC with high added
• isolating a compound of interest/​obtaining an isolate,
value and to fractionate extracts (e.g., separation of
e.g., anethole from star anise or from fennel, linalool
furocoumarins),
from rosewood, or ho (Cinnamomum camphora
• to deterpenate/​concentrate the citrus EOs (alternative
chemotype linalooliferum, lauraceae), etc.,
method to cold deterpenation),
• discarding one or more unwanted molecules for odorant
• to (at least partially) remove colour from the injected
or regulatory reasons, e.g., methyleugenol from rose EO,
extract (e.g., patchouli, hemlock, moss …).
furocoumarins (bp > 300 °C at atmospheric pressure), etc.,
• deterpenation by concentration (e.g., folded orange;
Crystallization is a process based upon the solubility of a spe-
Table 37.3).
cific compound of commercial interest in its primary extract,
most often an EO, e.g., menthol (freezing point 42 to 45 °C)
In comparison with cold deterpenation (Table 37.4):
from cornmint; this EO may contain up to 85% menthol; the
“flashed” EO (to discard water and off-​ odours) is put in a • some of the compounds of olfactory interest have a
container in which the temperature is steadily lowered, thus boiling point inside the terpenes’ range of boiling
decreasing menthol’s solubility in its EO and producing fine points; as this kind of distillation is under vacuum in
crystals, which are filtered and dried; the filtrate is the partially order not to break down fragile components, the boiling
dementholized mint EO (the one you can find on the market), point range is decreased (as a function of the applied
which still contains 40% to 45% of L-​menthol; it may be used vacuum); as a result, not only terpenes are distilled
as such or processed again by fractional distillation to give mint (more terpenes than expected remain in the pot),
Essential Oils 273

TABLE 37.6
Some Examples of Odorant Products
Type of material Type of material
to be extracted Example Extracts to be extracted Example Extracts
Needles Pine 1, 5, 6 Flowers Cananga 1 (0.8–1)
Fir tree 1 (0.1–1.2), 5, 6, 9 (maltol) Hyacinth 5 (0.01–0.02), 6 (50)
Jasmine 3, 4,5 (<0.2), 6 (50)
Aril Mace (shelling of 1 (8–13) Daffodil 5, 6
nutmegs)
Tuberose 3, 4, 5 (0.1), 6 (30)
Balsam (exudates) Benzoin 7 (85–95), 12 Mimosa 5 (1), 6 (25)
Peru 7 (43–55) Orange flowers 1 (0.08–1 = Neroli), 2, 10
Styrax 1 (5), 7 (50) Rose 1 (0.02–0.05), 2, 5, 6, 10
Tolu 1 (1.5–7), 8 (60) Ylang-ylang (alterna- 1 (1.5–2.5)
tive method to extract
Cananga)
Woods Amyris 1 (2–4) Rhizomes Ginger 1 (0.3–3.5), 8 (3.5–10.3)
Rosewood 1 (0.7–1.2), 9 (linalool) Orris 1 (0.2 = butter), 7
Birch 1, 9 (raspberry ketone), 11
Guaiac 1 (2.7–3) Fruits, pods, seeds, Dill 1 (2.5–4)
nuts, berries
Sandalwood 1 (4–6) Angelica 1 (0.6–1.5)
Anise 1 (1–4)
Flower bud Clove buds 1 (14–21), 9 (eugenol) Star anise 1 (2–3), 9 (anethole)
Cardamom 1 (1–10), 8 (10)
Bark Cinnamon 1 (1.6–3.5), 8 (7–12), 9 Carrot 1 (0.5–0.8)
(cinnamaldehyde)
Nutmeg 1 (2, 6–12), 8 (18–37) Caraway 1 (3–6)
Celery 1 (1.5–2.5), 8 (9–11)
Peel (zest, pericarps), Citrus 1 (0.1–1) Coriander 1 (0.1–1)
so-called bark
Cumin 1 (2,5-5)
Stamen Saffron 12 (0.5–1) Fennel sweet 1 (4–6)
Juniper 1 (0.5–2.5), 8 (36)
Leaves Basil 1 (0.1) Nutmeg 1 (2.6–12), 8 (18–37)
Cinnamon 1, 9 (eugenol) Parsley 1 (1.5–7)
Clove 1, 9 (eugenol) Pimento 1 (3.3–4.3)
Witch hazel 2 Pepper 1 (1–3.5), 8
(Hamamelis)
Laurel 1 (0.5–1), 8 (17–19) Vanilla 8, 12
Lemongrass 1 (0.2–0.3), 9 (citral)
Lemon balm 1 (0.01), 13 Gum-resins Elemi 1 (20–30), 7 (80–90)
Mints, menthol 1 (0.1–1), 9 (menthol) Incense 1 (5–9), 7 (50)
containing
Palmarosa 1 (0.7–1.3), 9 (geraniol) Galbanum 1 (20), 7 (35–50), 12
Patchouli 1 (3) Myrrh 1, 7, 8
Parsley 1 (0.05–0.2) Opoponax (opopanax) 1 (3–8), 7
Chili 1 (0.3–3), 8
Violet 5 (0.1), 6 (30) Herbs, herbaceous Dill 1 (0.3–1.5)
plants
Citronella 1, 9 (geraniol, citronellal)
Twigs and leaves Cistus 1 Geranium 1 (0.1–0.2)
Cypress 1 (0.2)
Eucalyptus 1 (1.8–2) Flowery tops Spike lavender 1 (0.5–1)
Mastic 5, 6 Tarragon 1 (0.3–1.5)
Petitgrain 1 (0.1) Lavender 1 (0.8–1.6), 5 (1.5–2.2), 6
Lentisque 1 (50–60)
(continued)
274 Eric Angelini, Laure Dziuba

TABLE 37.6
Some Examples of Odorant Products (Cont.)

Type of material Type of material

to be extracted Example Extracts to be extracted Example Extracts
Lavandin 1 (1–1.8)
Resins Labdanum 1 (1–2), 7 (70–85) Marjoram 1 (0.2–3.5)
Mastic 5, 6 Rosemary 1 (0.4–2)
Sage (officinalis) 1 (0.7–2), 8 (0.6–1.2)
Bulbs, tubers Garlic 1 (0.05–0.25) Clary sage 1 (0.07–0.15), 5, 6
Roots Angelica 1 (0.3–0.4) Thyme 1 (0.5–1.2)
Calamus 1 (1.5–4.8) Verbena 1 (0.07–0.2)
Curcuma 1 (2–7,2), 8 (7.9–10.4)
Vetiver 1 (2–3) Lichen Moss 5, 6

Note: 1. EOs. 2: floral water. 3: “pommade”. 4: absolue des châssis. 5: concrete. 6: absolute (the yield in absolute is calculated starting from the concrete).
7: resinoid. 8: oleoresin. 9: isolate (name). 10: absolute from waters. 11: dry distillation. 12: infusion, tincture. 13: alcoholate. The yield range, where
known, is given in parentheses.

• but pigments and other compounds are also concentrated
in these steps, which can be a nuisance for certain
applications;
• even if you work properly under vacuum, fragile
molecules tend to degrade; for instance, an increase in
terpineol alpha may be noticed;
• on the other hand, the concentrate obtained, having no
solvent, is readily usable as such.
Industrial Constraints
We have not yet discussed the quantities of VOC produced by
plants; it is, however, these quantities that are critical in the
choice of the extraction process, in parallel with the appreci-
ation of the fragility of the plant and/​or the kind of material to
recover.
Any part of a plant may produce scented compounds, and
not necessarily the same ones, according to the part. Table 37.6
shows some examples, without being exhaustive, of materials,
the extracts obtained, and the percentage yield where known.
Expected Yields as a Function of Time and Energy
We have seen that plants produce a great number of interesting
scented compounds, whose number and quantities are fixed for a
defined plant in a range depending upon the species variability,
the geological conditions and climate, etc.
The way the plants are collected (maturity, hour of the day,
period of sunshine) and the storage times and conditions before
processing are also to be taken into account, as well as how
long it takes to reach the production plant and the extraction
conditions.
The operating procedures enable the materials to be to
exhausted, and they take into account the time and energy neces-
sary to obtain the best value for money; it is of no use to try to
force the yield. Moreover, the parameter “users” (perfumers and
flavourists, for example) must be taken into consideration: they
are used to their standards!
REFERENCES
ASTM. 1973. Compilation of odor and taste thresholds value data,
American Society for Testing and Materials. http://​agris.fao.
org/​agris-​search/​search.do?recordID=US8232709.
Drouillard JB, Martins-​Gueunier M, Knauf-​Beiter G, Lebrihi A,
Mathieu F, Guérin L, Guérin-​ Schnieider R, Dumoulin M,
Riboulet JM, Arioli X, Treilhou M. 2005. Mouldy-​ earthy
aromas in wines: first practical results obtained by a multidis-
ciplinary partnership, Review of Viticulture and Oenology, 1.
Haarmann & Reimer. 1993. Le livre H&R du parfum, Gloss.
Le Guerer A. 2016. An Odyssey of Flavours and Fragrances,
Givaudan, Abrams.
Lide DR (ed.). 2005. CRC Handbook of Chemistry and Physics,
CRC Press.
Ohloff G, Pickenhagen W, Kraft P. 2012. Scent and Chemistry, The
Molecular World of Odors, Wiley-​VCH.
Proust B. 2006. Petite géométrie des parfums, Seuil.
Essential Oils: How to Safely Use Essential Oils
Eric Angelini1 and Laure Dziuba2
1 V.MANE FILS, 620, route de Grasse, 06620, Le Bar-​sur-​Loup, France
2 Consultant chromatographic analysis of fragrances and related compounds, 1955 chemin des vergers, 06620,
Le Bar-​sur-​Loup, France
Since the first uses of aromatic herbs, essential oils (EOs) have
been used as powerful odoriferous substances in food. A large
number of their compounds interact with taste and olfactory
receptors. Each of these compounds could saturate its respective
receptors, with a clear signal for the consumer when it is enough;
however, “enough” in terms of organoleptic properties may be
too much in terms of health, as being natural is not a guarantee
of safety. Here, we describe the methodology and information
on which the selection of EOs is based by users. The example of
making a reproduction of a bouquet garni is chosen.
Brief History of Safety Concerns
The self-​ limiting effect of odorant compounds was probably
a key element for appreciating the safety “under condition of
intended use” (Arctander, 1969), but a long history of uses is also
an important driving force for classifying odorant compounds
and EOs as “generally recognized as safe” (GRAS) by the Food
and Drug Administration (FDA, 2019). Today, however, it is
necessary to collect more data to allow a better safety evaluation.
A good knowledge of the composition of EOs is the entry point
of all safety assessments. This is easier now, due to advances in
analytical methods.
Once the question of the identity and quantity of substances
in an EO is answered, the next step is to define and classify EO
constituents in order to determine their toxicity.
A first difficulty is the variable composition of those materials
in relation to the natural variation due to the biosynthesis of EO
components in a plant (compared with the fixed composition of
chemically defined substances).
Plants produce substances necessary for their life. The
pollinator-​attracting function of some compounds is quite well
known, but there are also substances useful for protection. For
example, plants produce natural pesticides to defend against
various viruses, bacteria and insects, and others that help them
avoid being eaten by predators (Ames, 1990). The kind of pro-
tection has been selected as a function of the necessary energy
expenditure and of the zone to be protected.
We must be aware that the possible toxicity of plants to
human beings is not mainly a result of evolution, as humans and
mammals appeared well after the main plant metabolism was
established; it was not our “predator” activity that first influenced
plant metabolism.
Some of the possible metabolic pathways are (as a reminder
and in broad terms):
1. photosynthesis, by which plants use the sun’s energy to
create compounds, mainly small saccharides, which in
turn are used either to produce bigger saccharides or as
starting materials in other pathways; one of the concerns
with saccharides is the possible production of methanol
during degradation of polysaccharides (mainly pectin
molecules) (Dorokhov et al., 2018);
2. the acetate pathway, which produces lipids that are
often the origin of many substances used as attractants
by plants, e.g., saturated linear aldehydes (C8, C10),
unsaturated aldehydes (2-​ hexenal, 2,6-​ nonadienal,
decadienals), ketones (jasmone), etc.;
3. the shikimic acid pathway, which is necessary to
build amino acids, proteins, DNA, alkaloids and
phenylpropanoids such as elemicin, safrole and
myristicin, the function of which is clearly either to
repel or to kill the majority of predators (whatever their
size) and eventually to help with healing in the case of
injury;
4. the mevalonate and methylerythritol phosphate
pathways, which produce terpenoids and many other
compounds, for instance steroids; terpenoids are very
often EO constituents, supposedly due to their ability to
protect the plant from sunburn and evaporation.
Note, especially here, that toxicity is not neces-
sarily the first function of a given substance; not all
phenylpropanoids are necessarily poisons. We still have
a lot to learn about the functions of any kind of com-
pound in plants;
5. combination of different pathways to produce more
constituents.
Finally, via multiple different pathways in plants, vola-
tile and non-​volatile substances are produced.
275
276
But not all compounds used for plant protection are volatile.
For example,
• tannins (shikimic acid pathway) are astringent and/​or
bitter, so that predators are discouraged from eating
them (Selosse, 2019; Van Hoven, 1984);
• alkaloids (shikimic acid pathway, most often via amino
acids) are broad spectrum intoxicants (WHO, 2018).
These non-​volatile compounds are not present in EOs.
On the basis of active principles, we could consider all the
chemical constituents of the different extracts. Based on the
biosynthesis described earlier, the palette of different structures
appears a reasonable starting point. In addition to this starting
point, we must take into consideration all new substances coming
from simple water and oxygen reactions in the plant.
This has led different scientific bodies to refer to the most
active substances known in EOs.
North American Safety Approach by
Congeneric Groups
Safety evaluation has been mainly based on a constituent approach
(Taylor, 2016). This is linked to the fact that numerous toxico-
logical methods are not designed to manage complex mixtures
such as EOs. Toxicologists evaluate the toxicity of chemically
defined substances but are presented with difficulties in the case
of complex mixtures; they consider a complex mixture only in
terms of its main constituents with known toxicity. As a result
of this approach, toxicologists ask about the main metabolisms
that gave the complex mixture in question, and they group the
substances into biosynthetic families, from which they extrapo-
late, if they are familiar with the behaviour, the toxicity of one or
more molecules belonging to this family.
As can be seen, there is real difficulty in managing a mix-
ture of compounds with a testing protocol designed for a single
substance. So, combining variable compositions with a large
population of different substances pushed toxicologists to find
some alternatives to the standard approach, which led to the
“congeneric groups” (Cohen et al., 2018; 2019): “Since many
NFCs (natural flavor complexes) are products of common plant
biochemical pathways, the constituents can be organized into a
limited number of well-​established chemical groups referred to
as congeneric groups”. Detailed explanations can be found in
Cohen et al. (2018; Table 38.1).
European Safety Approach by Substances with a
Particular Concern
In Europe, a set of bioactive substances were identified in
different EOs and used to define a threshold of toxicological
concerns in the flavouring regulations (European Community,
2008). We could consider the safety of natural complex
substances (NCS), like EOs, by referring to key substances with
a known toxicological effect present in the NCS. In this way,
Eric Angelini, Laure Dziuba
we could refer to Annex III of the European Regulation # 1334/​
2008 of the European Parliament.
In this annex, Part A contains the list of substances that could be
present in extracts like EOs, for which the restriction is to forbid
any usage as such of those substances. Part B lists substances
found in some extracts, for which some restrictions have to be
applied to limit the exposure of the consumer. Parts A and B are
shown in their entirety in the sources given in the references.
Detailed explanations about how the new flavouring regulation
(Regulation EC # 1334/​2008) will replace the former Flavouring
Directive 88/​388/​EEC can be found in Demyttenaere (2012),
especially for the definition of flavouring substances and “food
ingredients with flavouring properties”, which will impact the use
of essential oils as flavouring ingredients for food products.
In Table 38.2, we have chosen to present the substances in
terms of their volatility (they may or may not be found in EOs)
and their pathway family, together with a non-​exhaustive list of
natural occurrences.
Applications
Leaves, fruits, nuts and even flowers are used as such for cooking.
Is it possible to replace them with EOs? We discuss this question
here using the example of a “bouquet garni”. Say, for example,
that it contains mainly laurel leaves, thyme flowering stems and
parsley leaves. What do these raw materials contain? Water (if
fresh), saccharides (cellulose and others, for instance pectins
and more complex sugars), lipids (as constituents of cell walls,
among other roles), proteins, peptides and amino acids, lignins
and tannins, alkaloids and, of course, volatile compounds from
which an EO may be obtained. It would be interesting to know
about the material balance of the plant, not only the yields of
EOs, but this is quite difficult, because those raw materials are not
used in sufficient quantities in food applications to justify this.
As for the EO yields, their ranges are quite large, because it
is never clear whether the raw material has been processed fresh
or dried. It must also be noted that, because of the loss of non-​
volatile constituents during the EO production, the sensorial
effects obtained by replacing raw herbs and spices by their EOs
may be different (loss of bitterness, for example).The quantities
proposed in the example represent traditional culinary usage; i.e.,
the “bouquet garni” is used to flavour about 1 kg of final food.
Let us first consider laurel (Laurus nobilis L.) leaves. The com-
position is given in Table 38.3.
The components of the laurel leaf EO are cineole-​1,8 (40%),
p-​cymene (20%), terpinyl acetate (7%), myrcene (4.5%), beta
pinene (3.5%), linalool and menthofuran (<0.2%), coming
from the terpene pathway, and eugenol, estragole (<0.1%) and
methyleugenol (<4%), coming from the phenylpropanoids
pathway.
The mass of a large laurel leaf being about 1 g, the mass of EO
that could be obtained from that laurel leaf will be a maximum of
0.03 g. This means that one should have, as a maximum, in 1 kg
of the final food: 30 mg/​kg (ppm) EO, or 12 ppm of cineole (con-
generic group 23), 0.06 ppm of menthofuran (congeneric group
25, Part B Annex III max level 200 mg/​kg), 0.03 ppm of estragole
Essential Oils: How to Safely Use 277

TABLE 38.1
Congeneric Groups
1 Saturated aliphatic, acyclic, linear primary alcohols, aldehydes, carboxylic acids and related esters
2 Saturated aliphatic, acyclic, branched-​chain primary alcohols, aldehydes, carboxylic acids and related esters
3 Aliphatic linear and branched-​chain alpha, beta-​unsaturated aldehydes and related alcohols acids and esters
4 Aliphatic allyl esters
5 Unsaturated linear and branched-​chain aliphatic, non-​conjugated aldehydes, related primary alcohols, carboxylic acids and esters
6 Aliphatic primary alcohols, aldehydes, carboxylic acids, acetals and esters containing additional oxygenated functional groups
7 Saturated alicyclic primary alcohols, aldehydes, acids and related esters
8 Saturated and unsaturated aliphatic acyclic secondary alcohols, ketones and related esters
9 Aliphatic acyclic and alicyclic alpha-​diketones and related alpha-​hydroxyketones
10 Alicyclic ketones, secondary alcohols and related esters
11 Pulegone and structurally and metabolically related substances
12 Aliphatic and aromatic tertiary alcohols and related esters
13 Aliphatic, alicyclic, alicyclic-​fused and aromatic-​fused ring lactones
14 Benzyl derivatives
15 Hydroxy-​and alkoxy-​substituted benzyl derivatives
16 Cinnamyl alcohol, cinnamaldehyde, cinnamic acid and related esters
17 Phenyl-​substituted primary alcohols, aldehydes, carboxylic acids and related esters
18 Phenyl-​substituted secondary alcohols, ketones and related esters
19 Aliphatic and aromatic hydrocarbons
20 Phenol derivatives
21 Hydroxyallylbenzenes and hydroxypropenylbenzene derivatives
22 Phenethyl alcohol, phenylacetaldehyde and related acetals and esters
23 Aliphatic and aromatic ethers
24 Furfuryl alcohol, furfural and related substances
25 Furan derivatives
26 Aliphatic and aromatic sulfides and thiols
27 Sulfur-​substituted furan derivatives
28 Sulfur-​containing heterocyclic and heteroaromatic derivatives
29 Aliphatic acyclic diols, triols and related substances
30 Aliphatic and aromatic amines and related amides
31 Nitrogen containing heterocyclic and heteroaromatic substances
32 Pyrazine derivatives
33 Anthranilate derivatives
34 Amino acids
35 Maltol derivatives
36 Epoxide derivatives

Source: Cohen et al. (2018)

(congeneric group 21, Part B Annex III max level 10–​50 mg/​kg),
and 1.2 ppm of methyleugenol (congeneric group 21, Part B
Annex III max level 15 mg/​kg); these quantities are well below
the regulatory threshold for the last three compounds.
If pure EO is used, it should be first diluted to 1% in ethanol
to ensure further solubilization in aqueous media; the application
level in the final product is then 1% to 3%.
Let us now do the same exercise for thyme flowering stems
(Thymus vulgaris L.), assuming that the thymol chemotype is
used. Here, the composition would be (%): proteins 5.56, lipids
1.68, saccharides 10.5 (polysaccharides 7.5), water 65.1, fibres
14, flavonoids and phenolic acids, including rosmarinic acid,
0.15–​2.6 (non-​volatile anti-​oxidant), triterpenes including ursolic
acid 1.88, volatile oil 1 to 2.5–​3%, containing thymol up to 50%,
carvacrol 0–​5%, p-​cymene 15–​20%, terpinene gamma 5–​10%,
presence of thujones alpha <0.25% and beta <0.1% (Stahl-​Biskup
and Venskutonis, 2004).
A thyme stem weighs about 2 g, which is equivalent to 0.06 g
EO at max; so, in 1 kg of finished product, there is 60 mg (ppm)
of EO, including 30 ppm of thymol (congeneric group 20, not
cibled (no regulatory restriction) in Europe), 0.15 ppm of thujone
alpha, 0.06 ppm of thujone beta (congeneric group 10, Part B
Annex III max level 0.5–​35 mg/​kg in beverages). Here, again,
the quantities are below the regulatory thresholds.
If pure EO is used, it should again be first diluted to 1% in
ethanol to ensure further solubilization in aqueous media. The
level added to the final product is then about 5%.
For parsley leaves (Petroselinum crispum Miller), the compos-
ition of the leaf is (%, from Aprifel, 2019): proteins 3.6, lipids 0.6,
carbohydrates 3.4, water 85, fibres 4.3, polyphenols (tannins) ~0.014.
278 Eric Angelini, Laure Dziuba

TABLE 38.2
Substances of Concern in 1334/​2008 Classified by Metabolic Families
Substances Natural occurrence Metabolic family
Menthofuran Peppermint, cornmint Terpenoid, volatile
Pulegone Pennyroyal, buchu, peppermint, spearmint Terpenoid, volatile
Thujones alpha and beta Artemisia species, sage, laurel, thyme, basil Terpenoid, volatile
Asarone beta Calamus, carrot seed, nutmeg Phenylpropanoid, volatile
Estragole/​1-​allyl-​4-​methoxybenzene Basil, chervil, tarragon, fennel, star anise, anise Phenylpropanoid, volatile
Methyleugenol/​4-​allyl-​1,2-​dimethoxybenzene Allspice, bay, tarragon, basil, laurel, nutmeg, caraway Phenylpropanoid, volatile
Safrole/​1-​allyl-​3,4-​methylene dioxybenzene Nutmeg, pepper Phenylpropanoid, volatile
Coumarin/​1,2-​Benzopyrone Cassia, lavandin, thyme Phenylpropanoid, semi-​volatile
Hydrocyanic acid Almond, apricot kernel, cassava, bamboo shoots Amino acid, volatile
Agaric acid Mushroom species Lipid, not volatile
Aloin Aloe species Lipid/​polyketide, not volatile
Capsaicin Capsicum species Lipid/​amino acid, not volatile
Hypericin Hypericum species (St John’s Wort) Lipid/​polyketide, not volatile
Quassin Quassia amara Terpenoid, not volatile
Teucrin A Teucrium species (Germandrée petit-​chêne) Terpenoid, not volatile

TABLE 38.3
Composition of Laurel Leaves
Source (1) (2) (3)
Water (%) 5.4
Lipids (%) 7 8.4
Proteins (%) 10 7.6
Saccharides (%) 35 49
Fibres (%) 26
Alkaloids Isoquinoline alkaloids (family of aconitine, cocaine, ergotamine, strychnine, etc) and aporphinic alkaloids such as
cryptodorin or actinodaphnin (in vitro cytotoxic activity, responsible for acute intoxications)
Flavonoids 18 flavonoids identified, some from kæmpferol
Essential Oil (%) 0.5 to 3

Source: (1) www.bio-​enligne.com/​produits/​148-​laurier.html; (2) www.passeportsante.net/​fr/​Nutrition/​EncyclopedieAliments/​Fiche.aspx?doc=laurier_​nu;

(3) www.bonneplante.com/​laurier_​sauce.php.

The equivalent of the aromatic in terms of EO is obtained Conclusion

from curly parsley (Petroselinum crispum Miller) with a yield of
0.04% to 0.4%; the volatile oil contains monoterpenes including
pinene alpha ~22%, 1,3,8-​ p-​
menthatriene ~21%, pinene beta
~17%, phellandrene beta ~12%, myrcene ~5% and myristicin 5%
to 17% (the real value may be lower because the high value for
myristicin is, rather, from seed oils).
Here, the safety evaluation must clearly be done with the con-
generic groups, since none of the constituents are in Annex III,
but myristicin has been considered for safety evaluation (JORF,
2014). About 2 g of parsley leaves are used in a “bouquet garni”,
which contains at max 8 mg of EO with about 80% (6.4 mg) of
monoterpenes (congeneric group 19) and at most 1.36 mg of
myristicin (congeneric group 23).
For EO use, dilution should first be performed to 1% in ethanol
to ensure further solubilization in aqueous media; the application
in the final product is then to be tested at 0.1–​0.8%.
The large palette of EOs obtained from aromatic plants is being
used more and more in the 21st century.
The natural trend associated with the search for well-​being
increases the direct use by consumers of different EOs for mul-
tiple purposes, other than to flavour food.
On the other hand, highly concentrated ingredients of this kind
need to be handled with care by users.
In this overview, we tried to highlight some concerns. We have
to taste any EO step by step, starting with the highest dilution, to
evaluate the sensorial impact before fixing a value in recipes. In
addition, the main rationale is to keep in mind the obtained yield
of an EO. If we are using a few milligrams of an EO, and this
small quantity represents two or three leaves, we keep in mind the
right proportion in the preparation.
Essential Oils: How to Safely Use
In this context of awareness, EOs will become among the key
elements of a sustainable gastronomy, with material extracted
from the field and concentrated to avoid losses and the carbon
footprint of transferring fresh herbs all around the globe.
REFERENCES
Ames B. 1990. Dietary pesticides (99.99 % all natural), Proceedings
of the National Academy of Sciences USA, 87, 7777–​7781.
Aprifel. 2019. Composition nutritionnelle du persil, www.aprifel.
com/​fr/​fiche-​nutritionnelle/​persil
Arctander SJ, Arctander S. 1969. Perfume and Flavor Chemicals
(Aroma Chemicals). Montclair, NJ (Currently available from
Allured Publishing Corp.).
Cohen SM, Eisenbrand G, Fukushi S, Gooderhamd NJ, Guengerich
FP, Hecht SS, Rietjens YMCM, Bastak M, Davidsen JM,
Harman CL, McGowen M, Taylor SV. 2018. Updated pro-
cedure for the safety evaluation of natural flavor complexes
used as ingredients in food, Food and Chemical Toxicology,
113, 171–​178.
Cohen SM, Eisenbrand G, Fukushi S, Gooderhamd NJ, Guengerich
FP, Hecht SS, Rietjens YMCM, Bastak M, Davidsen JM,
Harman CL, McGowen M, Taylor SV. 2019. FEMA GRAS
assessment of natural flavor complexes: Citrus-​ derived
flavoring ingredients, Food and Chemical Toxicology, 124,
192–​218.
Demyttenaere JCR. 2012. The new European Union Flavouring
Regulation and its impact on essential oils: Production of
279
natural flavouring ingredients and maximum levels of restricted
substances, Flavour and Fragrance Journal, 27, 3–​12.
Doroknov Y, Sheshukova EV, Komarova TV. 2018. Methanol in plant
life, Frontiers in Plant Science, 9, 1623.
European Community. 2018. Annex III of the regulation # 1334/​2008
of the European Parliament and of the Council of 16 December
2008 on flavorings and certain food ingredients with flavoring
properties for use in and on foods.
FDA. 2019. CFR –​Code of Federal Regulations, Title 21, 182 (10,
20, 40, 50).
JORF. 2014. Arrêté du 24 juin 2014 établissant la liste des plantes,
autres que les champignons, autorisées dans les compléments
alimentaires et les conditions de leur emploi, www.legifrance.
gouv.fr/​affichTexte.do?cidTexte=JORFTEXT000029254516
&categorieLien=id
Van Hoven W. 1984. Tannins and digestibility in greater kudu,
Canadian Journal of Animal Science, 64, 177–​178.
Selosse MA. 2019. Les goûts et les couleurs du monde. Une histoire
naturelle des tannins, de l’écologie à la santé, Actes Sud.
Stahl-​Biskup E and Venskutonis RP. 2004. Thyme, Handbook of
Herbs and Spices (Peter KV ed.), vol. 2, ­chapter 19, 310.
Taylor SV. 2016. Handbook of Essential Oils, Science, Technology,
and Applications, Baser KHC, Buchbauer G (eds.), CRC
Press, 8, 229.
WHO. 2018. Natural toxins in food, www.who.int/​news-​room/​fact-​
sheets/​detail/​natural-​toxins-​in-​food
Evaporation

Hervé This vo Kientza1,2

1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France

Water is the dominant compound of most foods, and the
vapour pressure of water is non-​ zero at room temperature,
increasing until it reaches 1 atmosphere at boiling temperature
(McQuarrie, 1997). Hence, it is not surprising that water evap-
oration occurs in almost all culinary circumstances. However,
other compounds from food ingredients can evaporate as well
during cooking, in particular when the temperature is increased.
From a thermodynamic point of view, evaporation is simple,
but the fact that most food is not stable thermodynamically but
only metastable, with different time constants, makes it unreal-
istic to stick to the first-​order equilibrium description. In this
chapter, we consider some examples of evaporation, ranked by
order of complexity. Many descriptions were found in articles,
which are quoted.
Ordering Systems
The dispersed system formalism (DSF; see chapter on this topic
in this book) can usefully indicate an order for discussing such a
broad question as evaporation from food (This, 2012).
Using a simple computer code for producing the list of all
possibilities, one would obviously get first the group G (gas), O
(“oil”, i.e., any liquid fat), S (solid), W (“water”, i.e., any aqueous
solution). But here only the W and O cases are of interest; they
correspond to simple solutions, with water and oil as solvents and
various solutes dissolved in them.
Hydrophobic Solutions: O
In physical chemistry, and in particular in food science and tech-
nology, the letter O stands for “oil”, i.e., hydrophobic liquids,
which can be mainly made up of either triglycerides or other
compounds, such as in “essential oils”.
Triglycerides have a low vapour pressure; for example, gly-
ceryl trioleate has a vapour pressure of 1.15 × 10−9 mm Hg at
25 °C (estimation US EPA), i.e., 600 billion times lower than
atmospheric pressure, and decomposition occurs before boiling.
However, odorant compounds dissolved in such liquid fat
can evaporate, because they often have small molecules weakly
associated with the matrix (here, O) through van der Waals bonds
(Figure 39.1). The flavour of food oils has been much studied, in
particular for mass-​scale products such as olive oil (Kiritsakis,
1998). The increase of the molecular kinetic energy with tem-
perature increases the evaporation of odorants, increasing as well
the intensity of odour, until chemical degradation occurs.
Essential oils, on the other hand, include either one primary
odorant or many, but all have small molecules with sometimes
high vapour pressure (for example, eugenol and octanol have
a vapour pressure of 0.01 and 0.08 mm Hg, respectively, at
20 °C) (Lide, 2003). Their extraction is discussed in the chapter
“Essential Oils” by Eric Angelini and Laure Dziuba. Here, let
it be enough to observe that in principle, the evaporation of any

Food Liquids: O, S
Food liquids are mainly aqueous solutions, including those with a
high concentration of ethanol, or oils, the latter dissolving various
hydrophobic compounds.
For all such systems, a “matrix effect” can be considered (see
the chapter “Food Matrices and Matrix Effect in the Kitchen”).
Here, we are interested in the various systems from which evap-
oration occurs.
FIGURE 39.1 A useful scale of energy for the study of evaporation.

281
282 Hervé This vo Kientza

FIGURE 39.2 The effects of evaporating essential oils on the indoor total volatile organic content in (a) a bedroom (volume 21.6 m3) and (b) an office
(volume 28.2 m3). Three hundred microlitres of each essential oil (Lavandula angustifolia, Eucalyptus globulus and Melaleuca alternifolia) was diluted with
50 mL of water for use in an incense evaporator with a burning candle.
(Su et al., 2007)

component of an essential oil increases with temperature as it
would do if this component were dissolved in ordinary oil. Of
course, the evaporation constants depend on the forces between
the molecules of interest and the solvent. This explains why
perfumers distinguish top notes, middle notes and base notes,
i.e., concepts that are also useful in the kitchen. Hence, the study
of the evaporation of essential oils in various environments is
helpful to predict the behaviour of such products as well as the
effect on consumers (see the two chapters on essential oils in this
book) (Figure 39.2).
gas and liquid phases is determined by the vapour֪–​liquid equi-
librium (VLE) of each individual odorant, which is characterized
by the partition coefficient (ki) between the vapour phase and the
liquid phase. This coefficient is expressed by the ratio between
compound concentrations in the gas phase and in the liquid phase
in the sample at equilibrium:
cigas (39.1)
ki =
ciliquid

Aqueous Solutions The partition coefficient can be expressed as a molar fraction,

The case of aqueous solutions includes beverages (e.g., fruits also called absolute volatility (Ki), or by the activity coefficient
juices, wines and beers), and also culinary systems such as (γi). Both parameters are related as follows:
broths or stocks. Here, we propose to rank the questions by
γ . P (T )
increasing order of use, as serving temperature can change the °
y
flavour perception considerably (Steen et al., 2017; Feron and Ki = i = i i (39.2)
xi PT
Salles, 2018).
For aqueous solutions, again, volatility and solubility depend
on the interactions between odorant compounds and solvents, but where xi and yi are the molar fractions in the liquid and gas phases,
also between the various solutes (some, such as proteins, can form respectively, Pi(T)◦ is the vapour pressure of pure component i at a
complexes). The distribution of odorant molecules between the given temperature T, and PT is the total pressure.
Evaporation
When volatile compounds in food and beverages are near
infinite dilution (e.g., concentrations lower than 10−4 mole
fraction), the activity coefficient can be considered constant and
is noted as γi∞. The product γi∞.P◦i(T) is Henry’s constant.
Partitioning of volatile substances between the liquid and gas
phases is mainly governed by odorant compound volatility and
solubility. These physicochemical properties are expected to be
influenced by constituents present in the medium, for instance
polysaccharides, polyphenols and proteins, among others (Pozo-​
Bayon et al., 2008). Consideration of the physicochemical
interactions that occur between odorant compounds and the
constituents of liquids is necessary to understand the perception
of ante-​or retronasal odour during consumption. The binding that
occurs at a molecular level reflects changes at a macroscopic level
of the thermodynamic equilibrium, such as volatility and solu-
bility, or changes in kinetic phenomena. Thus, thermodynamic
and dynamic approaches can be used to study the behaviour of
odorant compounds in simple (model) or complex (foods) media
(Pozo-​Bayon and Reineccius, 2009).
Aqueous Solutions at Room Temperatures or in
the Mouth
The sensory effect of food is based on various elementary
perceptions, i.e., name, shape, colour, visual texture, consist-
ency, taste, ortho-​and retronasal odour, trigeminal sensations,
temperature perception and others. It is said that the number of
elementary sensations and their interactions makes the formula-
tion of foods and drinks an art (Williams, 1992), but there is con-
fusion between the “art” of formulation, i.e., an empirical activity
based on scientific knowledge, and culinary art, i.e., the produc-
tion of “good” products, where “good” means “beautiful to eat”,
and this means that the production of good food has (almost)
nothing to do with the technical question.
Concerning the latter, the physical and chemical processes con-
trolling the release of odorant, trigeminal and taste compounds
from foods and drinks in the mouth has been studied, and advances
in sensory physiology have been a good basis for designing food.
In particular, taste, part of the trigeminal sensation (Guichard
et al., 2017), “calcium sensation” (Tordoff et al., 2012) and the
perception of long chain unsaturated fatty acids (sometimes
called oleogustation) (Laugerette et al., 2005) are determined by
the stimulation of receptors on the surface of the oral cavity (pri-
marily the tongue) by compounds dissolved in an liquid medium
(mainly water or oil). Odour and part of the trigeminal sensa-
tion, on the other hand, are determined by receptors in the olfac-
tory epithelium and require that odorant compounds be volatile
enough to be carried there in the gas phase during exhalation.
Modelling the odorant sensation requires considering the
physical effects of mixing and diluting the aqueous s­olution
with saliva at a (frequently) different temperature from that
of the food, swallowing and breathing, as well as possible
partitioning of odorant compounds with mucus in the throat and
losing odorant compounds into the lungs and blood. Since the
1990s, volatile release has been measured directly in the mouth
(Linforth and Taylor, 1993), and more and more data have been
acquired, in particular from various liquid systems, including
283
emulsions (Harrison and Hills, 1996), soft drinks (Harrison and
Hills, 1997) and protein solutions (Bakker et al., 1998; Andriot
et al., 2000).
In 2001, Wright et al. (2002) modelled periodic breathing
and the associated persistence effect often associated with par-
ticular odorant compounds, which can result in detectable odour
even after the material has been removed from the mouth by
swallowing. It was shown that, despite the complexity of phe-
nomena, a model can predict the breath-​by-​breath concentration
of odorants in exhaled air. The major limitation of the models is
reliance on empirical or experimental values for persistence ratio,
defined as the ratio of the odorant concentration in the second and
the first exhalation.
Aqueous Solutions Being Heated
Stock and Reductions
In the kitchen, evaporation occurs each time a liquid is heated,
such as during stock and broth making, but also in more complex
processes during which plant or animal tissues are heated in an
aqueous solution, such as daubes, sautés, ragouts, stews, etc. It is
the main phenomenon when the goal is to “reduce” (the volume),
as for sauce preparation or the making of fond, fumets, essences,
demi-​glaces and glaces.
It is indeed surprising that for a long time, the issue of losing
odorant compounds along with water (in particular by steam dis-
tillation) was not considered in kitchens, except practically by the
use of lids. However, it is now known that odorant compounds
can disappear partially or entirely from the food being heated.
Examples are many, but one is the comparison of the odorant in
fresh tomatoes and in tomato paste prepared by heating: Kelebek
et al. (2018) determined that the main volatiles are alcohols,
aldehydes, lactones, carboxylic acids, ketones, furans, esters,
volatile phenols, 13-​C norisoprenoid, terpenes and pyrroles. Via
aroma extract dilution analysis, a total of 21 and 13 key odorants
were detected in tomato and its pastes, respectively. The most
important differences of odorant compounds in tomatoes and
their pastes were in hexanal, (Z)-​3-​hexenal, 2,3-​butanediol, (Z)-​
3-​hexenol, (E)-​2-​octenal, benzaldehyde, (E,Z)-​2,6-​nonadienal,
methyl salicylate, β-​ionone, 5-​pentyl-​2-​(5H)-​furanone and
eugenol, which were not detected in tomato paste samples.
Many phenomena occur simultaneously during such processes,
including evaporation and steam extraction, such as in wine boiled
for sauces, stocks for fond, glaces or demi glaces, for example.
For processes of this kind, the phenomena seem to have been
known for a long time, with studies of the increase of the saturation
pressure, for example. However, in the out-​of-​equilibrium con-
text of cooking, the issue now is to know how far the phenomena
are different from those described by classical thermodynamics,
in particular when the solute concentration increases.
Here, physical phenomena are complemented by chem-
ical ones: during concentration processes, the thermal activa-
tion can lead to chemical changes (condensation or degradation
of solutes). For example, sucrose in acidic solution can be
hydrolysed into D-​glucose and D-​fructose, in particular when
thermal processing is performed for many hours (Cazor et al.,
284
2006), and odorant compounds such as 5-​hydroxymethylfurfural
can be produced through hexose dehydration (Fu et al., 2019).
For solutions obtained through the thermal treatment of animal
tissues in water, the modification of odorants was analysed (the
comparison of the same beef meat stock being concentrated ther-
mally, and later reconstituted with water, with the initial stock
shows the appearance of new chemical species, contributing to
flavour (This, 2019).
The preparation of meat stock involves cooking meat, bones,
vegetables and herbs in water (Bocuse, 1976). During this pro-
cess, compounds from the animal or plant tissues move to the
water phase, with or without chemical modifications; they are
finally responsible for the flavour and mouthfeel of the aqueous
solution called stock or broth (Cambero et al., 1992; Cambero
et al., 2000). These two kinds of products –​stocks and broths –​
differ because broth is any liquid that has had meat cooked in it,
whereas stock always involves bones, simmered for a long time
(roasting the bones makes a liquid with more colour and different
flavour) (Montagné, 1994).
The non-​volatile fractions extracted from animal tissues are
important taste compounds in stocks (Cambero et al., 2000);
some can also be precursors of odorant compounds (Hornstein
and Wasserman, 1987; Mottram, 1998). After separation of meat
and stock, the stock is often reduced (boiled down) in order to
enhance flavour and change consistency (Montagné, 2001). If
greatly reduced, the stock turns into a thick syrupy liquid, a so-​
called glaze, added to soups or sauces (Montagné, 1994).
The flavour of animal tissue thermally treated in water has
been much studied for a very long time (Brinkman et al., 1972;
Hornstein and Wasserman, 1987; Mottram, 1998; This, 2009).
Several studies have investigated the effects of different
cooking parameters, e.g., time and temperature, when preparing
beef stocks (Cambero et al., 2000).
As reported by This (1992), an increase in flavour intensity
takes place during reduction, despite the fact that odorants are
removed along with water, as evidenced by an intense smell in
the kitchen during reduction. Snitkjaer et al. (2010) studied such
development during the reduction process of beef (Bos taurus)
stocks.
The authors studied the effects of time and concentration
factors on the sensory properties and the volatile compounds
during the reduction process. They further related this to the
preparation techniques used by chefs. The volatile compounds
were monitored by gas chromatography-​mass spectrometry (GC-​
MS) at various times during the thermal processing of animal
tissues in water including dissolved sodium chloride. The relative
changes in peak area were used to study the formation and decay
of volatiles during the reduction process. In this way, 61 volatile
compounds were identified, including 20 aldehydes, 11 ketones,
12 alcohols, 9 furans/​furanones, 2 carboxylic acids, 6 compounds
belonging to other categories and 1 unidentified compound. Most
are known as typical beef odorant compounds (Hornstein and
Wasserman, 1987; Mottram, 1998; Melton, 1998).
The changes in odorant compounds with reduction time
were studied with principal component analysis, and clusters
(groups denoted by letters A–​F) of odorant compounds have been
identified.
Hervé This vo Kientza
Some volatiles (2-​ propanone, unknown ester) gener-
ally decrease during reduction, though with a small sudden
increase after approximately 16–​ 21 h of reduction. The
concentrations of odorant compounds in group B (2-​pentanone,
3-​hydroxy-​2-​butanone, 1-​octen-​3-​ol and 2-​ethyl-​1-​hexanol) also
decrease during reduction, but mainly after the first 16 h. The
concentrations of compounds in group C peak after 16 h of reduc-
tion, after which they decrease (e.g., 1-​butanol). There might be
compounds formed and lost within this time period that are not
detected. Another large group (E) of compounds has a similar
progression, but the concentration peaks after 20 h of reduction
(e.g., nonanal). The concentration of compounds in group D has
a broad peak between 16 h and 20 h of reduction, and the pro-
gression as a function of time is therefore similar to groups C
and E. Compounds in group F increase with time (e.g., 3-​methyl-​
butanal and cetic acid), the compounds of this group being the
most abundant in the most reduced samples.
The distribution of the chemical categories as a function of time
is the following: ketones and alcohols are dominant in groups A–​
D, since they either decrease from the initial concentration during
reduction or peak in concentration some time before 20 h of
reduction; the aldehydes, except for the two Strecker aldehydes
(2-​methyl-​butanal, 3-​methyl-​but-​anal), all peak in concentration
at some point during 16–​20 h of reduction; most furans peak in
concentration at 20 h of reduction; the two Strecker aldehydes
and the two acids (acetic acid and 3-​methyl-​butanoic acid) gener-
ally increase during the 22 h of reduction.
In order to investigate the importance of the reduction pro-
cess for the perceived flavour, a descriptive sensory analysis was
performed, with meat descriptors (MacLeod and Seyyedain-​
Ardebelli, 1980) and one for a tar odour. The intensities of nutty,
sweet, beef and chicken flavours decreased during the reduction.
Tar odour increased until 16 h, followed by a decrease. There was
a general increase in burned, bitter, astringent and sour flavour
intensities during the reduction. The main increase was from 0 to
16 h and for most descriptors, further, from 21 to 22 h.
Steam Extraction
Concerning steam extraction, one is invited to look at the chapter
on essential oils in this book, in which production processes of
odorant compounds are discussed. Complementary information
is in the chapter on distillation.
More Complex Systems Having a Continuous
Liquid Phase
For establishing the second level of complexity of food systems,
with biphasic systems, a ranking of operators has to be made.
This (2012) showed that, using free enthalpy as a criterion for
complexity, the order should be σ, @, /​and x, so that the list is
(the most unstable systems, such as W σ O, are not given):
• G σ O, G σ S, G σ W, O σ G, O σ S, O σ W, S σ G, S σ
O, S1 σ S2, S σ W, W σ S, W1 σ W2
• G@S, G@W, O@G, O@S, O@W, S@O, S1@S2,
S@W, W@O, W@S,
Evaporation
• G/​O, G/​S, G/​W, O/​G, O/​S, O/​W, S/​G, S/​O, S1/​S2, S/​W,
W/​G, W/​O, W/​S, W1/​W2
• GxS, OxG, OxS, OxW, S1xS2, SxW, WxS.
Of course, systems without water will not suffer water evapor-
ation, and the list can be reduced to:
• G σ W, O σ W, S σ W, W σ S, W1 σ W2
• G@W, O@W, S@W, W@O, W@S,
• G/​W, O/​W, S/​W, W/​G, W/​O, W/​S, W1/​W2
• OxW, SxW, W1xW2
Here, when water is a continuous external phase, the behaviour
should not be so different from that with a pure solution. This is
the case for
• W σ S, W1 σ W2
• G@W, O@W, S@W, W1@W2
• G/​W, O/​W, S/​W, W1/​W2
But for other systems, more thorough studies are still needed in
order to describe the phenomena correctly. For example, for an
SxW system, with two continuous phases intermixed, one could
ask whether the water evaporation is simply proportional to the
water contribution to the external surface of the whole system, or
how much the solid network is changing the water evaporation.
Such cases are important, because they are observed for gelatin
gels, for example, for which it is known that the flow of water is
not free.
Food science, and even more food technology, recognize so
well the importance of water evaporation from such systems
that a measurement of it was devised as “water activity” (Belitz
et al., 2009).
Water Activity
In thermodynamics, the activity a of a compound is a measure
of the “effective concentration” of a species i in a mixture, in
 ∂G 
the sense that the species’ chemical potential µ i =  ∂n 
P ,T , n
j ≠i
depends on the activity of a real solution in the same way as it
would depend on concentration for an ideal solution (here G is
the free enthalpy, and n the number of moles of the compound).
The expression of the activity of the species i is:
µ i − µ °i
ai = e RT (39.3)
where R is the constant of ideal gas law, and T the absolute
temperature.
The activity coefficient relates the activity to a measured
concentration c:
ci
ai = γ i (39.4)
c°
where C° is the standard amount concentration.
285
In 1952, Scott concluded that the storage quality of food does
not depend on the water content but on water activity (aw), which
was defined as follows:
aw = P / P0 = ERH / 100 (39.5)
where P is the partial vapour pressure of food moisture at tem-
perature T, P0 the saturation vapour pressure of pure water at T,
ERH the equilibrium relative humidity at T.
For a food system, the relationship between water content and
water activity in a food is represented by a “sorption isotherm”.
The desorption isotherm lies above the adsorption isotherm
pertaining to the storage of moisture-​sensitive food.
Decreased water activity retards the growth of microorganisms
(foods with aw values between 0.6 and 0.9 are “intermediate mois-
ture foods”, or IMF, largely protected against microbial spoilage),
slows enzyme-​ catalysed reactions (particularly involving
hydrolases EC 3) and retards non-​enzymatic browning. In con-
trast, the rate of lipid autoxidation increases in dried food systems
(see the chapter “Oxidation of Dietary Lipids”, by L. Eveleigh).
According to the Le Chatelier principle, any reaction occurring
with the elimination of a water molecule is also favoured.
Additives with high water binding capacities (humectants)
decrease the water activity. For example, the food industry uses
propylene glycol, hexylene glycol and butylene glycol, alpha
hydroxy acids such as lactic acid, egg yolk and egg white, gly-
ceryl triacetate, honey, molasses, polymeric polyols such as
polydextrose, sodium hexametaphosphate, sugar alcohols (sugar
polyols) such as glycerol, sorbitol, xylitol and maltitol, and urea.
Some are also sweeteners and would change the flavour, but
there is no real reason why the food industry could use them and
not chefs.
Evaporation from Emulsions
Let us now consider one of the first disperse systems in terms of
DSF complexity, namely emulsions. They are defined as “a fluid
colloidal system in which liquid droplets and/​or liquid crystals
are dispersed in a liquid. The droplets often exceed the usual
limits for colloids in size” (IUPAC, 1997). Formerly, the first
and the second liquids being considered would have been said to
be non-​miscible, but Ramsden systems can include one aqueous
phase dispersed into another aqueous phase (see the chapter on
Ostwald ripening and disproportionation).
The two liquid phases of an emulsion can release compounds in
a gas above the whole system. The oil phase acts as a flavour pre-
cursor, a flavour reservoir and carrier, and a flavour release mod-
erator. The fat content has been reported to influence qualitative,
quantitative and temporal perception of flavour in food products
(Tuorila et al., 1995). However, water-​soluble compounds can
also have a large flavour impact (e.g., ethanol). So, the question
is how the composition of both phases and the structure of the
emulsion determine the overall flavour.
The odour of emulsions is due to volatile molecules contained
in the vapour phase above them (Overbosch et al., 2009; Taylor
and Linforth, 1996), but conflicting results were reported on the
effect of the fat content on the release of odorant compounds
286 Hervé This vo Kientza

in and above emulsions and the resulting odour intensity. The Flavour Release for More Complex Systems
release of diacetyl, measured by headspace techniques on model
In order to further elucidate the release of odorants from com-
oil-​in-​water emulsions, appeared higher from high-​fat systems
plex food, it is necessary to combine instrumental analysis of the
than from low-​fat systems (Salvador et al., 1994). Wendin et al.
release of volatile compounds from the food matrix with sensory
(1997) reported that increasing the fat content of sour milks
evaluation of the food product.
increased the perceived flavour of polar odorant compounds but
As food is eaten, the release of the volatiles is influenced by
had no effect on the flavour of non-​polar odorant compounds.
a number of factors including mastication, mixing with saliva,
In the case of reduced-​fat mayonnaises, Wendin et al. (1997)
and changes in temperature and pH. Therefore, to gain an insight
reported the opposite effect of fat content on the flavour of either
into the kinetics of volatile release, in vivo techniques have been
polar or semipolar odorant compounds. And increasing the fat
developed, which are capable of monitoring real-​ time vola-
content of emulsions (700 to 820 g/​kg) was reported as having no
tile release. Atmospheric-​pressure-​chemical-​ionization mass
significant effect on the perception of flavour.
spectrometry (APCI-​ MS) and proton-​ transfer-​
reaction mass
The literature appears confusing, but the hydrophobicity of the
spectrometry (PTR-​MS) are used for this purpose. With these
odorant compounds, as expressed by the logarithm of the par-
systems, part of the participants’ breath is continuously sampled,
tition coefficient (log(P)), seems to be a key factor in the equi-
allowing sensitive and fast monitoring of volatile release (Taylor
librium between the emulsion and the vapour phase. Under the
and Linforth, 1996).
assumption that an odorant compound does not interact with the
In-​nose measurements of volatile compounds released from
interface, odorant partition coefficients between liquid emulsions
hydrocolloid gel systems have been the focus of many studies.
and air can be predicted, provided that air–​liquid and oil–​water
Using APCI-​MS release curves, Baek et al. (1999) confirmed
partition coefficients are available (Overbosch et al., 2009).
what chefs know, i.e., that softer gelatin gels released larger
At a constant bulk concentration of the odorant compound, an
concentrations of a volatile compound than harder gelatin
increase of the oil volume fraction should lower hydrophobic
gels. Flavour release from mixed phase gels demonstrated that
odorant compounds in the vapour phase and slightly enhance
both the properties of the volatile compound and the particular
polar components. Data from Guyot et al. (1996) dealing with
matrix influenced volatile release in vivo (Wright et al., 2002).
flavoured oil-​in-​water emulsions support this prediction, except
Hollowood et al. (2002) examined the release of volatiles from
for butyric acid. This compound is suspected to bind strongly to
a strawberry flavour mix in a hydroxypropylmethylcellulose
the interface of droplets, which results in reduced apparent par-
system, while Weel et al. (2002) analysed the release of diacetyl
tition coefficient. Wedzicha (1988) proposed that emulsifiers
and ethyl butyrate from whey protein gels. Both studies concluded
(surfactants or proteins) located at the interface could modify the
that gel texture did not influence in-​nose flavour concentration,
partition of odorant compounds in emulsions.
but that it had an impact on flavor perception. Lethuaut et al.
Charles et al. (2000) studied the effect of process and formu-
(2003) observed that the perceived firmness of model dairy
lation on sensory perception and odour release in salad dressing
desserts increased with increased sucrose concentration and that
models. Oil/​vinegar emulsions (volume fraction φ > 0.5, droplet
increasing the firmness of the dessert reduced sweetness percep-
size >10 μm) with thickeners and a whey protein concentrate were
tion. These authors also discovered that variations in sweetness
prepared with different fat droplet sizes and different distributions
perception could only partly be explained by changes in mechan-
of fat droplet size. The effect of the amount of emulsifier was
ical profiles and that gel strength was not the only factor leading
also tested. Sensory profile analysis was performed by a trained
to sweetness intensity modification.
panel, and flavour release was measured by dynamic headspace
However, questions remain: to what extent do interactions
analysis. When the droplet size was increased, the lemon and
between the texture and flavour of a food influence its texture
citrus odours significantly increased, whereas the egg, mustard
and flavour perception, and how do these interactions occur?
and butter odours significantly decreased. The concentrations
Boland et al. (2006) investigated how changes in gelatin and
of alcohols and acids significantly increased when droplet size
pectin gel texture influenced the static headspace concentrations
increased, whereas those of other compounds such as limonene
of strawberry volatiles, release of these volatiles in vivo, and sen-
or benzaldehyde significantly decreased. The dispersion of the
sory perception. In both gel systems and for all gel textures, as
droplet size had a small effect on odour perception, and the effect
the chain length of esters increased (ethyl iso-​pentanoate C-​7,
of the increase of the amount of emulsifier was noticed only by
ethyl hexanoate C-​8, cis-​3-​hexenyl acetate C-​8, methyl anthra-
instrumental analysis.
nilate C-​8, benzyl acetate C-​9, styrallyl acetate C-​10), there was a
In 2001, Brossard et al. (2001) showed that at equilibrium,
decrease in the air/​gel partition coefficients. These data show the
the oil volume fraction of the emulsion (from 0.12 to 0.48)
higher affinity of larger, more hydrophobic flavour compounds
does not influence the odour intensity of odorant compounds
for the gel matrices. This may be due to binding/​trapping of the
dissolved in the emulsion (benzaldehyde, ethylbutyrate, linalool
flavour compounds.
and acetophenone). It was confirmed that the perceived odour
In conclusion, both the rigidity of the gels and the type of
intensity is governed by the odorant concentration in the aqueous
hydrocolloids have a considerable effect on the gels’ flavour
phase at the time of the trial and not by the averaged apparent
release and/​or perception.
concentration in the emulsion.
Evaporation
Crust Formation
Water evaporation is important in drying, a process that is trad-
itionally used in order to preserve and store food ingredients
(in particular plant and animal tissues) for longer periods by
removing some of their moisture content. This process involves
simultaneous heat and mass transfer under transient conditions.
A number of internal and external parameters influence drying
behaviour. External parameters include temperature, velocity
and relative humidity of the drying medium (air), while internal
parameters include density, permeability, porosity, sorption–​
desorption characteristics and thermophysical properties of the
material being dried.
Studies that have investigated the effects of the above-​
mentioned parameters on the drying process of different fruits
and vegetables are too many to list, but one could cite the drying
of apples by Sacilik and Elicin (2005) and by Velic et al. (2004);
of tropical fruits by Karim and Hawlader (2005); of kiwis by
Simal et al. (2004); of potatoes and carrots by Srikiatden and
Roberts (2004); of vegetables by Krokida et al. (2003); of prickly
pear by Lahsasni et al. (2004); of figs by Babalis and Belessiotis
(2004); of chestnuts by Guine and Fernandes (2006); of pistachio
nuts by Kashaninejad et al. (2007); and of quince by Kaya et al.
(2007).
Often, the drying kinetics was followed as a function of
drying conditions, and the apparent diffusion coefficients were
evaluated, as well as the effects of the temperature, velocity
and relative humidity of the drying air on the drying kinetics.
However, dried systems are generally out of equilibrium, and a
crust is formed. Of course, the higher the temperature, the more
important the crust (a crude calculation of the thickness of the
crust of soufflés is proposed in the chapter “Simple Calculation
Based on Cooking” in Part II of this book).
Water Vaporization and Frying
In the drying studies, the temperature is often below the boiling
point of water, but with frying, phenomena are very different.
Frying is an operation used to create unique flavours and textures
in foods. Deep-​fat or immersion frying involves simultaneous
heat and mass transfer, resulting in counter-​flow of water vapour
(bubbles) and oil at the surface of the food, physicochemical
alterations of major food components, and significant microstruc-
tural changes (This, 2003). In fact, the desirable characteristics of
most fried foods are derived from the formation of a composite
structure: a dry, crisp and oily outer layer or crust, and a moist,
cooked interior or core.
The structure of the crust of fried potato products is the result
of several alterations, most of which occur at the cellular and
subcellular level. They take place in the outer cell layers of the
product, and the major changes include: physical damage due
to cutting (slicing), release of contents of broken cells, gelat-
inization of starch granules, softening of cell walls, fast dehy-
dration of tissues, and formation and release of vapour bubbles.
Microstructural changes in the core are much milder and similar
to those occurring during the cooking of potatoes, the major ones
being hydration and swelling of starch granules and softening of
287
the middle lamellae through pectin beta-​elimination. For a dis-
cussion of this process, see the chapter “Cooking”.
Conclusion
Not only are food ingredients out of equilibrium thermodynam-
ically speaking, but increasing the temperature of the assembled
systems during culinary processes increases the release of water
and other volatile compounds, including odorants. It is strange
that, in spite of numerous studies of distillation and, more gener-
ally, extraction of odorant compounds (in particular for the pro-
duction of essential oils), so few scientific studies have focused
on the phenomena responsible for the release of compounds
during culinary processes. Now, the time is ripe.
REFERENCES
Andriot I, Harrison M, Fournier N, Guichard E. 2000. Interactions
between methyl ketones and b-​lactoglobulin: sensory analysis,
headspace analysis, and mathematical modeling, Journal of
Agricultural and Food Chemistry, 48, 4246–​4251.
Babalis JS, Belessiotis VG. 2004. Influence of the drying conditions
on the drying constants and moisture diffusivity during the
thin-​layer drying of figs, Journal of Food Engineering, 65,
449–​458.
Baek I, Linforth R, Blake A, Taylor A. 1999. Sensory perception is
related to the rate of change of volatile concentration in-​nose
during eating of model gels, Chemical Senses, 24(2), 155–​160.
Belitz HD, Grosch W, Schieberle P. 2009. Food Chemistry (4th Ed),
Springer Verlag, Berlin-​Heidelberg, 723.
Bocuse P. 1976. La cuisine du marché, Flammarion, Paris.
Boland AB, Delahunty CM, van Ruth SP. 2006. Influence of the
texture of gelatin gels and pectin gels on strawberry flavour
release and perception, Food Chemistry, 96, 452–​460.
Brinkman HW, Copier H, de Leuw JJM, Tjan SB. 1972. Components
contributing to beef flavor. Journal of Agricultural and Food
Chemistry, 20(2), 177–​181.
Brossard C, Rousseau F, Llamas G, Dumont JP. 2001. Odor per-
ception over liquid emulsions containing single aroma
compounds: effects of aroma concentration and oil volume
fraction. Journal of Sensory Studies, 17, 511–​525.
Cambero MI, Seuss I, Honikel KO. 1992. Flavor compounds of beef
broth as affected by cooking temperature. Journal of Food
Science, 57(6), 1285–​1290.
Cambero MI, Pereira-​Lima CI, Ordonez JA, Gardia de Fernando
GD. 2000. Beef broth flavour: relation of components with the
flavour developed at different cooking temperatures. Journal
of the Science of Food and Agriculture, 80, 1519–​1528.
Cazor A, Deborde C, Moing A, Rolin D, This H. 2006. Sucrose, glu-
cose and fructose extraction in aqueous carrot root extracts
prepared at different temperatures by means of direct NMR
measurements. Journal of Agricultural and Food Chemistry,
54, 4681–​4686.
Charles M, Rosselin R, Beck L, Sauvageot F, Guichard E. 2000.
Flavor release from salad dressings: Sensory and physico-
chemical approaches in relation with the structure, Journal of
Agricultural and Food Chemistry, 48, 1810−1816.
Feron G, Salles C. 2018. Food oral processing in humans: Links
between physiological parameters, release of flavour stimuli
and flavour perception of food. International Journal of Food
Studies, 7, 1–​12.
288
Fu X, Hu Y, Zhang Y, Zhang Y, Tang D, Zhu L, Hu C. 2019. Solvent
effects on degradative condensation side reactions of fruc-
tose in its initial conversion to 5-​ hydroxymethylfurfural,
ChemSusChem, 13, 501–​512.
Guichard E, Salles C, Morzel C, Le Bon AM. 2017. Flavour, from
Food to Perception, John Wiley and Sons, Chichester, UK.
Guine RPF, Fernandes RMC. 2006. Analysis of the drying kinetics of
chestnuts, Journal of Food Engineering, 76(3), 460–​467.
Guyot C, Bonnafont C, Lesschaeve I, Issanchou S, Voilley A,
Spinnler HE. 1996. Effect of fat content on odor intensity of
three aroma compounds in model emulsions: d-​decalactone,
diacetyl, and butyric acid, Journal of Agricultural and Food
Chemistry, 44(8), 2341–​2348.
Harrison M, Hills BP. 1996. A mathematical model to describe fla-
vour release from gelatine gels, International Journal of Food
Science and Technology, 31, 167–​176.
Harrison M, Hills BP. 1997. Effects of air flow-​rate on flavour release
from liquid emulsions in the mouth, International Journal of
Food Science and Technology, 32, 1–​9.
Hollowood TA, Linforth RST, Taylor AJ. 2002. The effect of vis-
cosity on the perception of flavour, Chemical Senses, 27(7),
583–​591.
Karim MA, Hawlader MNA. 2005. Mathematical modeling
and experimental investigation of tropical fruits drying,
International Journal of Heat Mass Transfer, 48, 4914–​4925.
Kashaninejad M, Mortazavi A, Safekordi A, Tabil LG. 2007. Thin-​
layer drying characteristics and modeling of pistachio nuts,
Journal of Food Engineering, 78(1), 98–​108.
Kaya A, Aydin O, Demirtas C, Akgün M. 2007. An experimental
study on the drying kinetics of quince, Desalination, 212,
328–​343.
Kelebek H, Kesen S, Sonmezdag AS, Cetiner B, Kola O, Selli S.
2018. Characterization of the key aroma compounds in tomato
pastes as affected by hot and cold break process, Journal of
Food Measurement and Characterization, 12, 2461–​2474.
Kiritsakis AK. 1998. Flavor components of olive oil –​a review,
Journal of the American Oil Chemists’ Society, 75(6),
673–​681.
Krokida MK, Karathanos VT, Maroulis ZB, Kouris DM. 2003. Drying
kinetics of some vegetables, Journal of Food Engineering, 59,
391–​403.
Lahsasni S, Kouhila M, Mahrouz M, Jaouhari JT. 2004. Drying kin-
etics of prickly pear fruit (Opuntia ficus indica), Journal of
Food Engineering, 61, 173–​179.
Laugerette F, Passilly-​ Degrace P, Patris B, Niot I, Febbraio M,
Montmayeur JP, Besnard P. 2005. CD36 involvement in
orosensory detection of dietary lipids, spontaneous fat pref-
erence and digestive secretions, The Journal of Clinical
Investigation, 115, 3177–​3184.
Lethuaut L, Brossard C, Rousseau F, Bousseau B, Genot C. 2003.
Sweetness–​ texture interactions in model dairy desserts:
effect ofsucrose concentration and the carrageenan type,
International Dairy Journal, 13 (2003), 631–​641.
Lide DR. 2003. CRC Handbook of Chemistry and Physics, CRC
Press, Boca Raton (FL).
Linforth RST, Taylor AJ. 1993. Measurement of volatile release in
the mouth, Food Chemistry, 48, 115.
Litchfield BJ, Okos M. 1992. Moisture diffusion in pasta during
drying, Journal of Food Engineering, 17, 117–​142.
McLaughlin CP, Magee TRA. 1998.The determination of sorption
isotherm and the isosteric heat of sorption for potatoes, Journal
of Food Engineering, 35, 267–​280.
Hervé This vo Kientza
McQuarrie D. 1997. Physical Chemistry, a Molecular Approach.
University Science Book, Sausalito (CA).
Overbosch P, Afterof WGM, Haring PGM. 2009. Flavor release in
the mouth, Food Reviews International, 7(2), 137–​184.
Palipane KB, Driscoll RH. 1992. Moisture sorption characteristics
of in-​shell macadamia nuts, Journal of Food Engineering,
18, 63–​76.
Pozo-​Bayon MA, Biais B, Rampon V, Cayot N, Le Bail P. 2008.
Influence of complexation between amylose and a flavored
model sponge cake on the degree of aroma compound
release, Journal of Agricultural and Food Chemistry, 56(15),
6640–​6647.
Pozo-​Bayon MA, Reineccius G. 2009. Interactions beteen wine
matric macro-​components and aroma compounds. In Moreno-​
Arribas MV, Polo MC (eds.) Wine Chemistry and Biochemistry,
Springer, New York, USA.
Sacilik K, Elicin AK. 2005, The thin layer drying characteristics
of organic apple slices, Journal of Food Engineering, 73,
281–​289.
Salvador D, Bakker J, Langley KR, Potjewijd R, Martin A, Elmore JS.
1994. Flavour release of diacetyl from water, sunflower oil and
emulsions in model systems, Food Quality and Preferences,
5(1–​2), 103–​107.
Simal S, Femenia A, Garau MC, Rossello C. 2004 Use of expo-
nential, Page’s and diffusional models to simulate the drying
kinetics of kiwi fruit, Journal of Food Engineering, 66,
323–​328.
Snitkjær P, Frøst MB, Skibsted LH, Risbo M. 2010. Flavour devel-
opment during beef stock reduction, Food Chemistry, 122,
645–​655.
Srikiatden J, Roberts JS. 2004. Measuring moisture diffusivity
of potato and carrot (core and cortex) during convective hot
air and isothermal drying, Journal of Food Engineering, 74,
143–​152.
Steen I, Waehrens SS, Petersen MA, Münchow M, Bredie WLP.
2017. Influence of serving temperature on flavour percep-
tion and release of Bourbon Caturra coffee, Food Chemistry,
219, 61–​68.
Su HJ, Chao CJ, Chang HY, Wu PC. 2007. The effects of evaporating
essential oils on indoor air quality, Atmospheric Environment,
41(6), 1230–​1236.
Taylor AJ, Linforth RST. 1996. Flavour release in the mouth, Trends
in Food Science & Technology, Special Issue on Flavour
Perception, Elsevier Science, London, UK.
This H. 1992. Molecular Gastronomy Newsletter, July 1992.
This H. 2003. Matière grasse en cuisine: problème central de
gastronomie moléculaire. In Graille J (ed.) Lipides et corps
gras alimentaires, Lavoisier Tec et Doc, Paris, 189–​230.
This H. 2009. Histoires chimiques de bouillons et de pot-​au-​feu,
L’Actualité chimique, 346(1), 14–​16.
This H. 2012. Solutions are solutions, and gels are almost solutions,
Pure and Applied Chemistry, 85(1), 257–​276.
This H. 2019. New odorant compounds being produced during the
thermal treatment of animal tissues (Bos taurus) in water. To
be published.
Timoumi S, Zagrouba F. 2005. Water sorption and dehydration kin-
etics of Tunisian rosemary leaves, Desalination, 185, 517–​521.
Tordoff MG, Alarcon LK, Valmeki S, Jiang P. 2012. T1R3: A human
calcium taste receptor, Scientific Reports, 2(496), 1–​4.
Tuorila H, Sommardahl C, Hyvönen L. 1995. Does fat affect the
timing of flavour perceptions? A cade study with yoghurt, Food
Quality and Preferences, 6, 55–​58.
Evaporation
US EPA. 2011. Estimation Program Interface (EPI) Suite. Ver. 4.1.
January 2011, www.epa.gov/​oppt/​exposure/​pubs/​episuitedl.
htm
Velic D, Planinic M, Tomas S, Bilic M. 2004. Influence of airflow
velocity on kinetics of convection apple drying, Journal of
Food Engineering, 64, 97–​102.
289
Williams SD. 2003. Chemistry and Technology of the Cosmetics and
Toiletries Industry, Chapman and Hall, London.
Wright KW, Hills BP, Hollowood TA, Linforth RST, Taylor AJ.
2002. Persistence effects in flavour release from liquids in the
mouth, International Journal of Food Science and Technology,
38, 343–​350.
Expansion
Hervé This vo Kientza1,2
1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
During culinary processes, different kinds of expansion can be
observed. For example, bread fermentation (before baking) is
associated with a volume increase of the dough, because carbon
dioxide is formed through yeast metabolism (Belitz et al., 2014).
Also, during cooking, some cakes expand when they contain
leavening agents specifically selected to produce gases (see the
chapter on chemical leavening in this book).
Here, we focus, rather, on a different class of expansions that
occur during cooking, such as for soufflés, choux, puff pastry and
Durham popovers. For example, soufflés are preparations that
are traditionally made from a thick preparation (such as sauce
blanche, a cream patissière or a fruit purée) and whipped egg
whites; the two parts are mixed, and the mixture is poured into
ramekins and cooked in the oven at a temperature between 150 °C
and 250 °C for 20–​60 min depending on their size (Saint Ange,
1925). It is considered a success when the preparation expands.
But why does such an expansion occur?
Early Experiments
In the prehistory of molecular and physical gastronomy, i.e.,
before the scientific project of the discipline had not been clearly
identified and named (1988), the late Nicholas Kurti (Budapest,
1908–Oxford, 1998) studied soufflés experimentally, and he
even cooked a soufflé publicly during a lecture at the Royal
Institution of London. In the report of this lecture (Kurti, 1969),
he wrote:
The experiments with this roast leg of lamb taught me the
obvious thing, namely that continuous monitoring of the tem-
perature inside the dish one is preparing could be very useful.
So I started experimenting with various other dishes, not-
ably with soufflés, which are fairly tricky to prepare. There
is always some doubt when to take them out and I thought
that the normal method of peeping furtively into the oven, per-
haps shaking the dish a little bit to see whether the soufflé
still wobbles, or is firm enough, could be done away with if
one monitored the temperature. So I carried out quite a few
experiments with both savoury and sweet soufflés, and the
results of a typical run are shown in Figure 4, which gives the
temperature measured about 2 cm below the surface of the
mixture at the start. This is first of all a rapid rise, followed
by a levelling out, and in fact the temperature even drops a bit
because as the soufflé rises, cold parts from the bottom reach
the tip of the thermometer. After about 20 minutes, the rise in
temperature begins again and gets accentuated, and I found
that if one removes the soufflé when a temperature of about
70 °C is reached, it is done pretty well to perfection. About
ten minutes before the beginning of this lecture, we placed a
soufflé à la Chartreuse in the oven and connected the therm-
ometer to a chart-​recorder. When the right temperature is
reached we shall take it out of the oven and you will be able to
see whether the fact that we had a thermomoter in the soufflé
that we tried to judge the cooking time by a basic scientific
method in any way impaired its qualities or its taste. I think it
is a sad reflection on our civilization that while we can and do
measure the temperature in the atmosphere of Venus we do not
know what goes on inside our soufflés.
However, experiments that I did later in collaboration with
Kurti (in my laboratory, in Buc, France, or in Oxford, UK, after
1986) did not show the temperature behaviour that Kurti initially
observed (Figure 40.1), i.e., no “temperature drop” was detected.
For these cheese soufflés and for all the others that were
studied in the following years (This, 2002), the recipe and the
protocol were constant:
1. a “roux” was prepared with 75 g flour and 75 g butter;
2. the mixture was heated up to 100 °C, and, when a light
brown colour was obtained (this is also described more
technically in terms of heat transfer and heating time),
400 mL of milk was added;
3. after this “bechamel” was thickened by slow cooking,
50 g grated cheese and four egg yolks were added; four
egg whites were whipped separately and mixed into the
cheese béchamel; the soufflés were cooked in terracotta
ramekins 15 cm in diameter, 10 cm deep; they were
filled up to 8 cm and cooked in a De Dietrich P5447
oven with a K thermocouple (precision 0.1 °C, checked
with ice and water, plus with boiling water) inserted
4 cm above the bottom of the ramekins.
291
292
FIGURE 40.1 The temperature inside a soufflé during cooking. The uncer-
tainties are smaller than the size of the dots.
Figure 40.1 shows the temperature increase in the centre of the
soufflés as a function of time. In this particular experiment, a
slight slowing of the increase is observed after 10 min, but for
other experiments, this did not occur.
It is interesting to observe that, when Kurti did his first
experiments, and also when we did our first experiments with
soufflés, beginning on 16 March 1980, the expansion of soufflés
was said to be due to the dilatation of air bubbles from the egg
whites (Montagné, 1994).
However, the simple application of the ideal gas law showed
that this “theory” could not explain the doubling of volume for
soufflés. In my first calculation, the ideal gas law was applied
between the initial state, characterized by a pressure P1 = 1 atm,
a volume V1 and room temperature T1 = 20 °C (293 K); for the
final state, the maximum expansion was obtained for an inner
temperature of T2 = 100 °C (373 K) that was obviously too high
(because the soufflé would have been overcooked). But even with
these assumptions, and assuming that the pressure in the soufflé
was always 1 atm, the relationship:
P2 ⋅ V2 nRT2
= (40.1)
P1 ⋅ V1 nRT1
led to an expansion ratio equal to 1.27 only, which does not
explain the factor of 2 to 3 that can sometimes be observed.
Assuming a pressure higher than 1 atm in the soufflé (due to
the resistance of the crust, for example), the ratio is even lower
(indeed, the pressure inside soufflés was measured to reach about
0.24 mm Hg at the end of the cooking process, so that the expan-
sion ratio is not really changed). I proposed an improved theory
in 1994, after observing bubbles at the surface of soufflés being
cooked; this led to the assumption that the main mechanism of
soufflé expansion is water evaporation rather than bubble dila-
tation. This observation was validated by many experiments,
Hervé This vo Kientza
such as using a glass vessel for cooking, and also by weighing
a soufflé before and after cooking: soufflés with an initial mass
around 100 g (before cooking) lose 10 g after being cooked for
20 min at 180 °C. This means that more than 12 litres of steam are
formed during cooking, a result that explains the actual soufflés’
expansion. Moreover, this means that very large soufflés could
be obtained if the top crust of the soufflé is made impermeable
to steam. For a soufflé made from 3 eggs, 250 mL of milk, 80 g
of grated cheese and 40 g of flour, the total water mass is about
295 g; the maximum volume, i.e., the volume of a completely dry
soufflé, would be 185 litres!
Testing Culinary Precisions
For soufflés, culinary precisions were tested as well, generally in
terms of expansion. In particular, it was proposed by some chefs
that the expansion would be improved by using very firm whipped
egg whites, whereas other chefs proposed that bigger soufflés
would be obtained using quite soft whipped egg whites. After
preliminary experiments that I carried out in 1993 (Blanchet,
1993), I thoroughly tested the effect of firmness of egg whites
on soufflé preparation. The cheese béchamel was divided into
two equal parts, and foams of different firmness made from the
same egg whites (after pooling) were added to both parts; in the
first one, the whites were whisked until the first soft peaks were
observed, while in the second, the egg whites were so firm that an
egg in its shell (62 g) could stand on top of them without sinking.
The soufflés were placed in identical ramekins 15 cm diameter,
10 cm deep, filled up to 8 cm. They were cooked in the same
oven (De Dietrich P5447), for the same time, and the volume of
the two soufflés was compared (of course, the experiment was
replicated). The difference is shown in Figure 40.2.
The explanation of the difference is easy to understand, using
the more correct theory of steam formation at the bottom of the
ramekin; when a soufflé with soft egg whites is cooked, steam
bubbles formed at the bottom of the ramekin, where the tempera-
ture is high, rise up through the soufflé preparation and explode
at the upper surface of the soufflé; however, in a soufflé with
firm egg whites, the firmness of the foam prevents (partially) the
FIGURE 40.2 Pairs of soufflés having the same composition were cooked
simultaneously. They expand more (right) when the egg whites are whipped
until very firm.
Expansion
bubbles from rising through the soufflé preparation; the upper
layers are pushed upwards. Perhaps also the microstructure of
the upper crust better retains the steam bubbles, but this remains
to be determined.
As another validation of this theory, I tested in 1994 whether
soufflé expansion was different depending on the position of the
ramekins in the oven, and I showed that, in accordance with the
right mechanism of expansion (primarily water evaporation),
soufflés expand much better and faster when the ramekins are
put on the hot bottom of ovens. Later, during one of my public
monthly seminars, I even demonstrated that soufflés can expand
when soufflés are cooked in this way even if the egg whites
are not whipped (This, 2010), to the astonishment of the chefs
attending the experiment.
Other Expansions of This Kind
Soufflés are not the only preparation that expand through water evap-
oration (This and Kurti, 1995). Indeed, with Nicholas Kurti, we also
investigated “cannelés” and “Durham popovers”, but more recently,
experiments were also performed on choux, cakes and bread.
Cannelés and Durham popovers are made using pancake
dough (milk, egg, sugar and flour) cooked in small dedicated
vessels such as muffin tins, and choux are made from doughs
containing water, flour, butter and eggs, directly cooked as small
batches directly on a hot metallic plate in the oven. For cakes, the
composition can change, but the dough often also contains flour,
sugar and egg, and indeed, it was proposed by chefs that “eggs
make preparations expand” (This, 2009).
However, the observation of puff pastry can easily show some
expansion without eggs, and more when the number of layers is
increased. For example, Figure 40.3 shows the results obtained
from the same dough after various numbers of folds (1, 2, 3, 4, 5,
6, 7), during one of the French monthly seminars on molecular
gastronomy (see chapter “The monthly INRAE-​AgroParisTech
seminars on molecular gastronomy” in Part II of this book). For
such pastry, there can be considerable expansion, but the dough
contains no egg. Indeed, chefs in the past confused the presence
FIGURE 40.3 The same dough for puff pastry was folded (simple turns)
more and more, and samples were isolated at each step. After cooking, the
expansion is bigger for a larger number of foldings, and also the thickness of
the sheets decreases.
293
of eggs and the presence of water provided through egg addition,
and this is how they invented an “expanding principle” that never
existed.
One by One?
The preparation of choux is particularly interesting regarding how
molecular gastronomy can discuss culinary precisions (see also
the chapter “Culinary precisions and robustness of recipes”). For
making choux or “gougère” (choux with cheese, a specialty of
Burgundy, France), many recipes recommend boiling water and
butter in a pan and then adding flour. On very low heat, the batter
is kneaded, and, when this batter has cooled, whole eggs are added
to the dough, which is then cooked in the oven.
We have been interested in choux pastry puffs because they
are mentioned in many cookbooks (Carême, 1847; Escoffier
et al., 1903; Saint Ange, 1925; Pellaprat, 1936; Montagné, 1994;
Mathiot, 1995). Authors insist that the eggs have to be added
very specifically to the batter, some books even adding that the
mixing time must be equal during the introduction of each egg.
This seemed strange, since it might be thought that, as long as the
resulting paste was homogeneous, the mode of adding the given
quantity of eggs should have no influence.
For example, as early as 1739, Massialot (1717) gives one
recipe for “benoites, ou pets de putain”: they are small choux, and
it is advised to add the eggs two by two into the dough. In 1901,
Escoffier et al. (1903) also proposed to add the eggs two by two,
but they write that “more firmness is obtained by adding the eggs
three by three”. Gilbert (one author of the former book) gives
the same indication in one of his sole-​authored books (Gilbert,
1898). In 1919, Darenne and Duval (1919) indicated that the eggs
have to be added all together, but never by fractions, “i.e., two
by two or three by three”. In 1925, Madame Saint Ange wrote
(Saint Ange, 1925) that, in a dough for choux, the eggs have to
be added “with a rigourously constant interval”, and she adds that
the dough is lighter when the dough is aerated while mixing the
eggs. In 1924, de Pomiane proposes to add the egg differently
(Pomiane, 1924): “boil water, butter and salt; then add the flour in
rain; dry while eating, and add the eggs one by one”.
However, later, Prosper Montagné (1936) proposes a different
process: “Add 12 to 14 eggs, depending on their size, two by two”.
Much inconsistency thus appears in all this literature, and this
is an indication that the question was ill-​discussed. Indeed, even
if microscopic studies of the raw dough show the presence of
air bubbles, the expansion is mainly due to water evaporation.
This was tested at the November 1993 Séminaire of Molecular
Gastronomy; the water–​ butter–​
flour mixture (“panada”) was
divided into two equal parts, and eggs (four) were added dif-
ferently to these two batters, using a whisk. In the first half,
eggs were added one by one, the number of whisk turns being
counted; in the other half of the panada, the four eggs were added
all together. Because the assumption was that mixing was more
important than the means of egg introduction, the half with eggs
introduced together was mixed with twice the number of whisk
turns as in the first case. Then choux pastry puffs were formed
on a oven plate, with parallel rows of both preparations, and they
294
were cooked together. A blind test on more than 50 guests showed
that the more mixed preparation was preferred to the other.
Following this experiment, it was reasoned that, if air bubbles
were indeed the key to success in making choux pastry puffs, a
better expansion could be obtained if egg whites were separated
from the yolks before being introduced into the panada, whipped
separately and then mixed into the rest of the mixture. In a
second experiment, conventional choux pastry puffs and choux
pastry puffs made from panade added with whipped egg whites
expanded similarly.
In a third experiment, conventional choux pastry puffs were
compared with choux pastry puffs made by thoroughly mixing
panada and egg yolks and then adding whipped egg whites, in the
hope that the total number of air bubbles would be higher. In this
third case, the expansion was found to be about one-​third better
for the new kind of choux pastry puffs. From a culinary point of
view, however, the surface of the more foamy choux pastry puffs
was less appreciated than the classical choux pastry puffs because
they did not have the usual smoothness.
Finally, the number of untested culinary precisions about
soufflés and their relatives remains important, and it is to be
hoped that the culinary schools of today, now including science
and cooking approaches, can conduct tests so that correct theories
can be developed.
REFERENCES
Belitz HD, Grosch W, Schieberle P. 2014. Food Chemistry (4th Ed),
Springer Verlag, Berlin-​Heidelberg, 723.
Hervé This vo Kientza
Blanchet AM. 1993. TV News TF1, 13h00, 23 December 1993.
Carême MA. 1847. L’art de la cuisine française au XIXe siècle,
Carême, Paris.
Darenne E, Duval E. 1919. Traité de pâtisserie moderne, Flammarion,
Paris.
Escoffier A, Gilbert Ph, Fetu E. 1903. Guide culinaire, Flammarion,
Paris.
Gilbert Ph. 1898. La cuisine de tous les mois, Ollendorff éditeur,
Paris, 99.
Kurti N, The physicist in the kitchen, Proceedings of the Royal
Institution, 42 (199), 451–​467.
Massialot M. 1717. Nouveau cuisinier royal et bourgeois, La
Chapelle V (ed.), Claude Prudhomme, Paris.
Mathiot G. 1995. La cuisine pour tous, Livre de Poche, Paris.
Montagné P. 1936. Mon menu, Société d’applications scientifiques,
Paris, 316.
Montagné P (ed.). 1994. Larousse gastronomique, Larousse,Paris.
Pellaprat HP. 1936. L’art culinaire moderne, Comptoir français du
livre, Paris.
Pomiane E. 1924. Code de la bonne chère, Albin Michel, Paris, 382.
Saint Ange M. 1925. La cuisine de Madame Saint Ange, Larousse,
Paris.
This H. 2002. Molecular gastronomy, Angewandte Chemie,
International Edition in English, 41(1), 83–​88.
This H. 2009. Histoire de soufflés, L’Actualité chimique, 10, 334.
This H. 2010. Séminaire de gastronomie moléculaire, December
2010.
This H, Kurti N. 1995. Soufflés, choux pastry puffs, quenelles and
popovers, The Chemical Intlelligencer, 1, 54–​57.
Fats and Oils: Physicochemical Properties of Edible Oils and Fats
Sabine Danthine
Science des Aliments et Formulation, Gembloux Agro-​Bio Tech, University of Liège, Avenue de la faculté d’Agronomie 2B,
5030 Gembloux, Belgium
Oils and fats are highly abundant compounds in nature and are
widely used in food and other industrial products. Their behav-
iour heavily influences the microstructure and physical properties
of fat-​based products, such as chocolates, confectionery, mar-
garine and butter. Edible oils and fats are complex mixtures of
lipidic components. They mainly contain triacylglycerols (TAG):
three fatty acids esterified on a glycerol backbone (Figure 41.1).
The functionalities of fats in edible products are on the one hand,
to provide nutritional characteristics and on the other hand, to
provide structure. Both directly depend on the TAG composition:
from the nutritional point of view, an increased use of lipids made
of polyunsaturated fatty acids (liquid oils) is desirable, whereas
the physical structuring of food products requires saturated lipids
(solid fats). The properties of the fat crystal network formed in
the products determine their functional properties. The mech-
anical strength of the underlying fat crystal network determines
many of the sensory attributes, such as hardness, spreadability,
mouthfeel and also snap in the case of chocolate.
The physical properties of natural oils and fats vary widely,
even though they are generally made of similar fatty acids. The
relative proportions of these fatty acids and also their disposition
within the TAG are the reasons for these differences.
In general, the longer and the more saturated the hydrocarbon
chains, the more solid are the molecules. From a technological point
of view, this means that the saturated fatty acids constituting the
TAGs present in solid fats are ideal elements for structuring edible
fatty products. Liquid oils are, on the contrary, rich in unsaturated
fatty acids (free of crystallized material at room temperature).
FIGURE 41.1 Schematic representation of a triacylglycerol molecule. R,
R′ and R″ are fatty acids.
In a simplified way, we can say that tri-​ saturated TAGs
(containing three saturated fatty acids) have high melting points.
They mainly play the role of structuring and stabilizing agents;
in some cases, they also provide a physical barrier to water. The
di-​saturated TAGs (containing two saturated fatty acids and
one unsaturated fatty acid) melt at a temperature close to that
of the human body. They provide a structure to foods, play the
role of lubricant or barrier to water, and can also participate in
the aeration of the products that contain them. Tri-​unsaturated
TAGs (containing three unsaturated fatty acids) are liquid at room
temperature; their presence in seasoning oils makes them trans-
parent, which is a very important quality criterion for this kind of
product, for both appearance and functionality.
The technological functionality of edible oils and fats is mainly
governed by their physical properties. Understanding these phys-
ical properties and the mechanisms responsible for them is there-
fore an essential element in understanding the formulation of
many edible products. This is especially true if one is interested
in fat structured food products, of which the most prominent
example is the margarine family.
The complex physical behaviour of fats (melting properties,
crystallization behaviour, polymorphism, crystal morphology,
structure of the fat crystal network, …) is related not only to the
physical properties of each TAG present in the fat but also to the
behaviour of these triglycerides in mixtures, which is highly com-
plex. Lipids, like most long-​chain compounds, can exist in the
solid state under different crystalline forms; this property is called
polymorphism. When the same set of molecules can be stacked in
different ways during crystallization, depending on the process,
the substance is said to be polymorphic. Polymorphic forms gen-
erally have different physical properties, for example, different
melting properties, but after melting, the resulting liquid product
is the same. The polymorphism of glycerides has been widely
described in specialized literature and is the subject of numerous
scientific publications (Sato, 2001). Only the major elements are
mentioned below. In the case of lipids, polymorphism results
from the spatial organization of the aliphatic chains of TAGs and
also from the stacking of TAG molecules into strata.
A lipid crystal contains many methylene (CH2) groups. It is
therefore advantageous for crystallographers to consider repe-
tition elements of these CH2 groups, so as to specify the type
295
296
of lateral arrangement of the chains. Considered in this way, the
concept of the crystalline “sub-​cell” corresponds to the smallest
repetition unit.
Based on this, and thanks to the combination of results
obtained by X-​ray diffraction (XRD) and infrared spectroscopy,
inspired by the work of Lutton (1950) and Chapman (1957),
Larsson (1966) proposed as early as 1966 a classification (for
saturated monoacid TAGs) into three crystallographic types: the
form with the lowest melting temperature, called α, has a hex-
agonal type stacking, the intermediate form, β′, corresponds to
an orthorhombic type, and the most stable form, β, is of triclinic
type. This notation is still used today and has been extended to all
TAGs, although polymorphic varieties of unsaturated TAGs are
even more complex. Data regarding the crystalline sub-​cells char-
acterizing the polymorphism of lipids are obtained experimentally
mainly from powder XRD at wide angles; the values obtained in
this way are called “short-​spacings”. They are observed in the
2θ-​angular region ranging from 15° to 27° (which corresponds
to the d-​spacing region ranging from ~5.5 Å to ~3.5 Å) and are
independent of the chain lengths of the constituent fatty acids.
The three “basic” forms described for TAGs can be distinguished
unambiguously, since in this short-​spacing region, each “sub-​
cell” is characterized by a specific pattern. Indeed, the stacking of
the hexagonal “sub-​cell” of the α form is easy to identify, since
it has a single, very strong diffraction line, corresponding to a
lattice distance of ~4.15 Å. The β′-​stacking shape of the perpen-
dicular orthorhombic “sub-​cell” is characterized by two intense
lines located at ~3.8 Å and 4.2 Å. The parallel triclinic β-​“sub-​
cell” form is distinguished by a series of lines, one of which is
strongly dominant, located at 4.6 Å, and two other, less intense
lines located at 3.7 Å and 3.9 Å (Chapman, 1969). Since the poly-
morphism of triglycerides is essentially of monotropic type (α →
β′ → β) (except for the reversible sub-​α↔α transition), all the
varieties observed, except the most stable, which is generally β,
are metastable and not under thermodynamic equilibrium.
Sources of “natural” oils and fats are relatively limited. Among
these, we can distinguish plant sources and animal sources.
Among fats of plant origin, it is important to note that most are
liquid at room temperature, while only a few are solid or semi-​
solid under similar conditions. Palm oil, cocoa butter and coconut
oil are, for example, part of this latter category.
As described previously, the nutritional and techno-​functional
properties of oils and fats are due to their constituent TAGs.
Because of their specific TAG composition, each oil or fat is
therefore suitable for particular but also generally limited
applications.
There is, however, an important demand for particular melting
and/​or crystallization profiles according to the applications for
which the products are intended. Obtaining an adequate melting
profile is, for example, a key element for the production of
margarines that can be used in bakery, biscuits, puff pastries …
(see chapter on laminated bakery products by Detry et al.). This is
practically impossible to achieve today with only native products.
Food technologists have developed several ways to overcome
this problem and respond to specific needs. The simplest solu-
tion is the use of blends of two or more fats. However, unfor-
tunately, their properties are rarely additive because of the
Sabine Danthine
complex crystallization behaviour of the TAGs they contain,
due to intersolubility combined with polymorphism. Indeed,
due to intersolubility, the properties of the individual TAGs are
not always additive. This is very important when mixing several
oils and fats together, which is almost always the case in food
formulations. In addition, a mixture of fats in their native state is
often not sufficient; therefore, processes have been developed to
broaden the potential scope of application.
How Are Dietary Lipids Designed to Achieve a
Certain Functionality?
As just said, most of the natural oils and fats have only limited
applications in their original forms, as a consequence of their spe-
cific chemical composition. To extend their possible use in fat-​
based food products, oils therefore often undergo a chemical or a
physical modification.
Hydrogenation, interesterification (chemical or enzym-
atic) and fractionation processes have been developed by food
technologists and are widely used industrially. The main purpose
of these processes is to modify the physicochemical properties of
the oil or fat (by reducing the degree of unsaturation of the acyl
groups (hydrogenation), by redistributing the fatty acid chains
(interesterification) or by a physical separation of the TAGs
through selective crystallization and filtration (fractionation).
Hydrogenation
Hydrogenation overcomes the problem that most vegetable oils
are liquid at room temperature. Indeed, this technique makes it
possible to “harden” the oils chemically by linking hydrogen
to the double bonds of the unsaturated fatty acids constituting
the TAGs, thus reducing the degree of unsaturation of the acyl
groups. This reaction can be stopped before saturation of all the
double bonds; the process is then called “partial hydrogenation”.
However, in this case, some unsaturated fatty acids change their
configuration and move from a cis-​configuration (natural) to a
trans-​configuration, which leads to the appearance of non-​natural
isomers (industrial trans fatty acids (TFA)), currently criticized
from a nutritional point of view. Excessive consumption of this
type of isomer is now associated with an increased risk of car-
diovascular diseases (Mensink et al., 2003). As a result, food
technologists have been forced to replace partial hydrogenation
by other techniques that may give oils and fats the desired func-
tionality. Besides full hydrogenation (no remaining unsatur-
ated fatty acids and thus zero TFA), the other two modification
processes, interesterification and fractionation, make it possible
to generate new fats and diversified qualities without producing
trans isomers; they are therefore increasingly used.
Fractionation
Fractionation is a process allowing the separation of a fat into two
or more fractions with different physicochemical characteristics
(Braipson-​Danthine and Gibon, 2007). The dry fractionation pro-
cess is essentially based on the ability of fats to produce crystals.
Fats and Oils: Physicochemical Properties
This fractionation process makes it possible to obtain fractions
with properties different from those of the initial oil (or fat). This
is possible due to the variable crystallization temperatures of the
TAGs present in the bulk fat (Gibon, 2006).
The fractionation process (or fractional crystallization pro-
cess) is based on a simple principle, since it consists of a physical
separation of the TAG mixture (Timms, 2005). This is done by a
selective crystallization of some high-​melting TAGs, followed by
the separation of the crystals obtained from the liquid fraction,
which is generally carried out by filtration. The selective partial
crystallization of the melted oil is possible thanks to the appli-
cation of a controlled specific cooling programme. Due to the
continuous development of the technology, a whole variety of
products can now be obtained with a high degree of selectivity.
The operations can, moreover, be conducted in multiple steps,
giving rise to a wide range of fractions suitable for different
applications (Gibon, 2012, Danthine et al., 2017).
The process is applied to a large number of fats, for example
palm oil, which is by far the most fractionated oil in the world,
palm kernel oil, cottonseed oil, anhydrous milk fat, tallow, lard,
and also fish oils, etc.
Interesterification
The third method of modifying oils, used alone or in combin-
ation with fractionation, is called interesterification. The redistri-
bution of fatty acids on the glycerol backbone by catalytic action
is the basic principle of this technique; the redistribution of fatty
acids occurring between and within the different TAG molecules.
These redistributions can lead to significant changes in the func-
tionality of lipids without changing the nature of the constituting
fatty acids (De Clercq et al., 2012).
The level of unsaturation remains constant; there is no cis–​
trans isomerization as encountered during partial hydrogenation.
This process can be carried out directly on an oil, a fraction
resulting from the fractionation, or a mixture of lipids, using
chemical catalysis (random or directed) or enzymatic catalysis
(a lipase is used a biocatalyst). The interesterification technique
is thus used to modify the general melting profile or the melting
point of a mixture, improve the compatibility of the TAGs in the
297
solid state, and improve the plasticity of the solids obtained by
modifying their crystallization properties.
REFERENCES
Braipson-​Danthine S, Gibon V. 2007. Comparative analysis of
triacylglycerol composition, melting properties and poly-
morphic behavior of palm oil and fractions. European Journal
of Lipid Science and Technology, 109, 359–​372.
Chapman D. 1957. Infrared spectra and the polymorphism of
glycerides. Part III. Palmitodistearins and dipalmitostearins.
Journal of the Chemical Society, 2715–​2720.
Chapman D. 1969. Introduction to lipids. Sykes P (ed.). McGraw-​
Hill, London.
Danthine S, Lefébure E, Blecker C, Dijckmans P, Gibon V. 2017.
Correlations between cloud point and compositional properties
of palm oil and liquid fractions from dry fractionation. Journal
of the American Oil Chemists’ Society, 94, 841–​853.
De Clercq N, Danthine S, Nguyen MT, Gibon V, Dewettinck K.
2012. Enzymatic interesterification of palm oil and fractions:
monitoring the degree of interesterification using different
methods. Journal of the American Oil Chemists’ Society, 89
(2), 219–​229.
Gibon V. 2006. Fractionation of lipids for use in food. In Gunstone
FD (ed.) Modifying lipids for use in food, Woodhead Publishing
Limited, 201–​229.
Gibon V. 2012. Palm oil and palm kernel oil refining and fraction-
ation technology. In Oi-​Ming Lai OM, Tan CP, Akoh AC (eds.)
Palm oil, production, processing, characterization and uses,
AOCS Press, 329–​375.
Larsson K. 1966. Classification of glyceride crystal forms. Acta
Chemica Scandinavica, 20, 2255–​2260.
Lutton ES. 1950. Review of the polymorphism of saturated even
glycerides. Journal of the American Oil Chemists’ Society, 27,
276–​281.
Mensink RP, Zock PL, Kester ADM, Katan MB. 2003. Effects of
dietary fatty acids and carbohydrates on the ratio of serum total
to HDL cholesterol and on serum lipids and apolipoproteins:
a meta-​analysis of 60 controlled trials. American Society for
Clinical Nutrition, 77, 1146–​1155.
Sato K. 2001. Molecular aspects in fat polymorphism. In Widlak
N, Hartel R, Narine S (eds.) Crystallization and solidification
properties of lipids. AOAC Press.
Timms R. 2005. Fractional crystallisation –​the fat modification pro-
cess for the 21st century. European Journal of Lipid Science
and Technology, 107, 48–​57.
Fats and Oils: From Fat Droplets in Plant Seeds to Novel Foods
Juan C. Zambrano, Behic Mert and Thomas A. Vilgis
Max-​Planck-​Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany
What makes soymilk so special? For vegan-​oriented consumers and
people with lactose intolerance, it is a substitute for cow’s milk. For
East Asian consumers, it is a traditional beverage, which they have
been consuming for centuries, and a base for traditional soy-​based
products such as yuba and soy cheese (tofu). For other people, it is
just an unpleasant beverage, because it has a green, bean-​like flavor
that is difficult to get rid of. Finally, for some scientists, soymilk
conveys one of nature’s best hidden nanoparticles.
Soymilk is a colloidal dispersion system containing proteins,
lipids, and polysaccharides, which exists as a natural emulsion.
Usually, “man made” food emulsions are prepared by dispersing
oil into water in the form of tiny droplets of diameters generally
larger than 0.1 μm. Emulsions are thermodynamically unstable
systems, but they can be physically stabilized to prevent the
instant separation into oil and aqueous phases. This stability
results from the interplay of emulsifying agents (low-​molecular-​
weight surfactants and/​or polymers such as proteins) and appro-
priate processing conditions (e.g., mixing or homogenization). In
the case of soymilk, nature has developed native structures to dis-
perse oil into water in the form of emulsified droplets that confer
similar physical stability to that in “artificial” emulsions. It seems
that plant seeds have been one step ahead in solving the problem
of incompatibility between oil and water. What is nature’s trick,
which is hidden in soymilk and other oil seeds, such as hazelnuts,
flax seeds, and many other plant-​based foods?
The trick comes in the form of nanoparticles (200 nm in diam-
eter in soybeans), so called oleosomes or oil bodies. They are
ubiquitously present in oleaginous plants (almonds, peanuts,
hazelnuts, and safflower), mainly in their seeds or nuts, and their
size ranges from micrometers down into the nanoscale (Tzen
et al., 1993). These special particles are oil droplets, which are
naturally covered by phospholipids that act as emulsifiers. These
are compounds whose molecules contain a hydrophilic (water-​
loving) “pinhead” with two lipophilic (fat-​loving) “legs” that
are stuck in the oil droplet. These hydrophilic and hydrophobic
groups orient themselves at the oil–​water interface, which is key
to the formation of emulsions. However, phospholipids alone
would not achieve the trick of producing metastable oil droplets.
Therefore, oleosomes also contain on their shells an unusual type
of protein, so-​called oleosin (Figure 42.1).
Together, phospholipids and oleosins may stabilize oil droplets
with greater efficiency than oleosins or phospholipids alone
(Deleu et al., 2010; Karefyllakis, 2019). This type of architecture
is analogous to other remarkable structures that also occur nat-
urally: milk fat globule membrane, which stabilizes fat globules
through phospholipids and proteins, and animal lipoproteins
(high-​density lipoproteins (HDLs), low-​ density lipoproteins
(LDLs), and chylomicrons). Just like oleosomes, LDLs appeared
during biological evolution to disperse oil into water in the egg
yolk. They are widely used in mayonnaise and famous sauces
such as hollandaise and gribiche.
Here, the LDLs are surrounded by proteins (apolipoproteins)
and phospholipids. LDLs have been largely proved to play the
predominant role in the emulsifying properties of egg yolk. This
translates into a higher reduction in droplet size, higher surface
coverage, and higher stabilities than in emulsions stabilized by
granules or livetins (Aluko et al., 1998; Davey et al., 1969; Mine,
2000; Martinet et al., 2003). Nevertheless, oleosomes show
greater stability than LDL particles at temperatures up to 100 °C
(Zielbauer et al., 2018).
The reason for this is a different arrangement of their stabil-
izing proteins, in which apolipoproteins are located only on the
surface of LDL particles, whereas oleosins are deeply anchored
in the oil droplet in the form of little umbrellas (Figure 42.1).
The “umbrella stem” that stretches into the oil is highly lipophilic
and forms a “hairpin”-​like loop, which is key to the stability of
the oleosin. The umbrella-​shaped part is hydrophilic and remains
outside the oleosome, projecting into the water (Huang, 1992).
This protruding hydrophilic part provides steric repulsion forces
and is also electrically charged. These features prevent adjacent
oleosomes from agglomerating, which, together with other phys-
ical mechanisms such as Ostwald ripening and depletion forces,
causes droplet coalescence. This is the phenomenon behind the
development of oily lumps that is often seen in non-​stable food
emulsions.
When we look into this extraordinary stability, the role of
plant seeds becomes significant. Botanists affirm that oleosin
prevents individual oleosomes from coalescence, a danger that
is particularly imminent during demanding processes for the
seeds that accompany germination and growth (Kermode, 2003).
Therefore, oleosomes are considered a multifunctional organelle
participating in a host of cellular processes. These properties, all
together, have huge potential to be applied to a wide variety of
food products and recipes as natural emulsifiers. Phospholipids
299
300
FIGURE 42.1 Simplified illustration of an oleosome from oil seed, e.g., soybean. Phospholipids form a metastable layer but, unlike fat of animal origin
(LDL, HDL), they do so without cholesterol. Additional stability is achieved via oleosins. These proteins with a strongly lipophilic portion, which is bent into
the shape of a “hairpin” (top right), are also deposited and act to stabilize the oleosomes.
in oleosomes are a mixture of various phospholipids, amongst
which phosphatidylcholine (PC) is the major component (Tzen
et al., 1993). PC, predominantly available in commercial “leci-
thin”, is widely used in the food industry as a surfactant in a range
of food products, such as chocolate, bakery products, desserts,
and creams. Therefore, the use of intact and entire oleosomes
as a natural pre-​ emulsified oil-​ in-​
water emulsion can attract
the attention of curious cooks to try to experiment and develop
unusual creations with the aim of evolving and upgrading the
use of traditional emulsifiers such as egg yolk. In the following
sections, we will explore and discuss potential applications in
food products and culinary recipes making use of the manifold
properties of oleosomes.
Oleosomes Meet the Culinary World: Spreads
and Sauces
As mentioned earlier, oleosome droplet size can vary depending
on the plant species of origin, with a diameter of at least 0.1 µm;
for example, rapeseed shows a mean diameter of 0.7 µm,
soybeans 0.2 µm, maize 0.3 µm, mustard 0.73 µm, cotton
0.97 µm, hazelnuts 1.1 µm, peanut 1.95 µm, sesame 2.0 µm, and
almonds 2.6 µm (Nikiforidis, 2009; Beisson et al., 2001; Tzen
et al., 1993). This vast array of droplet sizes can be exploited to
combine oleosomes of different sizes. Therefore, oil droplet size
and fat content can be tuned to produce emulsions with different
compositions and thereby properties such as creaminess and vis-
cosity. These properties can be fully exploited in food products
that contain emulsion-​based formulations. For example, fluid-​
like or semi-​soft foods such as spreads will fully benefit from
these properties.
It is essential that spread products retain their shape when they
are removed from the refrigerator, but also that they spread easily
when a knife is applied. Thus, they must exhibit viscoelastic
properties, i.e., have yield stresses below which they are elastic
and above which they are viscous (shear thinning). This means
that when such a product is forced to flow, internal changes will
occur within the microstructure of the product, which will lead to
a faster flow of the product. In other words, the larger the force,
Juan C. Zambrano et al.
the faster the product will flow, and this is perceived as thinning.
These properties together are critical in spreads, since they will
influence their spreadability. Debon et al. (2017) proposed that all
these rheological properties can be tuned as desired by choosing
the appropriate combination of oleosomes from different plant
origins. Consequently, they prepared a spread comprising a blend
of rapeseed and soybean oleosomes, which possessed the same
shear-​thinning and yield stress properties as a regular fat spread
such as margarine. In addition, these authors observed that the
mixture presented a stable, constant thickness over time, which
is highly beneficial in spreads, since mouthfeel and other sen-
sorial properties are related to this physical property (Akhtar
et al., 2005). The use of oleosome blends in spreads also benefits
from the ability of oleosomes to be stable against lipid oxidation.
In several studies, it has been concluded that hydroperoxide and
secondary oxidation product formation is significantly lower in
natural emulsions based on oil bodies compared with technically
prepared emulsions formulated with other surfactants (Fisk et al.,
2008; Gray et al., 2010).
There now follow a few standard recipes for well-​ known
spreads and sauces in which oleosome blends can be exploited.
One could play with combining different types of oleosomes at
various ratios to obtain different viscosities, which can result in
interesting textures!
Fat Mayonnaise-​Type Spread
Mayonnaise is a good example in which the greater stability of
oleosomes can be exploited at low pH. Traditionally, mayonnaise
contain vinegar, which is basically an aqueous solution (5–​20%)
of acetic acid, and thus, it goes without saying that it is acidic (pH
< 4). Table 42.1 shows the formulation of an emulsion using an
oleosome preparation without the addition of egg yolk.
By regulation, it is compulsory that a spread should contain a
minimum of 6% egg yolk to be designated as mayonnaise (Codex
Commission, 2000). Therefore, we will limit ourselves to labeling
the emulsion as a fat spread “mayonnaise-​type”. The emulsion
can be produced by manually mixing the stated ingredients.
Deckers et al. (2000) used rapeseed oleosomes, which resulted
in a product that visually reflected a mayonnaise-​like texture
Fats and Oils: From Seeds to Novel Foods 301

TABLE 42.1 TABLE 42.3

Mayonnaise-​Type Recipe Béchamel Sauce Recipe

Sunflower oil 200 g Flour 50 g

Oleosome preparation 100 g Oleosome preparation 50 g
Vinegar 30 mL Milk 100 mL to 1 L

Source: Deckers et al. (2000) Source: Deckers et al. (2000)

bubbles begin to form at the edges) is gradually added until a

TABLE 42.2 sauce with the desired viscosity is obtained.
Table 42.3 shows a recipe for the preparation of a standard
Vinaigrette Recipe
béchamel-​like sauce using a preparation of rapeseed olesomes,
Sunflower oil 17.5 g in which the consistency can be as desired, depending on the
Mustard 0.4 g amount of milk that is added.
Vinegar 0.5 g The absence of hydrogenated fatty acids gives it an advan-
Oleosome preparation 7.7 g tage over a sauce prepared from common domestic margarine,
Source: Deckers et al. (2000)
because these fats are associated with cardiovascular diseases
(Perwaiz, 2014).

and was as stable as a commercial mayonnaise when stability

Liquid Creamers
was assessed using centrifugation. Since the spread is prepared
without egg yolk, it is free of cholesterol and ideal for vegan-​ The predominant lipid composition of oleosomes make them suit-
oriented people. able to be used as liquid creamers in beverages. Liquid creamers
are used as whitening agents in hot and cold beverages such as,
Mustard Spread for instance, coffee, cocoa, and tea. Additionally, they should pro-
vide mouthfeel, body, and a smooth texture. Normally, non-​dairy
A mustard spread can also be prepared by manually mixing creamers are added to coffee, because milk fat would generally
70 g of mustard plus 30 g of oleosome. Deckers et al. (2000) overpower the flavor and texture in coffee. For achieving the
used rapeseed oleosome to give an emulsion that visually ideal mouthfeel, akin to that of coffee cream, non-​dairy creamers
appeared as a mustard-​ like product, which may easily be often use vegetable-​based fats to which low-​molecular-​weight
spread and has creamier, yet less gritty, sensory characteristics emulsifiers, e.g., mono-​and diglycerides, are added to ensure
than mustard. stability of the oil-​in-​
water emulsion. Moreover, they contain
whitening agents (usually titanium dioxide), which are used to pro-
Vinaigrette vide the necessary whitening capacity when added to beverages.
Table 42.2 shows the ingredients for the preparation of a standard All these compounds may be perceived as artificial by consumers.
vinaigrette-​like emulsion comprising an oleosome preparation Accordingly, oleosomes can be used to replace oil in liquid
which was manually mixed with the other ingredients. creamers. Oleosomes can provide a comparable mouthfeel
Deckers et al. (2000) used rapeseed oleosomes. The resulting and smooth texture in a liquid cream while providing higher
product presented a visually vinaigrette-​ like texture that was whitening capacity. This allows the addition of artificial colors
stable for at least one day at 4 °C. such as titanium dioxide to the creamer to be avoided. The
effective emulsifying ability of oleosomes may provide good
physical stability, thus avoiding the need to use synthetic low-​
Béchamel Sauce
molecular-​ weight surfactants. In addition, the robust mixed
Another good example of applications where heat may be applied phospholipid-​oleosin layer at the surface of oleosomes makes
to oleosomes without affecting stability is in the preparation of the liquid creamer very stable when added to hot coffee. As a
a savory sauce such as a béchamel sauce. Béchamel is a basic result, a more natural product can be provided. Kapchie et al.
medium-​thick white sauce and one of the five “mother sauces” (2014) prepared a liquid creamer using an oleosome preparation,
of classical cuisine (i.e., béchamel, espagnole, velouté, holland- which was added to a coffee beverage. It was found that the liquid
aise, and tomato) (Ruhlman, 2007). It is also a base for other creamer had a good appearance without presenting creaming or
recipes such as soufflés. The main ingredients of béchamel con- de-​oiling, while also showing a good mouthfeel and smooth tex-
sist of butter, flour, and milk. Deckers et al. (2000) prepared a ture when added to coffee. Furthermore, the use of oleosomes in
béchamel-​like sauce using an oleosome preparation as a butter creamers is not limited to coffee applications. Oleosome-​based
substitute. The oleosome preparation can be heated at moderate creamers could also be used for other beverages, such as tea or
heat (50 °C), and an equal part of flour is added, heated with cocoa, or used with cereals or berries, as a creamer for soups,
the oleosome preparation, and stirred until a thick suspension to prepare ice creams, milkshakes, or other frozen food-​grade
is formed. Then, moderately heated milk (50 °C, or until little materials, or in a huge range of cooking recipes.
302 Juan C. Zambrano et al.

Ca2+

g
in
link
oss
cr
c
ni
: io
m
lciu
Ca
Tr
an
Ch

sg
lu
e m

am t
ica

in
lr

as
ea

e
ct
io
n

FIGURE 42.2 Simplified illustration of oleosome binding with Ca2+ (top) and microbial transglutaminase (MTGase) (bottom). The top right picture shows
the resulting gel with Ca2+, and the bottom right picture shows the resulting gel with MTGase.

What is even more fascinating about oleosomes is that their
functionalized surface can be exploited to conveniently design
a gamut of solid and soft-​solid materials, which can result in
unusual culinary structures with potential for use in experimental
kitchens around the world. Since lipids are the major constituent
of oleosomes, this offers the opportunity to modify the structure
of oil to create innovative textures, which may provide appealing
mouthfeel and alter flavor release in many recipes. In this regard,
the charged surface of oleosomes can be used to bind different
types of molecules using electrostatic attraction. For example, in
basic conditions (pH > 7), negatively charged amino acids are
predominant in the hydrophilic regions of the oleosins (Maurer
et al., 2013), which can therefore be linked to divalent ions such
as Ca2+ (Figure 42.2).
This will act like a bridge between oleosomes, which will lead
to their agglomeration, causing structured aggregates that result
in a gel-​like structure. Figure 42.2 shows some initial studies
carried out in our lab, in which a calcium-​mediated gel shows
a soft, almost melting texture. This process brings to mind the
formation of tofu, in which calcium ions bind to oleosomes, in
combination with soy protein particles and soluble proteins pre-
sent in soymilk, leading to globule aggregates that bind each
other, producing a gel network. However, the resulting textures
in oleosome-​calcium-​mediated gels may offer a more exciting
range of possibilities than in tofu, since the fat content could be
more precisely tuned.
The use of the enzyme microbial transglutaminase (MTGase,
EC 2.3.2.13) may even broaden the range of textural opportun-
ities in oleosome-​based gels. MTGase, broadly used in experi-
mental kitchens, is able to permanently cross-​ link oleosins
between adjacent oleosomes by catalyzing a reaction between
the carbonyl group of a glutamine residue and the amino group
of a lysine residue (Motoki and Seguro, 1998) (Figure 42.2).
Because MTGase induces covalent cross-​links, oleosomes are
likely stuck together, resulting in a gel with better toughness and
a texture reminiscent of jelly. Furthermore, MTGase combined
with calcium can also have a dramatic influence on the resulting
texture. Calcium acts as a cofactor (speeds up the reaction) for
the enzyme activity of MTGase, leading to an intensive formation
of cross-​links, which yields fragile, yet soft, gels (Figure 42.3).
Oleosomes could therefore be used to develop protein-​rich
foods with a high energy content that would be controllable
via their fat percentage. This would also be of interest in food
designed for elderly people for the prevention and treatment
of malnutrition, especially since the texture (and the ease with
which food could be swallowed in the event of dysphagia) could
be precisely modified (Vilgis et al., 2014).
In a similar fashion, oleosomes have the potential to be
combined with polysaccharides. This also brings a whole new
dimension to texture design with polysaccharides, because there
are a great number of these molecules, which can create a vast
array of structures. These remarkable macromolecules can be
long, short, rod-​or chain-​like shapes, branched, unbranched,
highly charged, slightly charged, or stiff or flexible backbones.
Therefore, polysaccharides can form many kinds of networks
between the oleosomes, depending on all these features. Imagine
the possibilities!
In our laboratory, we used xanthan and high-​methoxyl pectin
to bind oleosomes in water-​ continuous solutions via elec-
trostatic complexation. When the water-​ continuous phase is
removed, oleosomes can be physically entrapped within the
polysaccharide network. This results in two sorts of solid-​fat
structures. The samples made with xanthan show a brittle struc-
ture, whereas the ones made with pectin reveal a deformable and
Fats and Oils: From Seeds to Novel Foods
FIGURE 42.3 Increased stability by cross-​linking hazelnut oleosomes with calcium chloride, transglutaminase, and a combination of the two. The texture
is shown on the right-​hand edge of the figure. The calcium gels are very soft, almost melting; the oleosomes permanently cross-​linked with transglutaminase
exhibit a considerably higher shear modulus and have a texture reminiscent of jelly. The combination of MTGase and calcium (cofactor) leads to rapid cross-​
linking and the formation of fragile, yet soft gels.
FIGURE 42.4 Simplified illustration of oleosomes combined with the
polysaccharides xanthan (top) and pectin (bottom). The top right picture
shows the resulting gel with xanthan, and the bottom right picture shows the
resulting gel with pectin.
elastic structure. Understanding the molecular mechanisms that
yield these mechanical properties is still in progress. However,
one can dare to explain it by looking into the molecular struc-
ture of each of the polysaccharides. Xanthan is found in long,
straight, and stiff molecules, while pectin molecules are more
flexible/​semiflexible (Figure 42.4). These flexible chains ease the
formation of entangled networks between polymer chains, and
polymer physics has taught us that entangled networks lead to
elastic polymer gels.
303
The exciting point about Figure 42.4 is that the texture can be
controlled by the physical properties of the added hydrocolloids.
Xanthan appears as a relatively stiff polymer, which can be
viewed mainly as a rigid, highly charged polyelectrolyte
following a jamming transition under random orientation of the
rods (Vilgis, 2015). Pectin appears as a flexible, partially charged
block copolymer. Consequently, the gel prepared with xanthan
is more brittle, whereas the pectin-​bound gel becomes softer and
more deformable.
This manipulation of texture by the combination of extraor-
dinary molecules can be a rewarding experience in experimental
kitchens. With extra knowledge on the physical chemistry of these
molecules, we can try to use oleosomes from different sources
to vary the structure in everything ranging from chocolate to
comminuted meat products. Insights into the variables that make
this kind of tuning a success may lead to the creation of gels
varying in appearance from opaque to transparent gels, and in
hardness from soft to hard gels. This can impart new and exciting
sensory experiences yet to be experienced by the consumer!
REFERENCES
Akhtar M, Stenzel J, Murray BS, Dickinson E. 2005. Factors affecting
the perception of creaminess of oil-​in-​water emulsions. Food
Hydrocolloids, 19, 521–​526.
Aluko R, Keeratiurai M, Mine Y. 1998. Competitive adsorption
between egg yolk lipoproteins and whey protein on oil-​in-​
water interfaces. Colloids and Surfaces B: Biointerfaces, 10(6)
385–​393.
Beisson F, Ferte N, Bruley S, Voultoury R, Verger R, Arondel V. 2001.
Oil-​bodies as substrates for lipolytic enzymes. Biochimica et
Biophysica Acta, 1531(1–2), 47–​58.
Codex Commission. 2000. Proposed draft revised regional standard
for mayonnaise. CL2000/​17-​EURO.
Davey EM, Zabik ME, Dawson L. 1969. Fresh and frozen egg yolk
protein fractions: emulsion stabilizing power, viscosity, and
electrophoretic patterns. Poultry Science, 48(1), 241–​251.
304
Debon SJ, Moelants KR, Cabas Rodriguey Ld, Waschatko GM.
2017. Patent nr. WO 2017/​066569 A1.
Deckers HM, van Rooijen G, Boothe J, Goll,J, Mahmoud S, Moloney
M. 2000. United States of America Patent Nr. 6,146,645.
Deleu M, Vaca-​Medina G, Fabre JF, Valentin R, Mouloungui Z. 2010.
Interfacial properties of oleosins and phospholipids from rape-
seed for the stability of oil bodies in aqueous medium. Colloids
and Surfaces B: Biointerfaces, 80(2), 125–​132.
Fisk I, White D, Lad M, Gray D. 2008. Oxidative stability of
sunflower oil bodies. European Journal of Lipid Science
Technology, 110(10) 962–​968.
Gray D, Payne G, McClements D, Decker E, Lad M. 2010. Oxidative
stability of Echium plantagineum seed oil bodies. European
Journal of Lipid Science Technology,112(7) 741–​749.
Huang AH. 1992. Oil bodies and oleosins in seeds. Annual Review of
Plant Biology, 43(1), 177–​200.
Kapchie V, Bezelgues JB, Milo C, Leser ME, Sher AA. 2014.
Patent no. WO 2014/​154780 A1.
Karefyllakis D, van der Goot AJ, Nikiforidis CV. 2019. The behav-
iour of sunflower oleosomes at the interfaces. Soft Matter,
15(23), 4639–​4646.
Kermode A. 2003. Seed development. Physiology of maturation.
In Encyclopedia of Applied Plant Sciences (eds. Thomas B,
Murray BG and Murphy DJ), Academic Press, 1261–​1279.
Martinet VPS, Beaumal V, Courthaudon J, Anton M. 2003. Surface
properties of hen yolk low-​density lipoproteins spread at the
air-​water interface. Colloids and Surfaces B: Biointerfaces,
31(1–4), 185–​194.
Maurer S, Waschatko G, Schach D, Zielbauer B, Dahl J, Weidner T,
Vilgis TA. 2013. The role of intact oleosin for stabilization and
Juan C. Zambrano et al.
function of oleosomes. The Journal of Physical Chemistry,
117(44), 13872–​13883.
Mine YK. 2000. Selective displacement of caseinate proteins by hens
egg yolk lipoproteins at oil in water interfaces. Colloids and
Surfaces B: Biointerfaces, 18(1), 1–​11.
Motoki M, Seguro K. 1998. Transglutaminase and its use for food
processing. Trends in Food Science and Technology, 9(5),
204–​210.
Nikiforidis CK. 2009. Aqueous extraction of oil bodies from maize
germ and characterization of the resulting natural oil-​in-​water
emulsion. Journal of Agricultural and Food Chemistry, 57(12),
5591–​5596.
Perwaiz M. 2014. Trans fatty acids –​A risk for cardiovascular dis-
ease. Pakistan Journal of Medical Sciences, 30(1), 194–​197.
Ruhlman M. 2007. The Elements of Cooking: Translating the Chef’s
Craft for Every Kitchen. Simon and Schuster.
Tzen J, Cao YZ, Laurent P, Ratnayake C, Huang A. 1993. Lipid,
proteins, and structure of seed oil bodies from diverse species.
Plant Physiology, 101(1), 267–​276.
Vilgis TA. 2015. Gels: model systems for soft matter food physics.
Current Opinion in Food Science, 3, 71–​84.
Vilgis TA, Lendner I, Caviezel R. 2014. Ernährung bei
Pflegebedürftigkeit und Demenz. Springer-Verlag.
Zielbauer B, Jackson A, Maurer S, Waschatko G, Ghebremedhin
M, Rogers S, Vilgis TA. 2018. Soybean oleosomes studied by
small angle neutron scattering (SANS). Journal of Colloid and
Interface Science, 529, 197–​204.
Fats and Oils: Oxidation of Dietary Lipids
Luc Eveleigh
Sayfood (UMR 0782), INRAE, AgroParisTech, Université Paris-​Saclay, 91300, Massy, France
Occupying second place as an energy source in the Western diet,
lipids are important constituents of food. They are primarily pre-
sent as triglycerides, although the term “lipids” also includes
other constituents: sterols, phospholipids, etc. Here we are there-
fore interested in these triglycerides, i.e., tri-​esters of glycerol
and three fatty acids (Figure 43.1).
Table 43.1 shows the R-​COOH fatty acids most commonly
found in dietary fat.
In this chapter, we will abbreviate triglyceride compounds as
LH, where H designates one of the hydrogen atoms present in one
of the chains R1, R2 or R3.
Our experience shows us that lipids are subject to oxidation
(Table 43.2): they become rancid and denatured when heated. It
is oxidation that we will explore here.
The Autoxidation Mechanism
From a strict thermodynamic point of view, lipids, like most
other organic compounds, are not stable against oxidation and
therefore should not even exist in the biosphere. The normal
aqueous phase redox potentials of lipids are difficult to measure
(especially since the oxidation of a polyatomic molecule goes
through many steps). On the one hand, some values are known:
FIGURE 43.1 Structure of triglycerides. The chains R1, R2 and R3 are most often those of linear fatty acids with an even number of carbon atoms; these
fatty acid residues are saturated or have one, two or three, sometimes more, unsaturations.
E°(CO2/​CH4) = 0.17 V; E°(CO2/​C) = 0.21 V, to be compared
with E°(O2/​H2O) = 1.23 V.
On the other hand, lipids have been used as fuel since
antiquity, even prehistorically, as archaeological evidence shows
with stone lamps of the Palaeolithic age in which oil was burnt.
We are bathed in an oxidizing environment, since our atmos-
phere is composed of 20.95% oxygen, or an activity of 0.2 for the
average altitude at the surface of the Earth. However, our experi-
ence shows that the lipids around us, including those in our own
bodies, have a certain stability. Fat is sometimes even used as a
food preservation medium.
Indeed, due to its electronic structure, oxygen is extremely
unreactive: although it has 16 electrons, it has two unpaired par-
allel spin electrons. Compounds of this type, called triplets, are
extremely unreactive with respect to other species with an even
number of electrons, which almost all have only paired electrons
and are called singlet. This lack of reaction, i.e., an infinitely slow
reaction speed, can be interpreted in complicated quantum theor-
etical terms of preserving certain global properties of symmetry,
and is known as spin restriction.
However, oxidation phenomena can take place, in particular
via mechanisms involving intermediates that possess an unpaired
electron.
Thus, lipid oxidation reactions mainly follow a radical chain
mechanism. This mechanism requires first the presence of a lipid
305
306 Luc Eveleigh

free radical L•, i.e., a species that has an unpaired electron (a
so-​called doublet species) and is not subject to spin restriction.
L• comes from a bond cleavage in an L-​H lipid. In lipids that
have an unsaturated chain, it is most often an atom located at α
of a double bond, or in the so-​called malonic position in polyun-
saturated chains. Indeed, the L• radicals then have a particular
stability, as shown in Figure 43.2.
The energies of rupture of the L-​H bonds are of the order of
410 kJ/​mol for a carbon-​hydrogen bond in a CH2 group of a
saturated chain, 370 kJ/​mol for a hydrogen carried by the alpha
carbon of a double bond, and 305 kJ/​mol for hydrogen carried
by the carbon in the malonic position. Figure 43.3 illustrates the
general scheme of the process called lipid autoxidation. Rupture
takes place by capture of a hydrogen atom by any other radical
A• (pathway 1), by bond cleavage under the effect of ultraviolet
light (pathway 2) or in the presence of a metal ion (pathway 3).
The radical L• then enters the oxidation cycle:
• oxidation of L• to LOO• by oxygen.
• looping the chain : L• behaves as any radical in pathway
1) to form LOOH, a hydroperoxide called the primary
oxidation product.

TABLE 43.1 In fact, this chain is divergent: it generates more radicals than
Formula and Name of Fatty Acids Constituting Common Dietary Fats it consumes, since the great fragility of the O-​O bond of the
Short formula Name Origin (example) hydroperoxide can lead to the formation of two radicals, HO• and
C12:0 Lauric Coconut LO•. Once it is initiated, it is therefore very difficult to stop the
C14:0 Myristic Palm kernel lipid autoxidation process.
C16:0 Palmitic Palm Other oxidation mechanisms are possible. In particular,
C18:0 Stearic Tallow oxygen in an excited state (i.e., where the electron configuration
C16:1 Palmitoleic Macadamia is said to be singlet) proves to be highly reactive, especially, and
C18:1 Oleic Olive as expected, towards double bonds. The products of the action
C18:2 Linoleic Sunflower of singlet oxygen on LH lipids are also LOOH hydroperoxides
C18:3 α-​linolenic Walnut capable of decomposing into two radicals and thus launching the
C22:1 Erucic Rapeseed radical machine described earlier.
Several sources of singlet oxygen have been identified (Krinsky,
Note: For example, C18:1 is a fatty acid with 18 carbons and 1 unsat-
uration. There are many isomers, and the name given corres-
1977): in dietary lipids, it is probably the conjugated presence of
ponds to the isomer that is most abundant in nature. Oleic acid is triplet oxygen, light and a photosensitizer that causes the forma-
(9Z)-​octadec-​9-​enoic acid. tion of singlet oxygen (Usuki, 1984). This reaction requires little

TABLE 43.2
Oxidation products Identified by Frankel (1985) from the Oxidation of C18 Esters
Product Methyl oleate Methyl linoleate Methyl linolenate
aldehydes octanal pentanal propanal
nonanal hexanal butanal
dec-​2-​enal oct-​2-​enal but-​2-​enal
decanal non-​2-​enal pent-​2-​enal
deca-​2,4-​dienal hex-​2-​enal
nona-​3,6-​dienal
decatrienal

ester methyl heptanoate methyl heptanoate methyl heptanoate

methyl octanoate methyl octanoate methyl octanoate
methyl 8-​oxooctanoate methyl 8-​oxooctanoate methyl nonanoate
methyl 9-​oxononanoate methyl 9-​oxononanoate methyl 9-​oxononanoate
methyl 10-​oxodecanoate methyl 10-​oxodecanoate methyl 10-​oxodecanoate
methyl 10-​oxodec-​8-​enoate
methyl 11-​oxodec-​9-​enoate

alcohol heptan-​1-​ol pentan-​1-​ol

oct-​1-​en-​3-​ol

alkane heptane pentane ethane

octane pentane

Note: The mechanism was studied using methyl esters and not triglycerides, which explains the presence of methyl esters in the “esters” fields.
Fats and Oils: Oxidation of Dietary Lipids
FIGURE 43.2 Formation of L· from various aliphatic chains.
FIGURE 43.3 General scheme of lipid autoxidation.
energy, but the efficiency of the direct action of photons on the
triplet oxygen is extremely low. On the other hand, molecules such
as pigments (e.g., chlorophylls) easily capture light energy and are
thus promoted to an excited state. It is these excited molecules that
can, in turn, transmit energy to the triplet oxygen.
Final Oxidation Products
The hydroperoxides evolve, possibly via LO• radicals, into sec-
ondary oxidation products such as aldehydes, ketones, alkanes or
new shorter esters, as shown in Figure 43.4.
The detection of these products is the basis of a number of lipid
quality criteria: anisidine index, thiobarbituric acid (TBARS)
value and polar compound content.
Other molecules are produced at the same time as the oxida-
tion products: L• radicals can indeed react with another molecule
or, given the size of the molecules, with themselves. Polymers,
cyclic compounds and cyclic polymers may thus form, as shown
307
in Figure 43.5. The presence of these compounds increases the
viscosity of the oils and can cause the deposition of almost solid
matter. In the extreme, unsaturated oils show a drying behaviour,
which has been, for example, used in oil painting since the time
of Jan Van Eyck (15th century).
Degradation increases with temperature, according to
Arrhenius’ law. However, not all reactions have the same sen-
sitivity to temperature, and the proportions between different
oxidation products or other degradation pathways change with
temperature. In addition, the lighter degradation products (fewer
than 12 carbon atoms) are quite volatile and are likely to leave
the lipid phase that is undergoing oxidation. It is therefore dif-
ficult to accurately predict the composition of an oxidized fat:
time, temperature and the presence of oxygen come into play at
several levels. A large number of publications deal with oxidation
products obtained, for example, during frying (Choe and Min,
2007) or at room temperature (Guillén and Goicoechea, 2007). It
is, however, accepted that hydroperoxides, i.e., the first oxidation
products, do not accumulate in heated products: their decompos-
ition is too fast.
Most of the effects of oxidation are undesirable. Some products,
such as aldehydes, are toxic; others, such as hydroperoxides,
are involved in oxidative stress (Spikett and Forman, 2015).
Many volatile products, such as deca-​ 2,4-​
dienal, have an
unpleasant odour (Aladedunye, 2009), while the increase in vis-
cosity can make cooking operations or cleaning more difficult
(Sahasrabudhe, 2017).
How to Avoid or Limit Oxidation
In the kitchen, fats have to be managed with some care in order to
avoid oxidation. For example:
(1) Use stable oils; chains of unsaturated and especially
polyunsaturated fatty acids are more easily oxidized
than those of saturated fatty acids. Although the human
diet should not be too rich in saturated fatty acids, it
makes sense to use saturated oil such as hydrogenated
coconut oil for deep frying or clarified butter (Indian
ghee) for pan-​frying.
308
FIGURE 43.4 Products obtained by fragmentation of the two LO· radicals resulting from the oxidation of a radical in the malonic position.
FIGURE 43.5 Linear polymerization of a monounsaturated aliphatic chain (left) and cyclic polymerization of a di-​unsaturated aliphatic chain (right).
(2) Avoid over-​heating.
(3) Avoid the formation of radicals by storing lipids at
low temperature, away from light and air. For flat
frying, it is advisable to pour the oil into the hot pan
and not to heat the oil in the pan: as the temperature
will then be moderated by the presence of the food,
the initial formation of radicals by thermolysis will
be limited.
(4) The presence of antioxidants can delay the oxidation
phenomenon in several ways:
• by slowing down the initiation; this is, for example,
the case of metal-​chelating agents, which interfere
with reaction pathway (3) in Figure 43.3. Their
role is more interesting when lipids are in contact
with aqueous solutions where metal species are
dissolved, for example in emulsions. These may be
natural polyphenols or synthetic chelating agents,
such as EDTA (food additive E385).
Luc Eveleigh
• by inhibiting the activity of already formed
radicals: this is the role of the majority of natural
antioxidants, such as vitamin E, the polyphenols
of olive oil or essential oils (e.g., rosemary), or of
artificial antioxidants, such as BHA (food additive
E-​320) or BHT (E-​321). The principle of these
antioxidants is simple: if one of these molecules is
called AH, then the reaction
AH + L• → A• + LH
can occur, and the radical A• is relatively stable: it
takes a long time to react and enter the chain reac-
tion depicted in Figure 43.3. However, a massive
supply of antioxidant can cause the storage of
radical species in a lipidic medium and can pro-
mote oxidation in the long term (Zhou and Elias,
2013).
Fats and Oils: Oxidation of Dietary Lipids
• by consuming oxygen, i.e., by serving as a sacri-
ficial species vis-​à-​vis the oxidation phenomenon.
Most antioxidants in the previous category can play
this role, but they are then destroyed. The most
common of these antioxidants are water-​soluble,
like vitamin C, and are of less interest against lipid
oxidation.
• in dietary lipids, antioxidant concentrations are not
sufficient for the presence of an oil in a formulation
to protect the mixture. Thus, it has now been shown
(Blais, 2012) that the addition of a little olive oil
does not protect butter from browning, especially
since it is not a simple lipid oxidation (see the
chapter on glycation and Maillard reactions by
Tessier). At most, the dilution of all species when
mixing two ingredients can slow the reactions,
provided their kinetics are not of zero order.
Conclusion
Because of the problems caused by its products, especially
odours and toxicity, lipid oxidation must be limited. Knowing
that lipid oxidation is very difficult to avoid, especially during
non-​industrial cooking operations, it is highly advisable to use
the least oxidizable fats: saturated fats in moderate quantities,
and monounsaturated oils such as olive oil or oleic sunflower oil.
Polyunsaturated fats that can provide us with essential fatty acids
should be reserved, as far as possible, for seasoning.
REFERENCES
Aladedunye FA, Przybylski R. 2009. Degradation and nutritional
quality changes of oil during frying, Journal of the American
Oil Chemist’s Society, 86, 149–​156.
309
Blais C. 2012. www.ricardocuisine.com/​ chroniques/​ chimie-​
alimentaire/​436-​l-​huile-​empeche-​le-beurre-​de-​bruler-mythe-
ou-realite
Choe E, Min D. 2007. Chemistry of deep-​fat frying oils, Journal of
Food Science, 72, R77–​R86.
Frankel E. 1985. Chemistry of autoxidation: mechanism, products
and flavor significance. In Min DB and Smouse TH (eds.),
Flavor Chemistry of Fats and Oils, American Oil Chemists’
Society, 1–​34.
Guillén M, Goicoechea E. 2007. Detection of primary and secondary
oxidation products by Fourier transform infrared spectros-
copy (FTIR) and 1H nuclear magnetic resonance (NMR) in
sunflower oil during storage, Journal of Agricultural Food
Chemistry, 55, 26, 10729–​10736.
Krinsky NI. 1977, Singlet oxygen in biological systems, Trends in
Biochemical Sciences, 2, 35–​38.
Sahasrabudhe SN, Rodriguez-​Martinez V, O’Meara M, Farkas BE.
2017. Density, viscosity, and surface tension of five vege-
table oils at elevated temperatures: measurement and mod-
eling, International Journal of Food Properties, 20, S2,
S1965–​S1981.
Spickett CM, Forman HJ. 2015. Lipid Oxidation in Health and
Disease, CRC Press.
Usuki R, Endo Y, Kaneda T. 1984, Prooxidant activities of
chlorophylls and phenophytins on the photooxidation of edible
oils, Agricultural and Biological Chemistry, 48, 991–​994.
Zhou L, Elias R. 2013. Understanding antioxidant and prooxidant
mechanisms of phenolics in food lipids. In Logan A, Nienaber
U and Xiangqing P (eds.) Lipid Oxidation: Challenges in Food
Systems, AOCS Press.
Fats and Oils: Extra Virgin Olive Oil in Cooking –​Molecular Keys
for Traditional and Modern Mediterranean Gastronomy
Raffaele Sacchi
Department of Agricultural Sciences, Unit of Food Science and Technology, University of Naples Federico II,
Via Università 100, I-​80055 Portici (Napoli), Italy
Extra virgin olive oil (EVOO) is traditionally used in
Mediterranean countries, not only for oil dressing (raw use)
but widely for food cooking and frying. Phenolic antioxidants
and flavour compounds in EVOO show many molecular
interactions with other ingredients during cooking for traditional
Mediterranean food preparations. The use of raw EVOO added to
foods after cooking (or mixed without cooking, or as a salad oil)
is the best way to express the original flavour and to maximize the
intake of natural antioxidants and compounds related to positive
effects on human health (e.g., hypotensive, anti-​inflammatory
and anti-​carcinogenic activities), but a protective action is also
exhibited by EVOO components during cooking. The biophenols
and flavours, in particular, also interact (through hydrolysis, oil/​
water (O/​W) partitioning, complex formation, covalent and non-​
covalent binding) with other ingredients (tomato, meat, fish,
potato, vegetables and milk proteins) with a functional role in
cooked foods.
What Is EVOO?
Extra virgin olive oil (EVOO) is obtained from olive fruits exclu-
sively by physical-​mechanical technologies (olive crushing, cen-
trifugation, filtration, etc.), thus ensuring that this ‘lipid fruit juice’
has a particular composition, olive triacylglycerols becoming a
‘solvent’ for several natural fruit components (antioxidants, fla-
vour, pigments, etc.) responsible for its sensory and nutritional
quality (Sacchi, 2008) (Figure 44.1).
In the last few years, several papers have correlated the in
vitro and in vivo positive actions of EVOO on human health to
its chemical composition (Preedy and Watson 2010; Frankel
2011). Health properties of EVOO were attributed to both the
high level of oleic acid in triacylglycerols and to many different
minor compounds (squalene, tocopherols, pigments and volatile
compounds) and, in particular, to phenolic compounds such as
phenyl alcohols, secoiridoids and lignans (Figure 44.2) present in
the unsaponifiable fraction of EVOO (Frankel, 2011).
Antimicrobial, antioxidant and anti-​ inflammatory activities
were significantly associated with EVOO (Cicerale et al., 2012).
As regards anti-​ inflammatory benefits of EVOO, particular
attention was paid to the phenolic compound named oleocanthal
(Lucas et al., 2011). This was demonstrated to have ibuprofen-​
like activity, as both compounds inhibit the same cyclooxygenase
enzymes in the prostaglandin-​biosynthesis pathway (Bechaump
et al., 2005). However, a scientific demonstration that oleocanthal,
in the amount contained in EVOO, can be the only compound
responsible for EVOO’s anti-​inflammatory effect is still lacking
(Fogliano and Sacchi, 2006). In fact, mounting evidence in food
science and health research indicates that it is the complex mix-
ture of polyphenols in foods, rather than individual compounds,
that can synergistically act towards a final health effect. In the case
of EVOO, this concept has been demonstrated for pinoresinol
and colon cancer (Fini et al., 2008). Those authors reported that
pinoresinol-​rich EVOO extracts have potent chemo-​preventive
properties in colon cancer cells and that this result was achieved
at substantially lower concentrations in EVOO than with purified
pinoresinol (Fini et al., 2008).
The European Food Safety Authority (EFSA) Panel on Dietetic
Products, Nutrition and Allergies (NDA) in 2011 provided a sci-
entific opinion on the scientific substantiation of health claims
in relation to polyphenols in olive and protection of low-​density
lipoprotein (LDL) particles from oxidative damage, maintenance
FIGURE 44.1 Extra virgin olive oil (EVOO) can be considered a nat-
ural ‘lipidic fruit juice’ containing functional compounds (biophenols,
tocopherols, pigments, flavours, etc.) and, at the same time, a ‘funcional food’
and a ‘food flavour’.
311
312
FIGURE 44.2 Chemical structure of the main phenolic compounds in
EVOO: (1) hydroxytyrosol; (2) tyrosol; (3) dialdehydic form of oleuropein
aglycone; (4) dialdehydic form of ligstroside aglycone (oleocanthal);
(5) lignans; (6) oleuropein aglycone (oleacein).
of normal blood high-​ density lipoprotein (HDL)-​ cholesterol
concentrations, and maintenance of normal blood pressure: ‘anti-​
inflammatory properties’, ‘contributes to the upper respiratory
tract health’, ‘can help to maintain a normal function of gastro-
intestinal tract’ and ‘contributes to body defences against external
agents’.
The food constituent that is the subject of the health claims is
polyphenols in olive (olive fruit, olive mill waste waters or olive
oil, and Olea europaea L. extract and leaf). On the basis of the
data presented, the Panel concludes that a cause and effect rela-
tionship has been established between the consumption of olive
oil polyphenols (standardized by the content of hydroxytyrosol
and its derivatives) and protection of LDL particles from oxi-
dative damage. The Panel considers that in order to support
the claim, 5 mg of hydroxytyrosol and its derivatives (e.g.,
oleuropein complex and tyrosol) in olive oil should be consumed
daily (EFSA, 2011).
Along with chemical stability towards oxidation processes
(high oxidation stability) due to antioxidant capacity, phenolic
compounds in EVOO also contribute to its bitter and pungent
taste, influencing its acceptability (Cavallo et al., 2019). The
phenol compounds and the natural volatile compounds with the
‘olive fruity’ aroma make of EVOO, at the same time, a ‘func-
tional food’ and a ‘food flavour’ (Sacchi, 2008).
The large variation in chemical and sensory quality found
on the market among EVOOs, as well as in price, indicates for
Raffaele Sacchi
those products also a wide range of nutritional properties. In
particular, food chemists and panels of experts agree on the fact
that more bitter EVOOs have a higher quality, and thus, EVOO
taste may be a good driver for consumers to make a choice
towards products having healthy characteristics (Vitaglione
et al., 2015).
Changes in phenolic and volatile compounds also arise
during cooking of EVOO, as well as many interactions with
other ingredients used in Mediterranean food preparations. The
culinary aspects of EVOO, however, have been analysed in only
a few papers and already reviewed (Sacchi et al., 2014).
The aim of this chapter is to give a brief overview of the
interactions of EVOO in Mediterranean foods and to underline
some basic molecular keys to explain the ‘magic’ of EVOO and
its role in Mediterranean gastronomy.
EVOO in ‘Mediterranean Molecular Cuisine’
The use of raw EVOO added to foods after cooking (or as a salad
oil) is the best way to express the original flavour and to maxi-
mize the intake of natural antioxidants and EVOO compounds
associated with positive effects on human health (e.g., hypoten-
sive, anti-​inflammatory and anti-​carcinogenic).
EVOO, however, also exhibits its protective properties when
it is used in cooking. Chemical interactions between EVOO
biophenolic compounds and other food ingredients (water, milk
proteins, carotenoids of tomato, omega-​3 polyunsaturated fatty
acids in canned-​in-​oil fish, meat or fish proteins, etc.) have been
investigated. During cooking, EVOO exhibits strong antioxidant
properties and influences the overall flavour of cooked foods.
The physical (partitioning, emulsion) and chemical (hydrolysis,
covalent binding, antioxidant properties) phenomena occurring
when EVOO is cooked with other food ingredients also involve
changes in sensory (bitterness and fruity) and nutritional quality
of some traditional Mediterranean foods.
Molecular interactions also occur during cooking between
EVOO bitter phenolic compounds and other compounds in
different cooked food systems, thus increasing the healthy prop-
erties of cooked foods. These will be discussed in the following
sections.
EVOO in Fish Canning and Cooking
The interactions between EVOO used as filling oil and canned
tuna muscle allows the preservation of the intake of native n-​3
polyunsaturated fatty acids (PUFAs) present in fresh fish muscle
before tuna canning. The level of n-​3 PUFAs, measured by proton
nuclear magnetic resonance (NMR) spectroscopy after tuna can
sterilization, was significantly higher in tuna canned in EVOO
compared with tuna muscle canned and sterilized in soybean or
refined olive oil, or in brine (Medina et al., 1995; Medina et al.,
1998a). The level of oxidation found in lipids extracted from
canned tuna muscle was also lower when canned in EVOO. The
protective effect of EVOO during/​after the thermal treatment of
Fats and Oils: Extra Virgin Olive Oil
FIGURE 44.3 Mechanism of hydrolysis and partitioning in water of
EVOO secoiridoid aglycones during the thermal treatment of EVOO in
O/​W systems (canned tuna in EVOO/​brine, tomato/​EVOO sauces, etc.).
(1) Tyrosol; (2) hydroxytyrosol; (3) dialdehydic form of oleuropein aglycone;
(4) dialdehydic form of ligstroside aglycone; (7) elenolic acid.
the cans can be attributed to the properties of natural antioxidants
of EVOO, which are not present in other filling oils or brine
(Medina et al., 1998b; Fogliano et al., 1999). In particular, by
monitoring the level of phenol compounds in filling oils before
and after the sterilization of the cans, a strong partitioning from
the oil phase towards the water phase (muscle) was observed
after the thermal treatment, combined with the hydrolysis of
secoiridoid aglycones (Sacchi et al., 2002; Brenes et al., 2002)
(Figure 44.3).
The combination of these two phenomena (partitioning and
hydrolysis) leads to the accumulation of hydrophilic phenolic
antioxidants (hydroxytyrosol and tyrosol) on the muscle surface,
protecting the n-​3 PUFA from thermal oxidation.
Tomato–​EVOO Interactions
Fresh tomato and tomato products are characterized by healthy
value and anti-​ carcinogenic activity, especially towards pros-
tate cancer (Pannellini et al., 2010). The intake of carotenoids
(lycopene) has been related to anti-​ carcinogenic activity and
anti-​atherogenic effects both in vitro and in vivo (Omoni and
Aluko, 2005).
In the human diet, both fresh and transformed tomatoes are
consumed around the world, but typical (Italian) Mediterranean
gastronomy is characterized by a high use of tomato and olive oil
in combination with pasta, pizza, etc.
313
FIGURE 44.4 The slow cooking (80 °C for 8–​10 hours) of traditional
Mediterranean sauces (i.e., the Neapolitan Ragù) containing tomato purée
and EVOO is traditionally done in traditional earthenware pots.
FIGURE 44.5 Evolution of antioxidant activity (ABTS, mmol Trlox/​100 g)
in EVOO–​tomato systems during cooking at 80°C for 8 hours (nt, mixture not
treated thermally; 2h–​8h, 2–​8 hours of continous treatment). The oily phase
(O, yellow) and the tomato phase (W, red) were separated by centrifugation
after heating and analysed for their antioxidant activity (ABTS).
(From Pernice et al., 2007, modified)
In recent years, we studied the behaviour of different model
systems simulating the cooking of tomato sauces in combination
with EVOO. In particular, some traditional tomato sauce prep-
aration (i.e., the Neapolitan Ragù) requires long cooking (6–​10
hours) at medium-​low temperatures (70–​80 °C) in traditional
earthenware pots (Figure 44.4). What happens when a tomato–​
oil system is heated under these mild conditions? A loss in the
antioxidant content can be expected at the end of cooking, and
this has been verified in tomato sauce submitted to a heating
test (Pernice et al., 2007). Surprisingly, a significant increase in
the antioxidant properties of both water and oil phases during
cooking of tomato sauce with EVOO was observed (Figure 44.5),
with a protective action exhibited by phenolic compounds of
EVOO on tomato carotenoids. The chemical/​physical behaviour
of secoridoid aglycones present as main compounds in EVOO
314
FIGURE 44.6 Evolution of carotenoids (mg/​100 g) of a tomato purée/​
EVOO mixture during heating at 80 °C for 8 hours (nt, treated thermally;
2h–​8 h, 2–​8 hours of continous treatment) in the oily phase (O, yellow) and
tomato (W, red) separated by centrifugation after heating and separately
extracted and analysed for carotenoid content.
(From Pernice et al., 2007, modified)
when added to tomato sauce is characterized by partitioning
towards the acid-​water phase of tomato juice, and hydrolysis
of secoridoids in elenolic acid and phenyl alcohols (tyrosol
and hydroxytyrosol), similarly to other heated food systems, as
described earlier for tuna canning and sterilization (Figure 44.3).
These compounds, by partitioning towards the tomato phase and
ester bond hydrolysis, lost their bitter-​pungent properties and
exhibited, at the same time, their antioxidant action in the water
phase (Pernice et al., 2007).
On the other hand, the partitioning of carotenoids towards the
oil phase was also observed during cooking of tomato/​EVOO
mixtures (Figure 44.6) as well as that of some tomato flavonoids
(naringenin) (Sacchi et al., unpublished data). These molecular
phenomena can also explain the antioxidant efficiency, carot-
enoid bioavailability and in vivo effects of food preparations in
which EVOO is used (Lee et al., 2000).
Arranz et al. (2015) also evaluated the effect of adding olive
oil to tomato juice (not treated with heat) on the bioavailability
of plasma carotenoids and postprandial lipid response. In a
randomized, controlled, crossover feeding trial, volunteers were
assigned to receive a single ingestion of 750 g of tomato juice
(TJ) containing 10% of refined olive oil/​70 kg body weight (BW)
and 750 g of TJ without oil/​70 kg BW on two different days.
Levels of all lycopene isomers increased significantly in subjects
consuming TJ with oil, reaching a maximum concentration at
24 h. LDL-​cholesterol and total cholesterol levels decreased sig-
nificantly 6 h after the consumption of TJ with oil, which was
significantly correlated with an increase of trans-​lycopene and
5-​cis-​lycopene, respectively. Thus, the positive effect on bioavail-
ability and absorption of carotenoids of olive oil is expressed also
without cooking and can be considered in evaluating the nutri-
tional properties of the classical Mediterranean tomato salads or
fresh combinations of tomato juice like Gazpacho ‘funtionalized’
by the addition of EVOO.
Vallverdú-​Queralt et al. (2014) studied the effect of cooking
time (15, 30, 45 and 60 min) and the addition of extra virgin
olive oil (5% and 10%) on the phenolic content of tomato sauces
by using liquid chromatography coupled with tandem mass
Raffaele Sacchi
FIGURE 44.7 Meat and fish are traditionally marinated in EVOO before
roasting. This operation reduces the formation of carcinogenic heterocyclic
amines on the surface, giving more healthy roasted foods.
spectrometry (LC-​MS). The concentration of phenolics in the
tomato sauces decreased during the cooking process, with the
exception of caffeic acid and tyrosol. The main degradation
observed was the oxidation of quercetin, since the hydroxy-​
function at the C-​ring of this flavonoid is not blocked by a sugar
moiety, unlike rutin. Higher levels of virgin olive oil in tomato
sauce enhance the extraction of phenolic compounds from the
tomato, leading to higher phenolic contents in the sauces. These
data are in accordance with the trend shown in Figure 44.5 for
antioxidant activity, which increased in the tomato phase after 2
hours, but showed a slight decrease up to 8 hours. Another phe-
nomenon to be considered in these systems is Maillard reactions
with the newly formed melanoidins, which contribute to increased
antioxidant power of cooked foods (Pernice et al., 2007).
All these observations and data can be of interest in relation to
the day-​by-​day preventive action of some ‘traditional functional
foods’ such as Neapolitan pizza, pasta with tomato sauces (like
the traditional Neapolitan Ragù), ‘gazpacho’, ‘bruschetta’ and
other dishes in which EVOO and tomato are cooked or mixed
together. These traditional preparations can be considered to be
very healthy, as they allow very efficient absorption of functional
biomolecules (phenols, carotenoids and flavonoids) having anti-
oxidant and protective roles in vivo.
Marinating Meat in EVOO before Roasting
Another traditional process in the Mediterranean cooking style is
marinating meat and fish in oil, wine and herbs (oregano, rose-
mary) before roasting and the use of an oil–​lemon juice or oil–​red
wine emulsion during the cooking to wet the meat/​fish surface
(Figure 44.7). This tradition, in part lost today in home and res-
taurant practice, has been demonstrated to have a protective
effect against protein degradation during the thermal treatment
of cooking (Monti et al., 2001; Persson et al., 2003; Vitaglione
and Fogliano, 2004). Both phenol compounds and EVOO added
to model systems simulating cooking show a notable inhibition of
heterocyclic amine (HA) formation. Mutagenic HAs are formed
at low levels during cooking of meat and fish, and some of these
are considered to be possible human carcinogens. The formation
of HAs may be affected by the presence of synthetic or natur-
ally occurring antioxidants. Monti et al. (2001) studied the effect
of EVOO phenolic compounds, identified and quantified by
Fats and Oils: Extra Virgin Olive Oil
LC-​MS, on the formation of HAs in a model system. An aqueous
solution of creatinine, glucose and glycine was heated in the
presence of two samples of EVOO differing only in the com-
position of phenolic compounds. The addition of EVOO to the
model system inhibited the formation of HAs by between 30%
and 50% compared with the control. Fresh-​made olive oil, which
contained a high amount of dihydroxyphenylethanol derivatives,
inhibited HA formation more than a one-​year-​old oil did. The
inhibition of HA formation was also verified using phenolic
compounds extracted from virgin olive oil (VOO), demonstrating
that phenol-​rich EVOO can play an interesting functional role
during meat and fish cooking/​roasting in preventing the forma-
tion of potential carcinogenic molecules. This function should
also be applied in modern Mediterranean-​like food design and
development of novel foods in which the role of EVOO can be
used in cooked foods to protect against protein degradation.
Cooking in EVOO (Deep-​Frying, Pan-​Frying,
Sautéing, Sofrito)
EVOO is commonly used in traditional Mediterranean cooking
processes in different amounts and at different time–​temperature
combinations: from deep-​frying (a few minutes at 170–​180 °C)
(Figure 44.8), to pan-​frying (20 min at 120 °C) and sautéing (a
few minutes at 90–​100 °C in the presence of water). In these
cooking techniques, there are also exchanges of matter between
the oil and the cooked food, with complex chemical, physical and
chemical-​physical interactions.
Deep-​frying in EVOO showed some functional effects: (i)
the uptake of triacylglycerols rich in oleic acid and healthy
biophenolic antioxidants in the crust of fried-​in-​EVOO foods,
FIGURE 44.8 Deep-​frying in EVOO is a common traditional practice
applied in Mediterrranean countries to fry fish, meat, vegetables, potatoes, etc.
EVOO is also widely used for pan-​frying, sautéing, ‘sofrito’ of onion, garlic,
table olives and capers before addition of water, tomato sauce, etc. Fried-​in-​
EVOO foods absorb olive oil and its antioxidants and may be considered as
‘healthy functional fried foods’.
315
(ii) the minimization of the production of potentially toxic
compounds (acrylamide, hydroxy-​alkenals), (iii) the protective
action of EVOO antioxidants during frying on other food
ingredients (carotenoids in vegetables, fatty acids in meats), and
(iv) exchange of lipids, when fatty foods are fried with EVOO
and the meat lipids are lost in the frying bath (i.e., saturated fats
from meats, PUFAs for fish) and absorption of frying oil on the
crust with an increase in monounsaturated fatty acid (MUFA)
content in fried foods.
The biophenolic compounds of EVOO are quite stable during
deep-​frying, being found even after several hours of frying (Della
Medaglia et al., 1996; Ambrosino et al., 2002). They can interact
with the food matrix, inhibiting the formation of dangerous
compounds (acrylamide, toxic aldehydes).
The relationship between phenol compounds in EVOO and the
formation of acrylamide in potato crisps was first investigated by
Napolitano et al. (2008). The phenolic composition of 20 EVOO
samples was screened by LC-​MS, and four oils, characterized
by different phenol compound patterns, were selected for frying
experiments. Slices of potatoes were fried at 180 °C for 5, 10 and
15 min in EVOO, and the acrylamide content was determined by
LC-​MS. EVOO phenolic compounds were not degraded during
frying, and the crisp colour was not significantly different among
the four EVOOs. Acrylamide concentration in crisps increased
during frying time, but the formation was faster in the oil having
the lowest concentration of phenolic compounds. Moreover, the
EVOO having the highest concentration of ortho-​ diphenolic
compounds was able to efficiently inhibit acrylamide formation
in crisps from mild to moderate frying conditions. The use of
ortho-​diphenolic-​rich EVOOs was proposed as a reliable miti-
gation strategy to reduce acrylamide formation in domestic
deep-​frying. Interestingly, the presence of EVOO phenolic
compounds is also able to reduce the formation of acrylamide in
low-​moisture systems such as biscuits in oven cooking (Arribas-​
Lorenzo et al., 2009).
Hydroxy-​alkenals are other potentially toxic and carcinogenic
compounds arising from the thermal decomposition of PUFAs
during frying (Frankel, 1998). By high-​resolution proton nuclear
magnetic resonance (1H-​NMR) spectroscopy (400–​600 MHz),
hydroperoxide decomposition products in thermally oxidized oils
were quantitatively analysed (Sacchi et al., 2006). Different oils
(EVOO, sunflower and soybean) heated in a thermostatic bath
fryer (180 °C for 360 min) showed the NMR signals of alde-
hyde (n-​alkanals, trans-​2-​alkenals, 4-​hydroxy-​trans-​2-​alkenals
and alka-​2,4-​dienals) and were monitored in 15 oil samples (0,
60, 120, 240 and 360 min heating). The 4-​hydroxy-​2-​alkenals
were not detected (threshold 0.1 mM/​ L) in EVOO after 6
hours’ heating, but only in PUFA-​rich fried oils. The formation
of this compound, in fact, was related to the decomposition of
conjugated hydroperoxydienes arising from the oxidation of
PUFAs (Frankel, 1998). In olive oil, the low amount of lino-
leic (5–​10%) and linolenic (less than 1%) acids explains these
findings. For the same reason, alka-​2,4-​dienals were formed in
small amounts (1.1 mM/​L oil after 6 hours’ heating) compared
with polyunsaturated seed oils.
In terms of the uptake of phenol compounds in fried-​in-​EVOO
foods, the crust of French fries, when EVOO was used as frying
316
oil in continuous frying, was demonstrated to absorb a signifi-
cant amount of phenol compounds, which can be easily extracted
from the fried potatoes and quantified by LC-​ MS (Savarese
et al., 2006).
Kalogeropoulos et al. (2007a) studied the behaviour of finfish,
representing the most popular fish species in Greece, during pan-​
frying in VOO. Analyses for polyphenols, hydroxy pentacyclic
triterpene acids (HPTA) and α-​tocopherol were performed in the
fresh and fried oils and fish. Nine polyphenols were determined
in the frying oil samples; six of them were also found in fried
fish. The terpenic acids oleanolic, maslinic and ursolic acid were
also determined in frying oils and fried fish. Besides water loss
and oil absorption, pan-​frying caused the partial loss of all the
antioxidants studied in the fried oils, as well as their enrichment
in the fried fish. The polarity of the antioxidants studied affected
to some extent their partitioning between the frying oil and the
water-​containing fish.
The same authors (Kalogeropoulos et al., 2007b) studied
potatoes, green peppers, zucchini and eggplants shallow-​fried
in EVOO according to traditional Mediterranean culinary prac-
tice. Zucchini and eggplants were also blanketed with wheat
flour or batter prior to frying. Among 12 polyphenols determined,
tyrosol predominated in frying oils and zucchini samples, while
chlorogenic acid was the major phenolic species in the other
vegetable samples. Besides water loss and oil absorption, shallow
frying resulted in partial loss of all the antioxidants studied in
frying oils and enrichment of fried vegetables with olive oil
antioxidants, which was to some extent affected by the type of
vegetable fried and the culinary practice followed. The overall
retention of the antioxidants in oil and food ranged from 32%
to 64% for alpha-​tocopherol, from 25% to 70% for polyphenols
and from 35% to 83% for HPTA. Vegetables fried in EVOO pro-
vide an additional intake of alpha-​tocopherol, terpenic acids and
polyphenols such as tyrosol and chlorogenic acid.
The performance of VOO and a commercial vegetable
shortening was also investigated by Andrikopoulos et al. (2002a)
during ten successive pan-​fryings of potatoes at 180 °C for a total
period of 60 min and during ten successive deep-​frying cycles at
170 °C for a total period of 120 min. For both the oils tested, the
effect of pan-​frying was worse than the effect of deep-​frying. The
same was true for visible spectrum and total phenols in VOO.
Both oils performed similarly during pan-​frying, while VOO
performed better during deep-​frying. A very strong correlation
between octanoic acid formation and total polar artefacts in the
whole data set was observed.
Andrikopoulos et al. (2002b) also submitted VOO, sun-
flower oil and a vegetable shortening to deep-​frying and pan-​
frying of potatoes, for eight successive sessions, under the usual
domestic practice. The frying oil absorption by the potatoes was
determined to be within 6.1–​12.8%, depending on the oil type and
the frying process. The retention of total phenolics ranged from
70–​80% (first frying) to 20–​30% (eighth frying). Tannic acid,
oleuropein and hydroxytyrosol-​elenolic acid dialdehydic form
showed remarkable stability in all frying sessions for both frying
methods, while hydroxytyrosol and hydroxytyrosol-​elenolic acid
were more rapidly eliminated. The deterioration of the other
Raffaele Sacchi
phenolic species was 40–​50% and 20–​30% for deep-​frying and
pan-​frying, respectively, after three to four frying sessions, which
is most usual in the household kitchen.
The migration of health-​promoting micro-​constituents from
frying vegetable oils to French fries was also studied by Chiou
et al. (2012), who analysed the behaviour of vitamin E in this
thermal process. EVOO, water and a water/​oil mixture (W/​O)
were also used for frying, boiling and sautéing Mediterranean
vegetables (potato, pumpkin, tomato and eggplant) by Ramìrez-​
Ayala et al. (2019). Differences in antioxidant capacity (AC)
and levels of individual phenols in unused and used EVOO and
water were determined. The water used to boil tomatoes showed
the highest phenolic value, whilst the lowest was found in the
EVOO from the W/​O used for boiling potatoes. After processing,
the concentrations of phenols exclusive to EVOO diminished to
different extents. There was a greater transfer of phenols from
the vegetable to the oil when eggplant, tomato and pumpkin were
cooked. W/​O boiling enriched the water in most of the phenols
analysed, such as chlorogenic acid and phenols exclusive to
EVOO. The values of AC decreased or were maintained when
fresh oil was used to cook the vegetables (raw > frying > sautéing
> boiling). The phenolic content and AC of EVOO decreased
after cooking Mediterranean diet vegetables. Furthermore, water
content was enriched after the boiling processes, particularly
when oil was included.
Lozano-​Castellòn et al. (2020) studied how temperature, time
and the interaction between these affect the EVOO polyphenolic
profile during a domestic pan-​ frying process, simulating the
cooking conditions in a home kitchen without control of light or
oxygen. EVOO was processed at two temperatures (120 °C and
170 °C) for either a short time (30 min) or a long time (60 min),
and the polyphenol content was monitored. Temperature degraded
the polyphenols of EVOO during the sautéing, whereas time had
an effect on some individual phenols, such as hydroxytyrosol, but
not on the total phenol content. The polyphenol content decreased
by 40% at 120 °C and 75% at 170 °C compared with raw EVOO.
In a study by Vallverdú-​ Queralt et al. (2013), different
Mediterranean sofritos were also analysed for their content of
polyphenols and carotenoids. Simple phenolic and hydroxy­
cinnamoylquinic acids, and flavone, flavonol and dihydrochalcone
derivatives, were identified in Mediterranean sofrito. The quanti-
fication levels of phenolic and carotenoid compounds led to the
distinction of features among different Mediterranean sofritos
according to the type of vegetables (garlic and onions) or olive
oil added for their production.
All these studies confirm the active role of EVOO in
Mediterranean frying and cooking, with an exchange of
antioxidants with the cooked ingredients (O/​W partitioning), and
its protective role against thermal oxidation and preservation of
functional components during cooking.
EVOO and Milk Proteins
Interactions between EVOO compounds and milk proteins can
also occur, modifying the sensory perception and nutritional
Fats and Oils: Extra Virgin Olive Oil
FIGURE 44.9 Milk proteins with bitter biophenols and pungent flavour
compounds (aldehydes) from EVOO, when mixed (without cooking), interact
by forming covalent or non-​covalent bonds. This effect reduces the bitterness
and pungency of EVOO without altering the bioavailability of these bitter-​
pungent functional molecules (antioxidant, hypotensive, anti-​carcinogenic
and influencing the sense of satiety).
quality, not only in cooking but also when mixing EVOO with
milk, fresh cheeses or ice cream at room or low temperature.
Meynier et al. (2004) demonstrated that hexanal and t-​2-​
hexenal, volatile compounds found in high amounts in EVOO,
form covalent bonds with whey proteins and sodium caseinate,
causing changes in the amino acid composition of proteins.
A similar behaviour can be observed from the interaction
between EVOO phenolic/​volatile compounds and milk protein,
with a decrease of the perceived ‘bitterness’ ‘pungency’ and
‘green-​grass’ attributes in food systems containing milk, cream
and ricotta cheese, as well as in artisanal ice-​cream made with
EVOO (Sacchi et al., 2019) (Figure 44.9).
Another important issue is the perception of volatile and
non-​volatile (biophenols) compounds in relation to oil-​ in-​
water emulsions. In the case of emulsions, hydrophobic flavour
components can be perceived at lower concentrations in water
than in oil, and many of the lipid oxidation products show high
solubility in oil phase (Beltran et al. 2005). In food systems in
which EVOO and milk proteins are present, these interactions
cause strong modifications of the sensory properties of EVOO,
with an evident loss of bitterness and pungency and good accept-
ability of ice-​cream (Sacchi et al., 2019).
Conclusions
The scientific studies discussed here demonstrate an active func-
tional role of EVOO, not only as a source of antioxidants when
used as a raw ingredient in the Mediterranean Diet but also in
protecting other food components during cooking. The use of
raw EVOO added to foods after cooking (or as a salad oil) is
the best way to express the original flavour and to maximize the
intake of natural antioxidants and compounds related to posi-
tive effects on human health. EVOO, however, also exhibits
its protective properties during/​after cooking. Different chem-
ical interactions between EVOO biophenolic compounds and
other food ingredients (e.g., water, milk proteins, carotenoids
317
of tomato, omega-​3 PUFAs in canned-​in-​oil fish, meat or fish
proteins) occur and may positively change sensory and nutri-
tional properties of cooked Mediterranean foods.
Several positive interactions with other food components
(carotenoids, omega-​3 PUFAs and proteins) occur, and cooking
foods with EVOO can improve their nutritional quality and anti-​
carcinogenic effects, as has been well demonstrated for cooked
tomato–​EVOO mixtures (Lee et al., 2000).
The physical (partitioning, emulsion) and chemical (hydrolysis,
covalent binding, antioxidant properties) phenomena occurring
during cooking of EVOO, discussed here with an emphasis on
the changes in sensory (bitterness and fruity) and nutritional
quality of traditional Mediterranean foods, can contribute to
understanding the basic molecular keys of the ‘magic’ of EVOO
and its role in traditional and modern Mediterranean gastronomy.
Acknowledgements
This work has been supported in part by AGER 2 Project, grant
n. 2016-​0174 ‘Claims of Olive oil to iMProvE The market ValuE
of the product (COMPETiTiVE)’.
REFERENCES
Ambrosino ML, Della Medaglia D, Paduano A.2002. Behaviour
of phenolic antioxidants of virgin oil during oil heating and
frying. In Amado R (ed.) Bioactive Compounds in Plant
Foods: Health Effects and Perspectives for the Food Industry.
European Scientific Conference of COST Action 916. Vol EUR
20061. EC Office for Official Publications, 119–​120.
Andrikopoulos NK, Dedoussis GVZ, Falirea A. 2002b. Deterioration
of natural antioxidant species of vegetable edible oils during the
domestic deep-​frying and pan-​frying of potatoes. International
Journal of Food Science and Nutrition, 53, 351–​363.
Andrikopoulos NK, Kalogeropoulos N, Falirea A. 2002a.
Performance of virgin olive oil and vegetable shortening during
domestic deep-​frying and pan-​frying of potatoes. International
Journal of Food Science and Technology, 37, 177–​190.
Arranz S, Martínez-​ Huélamo M, Vallverdu-​ Queralt A. 2015.
Influence of olive oil on carotenoid absorption from tomato
juice and effects on postprandial lipemia. Food Chemistry,
168, 203–​210.
Arribas-​Lorenzo G, Fogliano V, Morales FJ. 2009. Acrylamide for-
mation in a cookie system as influenced by the oil phenol pro-
file and degree of oxidation. European Food Research and
Technology, 228, 311–​319.
Beauchamp GK, Keast RS, Morel D. 2005. Ibuprofen-​like activity in
extra-​virgin olive oil. Nature, 437, 45–​46.
Beltran G, Paz Aguilera M, Gordon MH.2005. Solid phase
microextraction of volatile oxidation compounds in oil-​ in-​
water emulsions. Food Chemistry, 92, 401–​406.
Brenes M, Garcia A, Dobarganes MC. 2002. Influence of thermal
treatments simulating cooking processes on the polyphenol
content in virgin olive oil. Journal of Agricultural and Food
Chemistry, 50, 5962–​5967.
Cavallo C, Cicia G, Del Giudice T, Sacchi R, Vecchio R. 2019.
Consumers’ preferences for bitterness in vegetable foods: the
case of Extra-​Virgin Olive Oil and Brassicaceae –​a narrative
review. Nutrients, 11 (5), 1164.
318
Chiou A, Kalogeropoulos N, Boskou D. 2012. Migration of health
promoting microconstituents from frying vegetable oils to
French fries. Food Chemistry, 133, 1255–​1263.
Cicerale S, Lucas LJ, Keast RS. 2012. Antimicrobial, antioxidant and
anti-​inflammatory phenolic activities in extra virgin olive oil.
Current Opinion in Biotechnology, 23, 129–​135.
Della Medaglia D, Ambrosino ML, Spagna Musso S. 1996.
Modification of phenols during the storage and heating of
extra-​virgin olive oil. Oil Process and Biochemistry of Lipids,
1st European Meeting of the American Oil Chemists’ Society,
University of Burgundy, Dijon (France), 19–​ 20 September,
B24.
EFSA. 2011. Scientific Opinion on the substantiation of health
claims related to polyphenols in olive and protection of
LDL particles from oxidative damage (ID 1333, 1638, 1639,
1696, 2865), ‘maintenance of normal blood HDL-​cholesterol
concentrations’ (ID 1639), ‘maintenance of normal blood
pressure’ (ID 3781), ‘anti-​inflammatory properties’ (ID 1882),
‘contributes to the upper respiratory tract health’ (ID 3468),
‘can help to maintain a normal function of gastrointestinal
tract’ (3779), and ‘contributes to body defences against external
agents’ (ID 3467) pursuant to Article 13(1) of Regulation (EC)
No 1924/​2006. EFSA Journal, 9 (4), 2033.
Fini L, Hotchkiss E, Fogliano V. 2008. Chemopreventive properties
of pinoresinol-​rich olive oil involve a selective activation of the
ATM-​p53 cascade in colon cancer cell lines. Carcinogenesis,
29, 139–​146.
Fogliano V, Ritieni A, Monti S. 1999. Antioxidant activity of virgin
olive oil phenolic compounds in a micellar system. Journal of
the Science of Food and Agriculture, 79, 1803–​1808.
Fogliano V, Sacchi R. 2006. Oleocanthal in olive oil: between myth
and reality. Molecular Nutrition and Food Research, 50,
5–​6.
Frankel EN. 1998. Lipid Oxidation. The Oily Press.
Frankel EN. 2011. Nutritional and biological properties of extra
virgin olive oil. Journal of Agricultural and Food Chemistry,
59 (3), 785–​792.
Lee A, Thurnham I, Chopra M. 2000. Comsumption of tomato
products with olive oil but not sunflower oil increases the anti-
oxidant activity of plasma. Free Radical Biology and Medicine,
29, 1051–​1055.
Lozano-​Castellón J , Vallverdú-​Queralt A, Rinaldi de Alvarenga JF.
2020. Domestic sautéing with EVOO: change in the phenolic
profile. Antioxidants, 9, 77.
Lucas L, Russell A, Keast R. 2011. Molecular mechanisms of inflam-
mation. Anti-​inflammatory benefits of virgin olive oil and the
phenolic compound oleocanthal. Journal of. Agricultural and
Food Chemistry, 59, 785–​792.
Kalogeropoulos N, Chiou A, Mylona A. 2007a. Recovery and dis-
tribution of natural antioxidants (a-​tocopherol, polyphenols
and terpenic acids) after pan-​frying of Mediterranean finfish in
virgin olive oil. Food Chemistry, 100, 509–​517.
Kalogeropoulos N, Mylona A, Chiou A. 2007b. Retention and dis-
tribution of natural antioxidants (a-​tocopherol, polyphenols
and terpenic acids) after shallow frying of vegetables in virgin
olive oil. Lebensmittel Wissenschaft und Technologie, 40,
1008–​1017.
Medina I, Sacchi R, Aubourg S. 1995. A 13C-​NMR study of lipid
alteration during fish canning: effect of filling medium. Journal
of the Science of Food and Agriculture, 69, 445–​450.
Medina I, Sacchi R, Giudicianni I. 1998a. Oxidation of fish lipids
during thermal stress as studied by 13C nuclear magnetic res-
onance spectroscopy. Journal of the American Oil Chemists’
Society, 75, 147–​154.
Raffaele Sacchi
Medina I, Sacchi R, Biondi L.1998b. Effect of packing media on
the oxidation of canned tuna lipids. Antioxidant effective-
ness of extra virgin olive oil. Journal of Agricutural and Food
Chemistry, 46, 1150–​1157.
Meynier A, Rampon V, Delgadarrondo M, Genot C. 2004. Hexanal
and t-​2-​hexenal form covalent bonds with whey proteins and
sodium caseinate in acqueous solution. International Dairy
Journal, 14, 681–​690.
Monti S, Ritieni A, Sacchi R. 2001. Characterisation of phenolic
compounds in virgin olive oil and their effect on the formation of
carcinogenic/​mutagenic heterocyclic amines in a model system.
Journal of Agricutural and Food Chemistry, 49, 3969–​3975.
Napolitano A, Morales F, Sacchi R. 2008. Relationship between
virgin olive oil phenolic compounds and acrylamide formation
in fried crisps. Journal of Agricutural and Food Chemistry, 56,
2034–​2040.
Omoni AO, Aluko RE.2005. The anti-​ carcinogenic and anti-​
atherogenic effects of lycopene: a review. Trends in Food
Science and Technology, 16, 344–​350.
Pannellini T, Iezzi M, Liberatore M. 2010. A dietary tomato
supplement prevents prostate cancer in TRAMP mice. Cancer
Prevention Research, 3, 1284–​1291.
Pernice R, Vitaglione P, Sacchi R. 2007. Phytochemicals in
Mediterranean Diet: the interaction between tomato and olive
oil bioactive compounds. In Hui YH (ed.) Handbook of Food
Products Manufacturing (2 Volume Set), Wiley-​Interscience,
­chapter 48, 53–​63.
Persson E, Graziani G, Ferracane R. 2003. Influence of antioxidants
in virgin olive oil on the formation of heterocyclic amines
in fried beef burgers. Food Chemistry and Toxicology, 41,
1587–​1597.
Preedy VR, Watson RR. 2010. Olives and Olive Oil in Health and
Disease Prevention. Academic Press, Elsevier Inc.
Ramírez-​ Anaya JdP, Castañeda-​ Saucedo MC, Olalla-​ Herrera M.
2019. Changes in the antioxidant properties of extra virgin
olive oil after cooking typical mediterranean vegetables.
Antioxidants, 8, 246.
Sacchi R, Paduano A, Fiore F. 2002. Partition behaviour of virgin
olive oil phenolic compounds in in oil-​brine mixtures during
thermal processing for fish canning. Journal of Agricultural
and Food Chemistry, 50, 2830–​2835.
Sacchi R, Falcigno L, Paduano A. 2006. Quantitative evaluation
of the aldehydes formed in heated vegetable oils using high
resolution proton-​NMR spectroscopy. Rivista Italiana delle
Sostanze Grasse, 82, 257–​263.
Sacchi R. 2008. Extravirgin olive oil in kitchen: the secrets of a pro-
tagonist of Mediterranean gastronomy (in Italian), Regione
Campania and Università di Napoli Federico II, SBR Ed.
(Portici, Napoli, Italy), 1–​64.
Sacchi R, Paduano A, Savarese M, Vitaglione P, Fogliano V. 2014.
Extra virgin olive oil: from composition to ‘molecular gas-
tronomy’. In Zappia V (ed.) Advances in Nutrition and Cancer,
Cancer Treatment and Research 159, Springer-​Verlag, Berlin,
Heidelberg, 325–​338.
Sacchi R, Caporaso N, Squadrilli G. 2019. Sensory properties, anti-
oxidant activity and flavours of an artisanal ice-​cream (‘gelato’)
functionalised with extra virgin olive oil. International Journal
of Gastronomy and Food Science, 18, 100173.
Savarese M, Parisini C, De Marco E. 2006. Detection of biophenols
from virgin olive oil in fried French-​fries potatoes by high-​
performance liquid chromatography tandem electrospray ion-
ization mass spectrometry (HPLC-​ESI/​MS). In 4th EuroFed
Lipid Congress (Oils, Fats and Lipids for a Healthier Future),
University Complutense of Madrid, Spain.
Fats and Oils: Extra Virgin Olive Oil
Vallverdú-​Queralt A, Rinaldi de Alvarenga JF, Estruch R, Lamuela-​
Raventos RM. 2013. Bioactive compounds present in the
Mediterranean sofrito. Food Chemistry, 141, 3365–​3372.
Vallverdú-​Queralt A, Regueiro J, Rinaldi de Alvarenga JF. 2014.
Home cooking and phenolics: effect of thermal treatment and
addition of extra virgin olive oil on the phenolic profile of
tomato sauces. Journal of Agricultural and Food Chemistry,
62, 3314−3320.
319
Vitaglione P, Fogliano V. 2004. Use of antioxidants to minimize the
human health risk associated to mutagenic/​carcinogenic het-
erocyclic amines in food. Journal of Chromatography B, 802,
189–​199.
Vitaglione P, Savarese M, Paduano A. 2015. Healthy virgin olive oil:
a matter of bitterness. Critical Reviews in Food Science and
Nutrition, 55 (13), 1808–​1818.
Fermentation: Kimchi
Weon-​Sun Shin
Laboratory of Food Chemistry & Molecular Gastronomy, Department of Food & Nutrition, Hanyang University,
Republic of Korea
For thousands of years, a dish based on fermented Chinese
cabbages and/​or radish with garlic, pepper and salt has been
produced in Korea and given the name kimchi. Various local-
ities and households use different side ingredients, such as green
onion, ginger, sesame, water parsley, fish sauce, fish, meat and
onion, giving specific flavours. These dishes will be explored in
this chapter.
Kimchi can be categorized by the main vegetables in the
recipe, including the Chinese cabbage (Brassica campestris
L. ssp. Pekinensis), radish (Raphanus sativus var. hortensis for.
acanthiformis Makino), cucumber (Cucumis sativus L.), cabbage
(Brassica oleracea L. var. capitata f. alba), green leafy shallot
(Allium ascalonicum L.) and garlic chives (Allium tuberosum).
Since the Korean peninsula has four seasons, kimchi has a season-
ality based upon the raw materials that can be produced in a local
place (Jang et al., 2015). In summer, watery kimchi is generally
made with small loose-​leaf type cabbage (Brassica campestris
L. ssp. pekinensis), coarsely ground fresh pepper (red and green),
garlic, fish sauce, salt, crushed ginger and rice flour slurry, for
which sweet rice flour with one cup of water in a small saucepan
is simmered over a low heat until it thickens to a thin paste, and
then set aside to cool. As kimchi is well fermented and chilled,
it gives a unique flavour and taste. In winter, dongchimi is a typ-
ical type of kimchi, made using whole radish, pear, whole green
onion and salted water. All the materials are put into a big cer-
amic jar and fermented for a while (33–​35 days) at a refrigerated
temperature (about 4–​ 6 °C). Well-​ fermented dongchimi is a
strong, carbonated, clear and chilled soup, which can be served
in one bowl as a side dish or a noodle soup as a winter speciality
(Hongu et al., 2017).
Lee et al. (1991) reported that kimchi’s shelf life at 4 and
28 °C was 33 and 3 days, respectively, based on kinetic modelling
of total acidity for quality prediction. Hong et al. (2016) reported
that the optimum ripening time for kimchi that was fermented at
4 and 20 °C was 35 and 2 days, respectively, after analysing vola-
tile compositions of kimchi.
Basically, the procedure of making kimchi starts with salting
the materials (cabbage, radish and cucumber). There are two cat-
egories: in one, the raw materials (e.g., a quarter-​piece or small-​
piece cut cabbage) should be salted first, because the water inside
the cabbage oozes out due to osmotic pressure, which makes the
texture of kimchi more crispy. In the other, no salt is added to
the materials (cabbage, radish and cucumber) prior to making
kimchi, which allows more water to be retained, resulting in a
less spicy and salty taste in the kimchi.
In the first case, the salted materials should be rinsed to remove
the excessive salt from the surface of the vegetables and then cut
into a customized size. Subsequently, the ingredients, such as
garlic, ginger, green onion, red pepper powder, salt, fish sauce
(optional) and sugar, should be prepared. Both garlic and ginger
are supposed to be crushed, and green onion should be cut into
5 cm lengths. Then, all the ingredients are put into a large-​sized
bowl and mixed thoroughly (Figures 45.1–​45.5).
During kimchi fermentation, sugar has a significant role in
promoting the growth of the micro-​organisms (lactic acid bac-
teria). During storage of kimchi at the correct temperature (about
4–​6 °C), various micro-​ organisms are involved in fermenta-
tion. During kimchi fermentation, aerobic and anaerobic micro-​
organisms grow and produce the organic acids (mainly lactic acid
and acetic acid). These organic acids decrease the pH of kimchi
to 3.5–​4.0.
A balance of aerobic and anaerobic micro-​ organisms has
been recognized to be very important during kimchi fermenta-
tion, because proper fermentation gives a well-​ripened taste and
flavour. Currently, more than 20 varieties of lactic acid bacteria
are known to be involved in kimchi fermentation; Leuconostoc
mesenteroides is a representative bacterium at the initial step
of fermentation, while Lactobacillus plantarum is dominant
under acidic circumstances as fermentation reaches the end. The
most savoury kimchi has a pH between 3.5–​4.5 and is in a well-​
fermented state (Park et al., 2019).
The flavour and taste of kimchi are mainly related to the
content of kimchi metabolites such as carbohydrates, amino
acids and organic acids. Their changes can be influenced by
the microbial community during kimchi fermentation (Jung
et al., 2014).
Lee et al. (2018), using non-​metric multidimensional scaling
plots and by analysis of similarity, found that microbial-​
community differences were strongly reflected in the season-
ality of kimchi samples. Additionally, the distribution patterns
of Leuconostoc, Lactobacillus and Weissella spp. were well
predicted by seasonality, demonstrating the importance of
321
322 Weon-Sun Shin

FIGURE 45.1 Standard cabbage kimchi-​making.

(Courtesy of WIKIM, 2018)

FIGURE 45.2 A seasonal kimchi, dongchimi.

(Courtesy of WIKIM, 2018)
Fermentation: Kimchi
FIGURE 45.3 A cucumber kimchi filled with green leafy shallot.
(Courtesy of WIKIM, 2018)
FIGURE 45.4 Bossam kimchi, cabbage leaf-​folded kimchi.
(Courtesy of WIKIM, 2018)
comprehensive correlations of the bacterial community with vari-
able environmental factors.
Various kimchi containers may be used, such as dok (cer-
amic jar); jungduri (small earthenware jar with full middle
part); bataenggi (a small jar with a fuller girth); and hangari
(ceramic pot), all depending on the type of kimchi and when
it is to be eaten. It is believed that a kimchi jar that was whole-​
heartedly made by the artisan helps, in a spiritual way, to make
tasty kimchi (World Institute of Kimchi, 2014). Various regions
and households have their own traditional know-​how on how
323
to keep kimchi from freezing and turning unfavourably sour
(Figure 45.6).
Kimjang (Intangible Cultural Heritage, UNESCO), winter
kimchi-​making, constitutes a significant part of Korean identity.
Despite urbanization, Westernization and commercialization, the
majority of Koreans still eat kimchi they make at home or kimchi
made by relatives and sent to them regularly, thereby showing
that kimjang is an important binding force of the family com-
munity in Korea. To make kimchi together and share it, espe-
cially with communities, strengthens social ties among Koreans
324
FIGURE 45.5 Yeolmu and leafy cabbage kimchi.
(Courtesy of WIKIM, 2018)
FIGURE 45.6 Kimjang kimchi, or kimchi stored underground after being
made during kimjang, offers a tasty chewy consistency throughout the winter.
(Courtesy of the Cultural Heritage Administration, 2013)
(Figure 45.7). Every year during the kimchi-​ making season,
regional communities and volunteer groups hold large-​ scale
kimchi-​making events for the less privileged, a good example of
kimchi-​making strengthening ties among Koreans. These efforts
Weon-Sun Shin
had a great impact on building up “kimchiology”, a systematic
way to integrate the scientific knowledge of kimchi in multiple
ways, from the humanities to science.
REFERENCES
Cultural Heritage Administration. 2013. Ministry of Culture,
Sports and Tourism: Korean Culture and Information
Service. Available at: www.kocis.go.kr/​koreanet/​view.
do?seq=2117(2013. 10.18).
Hong SP, Lee EJ, Kim YH, Ahn DU. 2016. Effect of fermentation
temperature on the volatile composition of Kimchi. Journal of
Food Science, 81, C2623–​C2629.
Hongu N, Kim AS, Suzuki A, Wilson H, Tsui KC, Park S. 2017.
Korean kimchi: promoting healthy meals through cultural trad-
ition. Journal of Ethnic Foods, 4, 171–​180.
Jang DJ, Chung KR, Yang HJ. 2015. Discussion on the origin of
kimchi, representative of Korean unique fermented vegetables.
Journal of Ethnic Foods, 2, 126–​136.
Jung JY, Lee SH, Jeon CO. 2014. Kimchi microflora: history, current
status, and perspectives for industrial kimchi production.
Applied Microbiology and Biotechnology, 98, 2385–​2393.
Lee KH, Cho HY, Pyun YR. 1991. Kinetic modelling for the predic-
tion of shelf-​life of kimchi based on total acidity as a quality
index. Korean Journal of Food Science and Technology, 23,
306–​310.
Fermentation: Kimchi 325

FIGURE 45.7 Kimjang, making and sharing kimchi.

(Courtesy of Cultural Heritage Administration, 2013)

Lee M, Song JH, Lee H. 2018. Effects of seasonal production on bac- World Institute of Kimchi. 2014. Humanistic understanding of
terial communities in Korean industrial kimchi fermentation. Kimchi and Kimjang culture. Kimchiology series No. 1 (2014–​
Food Control, 91, 381–​389. 05), https://​unesdoc.unesco.org/​ark:/​48223/​pf0000231307.
Park SE, Seo SH, Byun S, Na CS, Son HS. 2019. Changes of micro-
bial community and metabolite in kimchi inoculated with
different microbial community starters. Food Chemistry, 274
(15), 558–​565.
Fermentation: Fermenting Flavours with Yeast

Angela M. Coral Medina and John P. Morrissey

School of Microbiology, University College Cork, T12 YN60, Ireland

Think about the aroma of coffee in the morning, of freshly baked accumulation of these positive characteristics can be explained
bread when you walk outside the boulangerie, or of red wine or by the fact that humans have domesticated yeast from wild
a foamy beer on a Saturday night. Those are all aromas that we relatives. Some studies have revealed how industrial traits have
can sense by mentioning them; most of us feel related to them. been selected over time for the best quality of the final food
The interesting part is where these food and beverage flavours product. Therefore, yeasts with desirable traits have been selected
and aromas come from. Most people probably do not realise that as starters for other fermentations and slowly adapted to indus-
yeast is responsible for producing most of these during the pro- trial conditions, giving rise to improved yeast strains.
cess of fermentation. These characteristic aromas and flavours However, S. cerevisiae is not the only yeast species present
come from the action of microbes that, in fact, have the most in food fermentations; there is a diverse microbial community
important role during the fermentation process. In 1986, the well-​ found in the raw material of fermented beverages, for example, in
known author Lalli Nykänen stated that “the formation of the the grapes used for winemaking and in the winery environment.
most dominant compounds occurring in beverages depends more This microbial community includes other yeast species, moulds
on the yeast selected than the raw materials used in fermenta- and bacteria. These other yeast species, called non-​traditional
tion” and that “the body of flavour is formed during fermentation yeasts, are being explored by the brewing, winemaking and
by yeast”. Over the years, many scientific studies have focused baking industries, as some may have desirable properties, such
on aroma production by yeast, and these observations have been as resistance to different environmental constraints, production
supported. of compounds of interest such as higher alcohols, and the use of
different biological mechanisms to survive. Some of the commer-
cial yeasts have been developed purely on the basis of fast growth
and effective performance. In contrast, some other yeast strains
Yeast
have been developed to improve traits and increase production
The essential ingredient for the production of alcoholic beverages levels of desired compounds such as aromas.
and other fermented foods is yeast, but what is yeast? From the
food and beverage field, yeast is seen as a food enhancer. In gen-
eral, most people have a positive impression of yeast, as it converts
cheap materials to expensive goods. From a microbiologist’s point
of view, yeasts are unicellular microorganisms that belong to the
fungal kingdom. In other words, yeasts are related to organisms
such as edible mushrooms and moulds present in some types of
cheese. The yeast cell has an egg-​like shape and can only be seen
under a microscope (Figure 46.1). These microorganisms are nor-
mally found in different environments such as soil, plants, skin,
leaves, fruits, water and the human intestinal tract.
About 1500 species of yeast have been identified, but fewer
than 20 are typically associated with food fermentation. Of
these, Saccharomyces cerevisiae, which got its name from Greek
words meaning “sugar fungus”, is the most important and the
most widely studied yeast species. S. cerevisiae has desirable
characteristics for producing fermented foods and beverages, FIGURE 46.1 Yeast cells observed under the scanning electron micro-
such as high ethanol production and tolerance, the ability to com- scope. The image shows the food yeast Kluyveromyces marxianus.
plete fermentation efficiently, a lack of toxicity for the consumer,
(Courtesy of Arun Rajkumar and Suzanne Crotty, University College
and the production of desirable flavour and aroma molecules. The Cork)

327
328
Fermentation
Yeast is a microorganism that can convert raw materials such as
grape juice and malt into valuable products through fermenta-
tion. In principle, during fermentation, the yeast converts sugars
to alcohol and carbon dioxide (CO2). In the case of bread produc-
tion, the yeast consumes the starch in the flour and releases gas
that causes the bread dough to rise. Likewise, in an alcoholic bev-
erage process, the yeast converts the sugars present in fruit juice
or malt into alcohol, resulting in products with high amounts of
ethanol and sometimes bubbles of CO2 and foam. Other important
products of fermentation are volatile molecules that give different
flavours and aromas to the beverage.
Technically, the process of fermentation by yeast is carried
out anaerobically to enable the cell to maximise its fermenta-
tive abilities. The main biochemical reaction is the conversion
of glucose into pyruvate, followed by the cleavage of pyruvate
to produce CO2 and acetaldehyde, which later is converted to
ethanol by an alcohol dehydrogenase (Querol and Fleet, 2006).
The main reactions taking place during fermentation are shown
in Figure 46.2. Furthermore, the general fermentation yield of
S. cerevisiae in alcoholic beverages is such that one cell is able
to ferment its own weight of glucose per hour, producing up to
18% volume of ethanol. This is a valuable fact in the fermen-
tation industry, as cheap substrates are converted into expen-
sive products through a biological process carried out by yeast.
There are important factors that influence the fermentation and
FIGURE 46.2 Overview of metabolic routes during alcoholic fermentation inside the yeast cell. On the left, glucose conversion into ethanol and CO2. On the
right, the Ehrlich pathway and main reactions contributing to production of aroma volatiles.
Angela M. Coral Medina, John P. Morrissey
therefore, the quality of the final product; the main factors are the
yeast strain, temperature, sugars, nitrogen source, oxygen avail-
ability and pH.
For millennia, yeast has been used to produce alcoholic
beverages such as wine, beer, cider and other spirits. Originally,
the fermentation of food and beverages was a spontaneous pro-
cess that, over the years and the development of civilisation, was
replaced by a controlled process, in which known cultures of
yeasts were used to start the fermentation. Therefore, the fermen-
tation process has been improved in many ways, increasing the
quality of the products and the process itself.
Flavour and Aroma Properties of Yeast
When consuming fermented products, we find special flavours
and aromas that make the food or drink unique. The presence
of those flavours is an important matter, as the organoleptic fin-
gerprint of the product changes depending on the presence or
absence of the aroma compounds (Figure 46.3). Some of these
come from the ingredients used for fermentation, for instance,
the use of sweet or acid grapes in winemaking, or toasted or pale
malt in the case of beer; the choice of using one ingredient or
the other changes the final aroma profile of the product. Some
other aromas come from the ageing process, especially in cases
when the beverage is stored in wood barrels, a phase when chem-
ical exchange between the wood and the beverage takes place.
Fermentation: Flavours with Yeast
Almond Apple Roses Banana
O high amounts of these compounds can have the opposite effect on
OH Isoamyl acetate
O Ethyl hexanoate
H O
O
Benzaldehyde 2-Phenylethanol O compound produced by winery yeasts, which in concentrations
Yeast
FIGURE 46.3 Diagram of aromas produced by yeast in wine. Link between
the volatile metabolites and their respective aromas.
Finally, the part that most affects the flavour and aroma profile is
the type of yeast used and the fermentation conditions.
Each yeast has different ways to assimilate substrates and
yields different products depending on its genetic make-​up or
genotype. Thus, depending on its genotype, the yeast reacts dif-
ferently to the conditions that it finds in its environment. This
will cause different enzymes to be produced inside the cell and
ultimately, will lead to a distinct flavour and aroma profile. This is
the reason why the choice of yeast species to produce fermented
beverages is crucial. The fermentation factors with the highest
impact on flavour and aroma production are the yeast strain, the
temperature and the nitrogen source. Fermentations at low tem-
perature, for instance, result in aroma retention and increased
production of aroma compounds. Moreover, although the use
of a specific type of nitrogen source increases or decreases
the formation of specific aroma compounds, the difference in
the aroma production is attributed to the metabolism of each
yeast species.
The yeast takes amino acids and sugars to convert them into
aroma molecules that give different organoleptic properties to the
product: “fruity”, “floral”, “cheesy” or “rancid”, for instance. So,
when the label description of some fermented beverages has fruity
and flowery tones in the aroma profile, it is not referring to the
fact that roses, banana or pineapple were added to the beverage
329
but to the use of particular yeast species that produce those aroma
characters. While we depend on yeast to produce most of the great
aromas that we associate with our favourite fermented beverages,
excessive production of aroma volatiles is not always beneficial;
the aroma profile. A specific example is ethyl acetate, an aroma
of 25–​30 mg/​L gives a pleasant fruity aroma to the wine, but in
concentrations higher than that is perceived as giving a solvent-​
or nail-​polish-​like odour. A case of too much of a good thing!
The flavour and aroma compounds present in fermented food
products are volatile molecules with low molecular weight, small
enough to reach our sensory receptors and be recognised by the
olfactory and gustatory system. Most of these flavour molecules
are alcohols, aldehydes, esters and organic acids that are products
of fermentation (Dzialo et al., 2017).
How are the aroma molecules produced in yeast? What happens
inside the yeast cell is that some components of the raw fermenta-
tion material are slowly converted into higher alcohols and esters
through a mechanism called the Ehrlich pathway, described by
the German biochemist Felix Ehrlich over 100 years ago. The
Ehrlich pathway (Hazelwood et al., 2008) is a catabolic route
whereby the amino acids available in the media are converted
into higher alcohols through three main enzymatic steps, which
are illustrated in Figure 46.2. The Ehrlich pathway steps are
explained as follows:
• Transamination: The reaction consists of the transfer of
amines from the amino acid to another molecule to form
the respective α-​keto acid; there are four enzymes cata-
lysing the reaction, encoded by the BAT1, BAT2, ARO8
and ARO9 genes;
• Decarboxylation: This an irreversible step where the
a-​keto acids remaining from the transamination step
lose a carbonyl group to form the respective aldehyde
and release CO2. This reaction can be catalysed by five
decarboxylases, encoded by the PDC1, PDC5, PDC6,
ARO10 and THI3 genes;
• Reduction to higher alcohols: This is the last step of the
Ehrlich pathway. The alcohol dehydrogenase (ADH)
enzymes convert the aldehydes into their respective
higher alcohols, also known as fusel alcohols.
Besides the reduction step, there is a parallel oxidation of the
aldehyde to form fusel acids, which play a minor role in yeast.
The balance of these two reactions happening in the yeast cell
depends on the fermentation conditions; normally, in S. cerevisiae,
the production of fusel alcohols is 90% higher than that of fusel
acids, although it mainly depends on the redox status of the cell.
At the end of the formation of the fusel alcohols and fusel acids,
the cells release these compounds to the culture medium through
simple passive diffusion for the higher alcohols or with the action
of an exporter protein for the fusel acids.
Esters are another group of volatile compounds that sig-
nificantly contribute to the aroma in alcoholic beverages. In
wine and beer, esters define the final aroma depending on the
concentrations produced by the yeast strain (Holt et al., 2019).
330
A low level of these compounds is enough to generate pleasant
odours; on the contrary, when the compounds are overproduced,
there is a negative effect on the final aroma of beer and wine
(Pires et al., 2014; Querol and Fleet, 2006). Acetate esters are one
of the major groups of esters produced by yeast. Their formation
results from the transfer of an acetate group from acetyl-​CoA to
ethanol or a higher alcohol. The enzymes involved in this reaction
are alcohol acetyltransferases (AATases), which are encoded by
the genes ATF1 and ATF2. The availability of acetyl-​CoA in the
cell is important for the production of acetate esters in anaerobic
conditions.
Yeast in the Food and Beverage Industry
Currently, yeast is used for several purposes in the food and bev-
erage industry. The most common uses of yeast are alcoholic bev-
erage fermentation and baking. However, non-​alcoholic beverages
are also produced by yeast; in this case, the fermentation is
stopped before the drink becomes alcoholic, but the modification
of the beverage features is achieved. Thus, even non-​alcoholic
beer requires a yeast fermentation! Kefir and kombucha (see
chapter by Lavelle and Boulé in this book) are other examples
of non-​alcoholic yeast products, and in these cases, bacteria also
play a fermentation role. Moreover, yeast is used as a food addi-
tive; it is added to a large range of food products with different
purposes such as food flavouring, replacing ingredients for vegan
alternatives, and adding texture to the food. In other cases, yeast
is used as a nutrient supplement for dietary purposes, thanks to
Angela M. Coral Medina, John P. Morrissey
yeast being a source of vitamins, minerals and high-​quality pro-
tein. On top of that, some yeast species are commercialized as
probiotics, especially Saccharomyces boulardii, which is reported
to confer health benefits.
In conclusion, yeast plays a significant role in the food and
beverage industry, especially in the fermentation sector. Over the
years, increasing research and knowledge generation on different
strains of yeast has improved the quality of food products. It is
fascinating to understand how the desired flavour and aroma pro-
file can be accomplished by employing the right yeast strain.
REFERENCES
Dzialo, M.C., Park, R., Steensels, J., Lievens, B., Verstrepen, K.J.,
2017. Physiology, ecology and industrial applications of aroma
formation in yeast. FEMS Microbiol. Rev. 41, S95–​S128.
Hazelwood, L.H., Daran, J.-​M.G., van Maris, A.J.A., Pronk, J.T.,
Dickinson, J.R., 2008. The Ehrlich pathway for fusel alcohol
production: a century of research on Saccharomyces cerevisiae
metabolism. Appl. Environ. Microbiol. 74, 2259–​2266.
Holt S, Miks MH, de Carvalho BT, Foulquié-​Moreno MR, Thevelein
JM. 2019. The molecular biology of fruity and floral aromas
in beer and other alcoholic beverages. FEMS Microbiol Rev.
43, 193–​222.
Pires, E.J., Teixeira, J.A., Brányik, T., Vicente, A.A., 2014. Yeast:
The soul of beer’s aroma –​A review of flavour-​active esters
and higher alcohols produced by the brewing yeast. Appl.
Microbiol. Biotechnol. 98, 1937–​1949.
Querol, A., Fleet, G.H., 2006. Yeast in Food and Beverage. Springer-​
Verlag, Berlin Heidelberg.
Fermentation: A Short Scientific and Culinary Overview of Kefir

Christophe Lavelle and Jean-​Baptiste Boulé

Genome Structure and Instability, CNRS UMR7196 /​INSERM U1154, National Museum of Natural History,
Sorbonne University, Paris 75005, France

Introduction milk kefir grains cultivated in watery conditions or, as sometimes

claimed, it might have appeared spontaneously in the sugar-​
Fermented foods usually have long storage lives and interesting
rich juice of Opuntia cactus. It is made by inoculating a solu-
nutritional values, which makes fermentation a very popular
tion of water, sugar and added dry and fresh fruit (usually fig,
transformation technique all over the world, including for drink
lemon, etc.) with kefir grains and keeping it for a day or more at
production. Among these, kefir is a fermented drink produced
room temperature (Figure 47.1). The water kefir grains contain
by the action of a Symbiotic COmmunity of Bacteria and Yeasts
yeasts (including Saccharomyces and Candida species) and bac-
(SCOBY) contained in grains, made from a protein and poly-
teria (including Lactobacillus, Streptococcus and Leuconostoc
saccharide matrix. It is supposed not only to be tasty but also to
species), which also produce a sour, carbonated and slightly alco-
have many health benefits, including anticarcinogenic, antiviral
holic beverage.
and antifungal properties, although most of these claims still need
Once strained (to separate the grains from the liquid), kefir
further support from medical studies. However, even without
drinks can be stored for several days, undergoing a second
sound scientific proof, consumers are generally encouraged by
(slower) fermentation, which will enhance the sourness and
popular media to improve their health through fermented diets
amount of gas. Water kefir is an increasingly popular drink
that bring a good amount of useful probiotics, making kefir (like
throughout the world, mainly due to its ease of production, its
other fermented foods and beverages) increasingly fashionable
alleged health benefits and the fact that it is suitable for vegans.
(Debailly et al., 2018).

Origin, Production and Consumption Scientific Investigations

The microbial species diversity of water kefir consists of a con-
Two types of kefir exist today: “milk” and “water” (also called
sortium of mainly lactic acid and acetic acid bacteria and yeasts
“sugar”) kefir.
(Fiorda et al., 2017). Microbial analysis of kefir samples taken
Milk kefir, thought to have its origins in the north Caucasus
from different locations worldwide shows differences in micro-
mountains, is still consumed by a lot of Russians. It is tradition-
biological populations, which reflect the colonization of cultures
ally made by inoculating milk (cow, goat or sheep) with kefir
grains and keeping it in skin bags at room temperature for a
day or more, depending on the conditions and fermentation
level desired. The milk kefir grains contain yeasts (including
Saccharomyces and Kluyveromyces species) and lactic acid bac-
teria (including Lactobacillus, Lactococcus, Streptococcus and
Leuconostoc species), which produce a viscous, sour, carbonated
and slightly alcoholic beverage (Farnworth, 2005; Prado et al.,
2015). According to the Codex Alimentarius Standard for
Fermented Milks (CODEX STAN 243–​2003), milk kefir contains
milk protein (>2.8% w/​ w), milk fat (<10% w/​ w), titratable
acidity expressed as percentage of lactic acid (minimum 0.6% w/​
w), ethanol (not stated), and specific microorganisms constituting
the starter culture (minimum 107 colony forming units (cfu)/​g, in
total including as a minimum 104 cfu/​g of yeasts). FIGURE 47.1 (a) Home-​made orange/​fig water kefir; (b) Magnified image
Water kefir has a more mysterious origin and is not yet defined of water kefir grains (each is a few mm in size).
by the Codex Alimentarius: it might come from the evolution of (Photo in (a) courtesy of Guillaume Stutin)

331
332 Christophe Lavelle, Jean-Baptiste Boulé

by opportunistic yeasts and bacteria found in the environment

(Gulitz et al., 2011; Laureys and De Vuyst, 2014).
We collected about 100 different kefir grains in our lab and
started a thorough biochemical (using controlled fermen-
tation; Figure 47.2) and microbiological analysis (through
metagenomics; Figure 47.3) of each of them. The goal of this
ongoing study is to show (1) the link between the microbio-
logical content (bacteria, yeasts) and the physiological/​sensory
parameters (metabolites, taste) of the drink and (2) the high diver-
sity of kefir.
Previous studies have already shown that sucrose is mainly
converted into ethanol, lactic acid (which gives the sour taste),
polysaccharides (which create a new matrix for the grains), and,
to a lower extent, glycerol, acetic acid and mannitol (Laureys
and De Vuyst, 2014). More studies are needed to address many
unanswered questions, such as the influence of various parameters
(light, oxygen, ingredients added, liquid/​grains ratio, etc.) on the
final result, as well as to test some “fancy” statements (never use
metal tools, never use white sugar, etc.) often found in popular
papers and on social media (Debailly et al., 2018).
Regarding the microbial composition, some studies have
pointed out the differences between kefir strains (Nalbantoglu
et al., 2014), making it difficult to characterize what a “standard”
kefir could be. Indeed, we found totally different profiles in our
collection, with major species varying from one strain to another
(Figure 47.3). Again, further studies are needed to reach a more
complete “landscape” of what kefir is (or, indeed, are)!
Biochemical, metagenomic and metabolomic studies pro-
vide evidence of the large number of microorganisms in starting
strains and the variety of possible bioactive compounds that
could be formed during fermentation, and have to be followed
by medical/​clinical studies aimed at characterizing the potential
health effects of regular kefir consumption (Farnworth, 2005;
Lopitz-​Otsoa et al., 2006). This is especially true for water kefir,
for which fewer studies exist compared with milk kefir (Fiorda FIGURE 47.2 Water kefir grown in laboratory conditions. Various
et al., 2017). parameters are measured, including optical density, pH, RedOx and pO2 (y-​
axis, normalized units) during the time of fermentation (x-​axis, hours).

FIGURE 47.3 Genomic profile of three different kefirs. Each strain shows a different composition, with more than a dozen identified species, among which
some dominate, such as Bifidobacterium aquikefiri (red) for kefir A, Leuconostoc mesenteroides (orange) and Lactobacillus ghanensis (dark blue) for kefir B
and Oenococcus kitaharae (beige) for kefir C. Yeasts (mainly Saccharomyces species, various greens) are present in the three strains.
(Courtesy of Alexandra Joubert, Genome Structure and Instability unit)
Fermentation: Kefir
FIGURE 47.4 Kefir is increasingly used in restaurants; here it is seen on a
dessert at Flaveur restaurant in Nice, France.
(Photo courtesy of Sylvain Erhardt)
Culinary Uses
There is no need to wait for medical studies to enjoy kefir for its
taste! Milk kefir has been traditionally used for a long time in
various recipes from former Soviet Union countries (mainly in
soups like cold bortsch and okrochka). More recently, water kefir
has also become quite popular among chefs, who have started
using it in various preparations (breads, sauces, sorbets, etc.;
Figure 47.4) or directly serving it as a drink to accompany the
meal as a way to replace wine in food/​drink pairing menus. This
is probably the perfect occasion for collaborative work between
cooks and scientists!
333
Acknowledgements
We thank Pierre Renault (MICALIS-​ INRAE, Jouy-​ en-​
Josas,
France) for collaboration on metagenomics studies. Work at the
Genome Structure and Instability unit is supported by CNRS,
INSERM and MNHN.
REFERENCES
Debailly R, Lavelle C and Schultz E. 2018. Conserver un aliment
vivant. Entretien et circulation d’un ferment: le cas du Kéfir.
Techniques et Culture “Le temps des aliments” 69:180–​183.
Farnworth ER. 2005. Kefir –​a complex probiotic. Food Science and
Technology Bulletin: Functional Foods 2:1–​17.
Fiorda FA, de Melo Pereira GV, Thomaz-​Soccol V, Rakshit SK,
Pagnoncelli MGB, Vandenberghe LPS and Soccol CR. 2017.
Microbiological, biochemical, and functional aspects of sugary
kefir fermentation –​ A review. Food Microbiology 66:86–​95.
Gulitz A, Stadie J, Wenning M, Ehrmann MA and Vogel RF. 2011.
The microbial diversity of water kefir. International Journal of
Food Microbiology 151(3):284–​288.
Laureys D and De Vuyst L. 2014. Microbial species diversity,
community dynamics, and metabolite kinetics of water kefir
fermentation. Applied and Environmental Microbiology
80(8):2564–​2572.
Lopitz-​Otsoa F, Rementeria A, Elguezabal N, Garaizar J. 2006.
Kefir: a symbiotic yeasts-​bacteria community with alleged
healthy capabilities. Revista Iberoamerica de Micología
23(2):67–​74.
Nalbantoglu U, Cakar A, Dogan H, Abaci N, Ustek D, Sayood K and
Can H. (2014). Metagenomic analysis of the microbial com-
munity in kefir grains. Food Microbiology 41:42–​51.
Prado MR, Blandón LM, Vandenberghe LPS, Rodrigues C, Castro
GR, Thomaz-​Soccol V and Soccol CR. 2015. Milk kefir: com-
position, microbial cultures, biological activities, and related
products. Frontiers in Microbiology 6:1177.
Filtration Membranes for Food Processing and
Fractionation
Marie-​Laure Lameloise
UMR SayFood, AgroParisTech, INRAE, Université Paris-​Saclay, Massy 91300, France
Introduction
Membrane processes, mainly membrane filtration, from
microfiltration to reverse osmosis, have nowadays penetrated
almost all the sectors of the food industry, significantly modi-
fying the processing schemes and the products themselves.
Beyond the pioneer dairy industry, where membranes have
led to a real cracking of milk and whey into a multiplicity of
compounds or functional fractions, this chapter intends to review
the main current and potential applications of these technologies.
The Basics of Filtration Membranes
From Frontal to Tangential Filtration
Traditional filtration of liquid suspensions, i.e., dispersions of
solid particles in a liquid, relies on a transverse flow across a
porous medium under a driving force generally transmitted by
a pump. In cake filtration, particles retained at the surface of a
support progressively build a cake, which becomes the true
filtering medium and determines the characteristics of the oper-
ation (retention threshold). Cake filtration, also called separation
because the two phases (solid cake and filtrate) are recovered, is
limited by the size of the particles: below 20 μm, the cake resist-
ance strongly limits the filtration flow. In mass filtration, particles
as small as 0.2 μm are trapped within a thick porous medium,
sand bed, cartridge, cellulose sheet, sintered metal or filter-​aid
precoat. This type of filtration is only suited to low-​concentrated
suspensions with small particles considered as impurities,
namely, clarification: the particles are generally discarded with
the porous medium. Rapid fouling of the porous medium may be
rather costly and generate significant wastes. Traditional frontal
filtration is therefore intrinsically unsuitable for the recovery of
thin particles below a few micrometers in size and, a fortiori,
of macromolecules or molecules. Another operation should be
implemented: tangential filtration.
Tangential filtration associates (i) a membrane, namely a thin
medium, offering calibrated pores and acting as a screen and (ii)
a high-​velocity (several meters per second) feed flow parallel to
the surface of the membrane to limit particle or solute concentra-
tion or deposition at the surface and to maintain high permeate
flux over long times. Thus, in contrast to traditional filtration, tan-
gential filtration splits the feed flow into two liquid outputs, the
retentate and the permeate.
In search of high barrier properties and high mechanical
resistance combined with low hydraulic resistance, the design
of membranes has moved from thick homogeneous membranes
with narrow pores to asymmetrical membranes in which the
pores progressively enlarge throughout the membrane thickness.
Over the years, the functions of the superficial part (selectivity)
and the inner part of the membranes (permeate draining, mech-
anical resistance) have progressively been differentiated and
ensured by different structures and even materials (composite
membranes). Current polymeric reverse osmosis membranes
illustrate the concept of composite membranes, displaying an
(aromatic) polyamide active layer, co-​polymerized on an inter-
mediate polysulfone layer, itself deposited on a non-​woven poly-
ester support. The active layer represents only 0.2 μm over a
total thickness of 100 to 150 μm. Besides polymeric membranes,
mineral membranes, either alumina-​based or stemming from a
nuclear research program (isotopic enrichment of uranium), were
developed in the 1970s. Their superficial layer is constituted of
ultra-​thin metallic oxide particles (ZrO2, TiO2, Al2O3) fixed on
a macroporous metal, carbon or ceramic support. With a longer
shelf-​life than polymeric membranes but far more expensive,
their use is reserved for extreme conditions of pH and tempera-
ture or for liquids with high added value.
Membranes cover different filtration ranges according to their
structure and porosity. The cut-​off is defined as the diameter or
the molecular weight (MWCO) of the smallest solute retained
at 90%. Microfiltration (MF) membranes are microporous, with
pore diameters between 0.1 and 10 μm, able to retain very thin
particles or colloids, large proteins (lipoproteins), bacteria,
enzymes, etc. Removal of turbidity and bacteria are the main
issues. With pores of 10 nm to 1 micrometer, mesoporous ultrafil-
tration membranes (UF) are the favorite membranes for retaining
colloids, bacteria, viruses and macromolecules such as proteins.
Reverse osmosis membranes (RO) are dense membranes
without pores, in which mass transfer is controlled by diffusion
through the voids existing between the polymer chains. They
only let water and very small molecules (formic or acetic acid,
for example) pass through. Designed and commercialized for
335
336
sea-​water desalination, they are currently being considered with
much interest for the recovery of water from other resources,
such as wastewaters. The recent nanofiltration (NF) membranes
hold a special place between UF and RO; the tightest ones are
often considered as dense membranes, whereas the loosest can
be considered as porous.
Besides filtration membranes, which are the focus of this
chapter, there is a variety of other membrane-​based separation
processes. Electrically driven separation processes (electrodi-
alysis) use dense ion-​ exchange membranes and an electrical
field to separate ionized from non-​ ionized species. Osmotic
evaporation (OE) is based on a difference of water activity and
a vapor pressure gradient on either side of a hydrophobic micro-
porous membrane to concentrate heat-​ sensitive solutions. In
pervaporation, mostly used for ethanol dehydration, the driving
force results from the creation of a partial vacuum in the per-
meate side on a dense membrane.
Although size exclusion is the main factor of retention in
filtration membranes, it can be significantly modulated by the
properties of the membrane material. Polymer membranes
used in RO and NF (and UF to a lesser extent) display super-
ficial electrical charges according to the pH of the solution in
contact: generally positive at low pH, zero for pH between 3
and 5 (isoelectric pH) and negative above pH 5. Surface charge
is responsible for electrostatic exclusion or Donnan exclusion.
That is why NF membranes are able to reject divalent ions such
as Ca2+, Mg2+ or SO42− while letting monovalent species go
through. Chemical interactions between solutes and membrane
material (hydrophobic interactions, π–​π interactions, etc.) can
also modulate size exclusion: an NF membrane with a MWCO
of about 150–​300 g/​mol simultaneously retains xylose but lets
vanillin go through, two molecules with quasi-​similar MW
(150 g/​mol) but very different chemical structures and proper-
ties (Nguyen et al., 2015).
Driving Force, Permeate Flux and Retention
The driving force in tangential filtration is the pressure diffe- the bulk. Concentration polarization has several adverse effects;
rence between the two sides of the membrane, namely the trans- it increases osmotic pressure at the membrane surface on the
membrane pressure (TMP). However, as the high-​velocity flow retentate side and consequently decreases TMPeff and the permeate
is likely to cause a significant pressure drop ΔP on the feed/​ flux. The observed (measured) retention may be significantly
retentate side, TMP is conventionally calculated as the mean lower than the real retention at the membrane surface. Finally,
value between inlet and outlet, namely:
TMP =
( PRe + PRs )
− PP (48.1)
2 
where PRe and PRs are the pressure on the retentate side at the
inlet and at the outlet, respectively, and PP is the pressure of the
permeate.
The proportionality factor between pure water flux JW and
TMP at a given temperature represents the permeability of the
membrane to pure water AW. It is a reference value for analyzing
membrane behavior during filtration:
JW = AW .TMP  (48.2)
Marie-Laure Lameloise
with JW in (L/​h/​m2), AW in (L/​h/​m2/​bar) and TMP in bar.
As far as solutes are concerned, if their concentration is sig-
nificant, it may be necessary to consider an effective transmem-
brane pressure TMPeff in place of TMP to account for the osmotic
pressure difference Δπ between the retentate and the permeate
side (ΔR and ΔP, respectively) :
TMPeff = TMP − (Δπ)(48.3)
with Δπ = πR − πP(48.4)
The permeate flux JP generally follows the Darcy law:
JP = A. TMPeff(48.5)
with JP in (L/​h/​m2) and TMPeff in bar.
Besides permeation flux, the retention rate R for a given solute
reflects the performance of the membrane in operation. It is
defined as:
CP
R = 1− (48.6)
CR 
where CP and CR are the concentrations in the permeate and in the
retentate, respectively.
Intrinsic Limits and Challenges: Concentration
Polarization and Fouling
Concentration polarization is inherent in most membrane
processes, whether pressure-​driven (tangential filtration), elec-
trically driven (electrodialysis) or concentration-​driven (dialysis)
processes.
As a result of the convective permeation flux and of the reten-
tion, the retained species accumulate close to the membrane sur-
face. A retro-​diffusive flux then takes place, but concentration
at the vicinity of the membrane surface remains higher than in
it increases the risk of precipitation or gelation at the membrane
surface, resulting in increased fouling. Unlike fouling, concentra-
tion polarization takes place immediately and disappears as soon
as pressure is released. Increased turbulence in the membrane
module can reduce the polarization layer thickness.
Although tangential filtration is theoretically a continuous
operation, in contrast to traditional filtration processes, in practice
a fouling layer often builds up, which requires removal sequen-
tially either by water flushing (in the case of reversible fouling) or
by longer cleaning phases with appropriate chemicals (in the case
of irreversible fouling). Optimization of plant design and strict
control of operating conditions are essential to prevent or limit
this fouling. A cleaning frequency that is too high contributes to
an accelerated ageing of the membrane and a degradation of its
characteristics.
Filtration Membranes
Applications
Overview of the Interest in Filtration Membranes in
Food Processing
Membranes fundamentally split the feed into two fluxes, retentate
and permeate, according to the size of the solutes, with a possible
modulation by electrostatic or physical-​ chemical interactions.
Depending on the targeted output, membranes can ensure purifi-
cation, concentration/​extraction, and separation or fractionation.
Additionally, they can be diverted from these primary functions and
accomplish other functions classically devoted to specific reactors,
such as emulsification, phase contacting, crystallization (proteins)
or support for catalysts in catalytic reactors. As an example, emul-
sification of one phase into another through a porous membrane
can achieve more homogeneous emulsions with a lower range of
particle sizes than conventionally prepared ones made with high
stirring or high-​pressure homogenization (Ilić et al., 2017; Silva
et al., 2017; Hancocks et al., 2016; Charcosset, 2009).
Membranes can replace existing unit operations with benefi-
cial effects: (i) they are continuous operations, (ii) they reduce
or suppress the need for chemicals or process aids such as
regenerants, coagulants or filter-​aids, (iii) they operate at low
temperature (even if increasing temperature has a favorable influ-
ence on permeate fluxes) and do not involve energy-​consuming
phase changes, and (iv) they lead to simplified process schemes
(reduction of the number of operations).
Consequently, membrane techniques are proving competitive
at different levels: product quality, environmental impact, energy
consumption and productivity. Thanks to modularity and flexi-
bility, they can be integrated more or less intimately with other
operations to improve selectivity: as an example, UF membranes
combined with electrodialysis (EDUF) for the separation of bio-
active peptides.
Membranes are also a powerful driver of innovation. Indeed,
in food processing, small changes can induce great effects.
Replacing traditional processes by membranes may lead to
substantial product modification and even novel products, as
illustrated by the emblematic example of the MMV process for
cheese manufacturing: reversing the traditional sequence [coagu-
lation + draining] to [UF draining + coagulation] allowed the
recovery of whey proteins in the product, leading to the devel-
opment of several dozens of novel cheeses, different from trad-
itional products. Micro-​filtered milk is also a novel product, with
organoleptic and nutritional properties that are different from
those of thermally treated milk. In the not too distant future,
thanks to better understanding of the role of membranes and their
interactions with food composition, tailor-​made products could
be designed through a reverse engineering approach.
A rather recent but fast-​growing field of application is the
treatment of wastewater, either for the recycling or reuse of water
in the process or for the recovery of raw materials or valuable
molecules.
Purification
The objective here is to remove fractions or solutes that
affect the quality of a solution, its organoleptic, nutritional or
337
functional properties, or that could lead to its alteration over
time and adversely affect its conservation. It can also concern
the correction of composition to remove species detrimental for
downstream operations (softening) or to comply with regulations.
MF with mineral membranes eliminates particles and macro-​
solutes responsible for the turbidity of apple juice and replaces
a succession of traditional operations, including enzymatic
depectinization, fining, decantation, pre-​ filtration on diatom-
aceous earth, and final filtration on a cellulose sheet, before
bottling and pasteurization. MF definitely simplifies the whole
process and avoids the use of consumables and the production of
wastes. Processing time is reduced, a benefit in terms of hygiene.
Other beverages such as cider are now processed in the same way.
For fruit juices with a higher content of pulp particles, enzymatic
pretreatment can remove high-​molecular-​weight cellulose and
hemicelluloses responsible for viscosity issues (Baker, 2012) and
make the resulting juices easier to treat by membrane technology
(Vaillant et al., 2005). In the processing scheme of glucose
syrups from wheat or corn starch hydrolysis, macromolecules are
eliminated by mineral membranes with a 300 kg/​mol MWCO,
at the limit of UF and MF at 70 °C in multi-​stage plants. A high
volume reduction ratio (25) is achieved; volume reduction ratio,
or VRR, is the ratio between the initial feed volume (or flowrate)
and the final retentate volume (or flowrate). To limit the loss of
sugar in the retentate (as a non-​retained solute, glucose con-
centration distributes equally between retentate and permeate),
a diafiltration step is implemented: this consists of “washing”
off the retentate by adding (dia)volumes of water and extracting
equivalent amounts of permeate.
Stabilization and clarification by MF often leads simultan-
eously to the removal of microorganisms, reducing the process
effect on nutritional and organoleptic product properties (Vaillant
et al., 2005; Hongvaleerat et al., 2008; Polidori et al., 2018) and
representing a good alternative to heat pasteurization and ster-
ilization. Microbial stabilization of beer can be ensured by MF
after fermentation (Daufin et al., 2001). This technology is also
used for wine stabilization after fermentation (Lipnizki, 2017).
Cold (20–​50 °C) stabilization of milk with MF, also called the
“Bactocatch” process, allows fewer than 30 CFU/​ mL to be
reached for mesophilic microorganisms without significant
change in milk composition (Pouliot, 2008; Gésan-​ Guiziou,
2010). It preserves the nutritional factors and the genuine taste
of milk better than the thermal pasteurization process. Mineral
membranes with 1.4 µm pore diameter are used so as to let casein
micelles go through. High tangential velocity (8 m/​s) and low
uniform TMP (0.4 bar) are obtained thanks to the recircula-
tion of the permeate. Retentate represents less than 5% of the
volume processed (VRR > 20). It can be reused for cheese making
after thermal treatment or recycled in the centrifuge at the early
skimming stage of the process.
Alcohol reduction in red wines has been achieved with tight
NF membranes (MWCO between 150 and 300 g/​mol) as an
alternative to thermal processing. Ethanol content was decreased
from 12% (v/​v) to less than 7–​8% (Catarino and Mendes, 2011).
Nonetheless, it was suggested to implement a pervaporation
step first to improve the recovery of odorant compounds before
338
dealcoholizing (Del Olmo et al., 2014; Catarino and Mendes,
2011). In the brewing sector, around 2% of the total beer pro-
duction consists of low-​alcohol or alcohol-​free products . These
products can be obtained by RO treatment (Lipnizki, 2017). The
products obtained have alcohol contents as low as 0.5% (v/​v),
but their flavor needs to be corrected by the addition of hops or
syrups.
Concentration/​Extraction
The use of membranes for concentration is particularly relevant
when the compounds to be recovered are sensitive to heat or to
pH or ionic strength changes. Membranes have therefore under-
gone significant development for the concentration of proteins,
where preserving nutritional and functional properties is of the
utmost importance.
The concentration of milk and whey proteins is the main and
oldest application. Depending on the membrane type, a variety of
applications and new products have been developed and are cur-
rently produced at industrial scale. MF (0.1 μm) is used for concen-
trating the native phospho-​caseinates to be incorporated in cheese
manufacturing. Milk pre-​ concentration by UF covers a large
range of applications, from standardization (1.8 < VRR < 4) either
for drinking milk or for cheese manufacturing to the obtaining
of a real “liquid pre-​cheese”: the MMV process (Maubois et al.,
1969) made it possible to concentrate milk proteins (caseins
+ whey proteins) until VRR = 7 with a 10–​15 kg/​mol MWCO,
hugely improving cheese manufacture performances –​mass
yield was increased by 20% and rennet reduced by 80%, and the
net gain represented 8% of the value of the milk processed; in
2016, 600,000 tons of cheese in the world were manufactured by
this process. Concentration of coagulated milk by UF (MWCO
150 kg/​mol) at VRR between 3 and 7 was found to produce fresh
cheese with a smoother texture than the traditional centrifugation
process and to improve the yield by between 3% and 25%; 80% of
the fresh cheese produced in the world is based on this technology
(Aimar and Daufin, 2004). Thanks to membrane technology, whey
turned from a waste in the 1960s to the raw material of a new and
beneficial industrial sector. Whey proteins can be concentrated by
UF and incorporated in cheese manufacturing. Through a cascade
of membranes from MF to NF, they can also produce whey protein
concentrates (WPC) with high purity (up to 80–​90%) for various
diets and also lactose (Daufin et al., 2001); finally, the permeate of
NF can be reused as process water after a final RO stage.
Many other protein-​containing products, whether they are of
animal or plant origin, are concentrated by membranes, such as
egg products (the whole egg and the egg white are concentrated
industrially by UF) or in the plant sector, the proteins of soya or
alfalfa.
Concentration of liquid foods can be performed using mem-
brane technologies as alternatives to evaporation or distillation.
With energy consumption (in kWh per ton of water or permeate
removed) of 3.5, 5–​7 and 9 for UF, NF and RO, respectively,
membranes are competitive with evaporative concentration
(around 100 kWh per ton for a 5-​effect plant or 10–​20 kWh per ton
when equipped with mechanical vapor compression) and preserve
the product better, avoiding the effects of thermal degradation,
Marie-Laure Lameloise
such as browning, for example, for sugar syrups. However, due
to viscosity and osmotic pressure issues, membranes are prefer-
ably considered for pre-​concentration. Not limited by osmotic
pressure, osmotic evaporation (OE), at room temperature, allows
dry matter (DM) content up to 60% to be achieved. It has been
studied for concentrating sucrose solutions (Courel et al., 2000)
and fruit juices. In this latter case, it was proved to preserve the
sensory quality of the product better than evaporation. However,
evaporation fluxes remain low and show a marked decay at DM
content exceeding 45%. Therefore, combined processes associ-
ating RO (or NF) with OE have recently been proposed (Cissé
et al., 2011; Bhattacharjee et al., 2017).
Integrated Separation Processes
Separation or fractionation into several constituents with
adequate purity, recovery and concentration involves a cascade
of membranes with different MWCO, often associated with
diafiltration steps, as already illustrated in the case of whey
fractionation. Together with whey, the milk industry is also a
pioneer in ingredient fractionation by membrane technologies.
A number of different fractions and ingredients can be obtained
by pressure-​driven membrane technologies from whole milk. In
addition to the different applications mentioned earlier, recently,
fractionation of milk fat by MF allowed for separate fractions
of fat globules of different sizes in the retentate and in the per-
meate. Incorporating these fractions into dairy products allows
interesting textural characteristics to be achieved.
Diverse functional fractions are also purified from fruits and
vegetables and further used as food ingredients or nutraceuticals.
Phenolic compounds, such as tannins, anthocyanins or flavonoids,
can be isolated from fruits such as berries. Traditionally, these
compounds are obtained by solid–​liquid or liquid–​liquid extrac-
tion, depending on the raw material (must or clarified juice).
Membrane technologies present the advantage of avoiding
the use of polluting and toxic solvents. Different applications
for the recovery of polyphenolic compounds from raw plant
materials or food industry by-​products are presented by Conidi
et al. (2017a): for example, polymeric phenolic compounds are
concentrated from grapes in the retentate of UF membranes of
MWCO of 100 kg/​mol, while low-​molecular-​weight fractions
are recovered in the permeate. Winery effluents and sludges can
also be valorized, as they contain important residual amounts of
antioxidants and polyphenols.
Pomegranate juice can also undergo UF or NF, including a
diafiltration step, to separate phenolic compounds from sugars
and obtain antioxidant fractions (Conidi et al., 2017b). Phenolic
compounds can also be recovered by pressure-​driven membrane
technologies from olive mill wastewaters. In these processes, pH
control (at acidic values) is extremely important to avoid the oxi-
dation of antioxidant phenolic compounds. The MWCO of the
membranes used is often in the UF range (10–​25 kg/​mol) to allow
the recovery of polymeric fractions. Dense membranes are used
for the fractionation/​concentration/​purification of already prepared
extracts, generally hydro-​alcoholic ones. They are also used to
recover and reuse the extraction solvents (Conidi et al., 2017a).
Filtration Membranes
To improve selectivity, separation or fractionation can also
be obtained by the association of membranes either with other
membrane technologies, such as electrodialysis, or more highly
selective technologies, such as chromatography. The so-​called
EDUF process (Bazinet, 2005) based on the integration of UF
membranes and ion-​exchange membranes within a stack in an
electric field allows the improvement of peptide separation
obtained from the hydrolysis of proteins and the production of
fractions enriched in target peptides with controlled biological
properties. By combining the two types of membranes, the select-
ivity of the separation is improved.
Wastewater Treatment and Valuation
Following the development of membranes in the environmental
sector for potabilization of seawater or brackish water by RO or
NF, the improvement of potable water quality in the traditional
water plants (UF) or the introduction of a membrane bio-​reactor
(MBR) in municipal wastewater treatment plants, membranes are
attracting much interest for the recycling or reuse of the effluent
of food processing as process water and the recovery of raw
materials or valuable compounds. These objectives do not cur-
rently go hand in hand, for economic and regulatory reasons, and
the fractionation observed in milk or in whey industry remains an
exception. However, changes are expected due to water resource
issues.
In the refining process of vegetable oils, the neutralization
process releases very acidic wastewaters (soapstocks) with a
pH value of about 1. Physical pretreatment by decantation or
dissolved air flotation may not be efficient enough to remove
all the fats, and this severely impedes the subsequent biological
treatment (poor degradability, foaming). Interposing MF between
the physical and the biological steps was shown to enhance the
recovery of fats in the retentate and reduce the fat load sent to
the biological station without using chemicals (Decloux et al.,
2007). Ceramic membranes with pore size from 0.2 to 1.4 μm
were tested at very low TMP. With a 0.5 μm membrane, at 60 °C
and VRR values between 10 and 30, it was possible to maintain
a permeate flux of 100 L/​h/​m2 and achieve a reduction of 91% of
suspended solids, 96% of fats and more than 60% of the chem-
ical oxygen demand (COD). In this application, fat retention by
MF was proposed to be a matter more of chemical affinity than
of pure steric retention.
Evaporative concentration is often used to treat fluxes with
high COD (stillage in distillery, whey in the cheese industry,
etc.), allowing the separation of dry matter and the recovery of
good-​quality water as condensates. The recycling or reuse of
condensates as process water can necessitate further purification
treatments. In the starch industry, the condensates from the con-
centration of steep waters are used as boiler water after demin-
eralization by RO. In beet distilleries, recycling the condensates
from stillage concentration as dilution water at the fermentation
stage is challenging, because they contain organic species that
have been proven to inhibit fermentation: carboxylic acids (espe-
cially acetic acid), furan derivatives and phenolic compounds
(Morin Couallier et al., 2006). Loose RO, possibly coupled with
ion exchange, was demonstrated to remove them efficiently and
339
to produce fermentable water (Sagne et al., 2008; Lameloise
et al., 2015).
In the dairy industry, heat treatment and cleaning-​in-​place are
responsible for between 50% and 95% of the effluents produced.
In the last two decades, an efficient strategy of water treatment
and recycling within small loops was developed (RO treatment
of evaporation condensates, NF regeneration of alkaline and acid
cleaning solutions (Daufin et al., 2001)). In the malt industry,
barley steeping is the main effluent-​generating operation. MBR
treatment, complete with an NF step, was proven to produce water
that could be reused in the malting process (Guiga et al., 2007).
In cane sugar refinery, anion-​exchange resins are often preferred
to granulated active carbon to ensure the ultimate decoloring step
of the syrup before crystallization. However, resin regeneration
with NaCl 10% produces colored brines, which are difficult to
handle. NF brought a smart solution to this issue by allowing the
removal of the colorants from the brine. A first unit with low-​
cost spiral wound membrane was installed in 1997 in a French
refinery with the purpose of recovering brine and recycling it dir-
ectly. The most concentrated fraction of the brine was processed,
containing 93% of the total NaCl and 50% of the total COD. By
associating NF and diafiltration, it was possible to recover 98%
of the NaCl of the treated fraction (i.e., 91% of total NaCl) and
to concentrate 95% of the coloring matter in 3–​4% (discarded) of
the volume of the treated fraction (Cartier et al., 1997).
In addition to water recovery, valuable organic compounds
such as organic acids, phenolic compounds, sugars, peptides,
odorant compounds, etc. can be recovered from food effluents.
Winery effluents contain significant concentrations of colloidal
matter, polysaccharides and phenolic compounds exhibiting
antioxidant activities and are thus interesting for recovery with
successive MF, UF and NF separations (Giacobbo et al., 2017).
Blanching or cooking juices could be the source of natural
flavoring compounds sought after by the flavor industry. UF was
shown to efficiently remove the dark brown color of mushroom
blanching juices while preserving their odorant qualities, opening
the gate to further processing (Lameloise et al., 1997).
A recent research program associating academic and indus-
trial partners has developed methods for fractionation of cooking
juice from ham, through combined membrane processes, to
recover proteins and odorant compounds. These are reused
in the formulation of novel products and are commercialized
today. Recovery of odorant compounds from seafood cooking
juices is a bit more challenging because of the high concentra-
tion of salt. Concentration of shrimp cooking juice by RO after
a first desalination step by electrodialysis (to decrease osmotic
pressure) was proven to give a concentrate with satisfying sen-
sory quality while significantly decreasing the COD load of
the juice (Cros et al., 2006). More recently, it was proposed to
associate a pre-​concentration step by NF until VRR = 10 and a
final concentration step by osmotic evaporation (OE) (Jarrault
et al., 2017). Thanks to NF, the juice was partially concentrated
and desalinated, allowing the further OE step to run in optimal
conditions. A 52% DM concentrate was obtained with odorant
loss lower than 35%. This concentrate could be incorporated
at 2–​ 3% in food preparations, with sensorial characteristics
matching expectations.
340
Conclusion
Thanks to steric effects, and also to other selectivity effects
that still remain to be better understood for further control and
implementation, membranes allow a variety of liquid streams
and solutes to be clarified, purified, fractionated or concentrated
with significant benefits compared with traditional technologies:
they dramatically reduce the use of chemicals and the produc-
tion of waste; and they operate at low temperature and without
phase change, leading to better-​preserved organoleptic and func-
tional properties and to lower thermal energy expense. A constant
source of innovation in the food industry for decades, in terms
of processes and of products, membranes are expected to make
in the future a sound contribution to the design and elaboration
of novel food in different, non-​industrial contexts, provided the
technologies can be adapted (especially scaled down).
REFERENCES
Aimar P, Daufin G. 2004. Séparations par membrane dans
l’industrie alimentaire, Techniques de l’ingénieur Opérations
unitaires du génie industriel alimentaire, base documentaire:
TIB430DUO.
Baker RW. 2012. Membrane Technology and Applications. John
Wiley & Sons, Newark, California.
Bazinet L. 2005. Electrodialytic phenomena and their applications in
the dairy industry: a review, Critical Reviews in Food Science
and Nutrition, 45, 307–​26.
Bhattacharjee C, Saxena VK, Dutta S. 2017. Fruit juice processing
using membrane technology: a review, Innovative Food
Science & Emerging Technologies, 43, 136–​53.
Cartier S, Theoleyre MA, Decloux M. 1997. Treatment of sugar
decolorizing resin regeneration waste using nanofiltration,
Desalination, 113, 7–​17.
Catarino M, Mendes A. 2011. Dealcoholizing wine by membrane
separation processes, Innovative Food Science & Emerging
Technologies, 12, 330–​7.
Charcosset C. 2009. Preparation of emulsions and particles by mem-
brane emulsification for the food processing industry, Journal
of Food Engineering, 92, 241–​9.
Cissé M, Vaillant F, Bouquet S, Pallet D, Lutin F, Reynes M, Dornier
M. 2011. Athermal concentration by osmotic evaporation of
roselle extract, apple and grape juices and impact on quality,
Innovative Food Science & Emerging Technologies, 12,
352–​60.
Conidi C, Drioli E, Cassano A. 2017a. Membrane-​based agro-​food
production processes for polyphenol separation, purifica-
tion and concentration, Current Opinion in Food Science, 23,
149–​64.
Conidi C, Cassano A, Caiazzo F, Drioli E. 2017b. Separation and
purification of phenolic compounds from pomegranate juice
by ultrafiltration and nanofiltration membranes, Journal of
Food Engineering, 195, 1–​13.
Courel M, Dornier M, Herry JM, Rios GM, Reynes M. 2000. Effect
of operating conditions on water transport during the concen-
tration of sucrose solutions by osmotic distillation, Journal of
Membrane Science, 170, 281–​9.
Cros S, Lignot B, Jaouen P, Bourseau P. 2006. Technical and eco-
nomical evaluation of an integrated membrane process capable
both to produce an aroma concentrate and to reject clean water
from shrimp cooking juices, Journal of Food Engineering, 77,
697–​707.
Marie-Laure Lameloise
Daufin G, Escudier JP, Carrère H, Bérot S, Fillaudeau L, Decloux
M. 2001. Recent and emerging applications of membrane
processes in the food and dairy industry, Food and Bioproducts
Processing, 79, 89–​102.
Decloux M, Lameloise ML, Brocard A, Bisson E, Parmentier M,
Spiraers A. 2007. Treatment of acidic wastewater arising
from the refining of vegetable oil by crossflow microfiltration
at very low transmembrane pressure, Process Biochemistry,
42, 693–​9.
Del Olmo A, Blanco CA, Palacio L, Prádanos P, Hernández A. 2014.
Pervaporation methodology for improving alcohol-​free beer
quality through aroma recovery, Journal of Food Engineering,
133, 1–​8.
Gésan-​Guiziou G. 2010. Removal of bacteria, spores and som-
atic cells from milk by centrifugation and microfiltration
techniques, In Griffiths MW (ed.), Improving the Safety and
Quality of Milk, Woodhead Publishing, Cambridge, United
Kingdom.
Giacobbo A, Meneguzzi A, Bernardes AM, de Pinho MN. 2017.
Pressure-​driven membrane processes for the recovery of anti-
oxidant compounds from winery effluents, Journal of Cleaner
Production, 155, 172–​8.
Guiga W, Fick M, Ouarnier N, Boivin P. 2007. Malt house process
and equipment with steep water recycling, CNRS, IFBM and
SWAN Malt France. EP no. W0/​2008/​135222.
Hancocks RD, Spyropoulos F, Norton IT. 2016. The effects of mem-
brane composition and morphology on the rotating membrane
emulsification technique for food grade emulsions, Journal of
Membrane Science, 497, 29–​35.
Hongvaleerat C, Cabral LMC, Dornier M, Reynes M, Ningsanond
S. 2008. Concentration of pineapple juice by osmotic evapor-
ation, Journal of Food Engineering, 88, 548–​52.
Ilić JD, Nikolovski BF, Petrović LB, Kojić PS, Lončarević IS,
Petrović JS. 2017. The garlic (A. sativum L.) extracts food
grade W1/​ O/​W2 emulsions prepared by homogenization
and stirred cell membrane emulsification, Journal of Food
Engineering, 205, 1–​11.
Jarrault C, Dornier M, Labatut ML, Giampaoli P, Lameloise ML.
2017. Coupling nanofiltration and osmotic evaporation
for the recovery of a natural flavouring concentrate from
shrimp cooking juice, Innovative Food Science & Emerging
Technologies, 43, 182–​90.
Lameloise ML, Gavach M, Bouix M, Fargues C. 2015. Combining
reverse osmosis and ion-​ exchange allows beet distillery
condensates to be recycled as fermentable dilution water,
Desalination, 363, 75–​81.
Lameloise ML, Giampaoli P, Lewandowski R. 1997. Etude de la
décoloration de jus de blanchiment de champignons pour une
valorisation aromatique, Industries Alimentaires et Agricoles,
114, 372–​8.
Lipnizki F. 2017. 4.10 Basic aspects and applications of mem-
brane processes in agro-​food and bulk biotech industries, In
Comprehensive Membrane Science and Engineering (2nd ed.)
(eds. E Droli, L Giorno, E Fontanova), Elsevier, Oxford.
Maubois JL, Mocquot G, Vassal L. 1969. A method for processing
milk and dairy products. French Patent.
Morin Couallier E, Payot T, Pastore Bertin A, Lameloise ML. 2006.
Recycling of distillery effluents in alcoholic fermentation,
Applied Biochemistry and Biotechnology, 133, 217–​37.
Nguyen N, Fargues C, Guiga W, Lameloise ML. 2015. Assessing
nanofiltration and reverse osmosis for the detoxification of
lignocellulosic hydrolysates, Journal of Membrane Science,
487, 40–​50.
Filtration Membranes
Polidori J, Dhuique-​ Mayer C, Dornier M. 2018. Crossflow
microfiltration coupled with diafiltration to concentrate and
purify carotenoids and flavonoids from citrus juices, Innovative
Food Science & Emerging Technologies, 45, 320–​9.
Pouliot Y. 2008. Membrane processes in dairy technology –​From
a simple idea to worldwide panacea, International Dairy
Journal, 18, 735–​40.
Sagne C, Fargues C, Lewandowski R, Lameloise ML, Decloux
M. 2008. Screening of reverse osmosis membranes for the
treatment and reuse of distillery condensates into alcoholic
fermentation, Desalination, 219, 335–​47.
341
Silva PS, Starov VM, Holdich RG. 2017. Membrane emulsifica-
tion: Formation of water in oil emulsions using a hydrophilic
membrane, Colloids and Surfaces A: Physicochemical and
Engineering Aspects, 532, 297–​304.
Vaillant F, Cisse M, Chaverri M, Perez A, Dornier M, Viquez F,
Dhuique-​ Mayer C. 2005. Clarification and concentration
of melon juice using membrane processes, Innovative Food
Science & Emerging Technologies, 6, 213–​20.
Food Matrices and the Matrix Effect in the Kitchen
José M. Aguilera1 and Hervé This vo Kientza2,3
1 Department of Chemical and Bioprocess Engineering, Pontificia Universidad Católica de Chile, Avenida Libertador Bernardo
O’Higgins 340, Santiago, Chile
2 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
3 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
Introduction
Chemically speaking, dishes are mainly composed of water,
proteins and amino acids, lipids (mainly triglycerides) and
saccharides (mono-​, oligo-​or poly-​saccharides), but they also
contain many compounds (polyphenols, minerals, vitamins, etc.)
that are minor in terms of proportions but can be important in
terms of sensory appreciation, nutrition or toxicology (Belitz
et al., 2009). These compounds are not always homogeneously
dispersed in the dishes, or in a free molecular state; in most cases,
they are entrapped in a “matrix” that can take different forms.
These might be solid deposits or liquid droplets, a liquid phase
(water or oil), or glassy or rubbery domains, and they are also
sometimes adsorbed on various interfaces separating phases such
as gas, liquids and solids (Aguilera, 2013; This, 2016).
The breakdown of food during consumption can release some
of these compounds in saliva as dissolved molecular or ionic
species, or volatilized in the gas phase inside the mouth and
the nose. This release is not the same as it would be if the pure
compounds were sprinkled over the surface of a finished dish.
Moreover, during digestion, not all food compounds become
available as individual nutrients; thus, they are unavailable to
be carried to the cells of our bodies. Two key related concepts,
expressed in the terms “food matrix” (FM) and the “matrix
effect” (ME), are questions that we discuss in this chapter.
What Is a Food Matrix?
Dictionaries define a matrix as “something where other things
are embedded”. The term “matrix” is used in several scientific
disciplines to describe those parts of a whole that provide a spe-
cific functionality (scaffolding, stability, strength, diffusivity,
etc.). In cell biology, for example, the intracellular cytoplasmic
matrix corresponds to a gel-​like structure in the interior of cells,
based on filaments, microtubules and proteins, that restricts
the mobility of molecules in the cytoplasm. Thus, the FM may
be viewed as a part of the microstructure of foods, usually
corresponding to a spatial domain that contains, interacts with
and/​or gives particular functionalities to a specific constituent of
the food (e.g., a nutrient, odorant molecules, beneficial bacteria,
etc.) (Aguilera, 2018).
The organization of chemical compounds in dishes at the various
scales (macroscopic, microscopic, mesoscopic, nanoscopic and
molecular) determines their physical and biological (sensorial,
nutritional, etc.) properties (Aguilera, 2005; This, 2009; Heertje,
2014). For example, both the structural (physical) organization
and the chemical composition dictate the textural responses of
foods (Stanley, 1987) and their effects on the human body.
The term “matrix effect” has appeared in the food science and
technology literature to denote that compounds in dishes and in
other formulated products behave differently in isolated form
(e.g., in solution) than when forming part of food structures.
A first deduction from this definition is that the FM is component-​
specific, i.e., that different components in the same food may “see”
or interact with different matrices. A second deduction is that the
FM is scale-​sensitive, i.e., interactions of food components may
take place at various scales in the same food, hence involving
different matrices.
For example, the matrix in bread responsible for the textural
properties of the porous crumb consists of the protein-​starch
walls (a few hundred micrometres in thickness) surrounding the
air cells. Starch granules undergo gelatinization during baking
while constrained by the continuous gluten matrix (at a scale of
10–​100 micrometres); then, at the nanoscale, the hydrated and
swollen (or partially swollen) starch granules are the matrix on
which α-​ amylases act during digestion to release glucose or
oligosaccharides from amylose.
Classifying Food Matrices by the Dispersed System
Formalism
The diverse nature of FMs demands a classification of them at
different physical scales. The “dispersed system formalism”
(DSF) provides such a description (see the corresponding chapter
in this book) (This, 2007; This, 2016). In particular, food systems
343
344
can be classified in order of the number of phases (themselves
ranked in alphabetical order) and in order of “complexity” of
operators as ranked using the free energy (σ, @, x, /​). Here, we use
the microscopic scale to describe the main characteristics of some
food systems from the food science and technology literature.
Solid and Liquid Matrices (S)
Solids and solutions (aqueous solutions W, or “oils” O) are
among the simplest food systems; chemical interactions between
the molecules of a solid, or between solutes and the solvent
(in this case, the matrix), change the release of compounds as
compared with pure preparations of these compounds (Ammari
and Schroen, 2018).
For example, the matrix of wine corresponds to the aqueous +
ethanol phase, which contains organic acids, phenolic compounds
(including tannins), glycerol, minor quantities of proteins and
saccharides, and odorant compounds (Villamor and Ross, 2013).
Most of these chemical species, through their interactions, influ-
ence the flavour of wine (Jones et al., 2008).
One could assume that solutions are simple systems, but this
is not true, as even a dilute solution of sucrose in water contains
“aggregates” of sucrose (Kaufman and Dorman, 2008). When
solutions contain bigger molecules, more complex interactions
take place, and this determines the release of such nutrients as
vitamin C and bioactive compounds (carotenoids, flavonoids and
other polyphenols) from fruit juices, in particular when crushed
or homogenized (fruit “smoothies” contain a lot of fibres, i.e.,
cellulose or hemicellulose molecules) (Caswell, 2009).
Emulsions (L1/​L2) and Suspension (S/​L) Matrices
The next systems after solids and solutions (S) in terms of com-
plexity are emulsions and suspensions. Emulsions are described
by DSF formulas such as O/​W (oil-​in-​water), as in mayonnaise,
W/​O (water in oil) or even W/​W (water in water) in some par-
ticular Ramsden solutions (Nicolai and Murray, 2017). More
generally, a liquid L1 is randomly dispersed in another liquid L2,
hence the formula L1/​L2. The concept of the matrix in emulsions,
particularly in O/​ W emulsions, has various interpretations
depending on the scale. At the macroscopic scale, the matrix is the
continuous phase (L2) that contains the dispersed phase (L1), the
latter including the interface layer and the interior of the droplets.
This viewpoint has been important in studying the (meta)stability
of emulsions (e.g., by controlling the make-​up of the interfacial
layer and the viscosity of the continuous phase). However, at the
microscopic and nanoscopic levels, the architecture of the inter-
facial layer itself is also denominated as a “matrix” and plays
a key role in particle-​to-​particle interactions and protection of
the droplet contents against oxidation while allowing entrap-
ment of odorants, for example. Finally, the case of Ramsden
emulsions (sometimes wrongly called “Pickering emulsions”),
i.e., emulsions stabilized by solid particles that adsorb onto the
interface between the two phases, has to be considered, because
at this level, the DSF formula is different, which shows that DSF,
including the reference scale, could be usefully used for the defin-
ition of the FM (Ramsden and Oxon, 1904; Rayner et al., 2014).
José M. Aguilera, Hervé This vo Kientza
Gel Matrices (X/​D3(S), XxD3(S))
As shown in other chapters in this handbook, gel matrices are
important food structures that can hold large amounts of various
liquids (e.g., >80%), often within a three-​dimensional network,
providing a semi-​solid material with a specific soft texture behav-
iour. The network of many food gel matrices is fine-​stranded
(gelatine and pectin gels) or particulate (protein aggregates),
gel matrices possibly holding small elements dispersed in their
interior as well as particles (filled gels), oil droplets (emulsion
gels) and air bubbles (aerated gels) (Banerjee and Bhattacharya,
2012). Gels can be prepared from a single biopolymer (e.g., gel-
atin, agar-​agar, alginate and some carrageenans), from mixtures
of gelling agents (e.g., egg proteins in the white of a soft-​boiled
egg) and as part of many dishes (aspics, desserts) and confec-
tionery products (e.g., jelly beans). In particular, confectionery
gels have quite a complex FM that results from interactions of
high-​sugar components (sucrose and glucose syrup) and gelling
precursors such as starch, gelatin or pectin (Burey et al., 2009).
However, as can be seen in the chapter on gels in this book, plant
and animal tissues are also gels, owing most of their mechanical
properties to the connective tissues (collagen) or to the membrane
and cell wall surrounding and binding the gelled cell contents
(Vincent, 2008). Most of the time, the use of the word “matrix”
in fruit and vegetable studies refers to the entrapment inside cell
walls of microstructural elements relevant to foods (e.g., starch
granules, protein bodies, etc.) and organelles containing nutrients
and functional molecules (e.g., chloroplasts, chromoplasts, etc.).
In turn, the cell wall (around 100 nm in thickness) consists of a
matrix comprised of cellulose, hemicelluloses, pectins and some
structural proteins (Cosgrove, 2005).
Porous Matrices
Several foods are porous materials consisting of a continuous
matrix, which may be a solid (uncooked meringue) or a gel
(bread, marshmallows, etc.), enclosing open or closed gas cells
(bubbles or gas channels). Porous matrices may be formed by
fermentation and baking, extrusion, aeration or gasification, gas
release from chemical reactions, or freeze-​drying. In addition,
several fruits are naturally porous (most notably apples, which
have around 25% porosity). The dispersion of a gas phase within
a food matrix not only affects its texture (making the final product
lighter) but also changes the appearance, consistency, colour and
mouth-​feel. Foamed liquid matrices may be used as scaffolds
and folded in with sweet or salty fillers, as in soufflés. The tex-
ture of porous solid foods and liquid foams largely depends on
the properties of the matrix surrounding the dispersed gas phase
(Corriadini and Peleg, 2008).
More Complex Matrices
In food, there are often more than only two phases, and this leads
to more complex matrices (associated with a specific DSF for-
mula). For example, dense matrices are usually low-​moisture,
glassy, semi-​crystalline or crystalline structures. They are found
particularly in sugar-​ based confections and categorized as
amorphous (ungrained caramel), glassy (hard candy), crystalline
Food Matrices and Matrix Effect
(rock candy) or partially crystalline (fondants) (Ergun and Hartel,
2009). Food powders produced by spray-​drying (e.g., skim milk
or instant coffee), milling (flours of cereals or legumes, ground
dry spices) and starch flour also belong to the category of
amorphous complex matrices. Glassy matrices drastically reduce
the mobility of small solutes such as volatile odorants, which
become trapped during spray-​and freeze-​drying (e.g., in instant
coffee), or may be produced so as to encapsulate probiotic bac-
teria, bioactive compounds or flavourings.
FM and Cooking
There are several reasons why chefs and cooks should be aware
of FM and the ME. The following is a limited list.
Texture
Let’s take an egg as an example. Although the composition of an
egg does not change during heating in hot water, a soft-​boiled egg
has a texture quite different from that of a hard-​boiled egg. We
can say that the matrix changes from soft gelatinous (the white)
and liquid (the yolk) to a full gel consistency (the white) and a
crumbly solid (the yolk).
In the case of baked products, most of the desirable textural
properties are ascribed to a matrix formed by a stiff gluten-​starch
network that occludes a gas phase in different proportions.
Flavour Release
Of course, the release of specific compounds important for
flavour (including odour, taste, trigeminal sensation, etc.) is
linked to the breakdown of food during oral processing, which
comprises biting, mastication, comminution, mixing and
lubrication, bolus formation, and swallowing. During mas-
tication, solid and soft food matrices become reduced in size
to an extent depending on their physical properties and the
chewing behaviour of individuals. Most odorant, taste and
trigeminal compounds are detected sensorially or active bio-
logically only when they are released from the FM (Guichard
and Salles, 2016). Matrix hydration and breakdown in the oral
cavity favour the diffusion and mass transfer of molecules into
the saliva and through the gas phase to olfactory or trigeminal
receptors in the nose. The nature, amount and interactions of
different components liberated from the FM greatly influence
odorant and trigeminal perception (e.g., sensations of warmth,
coldness, tickling, pain, etc.).
Nutrition
In recent decades, nutritional sciences have become concerned not
only about the kind and amounts of nutrients required for good
health but also with the fraction of a given nutrient that is actu-
ally available to be utilized by our body. Molecular gastronomy
also came to consider such questions (This, 2013). The “com-
pound release” (fraction released for a particular physiological
process following mastication, digestion and perhaps fermenta-
tion in the lower intestine), bioaccessibility of nutrients (fraction
345
released during digestion) and bioavailability (fraction being actu-
ally absorbed) are directly related to the FM. Bioconversion and
bioefficacy have to do with biochemical transformations of food
components once released from the matrix, and their specific
physiological and health responses in the body. Bioavailability,
rather than the amount of food compounds ingested, has become
the criterion to assess the potential nutritional benefits derived from
bioactive compounds in foods and to sustain their health claims.
Food Analysis
Quite often, foods are analysed in laboratories for the presence of
minor components whose effects on our health could be positive
(e.g., micronutrients) or negative (toxic compounds, allergens,
pollutants and even pathogenic microorganisms). Laboratory
techniques are designed to recover and make available for quan-
tification most of (if not all) the molecules or entities under scru-
tiny. In the laboratory, drastic size reduction methods (mechanical
grinding, ultrasonic disruption) are used to break down the FM,
and a variety of solvents and enrichment procedures are used to
isolate molecular species (and microorganisms).
These procedures and conditions are quite different from
those experienced by foods in the oral cavity and during transit
in the gastrointestinal system. There is now ample evidence that
many molecules in foods are not released from the food matrix
during digestion, and hence are not absorbed into the body. As
a matter of fact, we eliminate around 130 grams of wet faeces
and 1.4 litres of urine per day, which mainly contain unabsorbed
material coming from foods (Rose et al., 2015). For example,
intact almond cells surviving mastication and digestion have been
found in faeces, and thus, their reported caloric content (based on
composition) may be greatly overestimated. In the case of pasta
products, extensive size reduction eliminates the encapsulating
effect of the protein matrix on starch granules, with a concomitant
overestimation of the glycaemic index. ME provided by intact
cells or complex matrices may thus be relevant to bioaccessibility
and bioavailability studies.
Conclusions
The concept of FMs is now extensively used by food and nutri-
tional scientists to try to explain why some compounds from
foods behave differently in a food than in isolated or pure form
(e.g., in a solution). Several types of matrices can be recognized
in foods that are also referred to in other disciplines: liquids,
emulsions, cellular tissues, polymer networks, etc. It follows
from this viewpoint that the FM is component-​specific and scale-​
sensitive. Analytical procedures to assess the bioaccessibility of
nutrients should preserve the MEs; otherwise, the results will
only represent the total amount present in a sample.
REFERENCES
Aguilera JM. 2005. Why food microstructure? Journal of Food
Engineering, 67, 3–​11.
Aguilera JM. 2013. Edible Structures: The basic Science of What We
Eat. CRC Press, New York, NY.
346
Aguilera JM. 2018. The food matrix: implications in processing,
nutrition and health. Critical Reviews in Food Science and
Nutrition, 59(3), 1–​43.
Ammari A, Schroen K. 2018. Flavor retention and release from
beverages: a kinetic and thermodynamic perspective (review).
Journal of Agricultural and Food Chemistry, 66(38),
9869–​9881.
Banerjee S, Bhattacharya S. 2012. Food gels: gelling process and new
applications. Critical Reviews in Food Science and Nutrition,
52(4), 334–​346.
Belitz HD, Grosch W, Schieberle P. 2009. Food Chemistry. Springer,
Heidelberg.
Burey P, Bhandari BR, Rutgers RPG, Halley PJ, Torley PJ. 2009.
Confectionery gels: a review on formulation, rheological and
structural aspects. International Journal of Food Properties,
12(1), 176–​210.
Caswell H. 2009. The role of fruit juice in the diet: an overview.
Nutrition Bulletin, 34, 273–​288.
Corriadini MG, Peleg M. 2008. Solid food foams. In Aguilera JM,
Lillford PJ (eds.) Food Materials Science: Principles and
Practice. Springer, New York, 169–​202.
Cosgrove DJ. 2005. Growth of the plant cell wall. Nature Reviews
Molecular Cell Biology, 6, 850–​861.
Ergun R, Hartel RW. 2009. Phase transitions in confections: the
graining factor approach. Manufacturing Confectionery,
89(10), 51–​55.
Guichard E, Salles C. 2016. Retention and release of taste and aroma
compounds from the food matrix during mastication and inges-
tion. In Etievant P (ed.) From Food to Behaviors, Wellbeing
and Health. Woodhead Publishing, Oxford, UK, 3–​22.
Heertje I. 2014. Structure and function of food products: a review.
Food Structure, 1(1), 3–​23.
Jones PR, Gawel R, Francis, IL, Waters EJ. 2008. The influence of
interactions between major white wine components on the
aroma, flavour and texture of model white wine. Food Quality
and Preference, 19(6), 596–​607.
José M. Aguilera, Hervé This vo Kientza
Kaufman SL, Dorman FD. 2008. Sucrose clusters exhibiting a
magic number in dilute aqueous solutions. Langmuir, 24(18),
9979–​9982.
Nicolai T, Murray B. 2017. Particle stabilized water in water
emulsions. Food Hydrocolloids, 68, 157–​163.
Ramsden W, Oxon MD. 1904. Separation of solids in the surface
layers of solutions and suspensions. Proceedings of the Royal
Society of London, 72, 156–​164.
Rayner R, Marku D, Eriksson M, Sjöö M, Dejmek P, Wahlgren M.
2014. Biomass-​based particles for the formulation of Pickering
type emulsions in food and topical applications. Colloids and
Surfaces A: Physicochem. Eng. Aspects, 458, 48–​62.
Rose C, Parker A, Jefferson B, Cartmell E. 2015. The charac-
terization of feces and urine: a review of the literature to
inform advanced treatment technology. Critical Reviews in
Environmental Science and Technology, 45(17), 1827–​1879.
Stanley DW. 1987. Food texture and microstructure. In Moskowitz HR
(ed.) Food Texture: Instrumental and Sensory Measurement.
Marcel Dekker, New York, 35–​64.
This, H. 2007. Formal descriptions for formulation. International
Journal of Pharmaceutics, 344(1–​2), 4–​8.
This H. 2009. Molecular gastronomy, a chemical look to cooking.
Accounts of Chemical Research, 42(5), 575–​583.
This H. 2013. Solutions are solutions, and gels are almost solutions.
Pure & Applied Chemistry, 85, 257–​286.
This H. 2016. Statgels and dynagels. Notes Académiques de
l’Académie d’Agriculture de France, /​Academic Notes from
the French Academy of Agriculture, 12, 1–​12.
Villamor RR, Ross CC. 2013. Wine matrix compounds affect per-
ception of wine aromas. Annual Review of Food Science and
Technology, 4, 1–​20.
Vincent JFV. 2008. The composite structure of biological tissue used
for food. In Aguilera JM, Lillford PJ (eds.) Food Materials
Science. Springer, New York, 11–​20.
Food Pairing: Is It Really about Science?

Hervé This vo Kientza1,2 and Christophe Lavelle3

1 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
2 UMR 0782 SayFood, AgroParisTech, INRAE, Université Paris-​Saclay, 91300 Massy, France
3 Genome Structure and Instability, CNRS UMR7196 /​INSERM U1154, National Museum of Natural History, Sorbonne

University, 75005, Paris, France

About the so-​ called “food pairing theory”, let’s tell the Refutation of the Food Pairing Theory: A Critical
story, because it helps to understand the issue. At the second Look at the Literature
International Workshop on Molecular and Physical Gastronomy
Food pairing is sometimes advertised as a scientific method to
(1995) in Erice, there were representatives of the flavouring
identify which foods and drinks go well together (The Food
industry who discussed the idea that one could associate two
Pairing Company, 2020). Briefly, it starts with a GC-​MS chemical
ingredients in a dish if they had odorant compounds in common.
analysis (gas chromatography coupled with mass spectrometry)
Later, some cooks collaborated with this flavouring company and
of an ingredient to identify the main odorant compounds that a
produced some dishes designed on the basis of odorant analysis
human will effectively smell. This enables the creation of a fla-
of food ingredients.
vour profile of the ingredient, which can be compared with other
At the third Workshop on Molecular and Physical Gastronomy
profiles by screening a database to build a tree graph based on the
(1997), when this question was discussed again, along with
so-​called food pairing hypothesis; ingredients that have aromas
tasting of some dishes prepared according to this principle, two
in common are said to match in a dish. Or, to put it another way:
observations were made: (1) not all participants of the workshop
“the more aromatic compounds two foods have in common, the
agreed that the proposed associations “worked well”; (2) when
better they taste together”. Now, the simplicity of this theory is
it was observed that the probability of having two odorant
as tempting and popular as it is scientifically unsound; indeed,
compounds in common was almost 100% for any pair of food
as a “scientific theory”, food pairing has been refuted for many
ingredients (for example, ethanol is present in most plant tissues,
reasons.
but also many compounds important for living systems are in
First, on an individual basis, many chefs are able to pair food
all living cells), the promoters of the theory modified it to say
ingredients given to them at random (Pierre Gagnaire, among
that these compounds had to be “preponderant”. Moreover, some
others, has played this game many times, in particular during
chefs attending the workshop claimed that talented cooks can
TV shows). Of course, food pairing does not apply every time
make a “good” dish with any association of food ingredients.
that only two ingredients are used, and this is fortunate, because
Since then, some companies have been set up to sell advice
dishes with only two ingredients on the basis of odour would
about pairing, and some chefs use it for inspiration. However,
probably lack interest in terms of consistency, taste, trigeminal
the analysis should be as follows: if culinary activity includes
sensation, calcium perception, fat perception, colour and so on.
an art component (and it does), the question of what is “good”
Then, when it is argued that “the major volatile molecules of two
is, in fact, the question of what is “beautiful to eat” (This, 2010).
foods are the same”, this is ambiguous, because “major” could
Also, the history of art shows clearly that all arts (music, painting,
be about quantity (but the probability that two ingredients have
etc.) evolved in order to escape the rules. For example, in music,
the same compounds in equal quantity is almost nil) or quality
it was said in the Middle Ages that a tritone (a musical interval
(but the human diversity in olfactory receptors does not make this
composed of three adjacent tones) was Diabolus in Musica (the
measurement possible).
interval of the devil), but many artists used it in order to produce
For a more systematic and comprehensive criticism of food
beautiful music. Also, the classical rules of perspective are cer-
pairing theory, let’s have a closer look at some assumptions made
tainly not respected in painting by Picasso and other artists from
and how the scientific literature addresses these issues.
the Cubism movement, while in theatre, the classic rule of unity
Using a sensory test with different ingredients mixed in all
of time, place and action was amply refuted by writers such as
combinations, the hypothesis that “if the major volatile molecules
Aristotle (Aristotle, 335 BCE).
of two foods are the same, they might taste (and smell) nice when

347
348
the foods are eaten together” was tested and rejected (Kort et al.,
2008). Other scientists interpreted this theory on the basis of cul-
tural heritage (Ahn et al., 2011). By building aroma networks
that linked and compared culinary ingredients, they claimed to
show that there is a very slight tendency to pair ingredients that
share compounds (hence matching the food pairing hypothesis)
in Western cuisine (the North American corpus: USA, Canada),
but that the effect was opposite (i.e., a slight tendency to avoid
ingredients sharing compounds) in Eastern cuisine (East Asian
corpus: Korea, China, Japan)!
This “anti-​ pairing” effect was confirmed in another study
addressing different regional cuisines in India; again, the authors
showed that each regional cuisine follows a negative food pairing
pattern. The greater the extent of flavour sharing between two
ingredients, the less their co-​occurrence in that cuisine (Jain
et al., 2015). This clearly emphasizes the fact that, if there is any
“rule”, it is cultural rather than physiological (de Klepper, 2011;
Perkel, 2012) and also influenced by the culinary educational
level (Traynor et al., 2020).
In a more “historical” but also methodological perspective,
Varshney et al. (2013) tested the food pairing hypothesis on a
collection of recipes from Medieval Europe. Although they did
indeed find some pairing tendency (incidentally, stronger in the
medieval period than in modern times) with a first dataset, they
also showed that this result depended on methodological choices,
in particular the quality of the raw data and of the data prepar-
ation. Hence, using a different (more complete) dataset drove
them to the quite confusing opposite conclusion, that ingredients
with shared compounds are not over-​ represented in recipes
(Varshney et al., 2013).
Some studies worked the opposite way, through addressing
the effectiveness of the theory in creating a new dish (Öztürk
and Zeyrekçe, 2019; Traynor et al., 2013), but nothing can be
concluded from so few trials.
Beyond artistic issues that might disrupt food pairing “rules”,
there is a good chance that this pairing is also influenced by nutri-
tional balance issues (sometimes referred to as “food combining”)
as a whole, as well as more particular preservative effects of some
ingredients, like spices (Ohtsubo, 2009).
Also, it should be noted that the pairing issue also concerns asso-
ciation of any food with a particular product, like olive oil (Cerretani
et al., 2007) or, even more importantly, with drinks; food and wine
pairing is quite a big issue in the scientific and gastronomic litera-
ture, since both the solid and the liquid should match key elements
(texture, flavours and trigeminal sensations) along with traditions
and other local/​cultural aspects (Harrington, 2006). But considering
that (1) both wine and dish are interacting with each other (the
wine changes the taste of the dish, and the dish changes the taste
of the wine), (2) the taste of wine evolves during the meal (due to
oxidation, change in temperature, etc.) and (3) above all, taste is a
personal matter, one can easily figure out how attempts to “ration-
alize” food/​wine pairing are quite desperate.
Conclusion
Although it is always interesting to “play” with big culinary
datasets and use systems and networks analysis tools, statistics or
Hervé This vo Kientza, Christophe Lavelle
even non-​equilibrium dynamics models (Kinouchi et al., 2008)
to extract some information from these data, it should always be
kept in mind that cooking contains artistic motivations, to which
it is quite hazardous to attribute universal “rules”. Also, there is
no need to look for the mechanisms of a phenomenon until this
phenomenon is demonstrated to exist. So, before trying to explain
the hypothetical mechanisms behind food pairing, one should
demonstrate that there are indeed some food pairing rules, which
is not clear so far!
Last but not least: when discussing a so-​ called “scientific
theory”, one has to understand whether the sciences in question
are natural sciences (physics, chemistry, biology and physiology)
or sciences of humans and societies. The art and science (if any)
of food pairing probably pertain to both.
REFERENCES
Ahn YY, Ahnert SE, Bagrow JP, Barabási AL. 2011. Flavor network
and the principles of food pairing. Science Reports, 1, 196.
Aristotle. 335 BCE. Poetics, https://​en.wikisource.org/​wiki/​The_​
Poetics_​translated_​by_​S._​H._​Butcher/​1.
Cerratani L, Biasini G, Bonoli-​ Carbognin M, Bendini A. 2007.
Harmony of virgin olive oil and food pairing: a methodological
proposal. Journal of Sensory Studies, 22, 403–​416.
de Klepper M. 2011. Food pairing theory: a European fad.
Gastronomica, 11(4), 55–​58.
Harrington RJ. 2006. The wine and food pairing process. Journal of
Culinary Science & Technology, 4(1), 101–​112.
Jain ANKR, Bagler G. 2015. Analysis of food pairing in regional
cuisines of India. PLoS ONE, 10(10), e0139539.
Kinouchi O, Diez-​Garcia RW, Holanda AJ, Zambianchi P, Roque
AC. 2008. The non-​equilibrium nature of culinary evolution.
New Journal of Physics, 10, 073020.
Kort M, Nijssen B, van Ingen-​Visscher K, Donders J. 2008. Food
pairing from the perspective of the “volatile compounds
in food” database. In Proceedings of Expression of
Multidisciplinary Flavor Science: Research Highlights from
the 12th Weurman Symposium, edited by I Blank, M Wüst, C
Yeretzian, 589–​592.
Ohtsubo Y. 2009. Adaptive ingredients against food spoilage in
Japanese cuisine. International Journal of Food Science and
Nutrition, 60(8), 677–​687.
Öztürk B, Zeyrekçe B. 2019. Utilization of flavor network analysis
for a new recipe in Marmara region. GIDA, 44(3), 396–​408.
Perkel JM. 2012.The new molecular gastronomy, or, a gustatory tour
of network analysis. Biotechniques, 53(1), 19–​22.
The Food Pairing Company. 2020. www.foodpairing.com.
This H. 2010. Cooking, the Quintessential Art, Columbia University
Press, New York.
Traynor M, Burke R, O’Sullivan M, Hannon J, Barry-​Ryan C. 2013.
Sensory and chemical interactions of food pairings (basmati
rice, bacon and extra virgin olive oil) with banana. Food
Research International, 54(1), 569–​577.
Traynor M, Moreo A, Cain L, Burke R, Barry-​Ryan C. 2020.
Exploring attitudes and reactions to unfamiliar food
pairings: an examination of the underlying motivations
and the impact of culinary education. Journal of Culinary
Science & Technology (in press). https://​ doi.org/​
10.1080/​
15428052.2020.1732253
Varshney KR, Varshney LR, Myers D. 2013. Flavor pairing in medi-
eval European cuisine: a study in cooking with dirty data.
ArXiv:1307.7982v1.
Freeze-​Drying

Yrjö H. Roos
University College Cork, Ireland

Freeze-​drying, also known as lyophilization, commonly refers
to freezing of water in a substance followed by subsequent sub-
limation of ice crystals in the frozen material (ASHRAE, 2018).
As water removal occurs at a low temperature (commonly
below −20 °C) in this technique, freeze-​ drying often retains
heat-​sensitive substances in a functional and biologically active
state. The use of freeze-​drying is common in the pharmaceut-
ical and food industries. Recently, small freeze-​drying equipment
has become available for use in home kitchens as well as food
outlets and restaurants. However, freeze-​drying requires a good
understanding of the physical chemistry of water and water
interactions in food materials at low temperatures.
Freeze-​drying occurs in nature, and partial freeze-​drying has
been used by many cultures. For example, freezing of potatoes
followed by mechanical and refreezing treatments was used by
Incas at the high altitudes of the Andes to produce Chuño, a dried
potato with a shelf-life of years (Romero, 2016). Freeze-​drying
of fish has been common in Arctic areas, and freeze-​dried cod
was transported long distances from Norway by the Vikings (Star
et al., 2017).
Altmann (1890) described freeze-​drying as a method to pre-
pare biological samples for laboratory studies, while the modern
industrial freeze-​drying technology emerged from laboratory-​
scale freeze-​drying reported by d’Arsonval and Bordas (1906).
Blood plasma was freeze-​ dried in large quantities during
the Second World War (Greaves, 1941), and Flosdorf (1945)
presented a comprehensive description of the early years
of industrial freeze-​ drying and its suitability for drying of
pharmaceuticals and foods. General Foods in 1954 started the
development of freeze-​dried instant coffee and launched Maxim
freeze-​dried instant coffee in 1964 (Copulsky, 1976). Nestlé
introduced another freeze-​dried instant coffee, Taster’s Choice,
in the United State in 1966. These international products were
produced in high volumes and introduced freeze-​dried foods to a
wide range of consumers.
Freeze-​drying has become a well-​established unit operation
in the pharmaceutical and food industries. Low dehydration
temperatures ensure the preservation of active substances of
pharmaceutical materials, the retention of viability of micro-
bial cultures as well as high quality of reconstituted materials.
In the food industry, freeze-​ drying has significant benefits,
providing years of shelf-life combined with little change in
product quality and high retention of flavour, including odorant
compounds. Typical freeze-​dried foods include berries, fruits,
meat, vegetables, and many prepared and formulated foods.
Interestingly, Oregon Freeze-​ Dry manufactures a wide range
of freeze-​dried foods and meals. Its Mountain House branded
meals, which are ready to eat by just adding water, have been
on the market since 1969. Some foods manufactured by Oregon
Freeze-​Dry come with >30 years shelf-​life guarantee based on
real-​time, long-​term quality records.
The overview in this chapter provides a description of the gen-
eral physicochemical and engineering principles of freeze-​drying,
provides underpinning information on the freezing of water, and
introduces typical foods available for industrial and consumer
uses as ingredients and ready-​to-​use or ready-​to-​eat foods.
The Principle of Freeze-​Drying
Three ubiquitous phases of materials in food are the crystalline
solid, liquid and gaseous states. A substance can normally exist
in two of the three states, although all three states can exist at
equilibrium at the triple point (Figure 51.1). Most consumers are
familiar with dry ice, i.e., solid CO2, which cannot melt to a liquid
at normal atmospheric pressure. The concept of freeze-​drying
utilizes the same principle. As dry ice can be “freeze-​dried” just
by waiting until its sublimation (evaporation directly from solid to
gas) is complete at normal ambient conditions, the corresponding
sublimation of water-​ice can be completed at conditions where
ice cannot melt to water. Such conditions can be found on the
phase diagram of water shown in Figure 51.1.
The phase diagram of water (Figure 51.1) shows that freeze-​
drying is possible when the ice temperature is lower than tem-
perature value of the triple point, i.e., the pressure–​temperature
point at which all phases (solid, liquid and gas) may co-​exist at
equilibrium. Freeze-​drying at normal atmospheric pressure would
be possible at ambient temperatures below 0 °C. Normal freeze-​
dryers, however, use low-​pressure conditions to keep water in the
solid ice state during sublimation.
Freeze-​Drying Equipment
The general principle of freeze-​drying is used in commercial
freeze-​dryers; that is, the need to maintain drying conditions such

349
350
FIGURE 51.1 Freeze-​drying requires pressure–​temperature conditions that support only solid and vapour states of water. Such conditions are shown as those
where the pressure is below 6.1 mbar or vapour pressure of ice <6.1 mbar (yellow shading). The vapour pressure of water in foods is lowered by solutes, and
melting of ice in foods occurs at low temperatures, as shown by the dotted line (red). Typical food freeze-​drying requires sublimation conditions, as defined by
the lower-​temperature shaded (blue) area.
FIGURE 51.2 Schematic operation principle of freeze-​drying equipment.
A frozen material is loaded on product shelves, which are heated during
drying to provide the heat of sublimation. Vacuum equipment is required to
lower pressure to <6.1 mbar (triple point) and to control ice temperature and
avoid ice melting during drying. Water sublimated from ice is transformed
back to solid ice on a refrigerated condenser surface.
that sublimation of ice occurs without ice melting. A schematic
principle of equipment required for successful freeze-​drying is
shown in Figure 51.2.
Although atmospheric freeze-​drying is possible, most existing
freeze-​dryers operate using the vacuum freeze-​drying principle,
whereby ice temperature can be controlled by pressure. In other
words, the pressure surrounding ice determines its temperature
according to the phase boundary relationship of pressure against
temperature (Figure 51.1). The drying chamber pressure, as shown
in Figure 51.2, is controlled by the vacuum pump equipment,
Yrjö H. Roos
while the refrigeration unit is used to maintain the condensing
surface at a lower temperature than that of ice sublimation.
Sublimation and Condensation in
Freeze-​Drying
Freeze-​drying differs greatly from common air drying, as there is
no drying medium for convective heat transfer. At a low drying
pressure and temperature well below the triple point of water, a
large volume of water vapour must be removed from the drying
material and condensed as ice within the dryer. For example, at
−40 °C, 1 kg of water has a vapour pressure of 0.1 mbar, and
water vapour has a volume of 11,670 L, while at 25 °C, 1 kg
water occupies only 1370 L. It may also be noted that the volume
of mass transfer in freeze-​drying is enormous and varies greatly
with sublimation temperature.
As shown in Figure 51.3, freeze-​drying as a unit operation is
simply based on the control of sublimation of ice at a temperature
below the ice melting temperature and condensation of ice from
the vapour state onto a cold surface.
Freezing of Water in Foods
Although pure water crystallizes as ice and melts at 0 °C, freezing
and melting of water in food and biological materials occur very
differently. The differences in food freezing and melting prop-
erties cause significant challenges to freeze-​ drying. Firstly,
water in foods crystallizes as pure ice, which results in freeze-​
concentration of dissolved substances. Secondly, the freezing and
melting temperature of ice depends on the solute concentration.
That is, the higher the level of freeze-​concentration, the lower the
temperature for freezing and ice melting.
Freeze-Drying
FIGURE 51.3 Sublimation and water vapour flow in freeze-​drying. The vapour pressure value of ice in the drying chamber, pice, is the same as that of
the drying chamber pressure, p. The condenser is refrigerated and cooled to a temperature at which the condenser surface water vapour pressure is pc. At
temperatures Tice > Tc, also pice > pc, and sublimation results in water vapour flow towards the condenser, where water vapour condenses as solid ice.
Raoult’s Law
Raoult’s law provides a fundamental relationship for water vapour
pressure and solute concentration for a dilute solution. Solutes
typically interact with water molecules by hydrogen bonding and
electrostatic forces. Such interactions between liquid water and
solute molecules and ions decrease water vapour pressure, pw, as
compared with that of pure water, p0, at the same temperature.
The resultant change in water vapour pressure depends on the
number of interactions and therefore on the number of interacting
molecules or particles. The number of molecules and ions can be
derived from mole fractions of interacting particles, which gives
Raoult’s Law (51.1).
pw = xp0 (51.1)
where x is the mole fraction of water.
The mole fraction of water, x, is given by equation (51.2).
 H2O 
x= (51.2)
 H2O  + [ S ]
where [H2O] and [S] are molar concentrations of water and
solutes, respectively.
Raoult’s law is derived from chemical potential, m, for the
liquid and vapour states. That is, at equilibrium, both phases
have the same chemical potential. Freezing of water in a solu-
tion requires that ice has a lower chemical potential than water.
At equilibrium, i.e., at the freezing or melting temperature, the
chemical potential of ice, water and water vapour must be the
same. Accordingly, the vapour pressure of ice and water in a solu-
tion must be equal at equilibrium freezing temperature.
Freeze-​drying requires the frozen state of water for sublim-
ation, while the capability of numerous solutes to depress the
freezing temperature of water is significant. Raoult’s law shows
that small solutes, which may have a significant mole fraction
in a food, may significantly depress the freezing temperature.
Furthermore, freeze-​concentration of solutes increases the mole
fraction of solutes in an unfrozen fluid. Successful freeze-​drying
351
of foods with small solutes, such as salts and sugars, requires
low sublimation temperatures, or freeze-​drying may not be pos-
sible in the chosen conditions because of excess unfrozen water.
Materials that may not be freeze-​dried or dried in general include
fruit juices, honey and formulated liquids with a high content of
monosaccharide sugars.
Ice Formation and Melting
As described by Raoult’s law, solutes in water decrease the equi-
librium freezing temperature. In other words, ice forms or melts
in the presence of solutes in water at a lower temperature than is
known for pure water (0 °C). This freezing point depression can
be explained by an equilibrium of the chemical potential of ice
and unfrozen water that occurs at a lower temperature. However,
the freezing or melting temperature is the highest temperature
at which ice crystals dissolve, and freeze-​drying requires signifi-
cantly lower temperatures.
The freezing properties of water in a solution or a food can
be explained by using a state diagram (Roos and Drusch, 2015).
Hydrophilic food constituent molecules form hydrogen bonds
with water molecules, and the level of such constituents has an
impact on the amount of water that remains unfrozen in a food at
maximum freeze-​concentration. Furthermore, crystallization of
components other than water is not likely during food freezing.
As a result of freeze-​concentration, a highly freeze-​concentrated
unfrozen fluid containing solutes and unfrozen water shows
increasing viscosity and stiffness with decreasing temperature.
Such an unfrozen phase vitrifies, i.e., solidifies and forms a non-​
crystalline solid glass-​ like structure at a solute-​ specific tem-
perature. This often occurs at conditions supporting maximum
freeze-​concentration. The water content of a maximally freeze-​
concentrated unfrozen phase may be given the symbol Wg′ and
the corresponding solute concentration is shown as Cg′.
Solute–​unfrozen water vitrification is important in freeze-​drying.
A solid structure of the unfrozen phase is required to carry the mass
of the dehydrated structure. Loss of structure in freeze-​drying is
known as ‘collapse’ and is caused by viscous flow of the unfrozen
phase as ice crystals disappear and pores left by ice are reduced in
352
FIGURE 51.4 State diagram of sucrose. The equilibrium ice melting temperature, Tm, decreases with increasing sucrose concentration. The maximum
amount of ice may form at Tm′, while cooling to lower temperatures results in vitrification (glass formation) of the unfrozen maximally freeze-​concentrated
fluid surrounding ice crystals. Maximally freeze-​concentrated solutes show onset of glass transition at Tg′.
size and filled by unfrozen fluid (Bellows and King, 1973; To and
Flink, 1978). The mechanism of collapse was described by Bellows
and King (1973) and was shown to lead to shrinkage and poor
dehydration. Bellows and King (1973) described collapse as a phe-
nomenon whereby a solid, glassy state of the freeze-​concentrated
amorphous phase surrounding ice crystals undergoes a glass transi-
tion to the liquid, viscous fluid state. As a result of this glass transi-
tion, the unfrozen phase shows significantly reduced viscosity and
flow. While the glassy state can support its own mass and maintain
the frozen state volume, a material above the glass transition can no
longer support its weight and shows collapse.
Freeze-​drying collapse temperatures agree with the onset of ice
melting temperatures in maximally freeze-​concentrated, binary
solutions of solutes in water. The state of solutes and water at
various concentrations and temperatures is commonly described
using state diagrams. State diagrams have been published for
numerous solutes and foods, although that of sucrose is most
frequently reviewed (Roos, 2010; Buera et al., 2011; Roos
and Drusch, 2015). The state diagram of sucrose is shown in
Figure 51.4. The sucrose state diagram shows that the glass tran-
sition in a maximally freeze-​concentrated solution occurs at Tg′
of −46 °C and ice melting occurs above Tm′ of −34 °C. The Tm′
coincides with or is slightly lower than collapse temperatures
reported for various solutes typical of foods (Bellows and
King, 1973; Roos,1993). Corresponding state diagrams can be
established for other binary solutions of solute and water. State
diagrams have also been published for a number of foods, such
as fruits, to describe the glass transition dependence of their pri-
mary components on water content (Sablani et al., 2010). Such
diagrams are useful for freeze-​drying, as ice melting properties
and required freeze-​drying conditions can be identified on the
basis of data shown for Tg′ and Tm′.
Yrjö H. Roos
Freeze-​Drying of Food Materials
Common freeze-​dried foods include almost any food that can
be frozen and enjoyed as a snack, used as an ingredient in a
dry formulation, or rewarmed as a meal after rapid reconstitu-
tion. Obviously, the compositions of such products vary substan-
tially. Typically, foods with a high content of monosaccharides,
such as berries and fruits, are more difficult to freeze-​dry than
foods composed mainly of large protein molecules (meat and
seafood) or starch and polysaccharides (vegetables and cereals).
However, food products also contain cellular structures, which
improve freeze-​drying properties. On the other hand, a food
intended for freeze-​drying must include a sufficient quantity of
water to form ice crystals. After freezing, the unfrozen solids
must retain a solid state to support the dehydrated structure in
order to avoid shrinkage. Voids left by ice crystals also need to
serve as channels for vapour flow (mass transfer) during further
dehydration.
Freeze-​drying is also an option for dehydration of formulated
foods. A traditional example of a space food is freeze-​dried
ice cream, which is available at the Kennedy Space Center
and also sold by various freeze-​dried food manufacturers. The
main challenge in freeze-​drying sugar-​containing desserts is ice
melting at low temperatures. For example, freeze-​drying a fruit
juice may not be possible because of the high content of small
sugars, fructose and glucose. As described by Raoult’s law, these
sugars lower the freezing temperature of water significantly, and
successful ice sublimation may need to be carried out at an ice
temperature below −50 °C. Such a low freeze-​drying temperature
may not be achievable when commonly available equipment is in
use, and it also increases freeze-​drying time. Therefore, freeze-​
drying such food products may not be economically feasible and
Freeze-Drying 353

FIGURE 51.5 Freeze-​drying occurs at conditions where no liquid state of water should exist (pressure <6.1 mbar; triple point). Heat transfer to ice leads to
sublimation. Condensation of the vapour state of water requires cooling and crystallization and heat removal using a refrigeration system. The change in heat
content, ΔH, between the solid and vapour states below the triple point is shown for the sublimation of ice within a food and subsequent condensation of the
vapour to ice at the condenser surface.

Heated Shelf Heated Shelf Heat transfer by

radiation
Mass transfer
Interface
through dry layer - sublimation point
Interface - pressure determines ice T
- sublimation point
- pressure determines ice T
Heat transfer by
conduction
Heat transfer by Heated Shelf Mass transfer
Heated Shelf conduction through dry layer

(a) (b)

FIGURE 51.6 Heat transfer modes of freeze-​drying. Conductive heat transfer (a) uses direct heat transfer from a heated shelf through the ice layer to the
sublimation interface. Heat transfer by radiation (b) uses trays loaded between heated shelves. Heat is absorbed and conducted through a dry product layer to
enhance sublimation at the ice interface.

may require reformulation to alter ice melting properties as well
as end product stability.
Heat and Mass Transfer
Freeze-​drying is direct transformation of ice to water vapour.
The change in state from solid ice to gaseous vapour requires
heat. Thus, a heat input of >2838 kJ/​kg of ice is required at
temperatures below 0 °C for the transformation. As sublimation
occurs at much lower temperatures, the amount of heat of sub-
limation increases. The requirements for heating and cooling in
vacuum freeze-​drying are explained in Figure 51.5.
As shown in Figures 51.1 and 51.5, water below the triple
point can exist only as a solid (ice) or vapour. The temperature
of ice is always <0 °C, while ice melting may occur at a lower
temperature depending on solutes. Heat transfer to the ice within
a product is required and commonly supplied using contact and
radiation heat transfer. The water vapour may show an increase
in temperature between ice and condenser, but all vapour needs to
crystallize as ice on the condenser as drying progresses.
Heat transfer for ice sublimation is often by conduction dir-
ectly to a frozen material from a heated plate on which a product
tray is loaded prior to drying. Large freeze-​drying equipment
in food applications may also use heat radiation as the primary
heat transfer mode. The difference in the supply of the heat of
sublimation between contact and radiation heating is shown in
Figure 51.6. The sharp interface between frozen and dried areas
of a material during freeze-​drying is also noteworthy.
Heat supplied by direct contact to a frozen material is rap-
idly transferred from a heated shelf through ice to the sublim-
ation interface because of the large thermal conductivity of ice.
Sublimation occurs at the sharp ice interface, and water vapour
354 Yrjö H. Roos

released from ice is transferred through the dried material to exhibits a very low water activity, aw, corresponding to almost
the outer surface and then towards condensation at the con- zero water content within the solids.
denser surface. The main limitation for the freeze-​drying rate
is mass transfer through the dried material. Freeze-​drying may
be accelerated by increasing heat input. However, the mass
Challenges
transfer must be sufficient to avoid extensive sublimation and
subsequent increase of pressure at the sublimation interface. An Freeze-​ drying requires a good understanding of food materials
increase in pressure increases ice temperature and may lead to science, freezing properties of water in foods, and stability of freeze-​
ice melting. dried structures in dehydration and post-​drying storage. Freeze-​
Freeze-​drying rates in radiation heat transfer are limited by drying of foods often lasts 24 hours or longer depending on drying
heat transfer through the already dry product layer. The dry equipment and materials. Drying of small particles, such as 5 mm
product has high porosity and poor heat transfer characteristics. cubes or granulated coffee concentrates, may provide an optimal
However, rapid initial drying rates can be achieved at the begin- size, while an increase in particle dimensions leads to an exponen-
ning of freeze-​drying as sublimation occurs on the surface of tial increase of drying time. Attempts have been made to accelerate
the drying material and the temperature of the material remains freeze-​drying by using agitation, microwave heating, puncturing or
low. The temperature at the beginning of drying is controlled by other means to improve heat and mass transfer during drying.
the freeze-​drying chamber pressure, according to the pressure–​
temperature relationship in Figure 51.1. Packaging
Freeze-​dried materials have a large volume with little mass, i.e.,
Freeze-​Drying Temperature a very low density. This low density results from the retention
Freeze-​drying conditions are controlled by the pressure–​ of the frozen state volume and the solute concentration on the
temperature relationships of Figure 51.1, which allow the use outside of ice crystals. Removal of ice crystals leaves a highly
of a high temperature to enhance sublimation (Figure 51.7). The porous structure with an enormously large surface area, as shown
temperature of ice in a vacuum chamber follows the chamber in Figure 51.8.
pressure. Materials requiring low sublimation temperatures The high porosity and relatively thin non-​ crystalline
because of ice melting and collapse require careful control of membranes of food solids provide a large surface area for sorption
heat input and a low condenser temperature. The main require- of gases and water. At the end of vacuum freeze-​
d rying, air or an
ment for successful freeze-​drying is that the temperature of the inert gas, such as nitrogen, is used for pressure release. The use
sublimating ice is lower than the temperature of the condensing of an inert gas delays exposure to oxygen and water, which are
surface (Figure 51.3). present in air and could otherwise enhance oxidation and water
An ideal freeze-​drying operation can hold the ice temperature sorption of the solids. The non-​crystalline freeze-​dried solids
at a required value throughout the sublimation stage of drying, as exhibit very high hygroscopicity and softening of the material
shown in Figure 51.7. At the end of freeze-​drying, the temperature (water plasticization). Various highly sensitive materials, such
of the heated shelves needs to be decreased to avoid overheating as pharmaceuticals, may be freeze-​dried in glass vials. The vials
of the dried material. At the end of drying, the material contains can be closed under vacuum inside the freeze-​dryer using rubber
no ice, and most unfrozen water is desorbed from the porous septa, while the material is left under vacuum at pressure release.
solids. Drying uniformity at the end of freeze-​drying is essential, Most freeze-​ dried foods are packaged immediately after
as any ice within a dry material leads to rapid local deterioration freeze-​d rying in hermetically sealed aluminium laminate pack-
of the otherwise dried food. A properly freeze-​dried food material aging, glass jars or tin cans. Such packaging ensures protection of

FIGURE 51.7 Differences of temperature profiles for freeze-​drying using conduction (left) or radiation (right) heat transfer. Freeze-​drying requires pre-​
freezing of material (externally or within the dryer), lowering of pressure (a), sublimation controlled by heat transfer according to heating of shelves (b–​d),
and final drying to ensure drying uniformity (e).
Freeze-Drying
log N
(a)
FIGURE 51.8 (a) Scanning electron microscopic image of freeze-​dried maltodextrin (Maltrin M100, DE10). (b) A 20% solution was frozen at −20 °C
before freeze-​drying. The relationships found for number of pores (N), pore diameter and wall thickness showed that increasing size of pores led to a linearly
increasing wall thickness.
(From Harnkarnsujarit et al., 2012)
FIGURE 51.9 Energy costs occurring in freeze-​drying of foods.
(Ratti, 2001)
the freeze-​dried solids against surrounding atmosphere and water
uptake. The shelf-life of freeze-​dried foods in thick Al laminate
packages or tin cans is several years and may exceed 30 years,
according to Oregon Freeze-​Dry (525 25th Ave SW, Albany, OR
97322, USA). However, typical shelf-​life may be dated as two to
five years from manufacturing.
Manufacturing Costs
Freeze-​drying includes several unit operations and a process using
equipment with a high investment cost. Freeze-​drying requires
pre-​freezing of materials as sufficiently small particles to enhance
heat and mass transfer so as to obtain rapid dehydration during
sublimation. The requirement to supply heat of sublimation and
concurrent refrigeration of water vapour leads to relatively high
energy requirements. The relative costs of a freeze-​drying process
divided into various unit operations are described in Figure 51.9.
Freeze-​Dried Foods
The benefits of freeze-​drying as compared with other dehydra-
tion technologies are numerous. The key benefits of high flavour
355
10 30
9
25
8
Wall thickness (µm)
7 20
6
5 15
4
10
3
2 5
1
0 0
0 100 200 300 400
Pore diameter (µm)
(b)
and colour retention are due to its non-​thermal nature; a material
for freeze-​drying is simply frozen and maintained in the frozen
state to reduce shrinkage until all ice and unfrozen water have
been removed. The dry material is a non-​crystalline, glass-​like
solid, which supports its own mass against gravity. The materials
are removed from the dryer and packaged immediately to avoid
water uptake from surrounding air. In general, all freeze-​dried
materials are highly hygroscopic and sensitive to oxygen and
light. Freeze-​dried foods need to be packaged in glass jars, as is
typical of freeze-​dried coffee, or metal laminate film packaging.
Such materials provide a strong moisture, oxygen and often light
barrier, extending product shelf-life to several years or even
decades.
Freeze-​Dried Coffee
Freeze-​dried coffee is one of the most common freeze-​dried
foods on the market. Manufacturing of freeze-​dried instant coffee
includes coffee extraction using hot water in extraction columns.
To ensure high product quality, coffee extract may be freeze-​
concentrated to remove ice crystals from a partially frozen fluid
followed by freezing, granulation and freeze-​drying of the con-
centrate. Freeze-​dried coffees are well known for their superior
flavour and solubility in comparison to other dried instant coffee
products.
Freeze-​Dried Berries, Fruits and Vegetables
Freeze-​drying of berries, fruits and vegetables is probably one
of the most obvious uses of freeze-​drying to produce ready-​to-​
eat fruit snacks or ingredients for the food industry. Well-​known
examples are whole or sliced strawberries, blueberries and
mangos. Freeze-​dried fruit are also sold with chocolate or other
hydrophobic coatings. Freeze-​dried berries and fruit have very
low water content, and they provide a flavourful choice for use
in chocolate manufacturing as crispy inclusions. Freeze-​dried
ingredients are also useful in breakfast cereals and other ready-​to-​
eat products or dessert mixes commonly found in supermarkets.
In general, freeze-​dried products have very low density. Solids in
356
plant materials present typically only around 10% of fresh mass
or less. In other words, dehydration requires >10 kg raw material
for each 1 kg of dry product.
Plant foods may also be powdered after freeze-​drying. A fruit
powder has reduced volume and provides an attractive and fla-
vourful food ingredient. Freeze-​dried plant powders are avail-
able from most freeze-​dried food manufacturers, and they can
be used as a decorative and flavourful powder in meals, desserts,
decorations and for other uses.
Herbs and Spices
Various aromatic herbs and spices retain their natural flavours
and aromas after freeze-​drying. High-​quality herbs and spices,
such as basil, green peppercorns, parsley and mint, are available
from numerous suppliers. Typically, freeze-​dried choices have a
higher cost than air-​dried materials, but they are superior in fla-
vour and rehydration characteristics.
Meat and Seafood
Meat and seafood, particularly low-​ fat meat products and
shrimp, are commonly freeze-​ dried. These materials contain
mainly proteins, and freeze-​drying can be carried out at higher
ice temperatures than freeze-​drying of foods containing sugars.
Freeze-​
dried food ingredient manufacturers produce various
diced or minced meats, poultry and prepared meats, freeze-​dried
egg products and freeze-​dried cheeses.
Ready Meals
Freeze-​dried ready meals, including pasta dishes, rice dishes,
sauces and desserts, ready to eat after reconstitution are available
from several manufacturers. These products have a long shelf-life,
and they are popular choices for camping and expeditions, to ensure
the availability of lightweight food where there are no services.
Novelties
Freeze-​drying even at a small scale can provide a means for prep-
aration of interesting and innovative flavours and structures in
food preparation. For example, freeze-​dried berries coated with
various types of chocolate are available in many markets. Dessert
products include freeze-​dried ice cream, smoothies and yoghurts.
Structures similar to those of meringues may be produced with
very different compositions and less sugar. Many innovative
products may not be available for home use, but freeze-​dried
materials combined with other foods are attractive alternatives to
traditional menus.
Besides foods, food supplements and starter cultures can be
preserved using freeze-​drying. Such products have been made
available to food manufacturers as well as individual consumers.
Conclusions
Freeze-​drying provides high-​ quality dehydrated foods, which
also require robust packaging. The shelf-life of freeze-​dried foods
may be as short as minutes on exposure to ambient conditions,
Yrjö H. Roos
while appropriate packaging may extend the shelf-life of the
same food to decades. Sublimation and condensation between the
solid and vapour phases of water, as well as vacuum equipment,
make the manufacturing costs of freeze-​dried foods quite high.
Moreover, the initial solids content of foods such as fruits is low,
and over 90% of their initial mass may need to be removed during
freeze-​
drying. Freeze-​ drying provides excellent opportunities
for food product innovation, particularly when high-​value func-
tional components need to be preserved for long-​time stability.
Furthermore, the bioactivity of food components may be retained
and delivered at the time of consumption. These opportunities are
possible at various scales, from food service to large-​scale food
ingredient and food manufacturers.
REFERENCES
Altmann R. 1890. Die Elementarorganismen und ihre Beziehungen
zu den Zellen. Leipzig: Veit.
ASHRAE. 2018. Refrigeration Handbook. Atlanta (USA): ASHRAE.
Bellows RJ, King CJ. 1973. Product collapse during freeze drying of
liquid foods. AIChE Symposium Series, 69(132), 33–​41.
Buera MP, Roos Y, Levine H, Slade L, Corti HR, Reid D, Auffert T,
Angell A. 2011. State diagrams for improving processing and
storage of foods, biological materials, and pharmaceuticals.
Pure and Applied Chemistry, 83, 1567–​1617.
Copulsky W. 1976. Cannibalism in marketplace. Journal of
Marketing, 40(4), 103–​105.
D’Arsonval A, Bordas F. 1906. De la distillation et de la dessiccation
dans le vide à l’aide des basses températures. Comptes rendus
de l’Académie des Sciences, 143, 567–​570.
Flosdorf EW. 1945. Advances in drying by sublimation. Blood
plasma, penicillin, foods. Journal of Chemical Education,
22(10), 470–​480.
Greaves RIN. 1941. The freezing of human serum and plasma in
Medical Research Council transfusion bottles, before drying
by sublimation from the frozen state. Epidemiology and
Infection, 41(5–​6), 489–​495.
Harnkarnsujarit N, Charoenrein S, Roos YH. 2012. Microstructure
formation of maltodextrin and sugar matrices in freeze-​dried
systems. Carbohydrate Polymers, 88, 734–​742.
Ratti C. 2001. Hot air and freeze-​drying of high-​value foods: a
review. Journal of Food Engineering, 49, 311–​319.
Romero S. 2016. A space-​age food product cultivated by the Incas.
New York Times, 11 August 2016, A5.
Roos Y. 1993. Melting and glass transitions of low molecular weight
carbohydrates. Carbohydrate Research, 238, 39–​48.
Roos YH. 2010. Glass transition temperature and its relevance in food
processing. Annual Review of Food Science and Technology, 1,
469–​496.
Roos YH, Drusch S. 2015. Phase Transitions in Foods, 2nd ed.
New York: Academic Press.
Sablani SS, Syamaladevi R, Swanson BG. 2010. A review of methods,
data and applications of state diagrams of food systems. Food
Engineering Review, 2, 168–​203.
Star B, Boessenkool S, Gondek AT, Nikulina EA, Hufthammer AK,
Pampoulie C, Knutsen H, André C, Nistelberger HM, Dierking
J, Petereit C, Heinrich D, Jakobsen KS, Stenseth VC, Jentoft
S, Barrett JH. 2017. Ancient DNA reveals the Arctic origin of
Viking Age cod from Haithabu, Germany. Proceedings of the
National Academy of Sciences, 114(34), 9152–​9157.
To EC, Flink JM. 1978. ‘Collapse’, a structural transition in
freeze dried carbohydrates II. Effect of solute composition.
International Journal of Food Science & Technology, 13,
567–​581.
Foams: Pickering Edible Oil Foam –​Toward New Food Products
Anne-​Laure Fameau
Research & Innovation, International Physical-​Chemistry Department, L’Oréal, Saint-​Ouen 93400, France
Introduction
Foams are widely applied in many industries, especially the food
industry (Cantat et al., 2013). Liquid foams are complex colloidal
systems with gas bubbles dispersed in a continuous liquid phase
containing surface-​active agents such as surfactants, polymers,
proteins or particles adsorbed at the gas/​liquid interface to form
and stabilize the bubbles (Hill and Eastoe, 2017). Two different
categories of liquid foams exist depending on the nature of the
continuous phase: aqueous and non-​ aqueous (Friberg, 2010;
Shrestha and Aramaki, 2012; Fameau, 2017). The most common
foams are aqueous foams. Products and research on non-​aqueous
foams have been very limited up to now in comparison to aqueous
foams, especially for edible applications (Heymans et al.,
2017). However, these non-​aqueous foam systems have enor-
mous potential (Fameau and Saint-​Jalmes, 2017). Fundamental
research in this area has been growing tremendously in the last
five years (Heymans et al., 2017; Binks, 2017). The industrial
potential of edible oil foams is high, since these systems could
have different potential benefits. First of all, oil foams could be
used to create food products with reduced fat content in combin-
ation with new textures (Gunes et al., 2017). The incorporation of
air bubbles in an oily continuous phase could provide a different
mouthfeel and appearance to oil products (Heymans et al., 2017).
Moreover, they could enhance flavour delivery by the inclusion of
air containing odorant compounds.
In the area of molecular gastronomy, edible oil foams are
already used, and these edible oil foams were produced before
the increased interest by researchers in these new types of foam
systems (Patel, 2017). A simple formulation to produce edible oil
foams based on olive oil and glycerin flakes has been described
on the website “MolecularRecipes.com” since 2012. However,
the formulation rules were not described in the literature at this
time. This chapter aims to describe how to produce edible oil
foams from vegetable oils based on Ramsden–​Pickering stabil-
ization by crystal particles.
Formation and Properties of Pickering Foams
Since the beginning of the 20th century, scientists have known
about the ability of some particles to impart greater stability to
foams and emulsions than molecular surfactant stabilizers. When
emulsions are stabilized by particles, they are commonly called
Ramsden emulsions or Pickering emulsions, after the pioneering
researchers in this field. Following this trend for emulsions,
foams stabilized by particles are referred to as Pickering foams.
Foam Stabilization by Particles
Liquids can form foams due to the presence of surface-​active
materials called foam stabilizers, such as surfactants, polymers,
proteins or particles. Particles are a very different class of foam
stabilizer in comparison to more common surfactants (Lam et al.,
2014). In particular, surfactants adsorb and desorb from interfaces
on a rapid time scale, whereas particles can remain almost irre-
versibly adsorbed at the interface. The driving force of the adsorp-
tion of particles at fluid interfaces is the free energy gained by
losing an area of interface. Particle adsorption is thermodynam-
ically favourable in the case of solid–​liquid–​air systems if the
sum of the solid–​air tension and the solid–​liquid tension is less
than the original liquid–​air tension. The main parameter required
for particles to adsorb is the wettability by both the fluid phases,
quantified by the contact angle (θ) that the particles adopt at the
interface and measured through the liquid phase (Figure 52.1)
(Binks, 2002).
The contact angle is given by the Young equation:
γ solid / air − γ solid / liquid
cos θ = (52.1)
γ liquid / air
Young’s equation suggests that the particle contact angle
could be changed by altering the respective interfacial tensions.
When the liquid phase is oil, the spherical particle is completely
wetted by this liquid when θ is 0°. The result is the dispersion
of particles inside the oil phase (Figure 52.1) (Fameau, 2017).
When θ is 180°, the particle is completely non-​wetted by the
liquid. In these two cases, no air entrainment is observed even
after vigorous mixing, since no particle adsorption occurs at
the air/​liquid surface (Figure 52.1). For intermediate θ values
between 0° and 180°, particles are more wetted by one of the
two phases, liquid or air, and they can adsorb at the interface.
Particles must be partially oleophobic in order to exhibit contact
angles between 0 and 180°. For θ < 90°, particles are partially
oleophobic, and stable oil foams should result after aeration of
357
358
FIGURE 52.1 Schematic of the effect of the increase of both contact angle
and surface tension on the position of the particle, from left to right. On the
left is an oil dispersion obtained with a zero contact angle. In the centre, oil
foams were obtained with intermediate contact angle values below 90°. On
the right, oil powder was obtained with contact angle values above 90°.
the mixture (Figure 52.1). For θ > 90°, particles are extremely
oleophobic, and the inverted case of oil-​in-​air droplets should be
stabilized (Binks et al., 2014).
Instability Mechanisms in Pickering Foams
So-​called liquid foams are formed of gas bubbles dispersed in a
liquid phase. They are thermodynamically metastable systems,
which tend to separate over time into the individual components
of which they are composed. Two main parameters are used to
describe liquid foam: the liquid fraction and the bubble size. The
liquid fraction of the foam corresponds to the volume of liquid
dispersed in the foam volume, ε = Vliquid/​Vfoam. The macroscopic
properties of a foam depend on the liquid fraction and on the
bubble size, which can evolve with time due to the foam destabil-
ization mechanisms (Cantat et al., 2013). Therefore, the control
of the foam stability is a major challenge in foam science and
for the food industry (Rio et al., 2014). The role of the foam sta-
bilizer is to stabilize the foam by slowing down the three main
mechanisms of foam destabilization: drainage, coarsening and
coalescence. The drainage comes from the gravity; the liquid
inside the foams drains rapidly until an equilibrium state is
reached between the gravity and the capillary force. In parallel to
drainage, a foam evolves due to the gas transfer between bubbles.
This phenomenon is called coarsening and leads to an increase of
the mean bubble size with time. Moreover, the film separating the
bubbles can rupture, a phenomenon called coalescence, the result
of which is an increase in bubble size with time.
Solid particles are an efficient foam stabilizer, since they
are adsorbed more or less irreversibly at the interface and they
cannot be displaced from the interface by random fluctuations
(Fameau and Salonen, 2014). The energy ΔG required to remove
a particle from the air–​liquid interface (with surface tension,
Anne-Laure Fameau
γ air − liq) is linked to the contact angle (θ) and the particle size
(r, radius):
∆G = πr 2 γ air − liq (1 ± cos θ )
2
(52.2)
This expression basically tells us that the closer the contact
angle of the particle at the interface to 90 ° and the larger the par-
ticle, the more energetically favourable the adsorption and more
difficult the detachment of the particles (Lam et al., 2014). For
example, for spherical homogeneous particles with a radius of
10 nm at the air/​water interface, ΔG is maximum when θ = 90°
and is about 104 kBT (kB is the Boltzmann constant and T is
the absolute temperature). Thus, ΔG is several orders of mag-
nitude higher than the thermal energy, which leads to irrevers-
ible adsorption of particles at the interface. In comparison, the
detachment energy for a surfactant molecule is only in the order
of kBT. The irreversible attachment of the particles at the interface
and the formation of solid-​like surface layers protect the bubbles
against coalescence and coarsening. Moreover, particles can sta-
bilize foams not only by adsorbing and packing at the interface,
but also by modifying the rheological properties of the liquid
phase (Fameau et al., 2015b). This change of the properties of the
liquid phase inside the foam liquid channels can slow down the
drainage. This is why Pickering foams are very stable foams, in
comparison to common surfactant foams, when the concentration
of particles is high enough for complete surface coverage.
Key Differences between Aqueous Pickering Foams and
Pickering Edible Oil Foams
In aqueous foams, the continuous liquid phase is based on water.
In edible oil foams, the liquid phase is an edible oil. Since both
aqueous and edible oil foams are biphasic dispersions of gas
bubbles dispersed in a continuous liquid phase, all the infor-
mation and concepts described in the literature for aqueous
foam formation and stability are useful (Cantat et al., 2013).
However, specific considerations must be taken into account
when replacing water by edible oil as solvent in the continuous
phase (Blázquez et al., 2014). The main difference between the
two systems comes from the surface tension at the liquid–​gas
interface (Friberg, 2010; Blazquez et al., 2014). Indeed, water is
unique, since this low-​viscosity liquid has a high surface tension
due to extensive hydrogen bonding: the water–​air surface tension
is around 72 mN.m−1 at 25 °C. In the case of edible oils, the sur-
face tension is relatively low, around 20 to 40 mN.m−1. In aqueous
foams, surfactants strongly adsorb at the interface, reducing the
surface tension to lower values ranging from 20 to 40 mN.m−1.
For oil foam systems, the oil already has rather low surface
tension, making the adsorption of hydrocarbon-​based surfactants
energetically unfavourable, in contrast to specific particles with
suitable wettability (Binks et al., 2010; Blázquez et al., 2014).
Moreover, the particle–​particle DLVO and dipole–​dipole repul-
sive forces are insignificant in non-​aqueous foams due to the
low dielectric constants of non-​aqueous phases (Heymans et al.,
2017). The properties of edible oil foam are mainly driven by the
physical properties of the oil phase: polarity, viscosity, density,
Foams: Pickering Edible Oil Foam 359

conductivity, dielectric constant and Hansen solubility parameter chains, such as are found in sunflower oil, high oleic sunflower
values (Fameau, 2017). These properties are directly linked to the oil, soybean oil and rapeseed oil (Fameau, 2017).
composition of the edible oils in terms of triglyceride compos- The main requirement to obtain oil foams from these systems
ition, fatty acids and other components. is the presence of crystalline particles, which is linked to the solu-
In terms of foam stabilizers, aqueous and non-​aqueous foams bility of the edible molecules in the oil (Binks et al., 2016). For
are different. In aqueous foams, various surfactants, polymers, an ideal solution, it is well known that the variation of solubility
proteins and particles are used as foam stabilizers. To stabilize of a chemical substance into a given solvent as a function of the
non-​aqueous foams, only three different foam stabilizer cat- temperature is given by:
egories have been described in the literature: specific surfactants,
solid particles and crystalline particles (Fameau, 2017). The ∆H fus  1 1 
only edible foam stabilizer category is crystalline particles ln ( x ) =  −  (52.3)
R  Tf T 
based on edible molecules such as fatty acids, monoglycerides,
diglycerides, triglycerides and sucrose esters (Shrestha et al.,
2006; Shrestha et al., 2007a; Shrestha et al., 2007b; Brun et al., where x is the mole fraction of solute in the saturated solution at
2015; Fameau et al., 2015a; Patel, 2017; Heymans et al., 2018). absolute temperature T, R is the gas constant, ∆Hfus is the enthalpy
of fusion and Tf is the temperature of fusion.
For example, Binks et al. (2016) used this equation to calcu-
late the variation of myristic acid solubility temperature in high
Formation and Properties of Pickering Edible oleic sunflower oil as a function of myristic acid concentration
Oil Foams (Figure 52.2) (Binks et al., 2016).
In the 1970s, Sanders et al. demonstrated that non-​aqueous foam In the first step, the myristic acid was dissolved in the oil by
formation was possible by using specific hydrocarbon surfactants heating. In the second step, the hot solution was cooled down.
with functional groups (i.e., alcohols, glycols, soaps, acids and Upon cooling, the solution becomes supersaturated. The satur-
amines) dispersed in mineral oil (Sanders, 1970; Pugh, 2016). In ation concentration of the myristic acid corresponds to the con-
the 1980s, Friberg et al. showed that non-​aqueous xylene foams centration at which the addition of more molecules does not
could be produced by using the surfactant triethanolammonium increase the molecular concentration in solution, and the excess
oleate in specific conditions (Binks et al., 2010). When the sur- amount of solute begins to self-​assemble. The supersaturation
factant formed an isotropic liquid phase, no foam was produced. leads to crystal formation occurring by nucleation and growth.
However, at the temperature at which the surfactant became Thus, below the solubility boundary, crystalline particles are pre-
insoluble and formed crystalline particles, foams were obtained. sent inside the oil phase, and above it, the crystalline particles
These preliminary reports highlighted that oil foams can be melt (Binks et al., 2016). Above the solubility boundary, no
obtained under conditions close to the phase-​separation boundary foam can be produced from the molecular solution, since no
when the oil affinity for the surfactant is low. The solubility of crystalline particles are present to stabilize the bubbles and the
long-​chain hydrocarbon molecules with particular functional molecular fatty acid is not surface-​a ctive. Below the solubility
groups (acid, alcohol or amine groups) in the oil phase seems to be boundary, a foam can be obtained. To produce oil foams from
the critical parameter to produce oil foams. Non-​aqueous foams these systems, both the appropriate concentration of myristic
can only be obtained when the hydrocarbon-​based surfactant acid and the temperature to produce the foam need to be chosen.
forms crystalline particles in the non-​aqueous phase, stabilizing The determination of the solubility boundary is the main param-
the gas bubbles. Therefore, the strategy in preparing stable edible eter to determine for these systems before studying the foaming
oil foams is to combine two main components: a vegetable oil properties. Differential scanning calorimetry (DSC) is a useful
for the continuous phase and edible oil-​soluble molecules, which technique to determine both the temperature at which the crystal
crystallize on cooling within the vegetable oil phase. When the formation occurs and the temperature at which crystals dissolve
concentration of the edible oil-​soluble molecule is high, leading to form a molecular solution (Fameau et al., 2015a).
to a high concentration of crystals, the mixture between these two As previously described for solid particles, only specific
components forms a gel called an oleogel in the literature (Rogers particles with the appropriate wettability can lead to the stabil-
et al., 2014; Singh et al., 2017). By the aeration of the mixture, ization of foams. Indeed, in the case of oil foams, the crystalline
the crystalline particles adsorb to the air bubbles, leading to the particles must be partially oleophobic in order to exhibit suitable
formation of edible Pickering oil foams (Fameau, 2017). contact angles and adsorb to the interface. In the case described
previously, the contact angle through the high oleic sunflower
oil by using discs of compressed myristic acid powder has been
Properties and Structure of the Crystalline Particles
measured as around 40° (Binks et al., 2016). This experimental
To obtain edible crystalline particles, commercial mono/​ result shows that the crystalline particles obtained from the myr-
diglyceride surfactants, long-​ chain alcohols, long-​ chain istic acid have the right contact angle to adsorb at the air bubble
carboxylic acids, triglycerides, sucrose esters and sunflower leci- surfaces dispersed in the oil phase (52.1). Similar contact angle
thin can be used. In the examples studied so far in the literature, values have been measured for other crystalline particles, such as
these systems have been mixed with liquid oils at room tempera- those obtained for the system based on monoglycerides and high
ture due to the high proportion of long unsaturated fatty acid oleic sunflower oil (Gunes et al., 2017). To measure the contact
360 Anne-Laure Fameau

FIGURE 52.2 Evolution of the solubility of myristic acid in high oleic sunflower oil over temperature, with points obtained from cooling at 0.1 °C. min−1
(squares) and heating at 0.1 °C. min−1 (circles). The full line without points corresponds to the solubility boundary and is calculated using (52.3). Inset:
photographs of (a) oil solution with molecular myristic acid, upright at 40 °C, and (b) oleogel containing crystalline particles from myristic acid inside the
liquid sunflower oil, inverted at 22 °C, for 10 wt.% myristic acid.
(Reproduced by permission of the Royal Society of Chemistry)

angle values, discs of compressed raw material powder can be
used or films prepared by spin coating.
The shape and size of particles are known to be important
parameters for Pickering aqueous foams, as they have a great
effect on the foamability and foam stability. However, even if
these parameters are well studied and described in the litera-
ture for oleogels based on these systems, almost no data are
available for oil foams. The shape and size of the crystalline
particles depend on both the mixture between the oil and the
components that crystallize and the process (shear rate, cooling
rate, time and crystallization approach). For example, myristic
acid and fatty alcohols crystallize into two-​dimensional plate-​
like crystals (Fameau et al., 2015a; Binks et al., 2016). The size
and polymorphs (α, β and β′) of triacylglycerols are linked to the
crystallization process (cooling and shearing rates) and lead to
different foam stability (Heymans et al., 2018).
Recently, to gain insight into the stabilization of the bubbles by
the crystalline particles, Mishima and coworkers used synchro-
tron radiation microbeam X-​ray diffraction. The oil foam system
studied was based on salad oil with triacylglycerol crystals (fully
hydrogenated rapeseed oil rich in behenic acid) (Mishima et al.,
2016). By using the X-​ray diffraction method, those authors
demonstrated that the lamellar planes (or faces) of triglyceride
crystals need to be aligned parallel to the air–​oil surface. The
surface energy of the faces composed of methyl groups is low.
The lamellar planes composed of methyl end groups are facing
the air phase, whereas the lateral planes composed of glycerol
groups are connected to each other through the crystals adsorbed
at the air–​oil surface. It is supposed that the same arrangement at
the interface takes place for plate-​like crystalline particles based
on fatty acids or fatty alcohols. The faces expose methyl groups,
which are in contact with air, whereas the edges expose mainly
methylene and carboxylic or hydroxyl groups interacting with
each other through these groups within the air–​oil surface (Binks
et al., 2016). However, the crystalline particles are randomly
arranged within the liquid oil.
Strategy and Process to Produce Edible Oil Foams
Two strategies are possible to produce edible Pickering oil
foams based on crystalline particles. The first strategy is the one
described previously, based on the mixture of a vegetable liquid
Foams: Pickering Edible Oil Foam
FIGURE 52.3 Schematic illustrating one of the strategies to produce edible oil foams. A hot liquid oil solution containing an edible high-​melting-​point addi-
tive is cooled down below the solubility boundary temperature during whipping. The crystalline particles are formed upon cooling and stabilize the air bubbles,
leading to the formation of a white creamy foam. The presence of the crystalline particles can be seen in the microscopy picture.
oil for the continuous phase and an edible high-​melting-​point
additive forming crystalline particles on cooling (Figure 52.3)
(Fameau et al., 2015a).
For the second strategy, only one component is required: an
oil/​fat composed of a mixture of triglycerides with different chain
lengths (Binks and Marinopoulos, 2017). Indeed, in a recent
study, additive-​ free edible oil foams have been produced by
using oil/​fat containing a high proportion of saturated medium-​
length fatty acid chains. When the oil is composed of different
triglycerides, as temperature changes, the triglycerides with high
melting point can be crystallized, and other triglycerides with
low melting point remain in a liquid state. Therefore, these oils
and fats are naturally composed of crystalline particles dispersed
inside a continuous oil liquid phase. For example, Binks et al.
(2017) have shown that coconut oil, shea butter and cocoa butter
give oil foams without the addition of other components (Binks
and Marinopoulos, 2017). The advantage of this second strategy
is to prepare oil foams directly, without additives. However,
the temperature at which the foam is produced needs to be per-
fectly controlled in order to have enough crystalline particles
dispersed in the liquid oil to stabilize the gas bubbles. It has been
demonstrated that to produce foams, the temperature of the oil
needs to be chosen to achieve a solid fat content around 30%. For
coconut oil, the optimum temperature is 23 °C, and the tempera-
ture at which all the triglycerides of the oil are liquid is 25 °C
(Binks and Marinopoulos, 2017).
Whatever the strategy used to produce oil foams, edible oil
foams can be produced by different processes: whipping, high-​
shear rotor-​stator and decompressing gas (Shrestha et al., 2006;
Fameau et al., 2015a; Gunes et al., 2017). Bubble size is linked to
the foam production method as well as the liquid volume fraction
(Drenckhan and Saint-​Jalmes, 2015). A parameter often used to
compare different foaming systems is the overrun, i.e., the per-
centage increase in volume due to incorporation of gas. The foam
overrun can be obtained by using (52.4), for which the volume
fraction of the air corresponds to Vt − V0:
 V − V0 
Overrun (%) =  t  * 100
 V0 
where Vt corresponds to the volume of foam after a time t, and
V0 corresponds to the initial volume of liquid used to produce
the foam.
361
Whipping is a highly popular foaming method, especially in
the food industry (Figure 52.3) (Drenckhan and Saint-​Jalmes,
2015). Various kitchen mixers have been described in the litera-
ture to produce oil foams. Foams can be produced directly by
whipping at room temperature from the oleogel (Brun et al.,
2015; Binks et al., 2016) or during the crystallization process
induced by decreasing the temperature below the solubility
boundary during shearing (Fameau et al., 2015a). In this foaming
technique, the production of foam is based on air entrainment
and systematic bubble breakup under shear (Drenckhan and
Saint-​Jalmes, 2015). For the same whipping technique, different
studies have shown that whipping invariably gave bubbles in
the same small size range, whatever the power input, the dur-
ation, etc. (Gunes et al., 2017). During shearing, the gas fraction
increases and the average bubble size decreases. Over time,
this process at the macroscopic scale leads to oil foam, which
becomes increasingly white and solid-​like. When the equilibrium
state is reached, the gas fraction and the bubble size can no longer
be changed by the beating device. Both the rheological properties
of the liquid and the beating speed control the characteristic gas
fraction and bubble size of a foaming solution (Drenckhan and
Saint-​Jalmes, 2015).
To further decrease the bubble size after whipping, Brun et al.
(2015) used a rheometer with a Couette cell geometry to shear
the foam. The applied shear decreased the average bubble size
by a factor of around two. Gunes et al. (2017) compared, for the
same foaming systems, three foaming processes: whipping, fast
cooling by using liquid nitrogen, and a depressurization method.
The tests performed by crash cooling using liquid nitrogen
yielded very small crystals, which could stabilize smaller bubbles
than the whipping method. However, the depressurization tech-
nique resulted in a larger average bubble size. Therefore, there is
a link between bubble size and crystal size, which in turn depends
on the foaming technique.
Edible oil foams have also been shown to be produced by an
industrial high-​shear rotor-​stator (Mondomix, Kinematica AG,
Germany). This industrial device enables the design of a con-
tinuous oil foam production line with high aeration level, as
(52.4)
mentioned in Gunes et al. (2017).
The reader should keep in mind that whatever the foaming pro-
cess, foam formation is only possible when crystals are present in
solution, i.e., below the solubility boundary.
362
Foamability and Oil Foam Stability
As described previously, Pickering foams are stabilized by the
presence of particles at the interface. In the case of the oil foam
systems, a dense layer of irreversibly adsorbed crystals around
the bubbles reduces gas diffusion and coalescence. The particu-
larity of these systems is that the non-​adsorbed crystals in excess
in oil can form a gel network in the continuous oil phase, i.e., an
oleogel. The presence of this network impends buoyancy-​driven
creaming of air bubbles within the foam (Binks et al., 2016).
The oil drainage is directly linked to the rheological properties
of the crystal network inside the liquid oil phase. Therefore, the
presence of the crystalline particles both at the interface and in
the continuous phase stabilizes the oil foam. The two parameters
governing the oil foam stability are the presence of crystalline
particles and their concentration, which are directly linked to the
solubility boundary.
At a fixed temperature, foam stability is directly linked to par-
ticle concentration. The rule is simple: the higher the crystalline
particle concentration, the higher the foam stability. For example,
in the case of foams generated by myristic acid dispersed in
high oleic sunflower oil, foams made with different myristic
acid concentrations have been kept at 22 °C for months (Binks
et al., 2016). For low myristic acid concentrations (<6 wt.%), the
solubility boundary is almost reached at 22 °C, as determined by
DSC, the temperature at which the foam is produced and kept
(Figure 52.2). Therefore, the foam stability is relatively low, with
around 30% of the initial oil volume draining after only one day.
A large amount of the crystals have melted to form the molecular
solution, and the amount of crystalline particles present to sta-
bilize the foam is low (Figure 52.4).
For a high myristic acid concentration (>8 wt.%), the solu-
bility boundary is much higher than 22 °C, and foams are stable
for months without any sign of destabilization due to the large
quantity of crystals present to stabilize the foam. This increase
in foam stability on increasing the myristic acid concentration at
a fixed temperature (22 °C) is linked to the solubility boundary
(Figure 52.2).
FIGURE 52.4 Schematic illustrating the foam stabilization mechanism by
the crystalline particles. Below the solubility boundary, the foam is stabilized
by the particles. When the temperature is increased above the solubility
boundary temperature, the crystalline particles located both at the interface
and in the bulk melt, resulting in foam destabilization. The result is a liquid
oil solution.
Anne-Laure Fameau
In the same way, the foam stability can be modified from low
to high by modifying the temperature at which the foam is kept
at a fixed particle concentration due to the boundary solubility
temperature of each system. Below the boundary solubility,
foams are always stable, but above this, the foam disappears
quickly due to the crystals melting (Figure 52.4). For example,
a system containing 10 wt.% of 1-​tetradecanol in sunflower oil
has a boundary solubility around 26.6 °C, as determined by
DSC (Fameau et al., 2015a). The foam was produced and kept
around 10 °C, below the boundary solubility, and was ultrastable,
whereas when the foam was kept above 26.6 °C, it was unstable.
To control the foam stability, it is very important to know the
solubility boundary of the foaming system. Very stable foams
lasting weeks and months can be easily obtained if the foam is
produced with a high amount of crystalline particles and kept
below the solubility boundary temperature. This ultrastability of
oil foam systems even above room temperature is very interesting
for commercial applications.
The foamability of these systems is also directly linked to the
particle concentration, which depends on the solubility boundary.
When the concentration of high-​melting-​point additives increases,
this leads to an increase in the boundary solubility temperature.
Moreover, below the boundary solubility limit, this increase in
concentration of the high-​melting-​point additives gives rise to an
increase in particle concentration. For example, to compare the
foamability of the 1-​octadecanol fatty alcohol in a sunflower oil
system at 20 °C below the solubility boundary, the overrun was
measured at two different fatty alcohol concentrations: 2 wt.%
and 10 wt.%. The overrun was increased from 20% to 75% by
increasing the concentration of the fatty alcohol, leading to an
increase in the concentration of crystalline particles (Marangoni,
2012; Fameau et al., 2015a). The foamability increased due to
the presence of more particles, and a larger area of air–​oil surface
could be stabilized. The same observations have been obtained
for various systems. For a fixed concentration of high-​melting-​
point additive, the temperature at which the foam is formed is
another crucial parameter in the foamability of these systems. For
example, in the case of 8 wt.% myristic acid dispersed in high
oleic sunflower oil, the foam volume decreased on increasing the
temperature from 20 to 35 °C (Binks et al., 2016). At 20 °C and
for 8 wt.% myristic acid, the system was below the solubility
boundary, leading to the presence of a large amount of crystalline
particles, which stabilize the air bubbles to reach an air fraction of
50 vol.% (Figure 52.2). Above 25 °C, the boundary solubility is
crossed, and the amount of crystalline particles decreases due to
progressive crystal melting. At 30 °C, the air fraction decreases to
reach 20 vol.%, since the amount of crystalline particles is lower
than at 20 °C. Above 35 °C, which is the solubility boundary,
it is only a molecular solution, and no crystalline particles are
present in the system (Figure 52.2). Therefore, no foam can be
produced. This decrease in foamability is linked directly to the
solubility boundary of the system. It is important to control and
tune the foamability, since most of the foam properties, such as
the sensorial properties, are linked to the air fraction. As a rule of
thumb, close to the boundary solubility, the foamability and foam
stability are low, but below it, the foamability and foam stability
are high (Figure 52.4).
Foams: Pickering Edible Oil Foam 363

For Pickering aqueous foams, it is known that the shape and the temperature, the triglycerides with high melting point can
size of the particles modify the foamability (Lam et al., 2014). To be crystallized, and other triglycerides with low melting point
change the size and shape of the crystalline particles, Mishima remain in a liquid state. Therefore, these types of oils naturally
et al. (2016) changed the crystallization and tempering pro- contain crystalline particles inside a continuous oil liquid phase
cess for a system based on long-​chain triglycerides dispersed in without additives.
salad oil. The overrun of the foams produced from the different From a fundamental point of view, a lot of questions remain
polymorphs is tuned by the tempering process highlighting the open on these systems. For example, the effects of the size, shape
influence of crystal size and morphology. The best foaming agent and concentration of crystalline particles on the foaming proper-
was tiny β-​fat crystals. Another recent study from Heymans et al. ties still need to be studied more deeply in order to perfectly con-
(2018) confirmed the effect of tempering on the foaming prop- trol them. The rheological properties of these interfaces stabilized
erties of these oil foam systems. However, many more studies by adsorbed particles also need to be understood.
are needed to better understand how the crystal shape and size From an applied point of view, studies on edible oil foams
modify the foamability and foam stability. for food applications are in their infancy. Recent patents on oil
foams highlight the interest in this subject, which will grow in
Structure and Rheological Properties of Edible the coming years (Chisholm et al., 2018; Gunes et al., 2018). The
potential of these systems to create food products with reduced
Oil Foams
caloric content, novel texturized food products or food products
These oil foam systems obtained from crystalline particles with new mouthfeel is very high. A major advantage of oil foams
exhibit non-​spherical bubbles. The crystals are clearly observed in comparison to aqueous foams comes from the fact that they
on the bubble surface using microscopic techniques showing do not contain water. Therefore, they are much less sensitive to
the textured surfaces of the bubbles (Figure 52.3). This non-​ microbial spoilage, and fewer or no additives are needed for pres-
spherical shape comes from the dense layer of adsorbed crystal- ervation (Heymans et al., 2017). However, the main disadvantage
line particles, which prevents the relaxation of the bubble into a in comparison to aqueous foams is their susceptibility to lipid
spherical shape. Until now, for all studies, the bubbles have been oxidation. For industrial aspects, before commercializing the
made of air. An important parameter not yet studied is the effect of edible oil foam systems, the foaming process, the shelf-​life and
the gas on the edible oil foam structure and properties. However, the sensory aspects have to be perfectly controlled.
it is known for aqueous foams that, by using various gases, such For molecular gastronomy, the current understanding is
as nitrogen and carbon dioxide, it is possible to modify the bubble enough to produce a large variety of edible oil foams by
size and the resulting texture of the foam (Achakulwisut et al., whipping. A recipe based on olive oil is described and avail-
2017; Heymans et al., 2017). able on the website “www.molecularrecipes.com/​ foams/​olive-​
Specific properties of these edible oil foam systems come from oil-​foam/​”. However, all the systems described in the literature
their rheological behaviour (Famaeau et al., 2015a). When the can also be directly used as recipes (Fameau, 2017). The new
excess of crystalline particles in the continuous oil phase forms texture obtained from these foams based on fatty components
a network, an oleogel is created. Thus, when the foam is turned can enhance the experience of gourmet cuisine and lead to new
upside down, the foam does not flow. These oil foams behave culinary experiences.
like a strong gel, which is why they can sustain their own weight
(Achakulwisut et al., 2017). The rheological behaviour of these REFERENCES
foams is strongly linked to the rheological behaviour of the con- Achakulwisut K, Tam C, Huerre A, Sammouti R, Binks BP, Garbin
tinuous phase. To tune the rheological properties of the foams and V. 2017. Stability of clay particle-​coated microbubbles in
the product texture in the case of food products, it is important to alkanes against dissolution induced by heating. Langmuir,
consider all the reviews and articles dealing with edible oleogel 33(15), 3809–​3817.
(Rogers et al., 2014; Singh et al., 2017). A lot of fundamental Binks B, Davies C, Fletcher P, Sharp E. 2010. Non-​ aqueous
work still needs to be done on these systems to understand their foams in lubricating oil systems. Colloids and Surfaces A:
properties and how to tune them. Physicochemical and Engineering Aspects, 360(1), 198–​204.
Binks BP. 2002. Particles as surfactants –​similarities and differences.
Current Opinion in Colloid & Interface Science, 7(1–​2), 21–​41.
Binks BP. 2017. Colloidal particles at a range of fluid–​fluid interfaces.
Conclusion Langmuir.
Binks BP, Garvey EJ, Vieira J. 2016. Whipped oil stabilised by sur-
Edible oil foams are Pickering foams stabilized by the presence factant crystals. Chemical Science, 7(4), 2621–​2632.
of crystalline particles at the interface. The formation of the Binks BP, Marinopoulos I. 2017. Ultra-​stable self-​foaming oils. Food
crystalline particles inside a liquid oil continuous phase depends Research International, 95, 28–​37.
on the solubility boundary. To produce edible oil foams, two Binks BP, Sekine T, Tyowua AT. 2014. Dry oil powders and oil foams
strategies are possible. The first involves combining a vege- stabilised by fluorinated clay platelet particles. Soft Matter,
10(4), 578–​589.
table oil for the continuous phase and an edible high-​melting-​
Blázquez C, Emond E, Schneider S, Dalmazzone C, Bergeron V.
point additive forming crystalline particles on cooling. The 2014. Non-​aqueous and crude oil foams. Oil & Gas Science
second involves directly using an oil composed of a mixture of and Technology–​ Revue d’IFP Energies nouvelles, 69(3),
triglycerides with different chain lengths. Thus, as a function of 467–​479.
364
Brun M, Delample M, Harte E, Lecomte S, Leal-​Calderon F. 2015.
Stabilization of air bubbles in oil by surfactant crystals: A route
to produce air-​in-​oil foams and air-​in-​oil-​in-​water emulsions.
Food Research International, 67, 366–​375.
Cantat I, Cohen-​ Addad S, Elias F, Graner F, Höhler R, Pitois
O, Rouyer F, Saint-​Jalmes A. 2013. Foams: structure and
dynamics, Oxford University Press, Oxford.
Chisholm H, Gunes ZD, Gehin-​delval C, Appolonia Nouzille C,
Garvey E, Destribats MJ, Chandrasekaran SN, Vieira JB,
German J, Binks BP. 2018. Aerated Confectionery Material.
US Patent App. 15/​559,994.
Drenckhan W, Saint-​ Jalmes A. 2015. The science of foaming.
Advances in Colloid and Interface Science, 222, 228–​259.
Fameau AL. 2017. Non-​aqueous foams based on edible oils. In Edible
Oil Structuring, ed. AR Patel, Royal Society of Chemistry,
London, UK, 275–​307.
Fameau AL, Lam S, Arnould A, Gaillard CD, Velev OD, Saint-​
Jalmes A. 2015a. Smart nonaqueous foams from lipid-​based
oleogel. Langmuir, 31(50), 13501–​13510.
Fameau AL, Saint-​Jalmes A. 2017. Non-​aqueous foams: Current
understanding on the formation and stability mechanisms.
Advances in Colloid and Interface Science, 247, 454–​464.
Fameau AL, Salonen A. 2014. Effect of particles and aggregated
structures on the foam stability and aging. Comptes Rendus
Physique, 15, 748–​760.
Fameau AL, Carl A, Saint-​ Jalmes A, Von Klitzing R. 2015b.
Responsive aqueous foams. Chemphyschem, 16(1), 66–​75.
Friberg SE. 2010. Foams from non-​ aqueous systems. Current
Opinion in Colloid & Interface Science, 15(5), 359–​364.
Gunes D, Murith M, Godefroid J, Pelloux C, Deyber H, Schafer
O, Breton O. 2017. Oleofoams: Properties of crystal-​coated
bubbles from whipped oleogels evidence for Pickering stabil-
ization. Langmuir, 33(6), 1563–​1575.
Gunes ZD, Schafer O, Chisholm H, Deyber H, Pelloux C, Binks BP.
2018. Lipid based foam, US Patent App. 15/​559,982.
Heymans R, Tavernier I, Danthine S, Rimaux T, Van der Meeren P,
Dewettinck K. 2018. Food-​grade monoglyceride oil foams:
The effect of tempering on foamability, foam stability and
rheological properties. Food & Function, 9(6), 3143–​3154.
Heymans R, Tavernier I, Dewettinck K, Van der Meeren P. 2017.
Crystal stabilization of edible oil foams. Trends in Food
Science & Technology, 69, 13–​24.
Anne-Laure Fameau
Hill C, Eastoe J. 2017. Foams: From nature to industry. Advances in
Colloid and Interface Science, 247, 496–​513.
Lam S, Velikov KP, Velev OD. 2014 Pickering stabilization of foams
and emulsions with particles of biological origin. Current
Opinion in Colloid & Interface Science, 19(5), 490–​500.
Marangoni AG. 2012 Organogels: An alternative edible oil-​
structuring method. Journal of the American Oil Chemists’
Society, 89(5), 749–​780.
Mishima S, Suzuki A, Sato K, Ueno S. 2016 Formation and
microstructures of whipped oils composed of vegetable oils
and high-​melting fat crystals. Journal of the American Oil
Chemists’ Society, 93(11), 1453–​1466.
Patel A. 2017. Stable ‘arrested’ non-​aqueous edible foams based on
food emulsifiers. Food & Function, 8(6), 2115–​2120.
Pugh RJ. 2016. Bubble and Foam Chemistry. Cambridge University
Press, Cambridge, UK.
Rio E, Drenckhan W, Salonen A, Langevin D. 2014. Unusually stable
liquid foams. Advances in Colloid and Interface Science,
205, 74–​86.
Rogers MA, Strober T, Bot A, Toro-​Vazquez JF, Stortz T, Marangoni
AG. 2014. Edible oleogels in molecular gastronomy.
International Journal of Gastronomy and Food Science,
2(1), 22–​31.
Sanders PA. 1970. Stabilization of aerosol emulsions and foams.
Journal of the Society of Cosmetic Chemists, 21(6), 377–​391.
Shrestha LK, Aramaki K. 2012 Non-​ aqueous foams: Formation
and stability. In Foam Engineering: Fundamentals and
Applications, ed. P Stevenson, John Wiley and Sons, Hoboken,
New Jersey, USA, 169–​206.
Shrestha LK, Aramaki K, Kato H, Takase Y, Kunieda H. 2006
Foaming properties of monoglycerol fatty acid esters in
nonpolar oil systems. Langmuir, 22(20), 8337–​8345.
Shrestha LK, Sato T, Aramaki K. 2007a. Phase behavior and self-​
organized structures of diglycerol monolaurate in different
nonpolar organic solvents. Langmuir, 23(12), 6606–​6613.
Shrestha LK, Shresth RG, Solans C, Aramaki K. 2007b. Effect of
water on foaming properties of diglycerol fatty acid ester-​oil
systems. Langmuir, 23(13), 6918–​6926.
Singh A, Auzanneau FI, Rogers M. 2017. Advances in edible
oleogel technologies –​A decade in review. Food Research
International, 97, 307–​317.
Frying
Franco Pedreschi
Departamento de Ingeniería Química y Bioprocesos, Pontificia Universidad Católica de Chile, P.O. Box 306,
Santiago 6904411, Chile
Introduction
The snack industry (which includes many fried vegetable products)
is an important part of the food industry worldwide, mainly in
Europe and the USA, with sales around the world of almost $374
billion annually, increasing by approximately 2% year-​on-​year
(Pedreschi et al., 2018). The current changes in eating habits and
the short time people allocate to eating have produced substan-
tial changes in both the frequency and the type of regular meals,
which has left a gap for the creation of new snacks with different
sensory, nutritional and technological characteristics. Frying, a
very old unit operation (Zeb, 2019), is increasingly used world-
wide in fast food outlets and at home, which has led to concerns
about its health implications (Pedreschi et al., 2018).
Frying is a food unit operation in which the liquid transfer-
ring heat is hydrophobic, yet becomes part of the food. Deep-​fat
frying consists of immersion of food pieces in hot vegetable oil.
It is a rather complex process, comprising simultaneous heat and
mass transfer with chemical reactions and textural changes taking
place. The intense heat and mass transfer achieved during deep
frying have led to innovative applications for food materials. As
a result of frying, the piece of food emerges sterile and dry at
the surface, with increased shelf-​life. The high temperatures of
frying (e.g., >120 °C) eliminate microorganisms from the food,
favouring its microbiological safety (Pedreschi, 2012).
Deep-​ fat frying also involves significant microstructural
changes. Most of the desirable characteristics of fried foods
are derived from the formation of a composite structure: a dry,
porous, crispy and oily outer layer (crust) and a moist cooked
interior (core) are formed during the process in the case of fried
potato strips (Bouchon et al., 2001). The high temperatures
of frying oils typically lead to a desirable surface colour; the
heating of sugars causes a complex group of reactions leading
to browning development, which defines the colour of the final
product. Additionally, toxic compounds (like acrylamide) may be
formed during this process, in particular as a result of Maillard
reactions. Moreover, some heat-​induced toxicants that can be
formed during frying include ethyl-​ carbamate, furan, hetero-
cyclic amines, 5-​hydroxymethylfurfural (HMF), polycyclic aro-
matic hydrocarbons and nitroso-​amines (Murkovich et al., 2018).
The characteristics of fried products depend not only on the
frying conditions but also on the type of oils and foods used during
the process (Aguilera and Gloria-​Hernandez, 1997). Immersed
frying or deep-​frying is a common multi-​functional unit oper-
ation for fast frying, texturing and cooking of foods. For decades,
consumers have desired fried foods because of their unique com-
bination of flavour, colour and texture (Saguy and Dana, 2003).
Oil Uptake
Over the last two decades, reducing the oil content of fried
products has had great appeal. The consumer trend is towards
healthier and less greasy products (Ziaiifar et al., 2010). The
majority of fried foods initially contain low levels of fat. The
mechanism of oil uptake has been studied primarily to under-
stand the phenomenon and to develop the know-​how and means
to reduce the final oil content (Dana and Saguy, 2006). However,
in spite of all these efforts, fried food products still possess signifi-
cant amounts of fat, exceeding 30% in some cases (Table 53.1).
The oil content in each fried product will depend on the frying
conditions, and principally on the original microstructure of the
raw material and the changes that it undergoes during deep-​fat
frying (Pedreschi, 2012).
Oil uptake is mainly a surface phenomenon, as confirmed
by many experimental observations (Baumann and Escher,
1995). Both the oil content and the moisture content are crit-
ical parameters that determine the quality and stability of fried
products (Pedreschi and Moyano, 2006). The final oil and mois-
ture content of the fried food will depend on its microstructure,
chemical composition, geometrical shapes and process conditions
(Saguy and Dana, 2003). Moisture content is an important prop-
erty in fried food product quality. It is desirable that the moisture
and oil content in fried foods are expressed on a dry basis (free of
oil), since both the oil and the water content of the sample being
fried change during the process (Moreira et al., 1999).
These issues have prompted a search for technologies that
reduce the absorption of oil, preserve the nutritional value of
raw materials, and mitigate the formation of potentially toxic
compounds such as acrylamide, furan and HMF while keeping
the desirable eating quality of fried products (Medeiros et al.,
2012; Pedreschi et al., 2018). Frying involves heat transfer by
convection (from the hot oil at 160 to 190 °C) and conduction (to
365
366
TABLE 53.1
Typical Mean Water and Oil Content of Some Fried Foods
Mean water Mean oil
Product content (%) content (%)
Potato chips 2.5 34.6
Corn chips 1.0 33.4
Tortilla chips 1.8 26.2
Doughnuts (plain) 20.8 22.9
Onion rings 28.5 18.7
Chicken breast-​breaded 45.7 18.1
Fish fillet-​battered or breaded 53.6 12.9
French fries/​par-​fried 39.5/​37.9 14.8/​7.6
Source: Reprinted from Saguy and Dana (2003) with kind permission of
Elsevier.
the interior of the piece) as well as mass transfer in the form of
loss of surface water (as steam bubbles) and oil uptake (Aguilera,
2018). During deep-​fat frying, the massive production of steam
bubbles at the evaporation front keeps oil from penetrating into
the forming crust. Concerning the time (moment) of oil absorp-
tion, it has been shown that oil does not enter the product to a great
extent during frying but is drawn from the oil film on the product
when it is removed from the oil bath (when steam inside the crust
condenses as it cools, and this “sucks in” some of the oil that
wets the surface) and relies on microstructural changes developed
during deep-​fat frying (Ufheil and Escher, 1996; Aguilera and
Gloria-​Hernandez, 2000; Bouchon et al., 2003; Pedreschi et al.,
2008; Ziaiifar et al., 2008).
Depending on the food microstructure, the oil will penetrate
the fried food to varying degrees, contributing to lubricity and
attractive flavour as well as adding calories. The quality of the
fried product depends not only on the frying conditions but also
on the type of oil and the microstructure of the food (Pedreschi,
2012). On the other hand, the high amount of oil retained inside
the fried products is incompatible with current health trends
(Moreno et al., 2010). In this sense, the high temperatures in
deep-​fat frying processes (around 160 and 190 °C) cause the
evaporation of water, which is transferred from the food towards
the surrounding oil (Mellema, 2003). Finally, not only is the total
amount of oil absorbed by the fried product important but also
how the oil is distributed in the crust, and the quality of this oil.
Oil uptake has been studied using different techniques, since it
is a very complex multifactorial phenomenon. Some researchers
have used dyed oil (Ufheil and Escher, 1996) or successive
washings with solvents (Moreira et al., 1997) to demonstrate that
most oil absorption takes place immediately during the cooling
period after the frying. Vitrac et al. (2000) confirmed these
observations by measuring the internal pressure inside a model
starch food during frying and cooling.
Chemical, Physical and Microstructural Changes
During frying, chemical, physical and microstructural transform-
ations of raw food material tissue are also induced (vegetable or
Franco Pedreschi
animal). These changes are responsible for the final attractive
sensorial attributes of fried products (Pedreschi, 2012). The study
and modelling of the kinetics of the changes in some important
physical properties in potatoes during frying have been reported
(Moyano and Pedreschi, 2006; Pedreschi et al., 2005; Pedreschi
and Moyano, 2005; Pedreschi et al., 2001). However, in par-
allel with the development of colour, texture and flavour of the
food pieces being fried, excessive heating in frying can produce
(according to the chemical composition of the raw material)
undesirable reactions and the formation of potentially toxic
compounds such as acrylamide, furan and HMF, among others.
As said, due to Maillard reactions, fried products develop not
only attractive sensory attributes, such as odour and colour, but
also some undesirable toxic contaminants such as acrylamide
and HMF, which are potential human carcinogens. The high
temperatures favour reactions between some substances naturally
present in the raw food (e.g., reducing sugars and asparagine),
inducing the formation of some toxic compounds such as acryl-
amide and furan, and in this way, negatively affect the chemical
safety of some starchy fried foods (Pedreschi, 2012; Murkovich
et al., 2018).
During atmospheric frying, fats and oils can reach much
higher temperatures than water at normal atmospheric pressure.
Therefore, the food sample is cooked quickly and acquires the
desirable and attractive sensorial attributes. If the frying time
is too long, the surface of the food could even be carbonized,
while some sugars are chemically modified. On the other hand,
in vacuum frying, the food is heated under reduced pressure in a
closed system that lowers the boiling points of frying oil and the
moisture in the food (Garayo and Moreira, 2002). Mariotti et al.
(2018) showed that vacuum frying is an effective technology for
furan and acrylamide mitigation in potato chips, since it reduces
the content of both contaminants and preserves the quality
attributes of fried snacks. Vacuum-​fried potato chips showed
reductions of about 81%, 58% and 28% of furan, acrylamide and
oil content, respectively, when compared with their atmospheric
counterparts. Vacuum frying can be considered as a promising
technology to produce healthier potato chips.
According to the food microstructure, geometrical shape and
frying process conditions, two principal kinds of fried products
could be formed after frying (Pedreschi, 2012; Pedreschi et al.,
2001). In the case of fried potatoes, there are: (a) thin fried slices
(chips), characterized by having a dehydrated and crispy region
where oil is located and some toxic compounds such as acryl-
amide could be distributed inside it; and (b) a composite structure
formed by two regions, such as an external dehydrated and crispy
region, where oil is located and where some toxic compounds
such as acrylamide could be distributed, and a humid and cooked
core free of oil and acrylamide, and other toxic compounds
formed in the surface, which could not migrate into the core.
Microstructural changes in the core are similar to those
occurring during the cooking of potatoes (Bouchon and Aguilera,
2001). Starch granules undergo gelatinization when the tem-
perature is higher than 60 °C, being swollen rapidly by intracel-
lular water, occupying the whole interior of the cell. The middle
lamellae that cement cells disintegrate within a narrow tempera-
ture range (60–​80 °C), and cells separate, giving a so-​called
Frying
FIGURE 53.1 Major microstructural changes occurring during potato frying.
(Adapted from Bouchon et al., 2003)
mealy texture. In addition, proteins may denature. A summary of
these microstructural changes appears in Figure 53.1.
On the other hand, the crust is the result of several alterations
that mainly occur at the cellular and sub-​cellular level, starting
from the outermost layers of the food, where the temperature
exceeds 100 °C. These chemical and physical changes include:
physical damage produced when the product is cut and a rough
surface is formed with release of intracellular material; starch
gelatinization and subsequent dehydration; protein denaturation;
breakdown of the cellular adhesion; water evaporation and rapid
dehydration of cells located in the forming crust; oil uptake; and
acrylamide formation. These changes give rise to the develop-
ment of a composite structure formed of two parts: (i) a crispy,
porous and oily outer layer or crust and (ii) an inner core with a
moist cooked interior, where the lamella media solubilizes and
starch gelatinizes. Bouchon et al. (2003) explained that three
different oil fractions can be identified as a consequence of the
different absorption mechanisms in fried potato cylinders: (i)
structural oil (STO), which represents the oil absorbed during
frying; (ii) penetrated surface oil (PSO), which represents the oil
sucked into the food during cooling after removal from the fryer;
and (iii) surface oil (SO), which is the oil that remains on the sur-
face (Figure 53.1). These authors showed that a small amount of
oil penetrates during frying, because most of the oil was picked
up at the end of the process, suggesting that oil uptake and water
removal are not synchronous phenomena. These findings have
been confirmed by Pedreschi et al. (2008) and Durán et al.
(2007).
Frying induces significant microstructural changes that can be
understood by using different modern visualization techniques. In
the case of potato tissue, most revealing has been a video micros-
copy study of in situ frying of potato cells showing that the starch
367
granules swell rapidly at a temperature of 65–​75 °C (Deslandes
et al., 2019), which is well below the temperature used for frying.
Steam bubbles leave the interior of the cells through pores in the
cell walls (probably through plasmodesmata) and find their way
to the product/​oil interface through many intercellular passages.
Swollen starch granules remained as a compact mass pressing
on the outer cell wall before becoming dehydrated at higher
temperatures (Bouchon and Aguilera, 2001). Cells heated in oil
to 180 °C remained largely intact and decreased marginally in
surface area without any evidence of oil in their interior (Aguilera
et al., 2001).
Confocal laser scanning microscopy (CLSM), a non-​invasive
technique that produces optical sections at increasing depths in
the specimen, demonstrated that oil is located in an “egg-​box”
arrangement surrounding intact potato cells. By using CLSM,
Pedreschi and Aguilera (2002) described the absorption of oil in
potato chips as seeming to wrap the surface of the parenchyma
cells of potatoes like an egg box (Figure 53.2).
Conclusions
Deep-​fat frying is a complex process comprising simultaneous
heat and mass transfer, with chemical reactions and textural
changes taking place. Deep-​fat frying processes can take place
at atmospheric pressure (conventional frying) or at pressures
below atmospheric (vacuum frying). Additionally, deep-​ fat
frying involves significant microstructural changes. The food
piece in contact with hot oil heats up, water is lost and oil content
increases; reactions between reducing sugars and amino acids
lead to browning and changes of texture, with softening at the
beginning of frying, and hardening of the external layers of food
368
(a) (b)
FIGURE 53.2 (a) Confocal images in fluorescence mode of oil distribution
in a potato chip fried in stained oil (170 ºC, 3 min) observed at 20× (each
square of the gallery is 640 × 640 mm). (b) Three-​dimensional reconstruction
from the serial sections in part (a) using Imaris software.
(Reprinted from Pedreschi et al., 1999, with kind permission from John
Wiley and Sons Ltd.).
with longer frying time. As a result, frying is often selected as a
method for developing unique flavours and improving texture in
processed foods that enhance their overall palatability.
Acknowledgements
Our research on frying was financially supported by FONDECYT
Projects 1190080 and 1150146.
REFERENCES
Aguilera JM. 2018. Relating food engineering to cooking and gas-
tronomy. Comprehensive Reviews in Food Science and Food
Safety, 17, 1–​19.
Aguilera JM, Cadoche L, Lopez C, Gutierrez G. 2001. Microstructural
changes of potato cells and starch granules heated in oil. Food
Research International, 34, 939–​947.
Aguilera JM, Gloria-​Hernández H. 1997. Determination of oil in
fried potato products by differential scanning calorimetry.
Journal of Agriculture and Food Chemistry, 45, 781–​785.
Aguilera JM, Gloria-​ Hernández H. 2000. Oil absorption during
frying of frozen parfried potatoes. Journal of Food Science,
65, 446–​479.
Baumann B, Escher E. 1995. Mass and heat transfer during deep
fat frying of potato slices. I. Rate of drying and oil uptake.
Lebensmittel-​Wissenschaft und-​Technologie, 28, 395–​403.
Bouchon P, Aguilera JM. 2001. Microstructural analysis of
frying potatoes. International Journal of Food Science and
Technology, 36, 1–​8.
Bouchon P, Aguilera JM, Pyle DL. 2003. Structure oil-​absorption
relationships during deep-​fat frying. Journal of Food Science,
68, 2711–​2716.
Bouchon P, Hollins P, Pearson M, Pyle DL, Tobin MJ. 2001.
Oil distribution in fried potatoes monitored by infrared
microspectroscopy. Journal of Food Science, 66, 918–​923.
Dana D, Saguy IS. 2006. Mechanism of oil uptake during deep-​fat
frying and the surfactant effect –​theory and myth. Advances in
Colloids and Interface Science, 128–​130, 267–​272.
Franco Pedreschi
Deslandes F, Plana-​fattori A, Almeida G, Moulin G, Doursat C, Flick
D. 2019. Estimation of individual starch granule swelling
under hydro-​thermal treament. Food Structure, 22, 1–​9. https://​
doi.org/​10.1016/​j.foostr.2019.100125
Durán M, Pedreschi F, Moyano P, Troncoso E. 2007. Oil partition in
pre-​treated potato slices during frying and cooling. Journal of
Food Engineering, 81, 257–​265.
Garayo J, Moreira R. 2002. Vacuum frying of potato chips. Journal
of Food Engineering, 55, 181–​191.
Mariotti-​
Celis MS, Martínez-​ Cifuentes S, Huamán-​ Castilla N,
PedreschI F, Iglesias-​ Rebolledo N, Pérez-​ Correa J. 2018
Impact of an integrated process of hot pressurized liquid
extraction-​macroporous resin purification over the polyphenols,
hydroxmethylfurfural and reducing sugars content of Vitis
vinifera “Carmenere” pomace extracts. International Journal
of Food Science and Technology, 53, 1072–​1078.
Medeiros R, Mestdagh F, De Meulanaer B. 2012. Acrylamide for-
mation in fried potato products –​present and future, a crit-
ical review on mitigation strategies. Food Chemistry, 133,
1138–​1154.
Mellema M. 2003. Mechanism and reduction of fat uptake in deep-​
fat fried foods. Trends in Food Science and Technology, 14,
364–​373.
Moreira R, Sun X, Chen Y. 1997. Factors affecting oil uptake in tor-
tilla chips in deep-​fat frying. Journal of Food Engineering, 31,
485–​498.
Moreira RG, Castell-​ Perez ME, Barrufet MA. 1999. Deep-​Fat
Frying: Fundamentals and Applications. Maryland: Aspen
Publishers.
Moreno MC, Brown C, Bouchon P. 2010. Effect of food surface
roughness on oil uptake by deep-​fat fried products. Journal of
Food Engineering, 101, 179–​186.
Murkovich M, Pedresch F, Ciezarova Z, 2018. Process Contaminants:
A Review. Reference Module in Food Science. Elsevier,
Oxford, UK, 1–​7.
Pedreschi F. 2012. Frying of potatoes: physical, chemical, and micro-
structural changes. Drying Technology, 30, 707–​725.
Pedreschi F, Aguilera JM. 2002. Some changes in potato chips during
frying observed by confocal laser scanning microscopy. Food
Science and Technology International, 8, 197–​201.
Pedreschi F, Aguilera JM, Arbildua JJ. 1999. CLSM study of oil loca-
tion in fried potato slices. Microscopy and Analysis, 37, 21–​22.
Pedreschi F, Aguilera JM, Pyle L. 2001. Textural characterization
and kinetics of potato strips during frying. Journal of Food
Science, 66, 314–​318.
Pedreschi F, Cocio E, Moyano P, Troncoso E. 2008. Oil distribution
in potato slices during frying. Journal of Food Engineering,
87, 200–​212.
Pedreschi F, Moyano P. 2005. Modelling water loss during frying
of potato slices. International Journal of Food Properties, 8,
289–​299.
Pedreschi F, Moyano P, Kaak K, Granby K. 2005. Color changes
and acrylamide formation in fried potato slices. Food Research
International, 38, 1–​9.
Pedreschi F, Saavedra I, Bunger A, Zuñiga R, Pedreschi R, Chirinos
R, Campos D, Mariotti MS. 2018. Tara pod (Caesalpinia
spinosa) extract mitigates neo-​ contaminant formation in
Chilean bread preserving their sensory attributes. LWT –​ Food
Science and Technology, 95, 116–​122.
Saguy S, Dana D. 2003. Integrated approach to deep fat frying:
engineering, nutrition, health and consumer aspects. Journal
of Food Engineering, 56, 143–​152.
Frying 369

Ufheil G, Escher F. 1996. Dynamics of oil uptake during deep-​ Ziaiifar A, Achir N, Courtois F, Trezzani I, Trystram G. 2008.
fat frying of potato slices. Lebensmittel-​Wissenschaft und Review of mechanisms, conditions, and factors involved in
Technologie, 29, 640–​644. the oil uptake phenomenon during the deep-​fat frying pro-
Vitrac O, Trystram G, Raoult-​Wack AL. 2000. Deep-​fat frying of cess. International Journal of Food Science Technology, 43,
food: heat and mass transfer, transformations and reactions 1410–​1423.
inside the frying material. European Journal of Lipid Science Ziaiifar A, Courtois F, Trystram G. 2010. Porosity development and
and Technology, 102, 529–​538. its effect on oil uptake during frying process. Journal of Food
Zeb A. 2019. Food Frying. John Wiley & Sons, Hoboken, New Process Engineering, 33, 191–​212.
Jersey, USA.
Gastrophysics: A New Scientific Approach to Eating
Charles Spence
Head of the Crossmodal Research Laboratory, Department of Experimental Psychology, University of Oxford,
Anna Watts Building, Oxford, OX2 6GG, United Kingdom
In recent years, a growing number of researchers, working in a
range of different scientific disciplines, have become increasingly
interested in the application of science to gastronomy. Several
terms or movements have been introduced to capture different
elements of this interface between science and food/​flavour: the
focus of molecular gastronomy research, for instance, is on the
science looking for the mechanisms occurring during the pro-
duction of dishes; the focus of neurogastronomy research, mean-
while, is on what is going on in the mind of the diner; finally, the
focus of gastrophysics (at least according to the definition used
here) is on the interface between gastronomy and psychophysics
(the measurement branch of perception science).
All three approaches, operating at the interface between
science and culinary artistry, can be seen as providing useful (and
non-​redundant) insights/​approaches that will hopefully help to
enhance the delivery of delicious food experiences. At the same
time, the focus for many of those working in these disciplines is
increasingly around how to nudge the consumer toward healthier
and/​or more pleasurable food choices that are more sustainable,
be that for the individual or the planet.
Introduction
The last half century has seen growing interest in science in the
kitchen (see Spence and Piqueras-​Fiszman, 2014 and Spence
and Youssef, 2018, for reviews). Three prominent strands
of this interaction/​collaboration are molecular gastronomy,
neurogastronomy, and gastrophysics, and these (non-​redundant)
movements are briefly summarized in the following.
Molecular Gastronomy
Molecular gastronomy has been defined as ‘the science of deli-
ciousness’ (e.g., Barham et al., 2010; Edwards-​ Stuart, 2012;
Youssef, 2013). Early advocates for the application of science
in the kitchen include Nikolas Kurti, Harold McGee, and Hervé
This (e.g., Kurti, 1969; McGee 1984/​2004, 1990; Kurti and This-​
Benckhard, 1994a, b). However, as highlighted by Spence and
Youssef (2018), while molecular gastronomy has undoubtedly
provided inspiration for chefs in terms of both novel preparation
methods as well as a range of new ingredients, its influence on
chefs has been more in terms of inspiration than, as is some-
times claimed, the adoption of the scientific method by chefs.
It is worth noting that molecular gastronomy has been taken up
more by some chefs/​countries than others. At the same time, the
scientific approach to cooking has been adopted for more than
two decades in cookery schools (e.g., in France) to help improve
chefs’ understanding of food transformation.
Neurogastronomy
A neologism apparently coined by Gordon Shepherd in 2006
(Shepherd, 2006). The term has become popularized following
the publication of the latter’s book Neurogastronomy (Shepherd,
2012). An annual neurogastronomy meeting, first held in North
America in 2015, has helped to cement interest in this ‘brain-​
based’ approach. The approach is undoubtedly important and
has helped shed light on the fundamental mechanisms under-
lying feeding behaviour (Chen et al., 2016), as well as shedding
light on a number of long-​running debates around such issues
as how branding and price information change our sensory-​
discriminative and hedonic responses to foods (Spence, 2016, for
a review). That said, it is worth highlighting the fact the majority
of research in this area involves participants lying by themselves
in an often-​noisy brain scanner with liquid flavour stimuli being
pumped into their mouth that they are periodically instructed to
swallow, meaning that the research context is far removed from
the more social aspects of everyday/​naturalistic dining on real
foods/​meals.
Gastrophysics
Gastrophysics has been defined in a number of ways since the
term’s original, fleeting first appearance in an article by Kathy
Parker (2004) in the journal Physical Education. Ole Mouritsen
and colleagues have stressed the links with physical chemistry
(e.g., Mouritsen, 2012; Mouritsen and Risbo, 2013, 2015; see
also article by Mouritsen on culinary science in the education
section of this book). Spence (2017), by contrast, in his book
Gastrophysics: The New Science of Eating, emphasizes the
371
372
(a) (b)
FIGURE 54.1 Three versions of the same plated ingredients (a 31 element salad) served to participants/​diners in a gastrophsyics study of plating reported
by Michel et al. (2014). Diners were willing to pay more for the Kandinsky-​inspired presentation (a) than for the regular tossed version (b) or the ordered (i.e.,
effortful) but unaesthetic plating arrangement (c).
combination of gastronomy and psychophysics that is at the heart
of an emerging body of research focusing on what is going on in
the mind (rather than the mouth) of the person doing the tasting.
While the former definition can be seen as close to molecular
gastronomy, that is, it is science related to the preparation of the
food itself, there are a growing number of researchers interested
in the science of the mind of the diner, too. That is, gastrophysics
highlights the scientific shift in focus from the kitchen (i.e., new
foams, spumes, cooking techniques, and ingredients, also known
as “molecular cooking” style) to the (mind of the) diner (i.e., the
person consuming that food or drink).
Importantly, much of the focus of gastrophysics, at least
according to this latter definition, is on ‘the everything else’ except
the food itself. There has, for instance, been growing interest in
the way in which cutlery (Spence and Piqueras-​Fiszman, 2014;
Welch et al., 2016) influences the diner’s tasting experience,
not to mention their eating behaviour. Similarly, there has been
a recent explosion of interest in the psychological and physico-​
chemical aspects of glassware (i.e., going beyond the wine glass;
e.g., Carvalho and Spence, 2018; Manska, 2018; Spence and
Wan, 2015). There is also growing interest in the multisensory
atmosphere in the places in which we eat and drink, and how the
ambient sensory cues may affect us (Spence et al., 2014; Spence,
2017), building on early work by the likes of Edwards (Edwards
et al., 2003).
One of the other areas where there have been significant
advances and interest in the world of gastrophysics is around
a scientific approach to the art of plating (Deroy et al., 2014;
Spence et al., 2014). It is clear that the visual appearance of
the food is playing an increasingly important role in people’s
food behaviours (Spence et al., 2016). In one example of the
gastrophysics approach, chef Charles Michel and his colleagues
scientifically investigated how the plating of a salad impacted
people’s willingness to pay and enjoyment of the dish (Michel,
Velasco et al., 2014). They were able to show, both in the
laboratory and in a real-​ world dining event (Michel et al.,
2015), that people rate the same food as tasting better, and are
Charles Spence
(c)
willing to pay more for it, when it is plated beautifully/​artistic-
ally (Figure 54.1).
Conclusions
After three decades or so of science being applied in the kitchen
(acknowledging, of course, that science has sporadically been
applied in the kitchen at various points over the centuries), there
is now a growing awareness that the pleasures of the table reside
as much in the mind as in the mouth. As such, a new scientific
approach is needed, one that moves beyond and builds upon trad-
itional sensory science techniques and/​or molecular gastronomy
approaches to the design of delicious food. This new approach,
here referred to as gastrophysics, emphasizes the importance of
managing (and understanding) people’s expectations as well as
their experiences and subsequent memories (Piqueras-​Fiszman
and Spence, 2015; Spence, in press). As highlighted here, this new
gastrophysics is equally interested in how ‘the everything else’
beyond the food or drink itself impacts the customer experience.
REFERENCES
Barham P, Skibsted LH, Bredie WLP, Bom Frøst M, Møller P, Risbo
J, Snitkjær P, Mortensen LM. 2010. Molecular gastronomy:
A new emerging scientific discipline. Chemical Reviews, 110,
2313–​2365.
Carvalho F, Spence C. 2018. The shape of the cup influences aroma,
taste, and hedonic judgements of specialty coffee. Food Quality
& Preference, 68, 315–​321.
Chen J, Papies EK, Barsalou LW. 2016. A core eating network and
its modulations underlie diverse eating phenomena. Brain and
Cognition, 110, 20–​42.
Deroy O, Michel C, Piqueras-​Fiszman B, Spence C. 2014. The plating
manifesto (I): From decoration to creation. Flavour, 3, 6.
Edwards JSA, Meiselman HL, Edwards A, Lesher L. 2003. The
influence of eating location on the acceptability of identically
prepared foods. Food Quality and Preference, 14, 647–​652.
Edwards-​Stuart R. 2012. Molecular gastronomy in the UK. Journal
of Culinary Science & Technology, 10, 97–​105.
Gastrophysics: A New Scientific Approach
Kurti N. 1969. The physicist in the kitchen. A transcript from the
weekly Evening Meeting of the Royal Society London Friday
14th March. Proceedings of the Royal Institution of Great
Britain, 42, 451–​467.
Kurti N, This-​Benckhard H. 1994a. Chemistry and physics in the kit-
chen. Scientific American, 270(4), 66–​71.
Kurti N, This-​Benckhard H. 1994b. The amateur scientist: The kit-
chen as a lab. Scientific American, 270(4), 120–​123.
Manska GF. 2018. Technical report –​Applying physics and sensory
sciences to spirits nosing vessel design to improve evaluation
diagnostics and drinking enjoyment. Beverages, 4, 93.
McGee H. 1984/​2004. On Food and Cooking: The Science and Lore
of the Kitchen (rev. ed.). New York: Scribner.
McGee H. 1990. The Curious Cook: More Kitchen Science and Lore.
New York: Collier Books.
Michel C, Velasco C, Fraemohs P, Spence C. 2015. Studying the
impact of plating on ratings of the food served in a naturalistic
dining context. Appetite, 90, 45–​50.
Michel C, Velasco C, Gatti E, Spence C. 2014. A taste of Kandinsky:
Assessing the influence of the artistic visual presentation of
food on the dining experience. Flavour, 3, 7.
Mouritsen OG. 2012. The emerging science of gastrophysics and its
application to the algal cuisine. Flavour, 1, 6.
Mouritsen OG, Risbo J. 2013. Gastrophysics –​do we need it?
Flavour, 2, 3.
Mouritsen OG, Risbo J. 2015. The emerging science of
gastrophysics –​do we need it? SMAG #04 2015.
Parker K. 2004. Recipe for success: Teachers get inspiration from
‘gastrophysics’. Physical Education, 39, 19.
Piqueras-​Fiszman B, Spence C. 2015. Sensory expectations based on
product-​extrinsic food cues: An interdisciplinary review of the
empirical evidence and theoretical accounts. Food Quality &
Preference, 40, 165–​179.
Shepherd GM. 2006. Smell images and the flavour system in the
human brain. Nature, 444, 316–​321.
Shepherd GM. 2012. Neurogastronomy: How the Brain Creates
Flavor and Why It Matters. New York: Columbia University
Press.
373
Spence C. 2012. Book review: ‘Neurogastronomy: How the brain
creates flavor and why it matters’ by Gordon M. Shepherd.
Flavour, 1, 21.
Spence C. 2016. The neuroscience of flavor. In Piqueras-​Fiszman B
and Spence C (Eds.) Multisensory Flavor Perception: From
Fundamental Neuroscience through to the Marketplace.
Oxford: Elsevier, 235–​248.
Spence C. 2017. Gastrophysics: The New Science of Eating. London:
Viking Penguin.
Spence C. In press. What role does memory play in our enjoyment
of meals? In Forrest B and de St. Maurice G (Eds.) Food in
Memory and Imagination: Place, Space and Taste. London:
Bloomsbury.
Spence C, Okajima K, Cheok AD, Petit O, Michel C. 2016. Eating
with our eyes: From visual hunger to digital satiation. Brain &
Cognition, 110, 53–​63.
Spence C, Piqueras-​ Fiszman B. 2014. The Perfect Meal: The
Multisensory Science of Food and Dining. Oxford:
Wiley-​Blackwell.
Spence C, Piqueras-​Fiszman B, Michel C, Deroy O. 2014. Plating
manifesto (II): The art and science of plating. Flavour, 3, 4.
Spence C, Velasco C, Knoeferle K. 2014. A large sample study on
the influence of the multisensory environment on the wine
drinking experience. Flavour, 3, 8.
Spence C, Wan I. 2015. Beverage perception & consumption: The
influence of the container on the perception of the contents.
Food Quality & Preference, 39, 131–​140.
Spence C, Youssef J. 2018. Assessing the long-​term impact of the
molecular gastronomy movement on haute cuisine. International
Journal of Gastronomy & Food Science, 14, 35–​44.
Velasco C, Jones R, King S, Spence C. (2013). Assessing the influ-
ence of the multisensory environment on the whisky drinking
experience. Flavour, 2, 23.
Welch W, Youssef J, Spence C. 2016. Neuro-​cutlery: The next fron-
tier in cutlery design. Supper Magazine, 4, 128–​129.
Youssef J. 2013. Molecular Cooking at Home: Taking Culinary
Physics out of the Lab and into Your Kitchen. London: Quintet
Publishing.
Gels
Hervé This vo Kientza1, 2
1 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
2 UMR SayFood, AgroParisTech, INRAE, Université Paris-​Saclay, 91300 Massy, France
Among colloids, gels have been used for changing the consist-
ency of solutions and also for differently releasing the compounds
dissolved in the liquid making up their liquid phase, giving new
flavour perceptions. More recently, they have been attracting much
interest because some are active materials, i.e., having properties
that can change according to the environment (Terech and Weiss,
1997; Abdallah and Weiss, 2000). Gels of various kinds are used
in cooking, but they have also been used in many other technical
fields (health, materials, energy, cosmetics, etc.) from which
innovative concepts could be transferred to the culinary arts
(Qiu and Park, 2001). For example, hydrogels formed from low-​
molecular-​weight compounds that respond to pH can be used for
oral drug delivery as well as for biosensor technology, especially
when biochemical components are involved; they could also be
made into interesting new foods. Hydrogels have been shown to
respond to ligand–​receptor molecular recognition (Zhang et al.,
2003, 2004) and redox stimuli (Kawano et al., 2004): the same
idea could be used for changing the consistency of food, based
on the recognition of specific compounds released in the mouth
during mastication.
The importance of gels is based on the fact that their liquid
phase can dissolve bioactive compounds, i.e., compounds that can
interact with receptors in the body of living beings (This, 2012),
slowing the release during mastication. This means that various
gels can be associated with specific “matrix effects” (how much
a system traps compounds; see the chapter on this topic in this
handbook). In particular, understanding the matrix effects of new
kinds of gels could be important for “note by note cooking”, i.e.,
the food preparation technique based on using pure compounds
as culinary ingredients instead of plant and animal tissues (see
the chapter on Note by Note Cooking, in Part III) (This, 2011;
This, 2013a).
One issue about gels is the various interpretations of the word
“gel” by different scientific and technical communities, in par-
ticular for food studies. A gelatin gel is certainly a gel, as are
agar-​agar gels, jams or cooked eggs; however, silica particles
simply packed into chromatographic columns are also said
to make “gels”, and for some rheologists (Ma and Barbosa-​
Canovas, 1995), mayonnaise and other densely packed emulsions
are considered as gels by some specialists of rheology.
However, this point of view is questionable, because it
considers only one property of colloidal systems (flow) (Mitchell,
1980) and not the structure itself, forgetting the various other
properties, such as electrical, optical, flow, bioactive compound
release, etc. (IUPAC, 2001).
The International Union of Pure and Applied Chemistry
advocates the following definition: “gels are non-​fluid colloidal
networks or polymer networks that are expanded throughout its
whole volume by a fluid” (IUPAC, 2007). Among the various gels
with networks of organic nature, “physical gels” have networks
made of subunits linked by bonds of energy that is lower than
for covalent bonds. They can be distinguished from “chemical
gels”, which are generally irreversible: the energy needed to dis-
rupt the network also disrupts the molecules making up the net-
work subunits. Interesting particular cases are when the energy
of bonds between the network subunits is close to kB T (where kB
is the Boltzmann constant and T the absolute temperature), as we
shall see later.
Using the Dispersed System Formalism for
Describing Gels
As the word “gel” is used in various scientific communities with
different meanings and also applies to very different systems,
it was proposed to describe colloids more precisely using the
“dispersed system formalism” (DSF) (This, 2007; This, 2009). It
was shown that this formal language is useful both for describing
the physical structure of existing colloidal systems and also for
predicting the existence of systems that were not envisioned
before. The DSF is generalizing a practice proposed by IUPAC
(IUPAC, 2001), but adding more operators than the sole “/​” used
for describing the random dispersion of a phase into another
(Figure 55.1).
Using the letters G, O, S, W for gas, oil, solid, water, respect-
ively, this “/​” operator can apply to emulsions (O/​W or W/​O),
foams (G/​W and W/​O), suspensions (S/​W, S/​O, S1/​S2), aerosols
(W/​G, O/​G, S/​G) and finally some gels made of liquid (W or O)
droplets dispersed in a solid (S). However, this operator does not
apply to gels, such as gelatin hydrogels, for example, where two
375
376 Hervé This vo Kientza

The introduction of the “x” operator leads to various questions,

such as: Are all gels described by the “x” and “/​” operators? How
many kinds of gels exist? How can gels be classified according to
their chemical and physical properties? Here, after this question
is theoretically discussed, it will be shown that an intriguing
line of development for physical chemistry, and in particular
for molecular gastronomy, relies on what is indistinctly called
a “gel”, of which one can distinguish two kinds (This, 2016):
those that are fixed (“statgels”), because of strong forces between
the subunits which make the solid network, and those that are
dynamic (“dynagels”), where the assembly of the network’s
subunits is of supramolecular nature (Lehn, 1995).

FIGURE 55.1 For foams, emulsions, aerosols, suspensions and some gels,
one phase X is randomly dispersed in another one Y; hence the DSF formula
X/​Y. One can observe that gelatin gels do not have this microstructure. In the
system shown in this picture, the Y continuous phase is three-​dimensional,
and the dispersed phase is zero-​dimensional (at least one order of magnitude
smaller than the continuous phase), so that the DSF formula can be made
more precise: D0(X)/D3(Y).
Simple Gels as Described Using DSF Operators
In order to envision all kinds of “simple” gels (i.e., gels
including only one liquid phase and only one solid phase), a
computer program can systematically produce all DSF for-
mulas (Figure 55.3). Programs for such purposes have to include
the three phases O (for “oil”), S, W (alphabetic order), the five
operators “/​”, “x”, “@”, “σ”, “+” (where “@” describes inclusion,
as recommended by IUPAC, “+” describes the coexistence of two
phases in another one, and “σ” describes superposition), and the
four possibilities for the dimensions of objects D0, D1, D2, D3 (for
objects of dimension zero, one, two and three, respectively); the
solid continuous network necessarily has three dimensions, and
it has to be used as the last term of the formula, in spite of DSF
preferably using alphabetical order.
The list of all possibilities is then:

D0(O)/​D3(S), D0(W)/​D3(S), D1(O)/​D3(S), D1(W)/​D3(S), D2(O)/​

D3(S), D2(W)/​D3(S), D1(O)xD3(S), D1(W)xD3(S), D2(O)xD3(S),
D2(W)xD3(S), D3(O)xD3(S), D3(W)xD3(S), D1(O)@D3(S),
D1(W)@D3(S), D2(O)@D3(S), D2(W)@D3(S).

Do all these systems exist? Indeed, there are impossible

proposals, such as when a three-​dimensional object is proposed
FIGURE 55.2 In a gelatin gel, the water (W) phase is not randomly
to be dispersed in a zero-​dimensional one (D3/​D0). The list has
dispersed inside a solid matrix (S) but makes a continuous phase, intermixed
to be pruned either manually or by using an improved program.
with the continuous solid phase. Hence the DSF formula SxW. In this case,
the two phases are three-​dimensional, so that a more precise formula can be
Anyway, many possibilities remain, and as envisioning all of
given: D3(S)xD3(W). them one after another is tedious, we shall now consider only
two of them.
As a first example, let us consider the first formula of the list,
phases, liquid and solid, are continuous in the three dimensions (D0(O)/​D3(S)): this corresponds to a random dispersion of oil
of the gel system. (O) droplets (D0) in a continuous solid (D3(S)), and this can be
For this reason, the DSF includes the operator “x”, which made in various ways. For example, one could freeze –​at a tem-
designates the inter-​dispersion of two continuous phases. For perature below 0 °C –​an O/​W emulsion, for which the “oil” (O) is
example, the SxW (the alphabetical order is to be used for this a compound that would not be miscible with water and would not
operator) formula corresponds to gelatin gels with a continuous solidify at the temperature at which water solidified (for example,
three-​dimensional aqueous phase W and a continuous three-​ carbon disulfide, for which the freezing point is −111.5 °C).
dimensional solid network S (gelatin); this SxW formula can be Another possibility, for food, is to freeze a W/​O emulsion using
made more precise, such as D3(W)xD3(S), when one adds the an oil fraction that solidifies below 0 °C (van Aken et al., 1999).
description of the dimensions of phases. Because of the continuity Of course, the formula holds only in a certain temperature
of the liquid phase S, such gels are said to be “connected”, which interval. This last example is a good opportunity to discuss the
means that a solute in the liquid phase can diffuse throughout the fact that the DSF formula can change upon heating or cooking,
whole system (Figure 55.2). because there can be phase transitions; when no particular
Gels
FIGURE 55.3 A simple program for generating gel descriptions using the DSF.
temperature is given, one assumes that it is room temperature. In
the food area, this is particularly important for systems including
fat phases, which can be liquid or solid at some degrees Celsius
apart (e.g., the solid–​liquid transition for chocolate is narrow,
near 34–​37 °C).
Another example formula in the produced list, D3(O)xD3(S),
describes two continuous networks, one of liquid oil and the other
solid. For example, this system can be obtained by cooking and
drying an O/​W emulsion for which the surfactants are proteins
containing cysteine residues: heating the system leads to protein
coagulation (D3(O)/​[D3(S)xD3(W)]), making a “gibbs” (This,
2009); drying it can lead to the desired gel (which I have named a
“graham”). “Natural” systems having this formula already exist,
such as chocolate at temperatures where a liquid continuous
“oil” phase is dispersed in a continuous solid network (Loisel
et al., 1997).
Of course, these objects could be termed using natural lan-
guage, ranking the information by order of “importance”: the
name “gel” is added by the indication on connection (connected,
non-​connected), and, when the liquid phase is organized as
“channels” (one dimension) or “sheets” (two dimensions), by
the indication “with channels” or “with sheets”, depending on
the dimension of dispersed objects (when objects of zero dimen-
sion are used, it is proposed not to give any particular indica-
tion, because the formula then corresponds to familiar systems);
finally, the nature of the dispersed objects is given (aqueous
solution, oil).
More Complex Gels
After this first list of simple gels has been made, it can be observed
that more than one phase (“water” or “oil”) can be dispersed in
the three-​dimensional, continuous, solid phase. This leads to the
possibility of connected or non-​connected gels with a dispersed
material which is polyphasic, such as an emulsion or a suspen-
sion. When gels have only one phase within the solid continuous
377
one, they are said to be of “class 1”; more generally, gels are of
“class k” (k being an integer) when k phases are in the continuous
solid network.
The whole list of all possible gels is infinite, because there is no
limit to the number of phases that can be included in the systems.
And some possibilities are really innovative, such as: connected
gels with sponge water–​oil phase (in sponge phases, the two-​
dimensional nature of the surfactant bilayer divides the space into
two volumes; the sponge phase itself corresponds to a symmetric
phase in which the two volumes are equal, the average curvature
of the surfactant film being exactly equal to zero) (Roux, 1995;
Wadsten et al., 2006); non-​connected gels with sponge water–​
oil phase; connected gels of aqueous channels; non–​connected
gels of aqueous channels; connected or non-​connected gels with
oil channels; etc. Sometimes, very long names such as “non-​
connected gel of liquid suspensions in oil sheets” are needed.
But using natural language to discuss the possible systems
is cumbersome. The DSF system makes it easier to envision
all possible gels. For example, the complex name given earlier
as an example (“non-​connected gel of liquid suspensions in oil
sheets”) corresponds to the formula [D0(S)/​D2(O)]/​D3(S). Here
again, a simple computer program (fewer than 50 lines) based on
DSF can produce all possibilities for the gels of the successive
classes. The simplest program (This, 2016), creating the formula
one by one, does not reject impossible formulas (such as D0(S)
xD3(W), with a phase of zero dimensions that would extend into
all three-​dimensional space), but again, the possible solutions can
be selected manually or digitally.
After this pruning, the list can be used to identify complex gels
in natural objects. For example, a sample of plant tissue (e.g.,
part of the root of Daucus carota L.) is made of parenchymatous
tissue and connective tissue (xylem and phloem) (Cazor et al.,
2006), and so the system has the formula [[D0(W1)/​]+[D1(W2)x]]
D3(S). In this formula, both “D0(W1)/​” and “D1(W2)x” operate on
the same D3(S) (Figure 55.4). Here, it is assumed that the cytosol
is a liquid (W1), but if it is more realistically considered that it is
itself a gel, then W1/​S′ could be used instead of W1.
378
FIGURE 55.4 This microscopic picture of part of an onion (Allium cepa L.)
bulb explains why it is a complex gel: the liquid content of parenchyma cells
is within a continuous solid network. Also, a channel (one-​dimensional struc-
ture) can be observed; it is full of the liquid sap (here colored blue).
Also, new gels can be produced using the formulas as guides:
for example, the gel D1(W)/​[D3(S)xD3(W)] can be made by dis-
persing cylinders of an aqueous gelatin gel in an agar-​agar gel in
which protease enzymes (EC 3.4) are dissolved; after the gelling
of agar-​agar, enzymes diffuse towards the gelatin gel, and they
destroy the solid gelatin network, leaving liquid channels in the
agar-​agar gel. Another example is the use of “chaotic mixers”
(Tabeling et al., 2004), where two solutions are injected, one an
aqueous solution and the other a solution of a gelling agent; after
dispersion and gelling, the result is a multi-​sheet gel.
Making new gels is only the first step. It is also interesting
to characterize them chemically, physically and biologically. It
has been proposed to use a general description called “compound
release”, along with the definition of a “matrix effect” (This,
2012; This, 2011; This, 2013b). This is useful for gels being used
in technology, in particular for formulation activities, because
compounds dissolved in the liquid phase can diffuse towards the
environment and vice versa.
For example, supramolecular interactions (such as van der
Waals forces, hydrophobic pseudo forces, hydrogen bonds and
disulfide bridges) between the solvent and the solid network
of gels can reduce the self-​diffusion coefficient of the solvent
molecules or of the dissolved compounds, so that the release
would be delayed (Matsukawa and Ando, 1996). Also, various
compartments can release compounds differently, such as plant
tissues (modelled as D1(W)xD3(S)+D0(W)/​D3(S) systems), for
which it was shown that the release of solutes in the sap is faster
than from parenchyma cells (This, 2012).
More than Two Phases: Dynagels/​Statgels
In all the previous sections, we considered gels with only two
phases, but many food systems are more complex dispersed
systems, with other phases than only solid and liquid. Such
systems can be easily envisioned using DSF, using the same
Hervé This vo Kientza
technique for building formulas as before. For example, one can
imagine connected gelled foams with aqueous sheets, or gelled
emulsions with non-​connected sheets, etc. For all such systems,
compound release can also be interpreted in terms of diffusion,
but also of chemical forces between species.
But there is more, as the previous analysis considered only
“static” gels, where solutes and solvent can diffuse in liquid
compartments (diffusion in the solid parts would be much
slower). They do not take into account the fact that another
analytical frame can introduce new phenomena and different
mechanisms. As said before, we can distinguish “physical gels”,
with a solid network made from the reversible assembly of con-
stituent molecules (for example, gelatin gels), and “chemical
gels”, with a fixed structure (This, 1996). But what if the bonding
energy between the subunits of the network is close to kB T? Here,
the network is no longer static but is, rather, dynamic, forming at
the same time as it deforms.
In chemistry, networks are often of polymeric nature, and this
observation calls for a discussion on dynamers, that is, dynamic
polymers that can reorganize because their subunits are linked
by supramolecular forces (Lehn, 2010). Dynamers are defined as
constitutional dynamic polymers, i.e., polymeric entities whose
monomeric components are linked through reversible connections
and which therefore have the capacity to modify their constitu-
tion by exchange and reshuffling of their components. They may
be of either supramolecular or molecular nature, depending on
whether the connections are non-​covalent interactions or revers-
ible covalent bonds.
For gels having dynamers as the polymeric network, the terms
“dynagels” and “statgels” have also been proposed by This (2016).
As for dynamers, dynagels can be defined as constitutional dynamic
gels, i.e., gels whose “monomeric” components of the solid network
are linked through reversible connections and which therefore have
the capacity to modify their constitution by exchange and reshuf-
fling of their components. They may be of either supramolecular or
molecular nature, depending on whether the connections are non-​
covalent interactions or reversible covalent bonds. It was recognized
that dynamers can be chemically dynamic, involving a reversible
chemical reaction, or physically dynamic, based on physical non-​
covalent interactions; the same idea holds for dynagels, as their
solid network is often a polymer. However, in this case, the descrip-
tion is made more complex, as the solvent, and possibly the solutes,
can interact with the solid network.
One example of dynagels is gelatin gels at a temperature
close to the critical melting point. Of course, the gelling of gel-
atin depends on the particular chemical composition of the
aqueous solvent in which the gelatin is dissolved (Bellini et al.,
2015), but a simple thermodynamical description can be made.
Let us assume, for example, that the continuous solid network
is obtained by the linking of a certain number n of subunits at
each node (for gelatin gels, n = 3) (Djabourov et al., 1988) with
a constant binding energy E for each subunit. In the assumption
of the Maxwell–​Boltzmann distribution, the proportion p(b,f) of
subunits being bound (b) on one end and free (f) at the other at a
certain absolute temperature T would be p (b, f ) = exp ( − E / kBT ),
kB being the Boltzmann constant. If the binding or release of a
particular end of a subunit does not depend on the bound or free
Gels
state of the other end, then the probability p(f,f) for a subunit to be
entirely free is simply proportional to exp ( −2 E / kBT ). The expo-
nential variation of this proportion as a function of temperature
explains why gelatin gel can be dynamic only within a narrow
range of temperatures. For example, when the temperature is
equal to 2E/​0.693kB, half the subunits would be free, which means
that the mechanical properties of the gel would be changed.
The compound released by statgels and dynagels having the
same DSF formula (i.e., the same physical organization) can be
very different, even with constant binding energy between the
solutes or the solvent, on the one hand, and the solid network,
on the other hand, because of a new component of diffusion of
monomers (with their surrounding solute and solvent molecules)
in the constantly reorganizing system. As for dynamers,
dynagels can exist in nature or can be created by intent. In order
to explore dynagels intentionally, one could first focus on the
non-​connected gels, so that diffusion of solutes through the
solvent could (by slow diffusion) progressively interact with
monomers and influence gelling. In this way, we would have
adaptive materials.
Not only do fixed systems having a particular DSF formula
need to be considered, but also the equilibrium between “limit”
DSF formula must be studied, creating new diffusion issues for
dynagels. Concerning compound release, how can the know-
ledge of the matrix effect for the limit formula allow the calcu-
lation of the matrix effect of the dynagel? Other questions may
also be asked, such as how the dynamic reorganization of such
colloidal systems changes the self-​diffusion coefficient of solutes
or solvent.
Conversely, one can also ask how the presence of solutes that
have specific chemical affinities for certain monomers would
change the equilibrium of monomer association–​dissociation. If
solutes do indeed change such equilibra, one can envision new
ways of controlling the formation of dynagels, as effectors can
change the equilibrium of dynamers (Lehn, 2005). Dynagels
are particular cases of dynamic materials called “dynamats”
(Lehn, 1999a), these systems being defined as materials whose
constituents are linked through reversible connections (non-​
covalent or covalent) and are able to continuously reorganize
through assembly/​ disassembly processes and exchange of
components in a given set of conditions, usually under thermo-
dynamic control (Lehn, 1999b), but eventually involving kinetic
bottlenecks or traps. As dynamers and dynagels are adaptable
and self-​assembling systems (Lehn, 1995; Atwood et al., 1996;
Philip and Stoddart, 1996) capable in principle of selecting their
components in response to external stimuli or to environmental
factors, so that they behave as adaptive materials (Lehn, 1999a;
Lehn, 1999b; Lehn, 2002a).
Like supramolecular dynamers, supramolecular dynagels
are defined as entities whose solid network is generated by the
poly-​association of molecular monomers bearing complemen-
tary binding groups capable of connecting through the usual
non-​covalent interactions implemented in supramolecular
chemistry: electrostatic, hydrogen bonding, donor–​ acceptor,
van der Waals as well as metal ion coordination. Of course, the
building of the solid network of dynagels imposes some par-
ticular characteristics on supramonomers, as for dynamers, and
379
in particular, the presence of more than two binding sites (other-
wise, the suprapolymer would be linear). Molecular dynagels,
like molecular dynamers, are reversible covalent systems, which
open a range of perspectives to polymer chemistry.
Finally, as for double dynamic polymers, one can envision
double dynagels, combining monomers that bear complementary
non-​covalent interaction units as well as complementary revers-
ible functional groups. This allows the generation of polygels
presenting double dynamic behaviour, gels that are dynamic on
both molecular and supramolecular levels (to our knowledge, this
has so far not been done).
In conclusion, the operation of the various new kinds of gels in
chemistry, and more generally in material science, confers new
physical and chemical characteristics on the gel entities. In par-
ticular, dynagels can have many of the properties of dynamers,
including the ability to respond to external stimuli and to envir-
onmental conditions, i.e., adaptability, a major tenet of consti-
tutional dynamic chemistry, which enables the development of
adaptive chemistry (Lehn, 2002a). The further exploration of
these features may be expected to open wide perspectives for
basic research in colloidal science as well as to give access to a
range of novel properties and applications in the culinary arts.
REFERENCES
Abdallah DJ, Weiss RG. 2000. Organogels and low molecular mass
organic gelators. Advanced Materials, 12, 1237–​1247.
Atwood JL, Davies JED, MacNicol DD, Vogtle F, Lehn JM. 1996.
Comprehensive Supramolecular Chemistry, Pergamon Press,
Oxford.
Bellini F, Alberini I, Ferreyra M, Rintoul I. 2015. Absolute determin-
ation of the gelling point of gelatin under quasi-​thermodynamic
equilibrium. Journal of Food Science, 80 (5), C-​935–​941.
Cazor A, Deborde C, Moing A, Rolin D, This H. 2006. Sucrose, glu-
cose, and fructose extraction in aqueous carrot root extracts
prepared at different temperatures by means of direct NMR
measurements. Journal of Agricultural Food Chemistry, 54,
4681–​4686.
Djabourov M, Leblond P, Papon P. 1988. Gelation of aqueous
solutions. II. Rheology of the sol gel transition. Journal de
Physique France, 49, 333–​343.
IUPAC. 2001. Manual of Symbols and Terminology for
Physicochemical Quantities and Units, http://​old.iupac.org/​
reports/​2001/​colloid_​2001/​manual_​of_​s_​and_​t/​node33.html
IUPAC. 2007. Definitions of terms relating to the structure and pro-
cessing of sols, gels, networks, and inorganic-​organic hybrid
materials (IUPAC Recommendations 2007). Pure and Applied
Chemistry, 79, 1801–​1806.
Kawano SI, Fujita N, Shinkai S. 2004. A coordination gelator that
shows a reversible chromatic change and sol−gel phase-​
transition behavior upon oxidative/​reductive stimuli. Journal
of the American Chemical Society, 126 (28), 8592–​8593.
Lehn JM. 1995. Supramolecular Chemistry: Concepts and
Perspectives, VCH, Weinheim.
Lehn JM. 1999a. Supramolecular Science: Where It Is and Where It
Is Going, Ungaro R and Dalcanale E (eds.), Kluwer, Dordrecht,
287–​304.
Lehn JM. 1999b. Dynamic combinatorial chemistry and virtual com-
binatorial libraries. Chemistry, the European Journal, 5 (9),
2455–​2463.
Lehn JM. 2002a. Supramolecular polymer chemistry –​scope and
perspectives. Polymer Internationa., 51, 825–​839.
380
Lehn JM. 2002b. Toward complex matter: Supramolecular chemistry
and self-​organization. Proceedings of the National Academy of
Sciences USA, 99, 4763–​4768.
Lehn JM. 2005. Dynamers: Dynamic molecular and supramolecular
polymers. Progress in Polymer Science, 30, 814–​831.
Lehn JM. 2010. Dynamers: Dynamic molecular and supramolecular
polymers. Australian Journal of Chemistry, 63, 611–​623.
Loisel C, Lecq G, Ponchel G, Keller G, Ollivon M. 1997. Fat bloom
and chocolate structure studied by mercury porosimetry.
Journal of Food Science, 62, 781–​788.
Ma L, Barbosa-​ Canovas GV. 1995. Rheological characterization
of mayonnaise. Part II: flow and viscoelastic properties at
different oil and xanthan gum concentrations. Journal of Food
Engineering, 25, 409–​425.
Matsukawa S, Ando I. 1996. A study of self-​diffusion of molecules
in polymer gel by pulsed-​ gradient spin-​echo 1H NMR.
Macromolecules, 29 (22), 7136–​7140.
Mitchell JR. 1980. The rheology of gels, Journal of Texture Studies,
11(4), 315–337.
Philip D, Stoddart JF. 1996. Self-​assembly in natural and unnatural
systems. Angewandte Chemie International Edition in English,
35, 1154–​1196.
Qiu Y, Park K. 2001. Environment-​ sensitive hydrogels for drug
delivery. Advanced Drug Delivery Review, 53, 321–​339.
Roux D. 1995. Sponge phases: An example of random surfaces.
Physica A, 213, 168–​172.
Tabeling P, Chabert M, Dodge A, Jullien C, Okkels F. 2004.
Chaotic mixing in cross-​channel micromixers. Philosophical
Transactions of the Royal Society of London A, 362 (1818),
987–​1000.
Terech P, Weiss RG. 1997. Low molecular mass gelators of organic
liquids and the properties of their gels. Chemical Review, 97,
3133–​3159.
Hervé This vo Kientza
This H. 2016. Statgels and dynagels. Notes Académiques de
l’Académie d’agriculture de France, /​Academic Notes from
the French Academy of Agriculture, 12, 1–​12.
This H. 1996. Can a cooked egg be uncooked. The Chemical
Intelligencer, 10, 51.
This H. 2007. Formal descriptions for formulation. International
Journal of Pharmacy, 344 (1–​2), 4–​8.
This H. 2009. Molecular gastronomy. Accounts of Chemical
Research, 42(5), 575–​583.
This H. 2011. De quelles connaissances manquons-​nous pour la
cuisine note à note? Actualité Chimique, 350, 5–​9.
This H. 2012. Solutions are solutions, and gels are almost solutions.
Pure and Applied Chemistry, http://​dx.doi.org/​10.1351/​PAC-​
CON-​12-​01-​01, pp. 1–​20.
This H. 2013a. Molecular gastronomy is a scientific discipline, and
note by note cuisine is the next culinary trend. Flavour, 2 (1).
doi: 10.1186/​2044-​7248-​2-​1.
This, H. 2013b. Bioactivité et effet de matrice. Actualité Chimique,
370 (1), 15–​19.
Van Aken GA, ten Grotenhuis E, van Langevelde AJ, Schenk
H. 1999. Composition and crystallization of milk fat
fractions. Journal of the American Oil Chemists’ Society, 76,
1323–​1331.
Wadsten P, Wöhri AB, Snijder A, Katona G, Gardiner AT, Cogdell
RJ, Neutze R, Engström S. 2006. Lipidic sponge phase crystal-
lization of membrane proteins. Journal of Molecular Biology,
364 (1), 44–​53.
Zhang Y, Gu H, Yang Z, Xu B. 2003. Journal of the American
Chemical Society, 125, 13680–13681.
Zhang Y, Yang Z, Yuan F, Gu H, Gao P, Xu B. 2004. Journal of the
American Chemical Society, 126, 15028–15029.
Heat Transfer in Culinary Sciences

Denis Flick
AgroParisTech, 16 rue Claude Bernard, Paris, France

Introduction Fluid flow can be caused by a fan (e.g., in a ventilated oven) or

Why is heat transfer so important in culinary science and tech-
nique? This is because temperature has a major influence on
chemical reaction rates, phase changes and biological transform-
ations. A rise of temperature above specific levels results in starch
gelatinization, protein denaturation, Maillard reactions, destruc-
tion of microorganisms and inactivation of enzymes. Conversely,
decreasing temperature below a certain level allows water or fat
crystallization, slows down microbial growth, etc.
Heat Transfer Modes
Heat can be transferred by conduction, convection and radiation
(see, for example, Transport Phenomena by Bird et al. 2002 for
further explanations).
Conduction occurs from a warm to a colder zone, even without
any movement, according to Fourier’s law:
∂T
in the x direction : j x = − k (56.1)
∂x
 
in three dimensions : j = − k∇T
   ∂T ∂ T ∂ T 
( )
where j = j x j y jz and ∇T =  , ,
 ∂x ∂y ∂z 
(56.2)
where j is the heat flux expressed in W.m-​2 and k is the thermal
conductivity, which is around 0.5 W.m−1K−1 for water and food
products containing high levels of water and around 0.03 W.
m−1K−1 for insulating materials or air.
Convection occurs when a fluid (liquid or gas) at temperature
Tf is flowing near a solid surface having a different temperature
Ts, according to Newton’s law:
in the direction normal to
(56.3)
(
the surface : j = h Ts − T f ) for very humid air or steam, because water vapor significantly
where j is the heat flux and h (in W.m−2K−1) is the heat transfer
coefficient (HTC), which depends on the geometry, the fluid
properties, the fluid velocity and the turbulence intensity.
by agitation (e.g., when stirring a sauce), and this is called forced
convection. However, even without mechanical agitation, a warm
fluid tends to flow upwards and a cold one downwards because
of density differences. For example, the liquid at the bottom
of a saucepan is warmer than the rest of the liquid, leading to
a fluid flow, which can eventually homogenize the temperature
if the viscosity is not too high. In an unventilated oven, when
hot air comes into contact with a colder product, its temperature
decreases and it tends to flow downwards. This is called natural
or free convection.
The heat transfer coefficient between a surface and air is typ-
ically between 10 Wm−2K−1 in free convection and 100 Wm−2K−1
in forced convection. For low-​ viscosity liquids (e.g., milk)
it is between about 200 Wm−2K−1 in free convection and
2000 Wm−2K−1 in forced convection.
The exchanges by radiation are like the exchange of visible
light. Light is electromagnetic radiation with wavelength between
400 nm (purple) and 800 nm (red). However, most of the radia-
tive heat transfers occur in the infrared (IR) range (wavelength
of some micrometers). Radiation is emitted by all solid surfaces
with an intensity and a range of wavelengths depending on the
surface temperature. The sun (about 5800 K at its surface) emits
visible light, whereas the wall of an oven (200 °C = 473 K) or
a warm product introduced into a refrigerator (20 °C = 293 K)
emits IR radiation. Radiation is partly reflected and partly
absorbed (converted into heat) by a surface. Light is reflected by
a mirror or scattered by a white surface, totally absorbed by a
black surface, and partly absorbed/​scattered by a grey surface.
The same is true for IR, as a polished aluminum sheet reflects
IR, while most food products are approximately black bodies for
IR; they absorb almost all the incident radiation. IR radiation can
also be absorbed by fluids, and even a thin layer of liquid absorbs
all the incident IR radiation; thus, liquids are like opaque bodies
for IR. The air in an oven absorbs part of the IR radiation along
a ray of IR (from the wall to the product). Over a short distance,
this effect is small for relatively dry air, but it becomes important
absorbs IR.
The surface to surface exchanges by radiation depend on
the geometry. For the simple case of two parallel planes (large
compared with the gap between them), one being a black body

381
382 Denis Flick

and the other having an emissivity ε (1 for a black body, 0 for an Heat has to overcome the external resistance of heat transfer
aluminum sheet), the net exchange of radiative flux is: by convection from the fluid to the product surface, Rth.ext, and
the internal resistance by conduction from surface to core,
j = εσ(T14 − T2 4 ) (56.4) Rth.int. These resistances are equal to 1/(hA) and R/(kA), respect-
ively (for a plate), where A is the product surface area. Therefore,
where σ = 5.67 × 10 −8 Wm−2K−4 is the Stefan–​ Boltzmann we can first consider two extreme configurations where one of the
constant. resistances is negligible compared with the other one. The Biot
Another kind of radiation occurs in the microwave range number compares these resistances:
(wavelength of 12 cm for a domestic microwave oven). This kind Bi = Rth.int/​ Rth.ext, so Bi = hR/​k for a plate. In general,
of radiation penetrates much further inside a food product than Bi = hLc/​k, with Lc the characteristic length of the product.
IR radiation. Thus, the absorption of microwaves allows a direct In case of external transfer limitation, Bi << 1 (typically Bi
heating of each part of the volume of the product through water < 0.2 for a first approximation, Bi < 0.1 for an accurate pre-
molecule vibration. diction), a global heat balance can be written. Indeed, in this
case, the surface/​core temperature difference is small compared
with the product/​fluid temperature difference: Ts –​ Tc << Tf –​ Ts.
So, the almost uniform product temperature T evolves according
Internal Energy Variation to (56.6):
The energy needed to increase by 1 °C (or 1 K) the temperature
of 1 kg of a food product (without freezing or evaporating water)
is called the specific thermal capacity: c (in J.kg−1K−1).
mc
dT
dt
(
= A.h T f − T ) (56.6)
Its value for pure water is 4180 J.kg−1K−1; for most food
products containing high levels of water, it is around 3500 If the fluid temperature Tf is constant:
J.kg−1K−1.
Thermal energy due to molecular agitation is called internal Tf − T  −t 
energy in thermodynamics. Without phase changes, when the = exp   (56.7)
T f − T0  τ
temperature of a mass m of product increases by dT, its internal
energy variation is: mc
where τ = is the characteristic time.
hA
dU = m.c.dT (56.5) In the case of internal transfer limitation: Bi >> 1 (typically Bi
> 10 for a first approximation, Bi > 50 for an accurate prediction),
Additionally, to melt 1 kg of ice, the so-​called latent heat a local heat balance has to be written. Indeed, in this case, the
of fusion is required: Lf = 333 kJ.kg−1. Similarly, to vaporize surface/​core temperature difference is large compared with the
1 kg of liquid water, the latent heat of vaporization is required: product/​fluid temperature difference: Ts –​ Tc >> Tf –​ Ts. The sur-
Lv = 2257 kJ.kg-​1 at 100 °C. Vaporization can occur by evap- face temperature is almost equal to that of the fluid. The product
oration or ebullition. Evaporation of pure water occurs at its temperature T depends on both time and position (r = 0 at core,
surface when saturated water vapor pressure at the liquid tem- r = R at surface). The local heat balance has the form of a partial
perature is higher than the partial pressure of water vapor in the derivative equation:
air. Evaporation occurs, for example, when an initially wet sur-
face dries naturally in non-​saturated air at ambient temperature. ∂T 1 ∂  β ∂T 
ρc = r k (56.8)
Ebullition (boiling with vapor bubbles inside the liquid) occurs ∂t r β ∂r  ∂r 
when the saturated water vapor pressure at the liquid tempera-
ture reaches the total pressure. This corresponds to 100 °C for
boiling at atmospheric pressure (at sea level): 1.013 × 105 Pa. At where β = 0 for a plate, 1 for a cylinder and 2 for a sphere, with
lower pressure (on a mountain, for example), ebullition occurs at T = T0 at t = 0 (initial condition) and T = Tf at r = R (boundary
lower temperature; at higher pressure (inside a pressure cooker, condition).
for example), ebullition occurs at higher temperature (~112 °C). For constant product properties:
The latent heat of vaporization slightly depends on temperature
(for example: Lv = 2454 kJ.kg−1 at 20 °C). ∂T α ∂  β ∂T 
= r (56.9)
∂t r β ∂r  ∂r 

k
where α = is the thermal diffusivity.
Heat Balance Equations ρc
Let us consider the heating of a food product of simple geom-
etry –​plate, cylinder or sphere of radius or half thickness R –​ by R2
The characteristic time is then : τ ’ = (56.10)
a hot fluid (water or air) of temperature Tf . Knowing the initial α
temperature of the product, T0, how much time is needed to reach
a given core temperature Tc? where R is the radius or the half thickness for a plate.
Heat Transfer in Culinary Sciences 383

The core temperature can be approximated by the following Process parameters (boiling water): h = 1000 Wm−2K−1,
expression: Tf = 100 °C
Question: time to reach Tc = 75 °C (to obtain starch
T f − Tc  4  − π2 t   gelatinization)?
≈ min  1, exp    for a plate and
T f − T0  π  4 τ '   Answer:
(56.11) Bi = h.R/​k = 62.5 > 50: internal transfer is limiting.
T f − Tc   t 
≈ min  1, 2exp  − π 2   for a sphere.
T f − T0   τ '  Thermal diffusivity: α = k / (ρc ) = 1.20 × 10 −7 m 2s −1
(56.15)
The time necessary for the core temperature to reach the mean
value of initial and fluid temperature (t at which Tc = (T0 + Tf)/​2) R2
is 0.38 τ′ for a plate, 0.20 τ′ for a cylinder and 0.14 τ′ for a sphere. Characteristic time : τ ' = = 7500 s (56.16)
α
In intermediate cases, where Bi ≈ 1 (typically 0.2 <Bi <10),
the local (partial derivative) heat balance equation is the same as T f − Tc 100 − 75  t
previously, but the boundary condition has to take into account = ≈ 2exp  − π 2 
T f − T0 100 − 20  τ
the convective heat transfer resistance: (56.17)
t = 0.188 τ ′ = 1410 s = 23 min
At r = R, − k
∂T
∂r
(
= h T − Tf ) (56.12)
For potatoes 2 cm in radius instead of 3 cm, time is reduced by
a factor of (2/​3)² = 0.44.
The exact solution of many convection/​conduction problems
can be found in Carslaw and Jaeger (1959), but often the charac-
teristic times can give a rough estimation of the necessary time
needed for a heat treatment. Their expressions also indicate how Introduction to Coupled Phenomena
these times depend on the different parameters. Baking Bread
Baking bread in an oven is not a simple problem of convection/​
conduction of heat in a solid. A major additional phenomenon is
Basic Applications the evaporation of water, which occurs not only at the surface but
also at the dough/​gas interface of each bubble inside the bread.
Refrigerating a Jelly
As the outer part of the product becomes dryer (crust formation),
Geometry: plate of e = 2 cm thickness, D = 20 cm in diameter water migrates from the core to the surface by liquid diffusion.
(e << D) Water migrates also by evaporation/​condensation; on the warm
Product parameters: k = 0.6 W.m−1.K−1, ρ = 1000 kg.m−3, side of a bubble, water evaporates, and the water vapor generated
c = 4000 J.kg−1.K−1, T0 = 20 °C diffuses through the gas of the bubble and condenses on the colder
Process parameters (refrigerator): h = 10 Wm−2K−1 (free con- side of the bubble. Since vaporization is highly endothermic, heat
vection), Tf = 4 °C and mass transfer are closely coupled.
Question: time to reach T = 8 °C (in order to gel)? Different mechanisms tend to inflate the bubbles; temperature
Answer: rises, water evaporates, and CO2 is produced and desolubilized.
Bi = h.R/​k = 0.16 < 0.2: to a first approximation, the external Inversely, the visco-​elastic stresses in the dough slow down the
transfer is limiting. expansion. This expansion modifies the heat and mass transfer.
In turn, the rheology of the dough evolves considerably because
Tf − T  −t  mc ρ (e / 2 ) c of starch gelatinization (depending on temperature) and dehydra-
= exp   with τ = = = 4000s
T f − T0  τ hA h tion. Thus, heat/​mass transfer is coupled with the deformation
(mechanics) and the product transformation. Finally, when the
(56.13)
temperature approaches 100 °C, boiling occurs, the gas pressure
 T f − T0  increases, and a flow of water vapor takes place from core to sur-
t = τln   = 5500s = 1.5 hours (56.14) face, as in a porous medium (Darcian flow). More or less real-
 Tf − T  istic/​complex models have been developed to take into account
these phenomena, which can be solved only numerically (Flick
For a thickness of 1 cm instead of 2 cm, time is reduced by a et al. 2015).
factor of two.
Freezing Ice Cream
Cooking Potatoes in Boiling Water The production of ice cream or sorbet usually begins in a scraped-​
Geometry: sphere of radius R = 3 cm surface freezer (batch for domestic production, continuous for
Product parameters: k = 0.48 W.m−1.K−1, ρ = 1090 kg.m−3, industrial production) to reach a temperature around −5 °C and
c = 3640 J.kg−1.K−1, T0 = 20 °C an ice fraction of about 30% before hardening at −18 °C. This
384
first step is very important to obtain a smooth texture. If the
ice crystals are too large, the texture is rather sandy. Ice crystal
appearance (nucleation) and growth depend on the difference
between the actual temperature and the saturation temperature,
which depends on the local sucrose concentration. Therefore, to
predict ice crystal size distribution, one way is to couple fluid flow,
heat transfer (including phase change) and a population balance
model that predicts the evolution of the number of ice crystals by
volume of different sizes (Benkhelifa et al. 2008). The problem
is difficult to solve numerically, notably because the apparent vis-
cosity increases by a factor of 50 when the ice fraction passes
from 0% to 30%. Another important aspect is the air incorpor-
ation and the bubble fragmentation, because the bubbles also par-
ticipate in the smooth texture of ice cream and sorbets.
Denis Flick
REFERENCES
Benkhelifa H, Haddad Amamou A, Alvarez G, Flick D. 2008.
Modelling fluid flow, heat transfer and crystallization in a
scraped surface heat exchanger. Acta Horticulturae, 802,
163–​170.
Bird RB, Stewart WE, Lightfoot EN. 2002. Transport phenomena.
John Wiley.
Carslaw HS, Jaeger JC. 1959. Conduction of heat in solids. Oxford
Science Publications.
Flick D, Doursat C, Grenier D, Lucas T. 2015. Modelling of baking
processes. In Modelling food processing operations. Bakalis
S., Knoerzer K. Fryer P. eds. Woodhead Publishing.
Hydrocolloid Usages as Gelling and Emulsifying Agents for Culinary
and Industrial Applications

Rachel Edwards-​Stuart1 and Reine Barbar2,3

1 Culinary Science Department, Westminster Kingsway College, 76 Vincent Square. London SW1P 2PD, United Kingdom
2 UMR IATE Univ. Montpellier, INRAE, Institut Agro, 2 Place Pierre Viala, F-​
34060 Montpellier Cedex, France
3 Department of Agriculture and Food Engineering, School of Engineering, Holy Spirit University of Kaslik, P.O. Box 446

Jounieh, Mount Lebanon, Lebanon

Introduction Besides functional attributes, they are of current interest due

largely to their dietary fiber aspect (Chawla and Patil, 2010;
Structuring foods during manufacturing is mainly done by a
Brownlee, 2011). These components provide viscosity and play
complex assembly of various food materials such as water,
a role in developing foods with satiating capacity. The satiety
proteins, saccharides, lipids, emulsifiers, minerals and other
effect is related to slowing down enzyme action efficacy and/​or
minor components. Among these components, proteins and
delaying gastric emptying (Morell et al., 2014).
polysaccharides are the most utilized materials in food produc-
Food colloids have an impact on food properties when used at
tion (Figure 57.1). They play a synergistic role in structuring the
levels ranging from a few parts per million for carrageenans in
bulk phases of colloidal systems (and are therefore referred to
heat-​treated dairy products to high levels of acacia gum, starch
as hydrocolloids) while also providing interface-​stabilizing prop-
or gelatin in jelly confectionery. The primary reason behind the
erties through a combination of various molecular interactions
widespread use of hydrocolloids in foods is their ability to modify
with/​in the dispersed phases (Wijaya et al., 2017). They
the rheology of food systems. This includes two basic properties
function as thickeners, gelling agents, emulsifiers, stabilizers, fat
of food systems, namely flow behavior (viscosity) and mech-
replacers, clarifying agents, flocculating agents, clouding agents
anical solid property (consistency) (Milani and Maleki, 2012).
and whipping agents; additionally, they have applications in the
The modification of rheological characteristics is also helpful in
areas of edible films encapsulating odorant and taste molecules,
modifying foods’ sensory properties (Valdez, 2012). Hence, they
and inhibiting crystallization (Li and Nie, 2016).
are employed as significant food additives. Examples include
A hydrocolloid can be defined as a hydrophilic (polymeric)
alginic acid (E400), sodium alginate (E401), agar-​agar (E406),
material with colloid-​like character, or a highly water-​soluble (or
carrageenans (E407), gum arabic (E414), xanthan gum (E415)
water-​dispersible) material that readily dissolves (or disperses) to
and gellan gum (E418), among others (Branen et al., 2002).
form highly hydrated entities of colloidal (nanoscale) dimensions
Most hydrocolloids are polysaccharides and are grouped
(Dickinson, 2017). A colloid can be classified as a dispersion
according to their source. Therefore, gum karaya, gum traga-
of one material state in another, in which the dispersed compo-
canth, gum ghatti, gum arabic and other acacia gums are
nent typically has dimensions in the micrometer length scale.
assembled under the tree exudate group, while agar-​agar, alginate,
Hydrocolloids are commonly used by the food industry and
carrageenans, furcellarans and fucoidans are classified as the sea-
are a widely encountered structural component present in many
weed group. Additional gum-​like substances, such as pectins and
different food systems (Li and Nie, 2016). They include a range
psyllium, are categorized in separate plant groups, while gelatin
of polysaccharides and proteins that are now also parts of the
and chitin are included in the animal group (Li and Nie, 2016).
culinary cupboard. In this chapter, we will be mainly focusing on
Another practical approach to classification would be the use of
polysaccharides.
chemical structure to classify hydrocolloids (Table 57.1).
According to a Zion (2017) market research report, the global
In this chapter, the role of hydrocolloids as gelling agents and
food hydrocolloids market was valued at around USD 6.57 billion
also as emulsifiers will be detailed. By interacting with water
in 2017 and is expected to reach approximately USD 9.12 billion
mainly via hydrogen bonds, hydrocolloids hydrate and tend to
by 2024, growing at a compound annual growth rate of around
interact with each other to form entangled networks and thus act
4.8% between 2017 and 2024. Currently, there is an increasing
as effective gelling agents. This can be achieved through varying
demand for “natural” food products by health-​conscious con-
mechanisms of interaction, including hydrogen bonding (such
sumers, and this became one of the driving factors that boosted
as for agar-​agar), electrostatic cross-​linking (low methoxylated
the development of the hydrocolloids market (Li and Nie, 2016).

385
386 Rachel Edwards-Stuart, Reine Barbar

pectin, sodium alginate in combination with calcium ions), and
hydrophobic interactions (hydroxypropylmethylcellulose) (Day
and Golding, 2016).
FIGURE 57.1 Hydrocolloids as structuring agents contribute to the texture,
nutritional and functional properties of foods.
(From Gao et al., 2017)
However, this characteristic as a viscosity modifier can also
add to their role as stabilizing agents for emulsions. A stabil-
izing agent is an ingredient that confers long-​term stability
on an emulsion, possibly by a mechanism involving adsorp-
tion, but not necessarily so. In oil in water (O/​W) emulsions,
the stabilizing action of hydrocolloids such as xanthan,
carboxymethylcellulose, carrageenans, etc. is traditionally
attributed to the structuring, thickening and gelation of the
aqueous continuous phase (Dickinson, 2009a; 2009b). By
enhancing the viscosity of the medium, hydrocolloids facili-
tate effective immobilization of dispersed oil droplets, thereby
inhibiting gravity-​induced creaming and any accompanying
serum separation during extended emulsion storage (Dickinson,
2004; 2009a; 2009b).
The emulsifying potential of some hydrocolloids, such as gum
arabic, is also due to the presence of a surface-​active compo-
nent in their structure, such as a proteinaceous part. An emul-
sifying agent is a surface-​active ingredient that adsorbs at the
newly formed oil–​water interface during emulsion preparation,
and it protects the newly formed droplets against immediate
recoalescence (Dickinson, 2009a,b).
This chapter was written from both culinary science and
food-​processing perspectives. The first part deals with the main
polysaccharides that act as gelling agents, and the second part
focuses on some of their emulsifying usages as surface-​active
agents.

TABLE 57.1
Classification of Hydrocolloids
Basis Class Example
Origin Plant Pectins, inulin, gum arabic, gum ghatti, gum tragacanth, gum karaya, cassia seed gum, basil seed gum,
mesquite seed gum, fenugreek gum, chicle gum, oat gum, rye gum, konjac, psyllium, guar gum,
locust bean gum, flaxseed gum, wattle gum, starches
Animal Chitin, chitosans, gelatin, caseinate, whey protein, soy protein, egg white protein
Seaweed Agar-​agar, carrageenans, alginic acid, alginate, furcellarans, ulvans, fucoidans, red alga xylans
Microbial Xanthan, gellan gum, tara gum, dextran, pullulan, welan gum, curdulan, levan
Synthetic Methylcellulose, methylethylcellulose, carboxymethylcellulose, hydroxyethylcellulose,
hydroxypropylcellulose, hydroxypropylmethylcellulose, microcrystalline cellulose
Structure Glucans Starch, oat gum, barley gum, curdulan, welan gum, pullulan, dextran
Fructans Inulin, levan
Xylans Red alga xylan
Rhamnans Ulvans
Galactomannans Guar gum, locust bean gum, tara gum, cassia seed gum, basil seed gum, mesquite seed gum,
fenugreek gum
Glucomannans Konjac, alginate
Arabinoxylans Psyllium, flaxseed gum (containing another galacturonan fraction), rye gum, wheat gum
Galactans Agar-agar, carrageenans, fucoidans, furcellarans
Arabinogalactans Gum arabic
Galacturonans Pectin
Glycano-​rhamnogalacturonans Gum karaya, gum tragacanth (containing another arabinogalactan fraction)
Glycano-​glucuronomannoglycans Gum ghatti
Glucosamine polymer Chitin, chitosans
Proteins Gelatin, caseinate, whey protein, soy protein, egg white protein

Source: Williams and Phillips, 2009; Li and Nie, 2016.

Hydrocolloid Usages
Use of Hydrocolloids as Gelling Agents
According to the International Union of Pure and Applied
Chemistry, a gel is defined as “a non-​fluid colloidal network or
polymer network that is expanded throughout its whole volume
by a fluid” (IUPAC, 2019). For many years, gelatin has been the
gelling agent used most commonly by chefs and home cooks to
set liquids. Gelatin is often quoted as having the same melting
point as body temperature, but varying both the type of gelatin
and the gelling conditions has been shown to reduce the melting
temperature of gelatin gels by several degrees (Osorio et al.,
2007). When placed in the mouth, gelatin gels will melt back
to a relatively thin liquid (but one with body), creating a unique
“melt-​in-​the-​
mouth” experience, which is thought to account
for its popularity within the world of gelling agents. However,
due to gelatin’s unsuitability for a number of different dietary
and ethnic groups, there has been a demand recently to develop
non-​meat-​derived gelling agents. These include agar-​agar, gellan
gum, carrageenans, sodium alginate (when coupled with cal-
cium) and modified celluloses. Cellulose and its derivatives will
be discussed in a separate chapter in this book.
These vegetarian gelling agents have been used in the food
industry for many years for a number of different applications,
including the prevention of moisture migration in bakery fillings,
creating the pimento gel that is used to stuff olives, and providing
texture in dairy desserts. A number of years ago, they started
to appear in the kitchen (although agar-​agar and carrageenans
have been used for some time in their natural form in Asian and
Irish cooking). Nowadays, chefs are experimenting with them to
create more unusual textures in their cooking, such as to make
fake caviar and fluid gels, as well as exploiting their ability to
produce gels that can be served hot (see chapter on Molecular
Cooking). However, due to their elevated melting points, which
in itself can offer benefits, they do not offer the same desirable
FIGURE 57.2 Mechanisms of action of hydrocolloids as thickening (top) and gelling agents (bottom). Blue circles indicate water molecules, and black lines
represent the hydrocolloids.
387
melt-in-the-mouth sensation that gelatin does, so they will
struggle to completely replace its unrivalled success as a gelling
agent.
In the presence of water, hydrocolloids will form associations
with water molecules –​this is how they act to thicken solutions,
and at certain concentrations or under certain conditions the
hydrocolloid molecules will start to form a network with each
other, trapping the water molecules completely and resulting in
the formation of a gel (Figure 57.2).
Unlike the traditional gelling and thickening agents gelatin and
flour, these hydrocolloids have the ability to thicken and form
gels at very low concentrations, often as low as 0.2%.
Agar-​Agar
Agar-​agar is derived from red algae, or Rhodophyceae. It is per-
mitted for use as a food additive in Europe and has the E-​number
of E406 (European Union, 2006).
Chemical Composition/​Structure
Agar-​agar is a mix of two components –​agarose, a firmly gelling
polysaccharide, and agaropectin, a weakly gelling polymer. When
we talk about the gelling properties of agar-​agar, we are generally
referring to agarose. Agarose is a linear polymer composed of
repeating units of D-​galactose and 3,6-​anhydro-​L-​galactose, as
shown in Figure 57.3.
The Gelation of Agar-​Agar
When hydrated in boiling water, agar-​agar molecules take the
form of random coils that are distributed homogeneously in solu-
tion. On cooling, they form helices between two single agarose
chains (Vilgis, 2015). However, the formation of these helices is
not enough to explain gelation –​a gel will only form when these
helices associate further, as shown in Figure 57.4.
388
A two-​step gelation mechanism for agarose has therefore been
proposed: firstly, the formation of a gel by joining the randomly
distributed coils into a double helical association by hydrogen
bonds, followed by the aggregation of the double helices into a
tight three-​dimensional network (Vilgis, 2015). The aggregated
double helices are sufficiently large to refract light, which is why
even the purest of agar-​agar gels are slightly opalescent/​turbid
(Coultate, 2009).
Unlike most of the other gelling agents described in this
chapter, agar-​agar molecules gel well without any cations pre-
sent. The texture of agar-​agar gels is therefore not affected by
variations in cation levels in the food, such as calcium in dairy
and fruit products (Imeson, 2010).
Agar-​agar gels are easy to prepare. Agar-​agar is insoluble in
cold water, so the liquid that is to be gelled needs to be heated
to above 85 °C in order to fully hydrate the agar-​agar (Vilgis,
2015); firm and brittle gels will form on cooling to below about
30–40 °C. Agar-​agar gels can be formed at concentrations as low
FIGURE 57.3 Structure of the structural unit of agar-​agar: agarose, or
poly(-​(β1-​4)-​3,6-​anhydro-​L-​galactose-​(α1-​3)-​D-​galactose-​).
FIGURE 57.4 Mechanism of agar-​agar gelation.
Rachel Edwards-Stuart, Reine Barbar
as 0.2% (Vilgis, 2015), but levels in food products are typically
between 0.5% and 2.0% (Imeson, 2010). In the experience of
one of the authors (Edwards-​Stuart), the most common reason
an agar-​agar solution fails to gel is because it has not been heated
to a high enough temperature. Not adding enough gelling agent
or insufficient mixing could potentially be the cause, but since
agar-​agar gels can set at a very low concentration (although the
exact concentration depends on its source) and they are readily
hydrated by stirring or whisking, heating is often the most likely
cause for a gel not setting. In order to ensure that it has been
heated sufficiently, one could bring the mixture to the boil for a
couple of minutes while thoroughly stirring and then cool in the
required mould.
External Effectors and Synergies
Agar-​agar is a versatile gelling agent and can be used with a wide
range of foodstuffs, although significant quantities of tannic acid
(which can be found in fruits like quince, some apple varieties
and plums) can markedly inhibit the gelling process (Imeson,
2010). Agar-​agar forms gels over a wide pH range, and the neu-
tral polymer chain confers good resistance to acid hydrolysis at
the normal pH values found in foods, such as in fruit products,
although culinary websites have reported that acidity weakens the
gel texture (Science of cooking, 2016). However, agar-​agar can
be hydrolyzed by acid at high temperatures, so it is suggested
that when the pH value of the solution to be gelled is below 5,
the pH of the food should be lowered just before cooling in order
to allow the gel to form (Imeson, 2010). This allows hydration to
occur in more neutral conditions.
Agar-​agar forms a synergy with locust bean gum. The inclu-
sion of locust bean gum has a marked effect on gel texture and
produces more elastic, cohesive gels than those obtained with
agar-agar alone; however, it should be noted that this only
appears to occur with agar-​agar extracted from Gelidium species,
not Gracilaria agar-​agar (Imeson, 2010).
Regarding the heat stability of agar-​agar gels, for the purposes
of culinary applications, gels made from agar-​ agar display
thermo-​reversible behavior, so they will melt on heating up to
above 85 °C (McGee, 2004) but will reform on subsequent cooling
Hydrocolloid Usages 389

to about 30–40 °C; the difference between gelling and melting

temperatures is called gel hysteresis, and agar-agar, compared
with the other gelling agents, has relatively large hysteresis.

Applications of Agar-​Agar
Agar-​ agar has been used in foods for over 300 years. It is
thought to have been discovered in Japan in the mid-​17th cen-
tury (Armisén et al., 2009), 200 years before it was introduced
to the West. Agar-​agar gels have also been used in microbiology,
where they are prepared with various nutrients and used to grow
colonies. However, agar-​agar has not always been used for this
application – in the mid-​19th century, scientists were still using
gelatin. It was actually Fannie Hesse, whose husband worked
with the German scientist Robert Koch, who suggested he used
agar-​agar instead of gelatin, being aware of the efficiency of agar-​
agar in culinary jellies (Hitchens and Leikind, 1939). As well as
remaining solid at 38 °C (a good temperature for bacterial growth
yet one at which gelatin gels start to melt), agar-​agar gels also
offer an additional advantage over gelatin in that very few bacteria
can digest the unusual agar-​agar polysaccharide, so gels remain
whole –​with gelatin gels, the bacteria will digest the protein and
produce a liquid mess. Food industry applications of agar-​agar
include water dessert gels, aspics, confectionery jellies, canned
meats, icings, piping gels and flan desserts (Imeson, 2010). The
main chef applications are as a gelling agent, where the gel can
be served hot, and using agar-​agar to make gelled beads (drops of
hot agar-​agar solution can be pipetted into cold oil to form instant
spheres).

Carrageenans (Iota and Kappa)

Carrageenans, like agar-​agar, are obtained from red algae, or
FIGURE 57.5 Structure of some carrageenans.
Rhodophyceae. They are permitted for use as a food additive
in Europe and have the E number of E407 (European Union,
2006).
aggregate to form the gel (Figure 57.6) –​as with agar-​agar, the
Chemical Composition/​Structure formation of the double helices alone is not sufficient to explain
Different species of seaweed contain different types of gelation. While agar-​agar molecules gel well on their own because
carrageenans. Every carrageenan has a backbone of D-​galactose of their low sulfate content, the carrageenans, which have a higher
and 3,6-​anhydrogalactose monomers, but the various carrageenans sulfate content, require cations to neutralize the negative charges
differ in the proportion and location of ester sulfate groups and on the sulfate groups in order for the helices to aggregate (Coultate,
the proportion of 3,6-​ anhydrogalactose within the backbone. 2009). Kappa carrageenans, with their relatively low sulfate con-
There are three main types, two of which are able to form gels. tent compared with the other carrageenans, require potassium
Kappa carrageenans (22% sulfate groups, approximately 33% ions (K +) for the helices to associate, whereas iota carrageenans,

3,6-​anhydrogalactose content) form firm and brittle gels, whereas with their higher sulfate content, need calcium ions (Ca ) for the
2+

iota carrageenans (32% sulfate groups, approximately 26% 3,6-​ helices to associate. According to Imeson (2010), during the gel-
anhydrogalactose content) form soft elastic gels; the two can ation of kappa carrageenans, the K + ions neutralize the negative

be mixed to form more intermediate textures. The third type is sulfate charges, allowing the helices to get close enough together
lambda carrageenan (37% sulfate groups, with little or no 3,6-​ to hydrogen bond with each other, resulting in firm, brittle gels.
anhydrogalactose content), which is not able to form gels and is With iota carrageenans, however, the double helices cannot stack
used instead to thicken instant drinks and dairy desserts (Imeson, as they can with kappa due to steric hindrance of the sulfate groups
2010). The structure of the three types of carrageenan and their (which are more abundant than in kappa); this creates a bigger
properties are shown in Figure 57.5 and Table 57.2. gap between chains, allowing divalent bridging to occur with the
calcium ions but not hydrogen bonding. Soft, elastic gels form
The Gelation of Carrageenans as a result (Imeson, 2010). Lambda carrageenan molecules lack
The mechanism of gelation of carrageenans is similar to that of anhydrogalactose content on their structure, which makes them
agar-​agar. The galactan chains on the carrageenan molecules unable to form the initial helices, and this therefore prevents them
first wind round each other to form double helices, which then from being able to form a gel (Coultate, 2009).
390 Rachel Edwards-Stuart, Reine Barbar

TABLE 57.2
Table Outlining the Different Properties of the Carrageenans
Solubility Lambda Iota Kappa
Hot (80 °C) water Soluble Soluble Soluble
Cold (20 °C) water All salts soluble Na+ salts soluble Na+ salt soluble
Ca2+ salt gives thixotropic swollen Limited swelling of K+, Ca2+ salts
particles
Hot (80 °C) milk Soluble Soluble Soluble
Cold (20 °C) milk Thickener Insoluble Insoluble
50% sugar solution Soluble Insoluble Soluble hot
10% salt solution Soluble hot Soluble hot Insoluble

Gelation
Effect of cations Non gelling Strongest gels with Ca2+ Strongest gels with K+
Gel texture –​ Elastic Brittle
Syneresis –​ No Yes
Hysteresis –​ 5–​10 °C 10–​20 °C
Freeze–​thaw stable Yes Yes No
Synergy with locust gum bean No No Yes
Synergy with konjac flour No No Yes
Synergy with starch No Yes No
Shear-​reversible Yes Yes No
Stability in acid Hydrolysis Hydrolysis accelerated by heat, low pH Hydrolysis accelerated by heat, low pH
and time and time
Gels are stable Gels are stable
Protein reactivity Strong protein interaction in acid Strong protein interaction in acid Specific reaction with kappa casein

Source: Imeson (2010).

FIGURE 57.6 Gelation mechanism of carrageenans.

(Reproduced with kind permission from DuPont Nutrition and Health)

While the sodium salts of all three carrageenans discussed here
are soluble in cold water, various methods may be used to ensure
that the carrageenans are fully dispersed and do not form clumps
before the onset of hydration. Proper dispersion can be achieved
by either agitating the liquid with a hand blender to create a
vortex before adding the powder; pre-​blending the hydrocolloid
with 5–​10 times its weight of a bulking agent such as table sugar
(sucrose) or maltodextrin before adding to the liquid; or slurrying
the carrageenans in oil to provide a hydrophobic barrier around
each particle.
As a carrageenan dispersion is heated, there is no significant
particle swelling or hydration until the temperature exceeds
40–​60 °C. As the particles hydrate, the viscosity rises, as the
swollen particles offer more resistance to flow. Further heating
Hydrocolloid Usages
to 75–​80 °C results in a drop in viscosity. On cooling, the solu-
tion shows a marked increase in viscosity, followed by gelation
below temperatures of 40–​50 °C. The exact hydration and gel-
ation temperatures are strongly dependent on the salts associated
with the carrageenans (Imeson, 2010).
From the experience of one of the authors (Edwards-​Stuart),
the most common reason why a carrageenan gel may fail to set
is due to inadequate dispersion, which prevents proper hydra-
tion. Effective dispersion is most easily achieved by using a hand
blender to prepare the solutions.
Kappa carrageenan gels display high syneresis levels; water is
eliminated from the gel as the gel structure tightens and contracts.
Iota gels, however, show no syneresis and display good freeze–​
thaw stability. They are unaffected by freezing, whereas the
kappa gel structure is further tightened irreversibly upon freezing
(Table 57.2).
External Effectors and Synergies
Kappa carrageenans interact synergistically with mannans
such as locust bean gum and konjac to produce stronger gels,
which are less brittle, more cohesive and less prone to syneresis.
Kappa carrageenans are often used to stabilize dairy-​ based
applications because they interact specifically with kappa casein
(Imeson, 2010).
Carrageenan solutions lose viscosity and gel strength when
subjected to pH values below 5.5, but the effect is not significant
until about pH 4.5 and is still manageable down to pH 3.5. The rate
FIGURE 57.7 Structure of alginate.
(Reproduced with kind permission from DuPont Nutrition and Health)
391
of acid degradation increases significantly at high temperatures,
so when working with acidic products, the carrageenans should
be added at the last moment to avoid excessive acid degradation
(Imeson, 2010).
Regarding the heat stability of carrageenan gels, similarly to
agar-​agar, kappa and iota carrageenan gels are thermo-​reversible,
yet they melt on heating to above only about 60 °C, although the
exact melting temperature depends on the type of carrageenan, its
concentration and the presence of cations. Therefore, the hyster-
esis value of the carrageenans is much lower than for agar-​agar.
Applications of Carrageenan Gels
Food industry applications of the carrageenans include injected
meats, canned meats, water dessert gels, dairy desserts and
instant drinks (for lamba carrageenan) (Imeson, 2010).
Sodium Alginate
Alginates are isolated from the cell walls of various species of
brown algae found off the coasts of the North Atlantic, South
America and Asia. Sodium alginate has an E-​number of E401
(European Union, 2006).
Chemical Composition/​Structure
Alginates are linear polymers of two different building blocks:
β-​D-​mannuronic acid (M) and α-​L-​guluronic acid (G), linked
together with (1,4)-​glycosidic bonds (Figure 57.7).
392 Rachel Edwards-Stuart, Reine Barbar

Alginate Solutions
Alginic acid itself is insoluble, but the alkali metal salts of alginic
acid such as sodium alginate are freely soluble in both hot and cold
water; however, they require proper dispersion to hydrate effect-
ively. This can be done in a similar way as with the carrageenans:
either using a hand blender to create a vortex; pre-​blending the
powder with a highly soluble solute such as sugar; or suspending it
in a hydrophobic solvent like vegetable oil. When added to water,
and in the absence of divalent ions such as Ca2+ or Mg2+, sodium
alginate will act to thicken the liquid due to its high molecular
weight (~10–​600 kDa); the viscosity of a particular alginate solu-
tion will depend on the chain length of the alginate molecules
within it, with higher molecular weights increasing its ability to
thicken (Imeson, 2010). Due to its efficient thickening ability, agi-
tating an alginate-​containing liquid with a hand blender in order
to hydrate the polymer can cause air bubbles to get trapped within
the solution, so often the solution will need to be left to stand for
a few hours (ideally overnight) before being used. Alternatively,
the mix can be vacuum sealed and chilled for two hours: vacuum
sealing removes air from the solution and helps prevent bubbles
from being present in the final gel (Myrhvold et al., 2011).
A viscous solution of alginate displays shear-​thinning properties
due to the long polymer chain and the stiffness of the hydrated
molecules; at low shear rates, the alginate molecules will arrange FIGURE 57.8 Gelation of alginate in the egg-box arrangement (the blue
themselves more or less randomly, but as you increase the shear disks are calcium ions).
rate, the molecules will start to line themselves in a more par-
(Reproduced with kind permission from DuPont Nutrition and Health)
allel way. Therefore, the resulting viscosity will decrease when
the shear rate is increased above a certain point (Imeson, 2010).
varies from 68:32 in the stem of Laminaria hyperborea to 29:71
The Gelation of Alginate in Durvillea antarctica. Since the M/​G ratio has such a major
In the presence of calcium and some other divalent ions, alginate impact on gelation, alginates from different sources will there-
molecules will form a strong gel. The interaction between sodium fore behave differently, and this may explain why properties
alginate strands and calcium ions is commonly represented as an such as minimum gelling concentration, and how the alginate
“egg-​box” arrangement, where the calcium ions fit in between the may behave under certain conditions such as the value of pH,
alginate strands like eggs in an egg box. Unlike the other gelling vary with alginates purchased from different suppliers. This can
agents discussed in this chapter, gelation can occur at room tem- be frustrating for a user when first experimenting with sodium
perature without the need for heating. alginate, since recipes designed for use with an alginate from one
In order for the alginate to form a gel with calcium ions, it source may not always apply to alginates purchased from some-
needs to contain a certain proportion of G units on the strand, and where else. This is generally true for most of the gelling agents,
these need to occur in blocks. The G units on one alginate chain where gelling conditions will vary depending on the grade.
will then link to the G units on the other chain via interactions The ability of sodium alginate to gel in the presence of cal-
with the calcium ions to form a junction zone (Figure 57.8). cium ions makes it ideal for use in the making of spheres, a con-
Calcium is not the only cation to interact with alginate: Ba2+, cept that has been exploited for years in the food industry and by
Zn2+, Cd2+ and Al3+ can also combine with the G blocks of chefs for about 20 years. When solutions of sodium alginate are
alginates (Tsai et al., 2017). pipetted or spooned into a calcium “bath” –​a calcium solution
Parts of the alginate molecule containing blocks of M units, made with a calcium salt such as calcium chloride –​the alginate
or MG blocks, will not participate in the junction zones but form solution will instantly gel on the outside as it comes into con-
elastic sections to the gel network, so while M blocks provide tact with the calcium, creating either small beads (known as fake
flexibility and elasticity, G blocks result in the strengthening and caviar) or spheres with a liquid core. Surface tension gives the
stiffness of the polysaccharide, and a high G content and long beads their distinctive spherical shape. This technique is called
blocks of G monomers within an alginate molecule will result in direct spherification, and it can be used to make small beads
strong gel-​forming potential. Therefore, the M/​G ratio has a major (by using either a syringe or pipette, or a peristaltic pump for
impact on alginate gelation characteristics (Tsai et al., 2017). making larger amounts) or large beads (which requires the use of
Alginates from different species, or even from different parts a tablespoon or half-​spherical spoon to gently drop spoonfuls of
of the same plant, differ in the ratio of these two acids. For the mix into the calcium bath). Alternatively, a liquid containing
example, Coultate (2009) lists the guluronate:mannuronate ratio calcium (either naturally or via the addition of a calcium salt)
in a number of different species of seaweed and reports that it can be dropped into the setting bath containing the alginate to
Hydrocolloid Usages
Alginate containing
solution
Calcium setting bath
(a)
FIGURE 57.9 Schematic of direct (a) versus reverse (b) spherification.
form a sphere. This technique is known as reverse spherification
(Figure 57.9).
When compared with preparing the alginate solution
(described earlier), where care must be taken to properly dis-
perse the alginate and prevent excessive bubble entrapment, the
preparation of the calcium solution is straightforward: the cal-
cium salt is added to water and stirred to dissolve. When pre-
paring the calcium solution to be used as the bath for direct
spherification, corn syrup or a sugar syrup (up to 22% sugar
concentration) can be added to the bath so that its density
matches that of the liquid to be gelled: this helps force the
drops into spherical shapes. Without this addition, the droplets
can sink to the bottom of the setting bath and become lop-
sided (Myrhvold et al., 2011). When preparing the setting bath
for reverse spherification, distilled or deionized water should
be used to hydrate the alginate, or where this is not available,
bottled water with a low calcium content can be used; tap water
should be avoided, especially in areas where the water is hard,
since the natural calcium present in the water will cause the
alginate to start setting prematurely in the bath.
Different calcium sources can be used for spherification: cal-
cium chloride, calcium lactate and calcium gluconate. The use
of these in spherification has been studied in detail by Lee and
Rogers (2012). Calcium chloride rapidly dissociates when added
to a solution because of its high solubility, making it an attractive
calcium source for spherification. Although it is commonly used
in direct spherification, its applications in reverse spherification
are limited due to the unpleasant bitter taste it contributes to the
food, which can be hard to mask.
Alternatives such as calcium lactate and calcium gluconate
give a nearly flavorless alternative and so are often preferred
(Vega, 2012). However, many recipes for reverse spherification
make use of the calcium naturally present in the ingredient to
be gelled, assuming it is present in a high enough concentration,
eliminating the need to add a calcium salt altogether. The rate
of gelation is fastest when using calcium chloride and slowest
when using calcium gluconate. Although the gelation kinetics are
affected by the source of calcium, neither the final alginate gel
393
Calcium containing
solution
Alginate setting bath
(b)
strength nor the heterogeneous structural resistance to calcium
diffusion is altered (Lee and Rogers, 2012).
The gel strength of alginate spheres increases with both
increasing alginate and calcium concentration. Small angle X-​
ray scattering analysis by Stokke et al. (1997) showed that the
cross-​sectional thickness of the junction zones actually increases
when either the calcium concentration and/​or the proportion of
G blocks increases, forming multilayer junction zones, as shown
earlier in Figure 57.8.
As explained earlier, minimum gelling concentrations differ
depending on the source of alginate, but as a general guide,
beads can be made (by the method of direct spherification) using
alginate concentrations of between 0.5% and 1% and a calcium
bath concentration of 1%. For reverse spheres, an alginate con-
centration of 0.5–​0.7% in the setting bath tends to work well,
while the calcium in the liquid to be gelled should be between
1.5% and 3% if using calcium lactate.
External Effectors and Synergies
Acidity
When sodium alginate is mixed with acidic ingredients (such as
apple juice) the mixture becomes very thick. It is hypothesized
that this is due to the creation of alginic acid, the insoluble com-
pound from which sodium alginate is derived, which causes
the solution to thicken (Vega, 2012). The alginate solution can
become so thick with time that it will form an almost gel-​like
structure when mixed with acidic liquids, making subsequent
spherification almost impossible. Adding a weak alkali such as
trisodium citrate can increase the pH of slightly acidic solutions
and reduce the thickness of the liquid that ends up inside the
sphere.
Calcium
When preparing base mixes to use for direct spherification, dairy
products should be avoided, as the calcium present in milk and
cream, for example, will cause the alginate to set before the solu-
tion can be pipetted into the calcium bath. When alginate is added
394
to a milk-​based sauce, the mix will gel instantly (Vega, 2012), but
these gels will often be messy and uneven. For similar reasons,
when using water in the alginate base to create spheres in direct
spherification, you need to use distilled water or bottled water
with a low calcium content, especially in areas of hard water, as
you would do when preparing the alginate setting bath for reverse
spherification. Furthermore, many food ingredients other than
dairy products naturally contain calcium, so you may need to
add a calcium sequestrant when using these ingredients for direct
spherification. Using dairy products or ingredients naturally
containing calcium is not an issue with reverse spherification –​
moreover, the natural calcium in the dairy product, or any other
ingredient, can be beneficial, as if it is present at a high enough
concentration then it can be used as the calcium source without
having to add extra calcium.
Salt
Direct spherification also does not work well with very salty
ingredients. This is possibly because the reaction is favored in the
direction of the sodium rather than the calcium, and since monova-
lent ions cannot link the G-​rich areas of the alginate strands, no gel
can be formed. In the work of Ouwerx et al. (1998) on the effect
of ionic strength on spherification, beads were formed in solution
with a constant concentration of either calcium (0.1 M) or copper
ions (0.05 or 0.1 M) with increasing concentrations of sodium ions
(0–1 M). Young’s modulus (which describes mechanical properties)
of Ca alginate beads is constant up to a molar fraction of 0.6 and
decreases strongly above this value. Above this value, the balance of
Na+/​Ca2+ goes in favor of the exchange of calcium ions by sodium
ions in the network, preventing the junctions between the chains.
Therefore, direct spherification has limited use as a tech-
nique for several ingredients, such as those with low pH values
and those naturally containing calcium, because alginate gel-
ation may occur within the ingredient mix before the alginate
reacts with cations in the setting bath. However, during reverse
spherification, the ingredients to be gelled only make contact
with the alginate during gelation; thus, reverse spherification
can be used with a much wider range of ingredients than direct
spherification (Tsai et al., 2017), such as those that are acidic,
salty and made of dairy products.
Other important factors can affect the gel texture:
(1) Time: with direct spherification, spheres will form
the instant the alginate solution makes contact with
the calcium bath. Spheres are normally left to set for
up to 3 minutes (the exact time depends on the size
of the sphere) to allow the skin to form. The longer
they are left, the firmer the beads will become: a short
dip gives a thin tissue-​like skin, while a longer stretch
in the bath produces a more chewy bead (Myrhvold
et al., 2011). When using the technique of reverse
spherification, the spheres need to be in contact with
the alginate bath for a long enough time for them to
form a sufficiently strong outer shell that allows them
to be removed without breaking, yet short enough that
the gelled exterior does not become too thick. In the
Rachel Edwards-Stuart, Reine Barbar
experience of one of the authors (Edwards-​Stuart),
with a calcium lactate concentration of 1.5–​2% and
an alginate concentration of 0.5–​0.7%, the optimum
time required to make a sphere is approximately
1.5–​2 minutes. However, the exact time will depend
on a number of factors, such as the type of alginate
(which affects its M:G ratio), the size of the sphere,
the exact calcium concentration and the exact alginate
concentration.
(2) Lifespan: during direct spherification, external gel-
ation results in the Ca2+ cross-​linking with the alginate
molecules at the bead surface, drawing the polymer
chains close together. This results in the formation
of a less permeable surface, slowing the subsequent
diffusion of Ca2+ ions (Lee and Rogers, 2012). So if
the beads are removed, rinsed and consumed imme-
diately, one can get droplets of liquid surrounded by
a skin of gel. However, while the calcium-​induced
cross-​linking at the surface will act to slow down
diffusion of further calcium into the center of the
bead, calcium ions will continue to move inside the
sphere as the alginate becomes saturated with ions
on the surface, so if the beads are either left in the
calcium bath or held for any length of time after-
wards, even if they are thoroughly rinsed, calcium
will continue to penetrate the spheres, and the beads
will continue to gel until the whole inside has solidi-
fied. While rinsing the gelled beads twice slows the
gelling process and washes away any bitter flavor
left from the calcium bath, only heating the beads
to 85 °C for 10 minutes will stop the gelation pro-
cess and prevent further solidification (Myrhvold
et al., 2011).
  With reverse spherification, it is possible to keep the
center of the bead liquid. This is because the calcium
moves outwards and is available in limited amounts,
i.e., whatever was the amount that was contained
initially in the single sphere. Liquid centers can be
achieved because there will not be any hydrocolloid
present in the middle of the sphere –​presumably the
large alginate molecule is unable to penetrate the skin
in the way that the small calcium ion can in direct
spherification. Since the bath does not contain calcium
salt, spheres made using the reverse spherification
technique do not need to be rinsed.
  Once the spheres are made, you can store them in
fresh water, infusions or flavored oil for up to three
days, and they can be served hot or cold (Myrhvold
et al., 2011). As explained earlier, the film formed
around the outside of an alginate sphere is perme-
able. This permeability allows the transport of small
enough molecules across the film, as explained earlier
for the diffusion of calcium. This could also cause any
molecules inside the sphere, such as sugar molecules,
to diffuse out, which will result in, for example, a
loss of sweetness inside the sphere (Vega, 2012).
Therefore, recipes often recommend that spheres are
Hydrocolloid Usages
stored in the same liquid used to make the spheres to
help them retain their flavor until they are used.
(3) Viscosity of liquid inside the sphere: since sodium
alginate is a thickening agent, beads made using the
direct method, even if consumed immediately after
gelation, will have a slightly thickened, albeit liquid,
interior. With reverse spherification, the central core
will have the viscosity of the calcium-​ containing
liquid, which does not contain the thickening agent
alginate so technically can be much thinner. However,
with reverse spherification, if the calcium-​containing
liquid is not dense or viscous enough, it will not sink
in the alginate bath and therefore, it will be difficult
to form spheres. Xanthan gum can be added to alle-
viate this difference, making the thickness of the
calcium solution and alginate bath more equal and
thus allowing spheres to form (Vega, 2012). Having
a thicker liquid interior can also be beneficial if the
user wants to add solid shapes to suspend in the
center of the sphere. However, adding xanthan gum
or another thickener will result in a thick interior to
the sphere.
  In order to create a sphere with a truly liquid
center, which would explode in the mouth as the con-
sumer bites into it, the calcium-​containing liquid can
be frozen into half spheres first using spherification
molds, which can then be dropped into the alginate
bath in their frozen state as a pre-​formed shape. As
they start to defrost, the outer layer will gel, and if
they are removed from the bath after a few minutes
and consumed once defrosted, “true” liquid centers
can be achieved. Heating the alginate bath to about
50 °C before dropping in the frozen shapes can speed
up the defrosting process and help in the creation of
these spheres.
Gels formed between calcium and alginate are very heat stable
and can be heated up to 100 °C without melting. Therefore, they
are often used in the food industry to make heat-​stable bakery
inclusions such as synthetic cherries, or the fake fruit pieces
sometimes found in cheap pie fillings, as well as cold setting
bakery cream fillings and also reformed food such as onion rings
and olive fillings (Imeson, 2010).
Applications of Alginate
In the food industry, internal setting calcium alginate gels (rather
than the diffusion setting mechanisms explained earlier) can be
formed by using a suitable calcium sequestrant, such as a phos-
phate or citrate, and a slowly soluble calcium salt. Both these
ingredients are directly added to the alginate mix from the start.
The sequestrant will bind any free calcium in the mix and pre-
vent pre-​gelation of the alginate during the time required to mix
the food with the alginate and process it into its desired shape
(Imeson, 2010). The slowly soluble calcium salt will then provide
the calcium needed for the alginate to set.
Sodium alginate also has other functions in the food industry
aside from its ability to form gels: alginates can act as stabilizers
395
in oil in water emulsions, like salad dressings, and suspensions of
solids in water, such as fruit juice. The alginate acts as a stabil-
izer by increasing the viscosity of the aqueous phase, preventing
separation. Also, drying a thin layer of alginate solution or gel
can form a film, which can be used for a number of functions,
such as preventing moisture from fillings passing to the rest of the
product in pastries, protecting frozen fish from oxidation, binding
herbs and spices to the surface of meat products, and creating rap-
idly dissolving films such as menthol breath strips. Alginate has
been used as the gelling and structuring agent for artificial bird’s
nests and shark fins in certain countries in Asia over many years
(Imeson, 2010).
Gellan Gum
Gellan gum is a polysaccharide produced by the bacteria
Sphingomonas elodea, and is manufactured in large fermenta-
tion vessels, which allow the bacteria to convert simple sugars
and other nutrients into the gellan gum polysaccharide (Imeson,
2010). It is permitted for use as a food additive in Europe and has
the E number of E418 (European Union, 2006).
Chemical Composition and Structure
The structure of gellan gum is an unbranched polymer chain
consisting of a repeating unit made of four monosaccharide
residues: glucose –​glucuronic acid –​glucose –​rhamnose.
The total length of the chain is approximately half a million
residues.
In its native form, referred to as “high acyl”, two acyl
substituents –​acetyl and glyceryl –​are present on the first glu-
cose molecule within the repeating unit. On average, there is one
glyceryl present per repeating unit, and one acetyl present per
every other repeating unit. If native gellan gum is treated with
alkali, these glyceryl and acetyl groups are removed. This gives
the low acyl form (Coultate, 2009), as shown in Figure 57.10.
Hydration of Gellan Gum
The hydration temperature of low acyl gellan gum strongly
depends upon the type and concentration of ions in solution. Gum
hydration is inhibited by the presence of the divalent ions usually
found in most water supplies, and in the absence of sequestrants,
low acyl gellan gum will need temperatures of around 75 °C to
fully hydrate. The presence of sequestrants can allow hydration in
cold water, if the water being used is soft. High acyl gellan gum,
however, always requires heat to hydrate (CP Kelco, personal
communication, 2018; Imeson, 2010).
The pH must be above 4 for good hydration, and like other
polysaccharides, gellan gum will undergo hydrolytic degrad-
ation at high temperatures, especially in acidic conditions. While
a low acyl gellan gum solution can be maintained at 80 °C and
at pH 3.5 for up to one hour with minimal degradation in the
quality of the gel that forms (Sworn, 2009), high acyl gellan
gum is more susceptible to degradation, so long holding times
in acidic conditions should be avoided (CP Kelco, personal
communication, 2018).
Hydration is also affected by sugar levels of greater than 25%.
Therefore, as a rule, gellan gum should be added to solutions that
396
FIGURE 57.10 Diagram showing the building blocks of low acyl and high acyl gellan gum.
initially contain low levels of calcium, moderate pH and dissolved
sugar of less than 25%. Once it is hydrated, salts, acids and add-
itional sugars can be added to form the final food (Imeson, 2010).
The Gelation of Gellan Gum
Solutions of gellan gum will gel on cooling at low concentrations.
The exact gelling temperature will depend on the type of gellan
gum, the presence and concentrations of ions, and the presence
of other dissolved solids. Gellan gum gels set very quickly, as
soon as the setting temperature has been reached. This is known
as “snap setting” (CP Kelco, personal communication, 2018).
High acyl gellan gum solutions set quickly once the tempera-
ture falls below 70 °C, whereas solutions of low acyl gellan gum
do not set until the temperature drops into the 30–​50 °C range
(Coultate, 2009).
Low Acyl
Solutions of low acyl gellan gum, often known as gellan F, require
cations, acid or a combination of both to form a gel. They give
firmer and more brittle gels than high acyl gellan gum, with excel-
lent heat stability –​gels made with low acyl gellan gum will not
Rachel Edwards-Stuart, Reine Barbar
melt and are retort and bake stable (CP Kelco, personal commu-
nication, 2018). Divalent cations (Mg2+ and Ca2+) are much more
effective in promoting gelation than monovalent cations (Na+ and
K+), with K+ being more effective than Na+. The cation content of
gellan samples is therefore crucial –​the concentration of divalent
cations required to induce gelation of gellan is far lower than for
monovalent cations, and the resulting gels have greater thermal
stability (Morris et al., 2012). Acid can also promote gelation
with low acyl gellan gum: when the pH is below 3.0, gellan gum
gels can be made with hydrogen ions (Sworn, 2009). This will be
discussed in more detail in the next section. Commercial low acyl
gellan gum allows the creation of gels that are strong enough to be
characterized by compression testing at polymer concentrations as
low as 0.2% weight (Morris et al., 2012). However according to
Imeson (2010) and Sworn (2009), demoldable gels can be formed
with low acyl gellan gum at concentrations as low as 0.05% when
the conditions are right (see later). This is at a lower concentra-
tion than all the other gelling agents discussed here. It has been
reported that despite agar-​agar being a very efficient gelling agent
in traditional Japanese food products, all its physical properties,
such as modulus, hardness, brittleness, elasticity and cohesiveness,
Hydrocolloid Usages
could be matched well by using gellan at two to three times lower
concentrations (Morris et al., 2012).
High Acyl
By contrast, solutions containing high-​ acyl gellan gum, also
known as gellan LT, set quickly on cooling to give a soft and
elastic gel. Their properties are much less dependent on the
concentration of ions in solution –​they simply gel on cooling,
unlike low acyl, which requires either cations or acid. Unlike gels
made with low acyl gellan gum, high acyl gellan gum gels are
not heat stable but are thermo-​reversible: they will soften with
heating and will melt with prolonged heating (CP Kelco, personal
communication, 2018).
Combinations of high acyl and low acyl gellan gum can be
used to form a range of textures, from soft and elastic to firm and
brittle.
Gelling Mechanism
As has previously been described for agar-​agar and carrageenen,
the gelation of gellan gum is generally considered to be a two-​
step process. On cooling, the molecules that have been in a
disordered single-​chain coil when heated transform into ordered
double helices, followed by associations of the helices via weak
interactions such as hydrogen bonds and van der Waal forces,
which results in the formation of a weak gel (Noda et al., 2008).
In the case of low acyl gellan gum, the electrostatic repulsion
between the helices can be suppressed by the addition of either
acid or salts, which then allows gels to form. The simple anions
and cations from the dissolved salt screen electrostatic repulsion
between the gellan helices, promoting aggregation (Morris et al.,
2012). Gel strength increases with increasing ion concentration
until a maximum is reached. Further addition of ions results in
a reduction of gel strength due to “over-​conversion” of the low
acyl gellan gum with excess ions (Sworn, 2009); i.e., the exces-
sive screening of the helices by the salt actually prevents gelation.
Where the pH of the solution is low, the acid present reduces
the charge on the helices by converting the negatively charged
COO− on the glucuronate carboxyl groups to the uncharged
COOH form (Morris et al., 2012), allowing helical association.
Since most of the carboxyl groups on the gellan gum mol-
ecule will then be protonated, they cannot interact with cations.
Therefore, acid-​induced gellan gums do not need cations in order
to set (Imeson, 2010).
Two models have been proposed to suggest how the gellan
helices associate to form a gel. The first model, known as the
conventional method or domain model, proposes the formation of
distinct junction zones, with disordered flexible polymer chains
connecting them that provide the elasticity (Noda et al., 2008).
The other model, the fibrous model, suggests that the network
structures develop through the formation of non-​associated fibers
or strands via either elongation or branching, which percolates
throughout the entity and gives rise to elasticity, as shown in
Figure 57.11 (Noda et al., 2008).
The acyl groups present on high acyl gellan gum significantly
affect the characteristics of the gel. According to the domain
model, after the random coils have formed helices on cooling,
further aggregation of the helices is limited by the presence of
397
FIGURE 57.11 Schematic model for the gelation of gellan gum.
(a) Conventional model; (b) fibrous model.
(Noda et al., 2008).
the acetyl group. According to the fibrous model, the acyl groups
inhibit end-​ to-​
end type intermolecular associations through a
type of steric hindrance, resulting in a decrease in the degree of
continuity and homogeneity of the gelled system (Sworn, 2009).
As a result, a soft and elastic gel is formed.
When gels are made of low and high acyl gellan gum together,
the resulting gel appears to be composed of interpenetrating
rather than cooperative networks of polymers. On cooling, first,
the high acyl form will set, at higher temperatures than it usu-
ally would. With further cooling, the low acyl form sets itself
up within the high acyl network: this “gel within a gel” system
allows one form to smoothly transition into the other form as the
ratio of high acyl to low acyl changes (Imeson, 2010). By varying
the ratio of low acyl and high acyl gellan gums, it is possible to
obtain textures close to those of carrageenan and gelatin gels (CP
Kelco, personal communication, 2018). A ratio of 80% high acyl
gellan gum to 20% low acyl gellan gum results in a texture that
is similar to the texture of a gelatin dessert gel (Imeson, 2010).
External Effectors and Synergies
Several factors influence the texture of gellan gels. These include
gum concentration, pH, temperature, the presence of metal ions,
and whether a co-​solute like sugar is present.
In its “simplest” form, i.e., in conditions of low sugar and low
acid, low acyl gellan gum is a very effective gelling agent, with
demoldable gels being formed at as low as 0.05%. As sugar levels
increase, gels become softer, less brittle and more elastic, and
higher gum levels are needed to form demoldable gels (typic-
ally 0.3% or higher). In general, sugars have a plasticizing effect
on the gel –​associations between molecular chains of gellan
gum are slower to form and are not as extensive once they have
formed. Gelation time is slower, and low acyl gels no longer
exhibit “snap” setting in the presence of high-​sugar solids –​it can
398
take weeks for the gellan gum gel to develop its full structure.
Sucrose has a more pronounced influence on the set time and gel
texture than fructose or glucose (Imeson, 2010).
In an acidic environment, low acyl gellan gum is able to form
a gel without the need for inclusion of mono-​and divalent metal
ions. Demoldable gels can be made at concentrations as low as
0.05% without the need for cations. While the gels formed are
stronger and firmer than those made with cations, they are quite
brittle (however, the brittleness of these gels is often perceived
as soft and mushy, because the gels break under low levels of
strain). Addition of other gelling ions, such as sodium or cal-
cium, generally results in a reduction in gel strength of the acid
gels. Optimum gel modulus occurs for these acid gels at approxi-
mately pH 2.8–​3 regardless of the acid used. This optimum is
not affected by the presence of sugars to the same extent as ion
requirements. For example, in the presence of 60% sucrose, the
optimum shifts to approximately pH 2.5–​2.7 (Sworn, 2009).
Gels made at low pH are also more likely to synerize than their
counterparts at neutral pH. As with the neutral low acyl gellan
gum gels, gelation is quick; however, gelation tends to occur at a
lower temperature when induced by H+ rather than other cations,
and the resulting gels are not as thermostable and will not reheal
after shearing (Imeson, 2010).
High acyl gellan gum is not as sensitive to cations or dissolved
solids as low acyl gellan –​slightly firmer gels are formed as cal-
cium concentration increases, but overall, the changes to gel prop-
erties are slight. Adding dissolved sugars to the gel also appears
to increase gel strength, but again, the effect is not significant.
However, when sugars are present at greater than 50%, they inter-
fere with gelation and plasticize the gel (as they do for low acyl),
as well as raising the setting temperature, so higher levels of
gellan gum are needed to form a demoldable gel in environments
of high solids. Very low pH values below 3.5 seem to soften gels
made with high acyl gellan gum. Unlike for low acyl gellan gum,
there is no evidence of an acid-​induced gel mechanism for high
acyl gellan gum. The gum simply becomes less able to form a gel
as the pH drops (Imeson, 2010).
Depending on the amount added, water-​ miscible organic
solvents such as alcohols precipitate gellan gum from solution.
However, moderate levels of these solvents can be tolerated.
Gellan gum also shows synergies with other thickening or gelling
agents, such as starch, guar gum, locust bean gum, cellulose
derivatives or xanthan gum, which may be used with gellan gum
in order to obtain the desired texture (CP Kelco, personal com-
munication, 2018).
Properties of Gellan Gum Gels
In general, gellan gum gels offer good heat stability, although
melting temperatures, along with setting temperatures and gel
strength, are dependent upon the type and concentration of ions
present and gum level (CP Kelco, personal communication,
2018) as previously detailed.
Gels formed by gellan gum are characterized by their spark-
ling clarity, good flavor release, rapid setting behavior and
the fact that they do not require high gelling concentrations
(Banerjee et al., 2013). It is generally accepted that gelled
Rachel Edwards-Stuart, Reine Barbar
products have lowered perceived intensities of flavor compared
with fluid products, but gellan causes less suppression than
most other gelling agents and is often described as having out-
standing “flavor release properties”, with flavors carried by this
type of gel being released in the mouth much more quickly than
from other types of gel (Coultate, 2009). One possible explan-
ation for its excellent flavor release properties is that water is
released from the gels during mastication, carrying with it taste
compounds (Morris et al., 2012). Also, the brittleness of low
acyl gellan means that it tends to crumble in the mouth, giving
the impression of melting (Coultate, 2009).
As described earlier, gellan gum gels are normally formed by
cooling solutions from the disordered state at high temperatures.
Gelation, however, can also be induced by diffusion of cations
into solutions of gellan at ambient temperature, or by the
“internal set” procedure developed for alginate, in which Ca2+
ions are released into solution at a controlled rate by dissociation
of an insoluble calcium salt on reduction in pH by glucono delta-​
lactone (GDL) (Morris et al., 2012).
Gellan Gum Fluid Gels
At levels too low to form a demoldable gel (which, according to
Coultate (2009), is around 0.03%), gellan gum can be used to
make “fluid gels” –​even though it is below its gelling concentra-
tion. The gellan gum molecules still associate and form a long-​
range network, yet the system remains fluid (Imeson, 2010). This
fluid gel will appear like a liquid in terms of its viscosity, yet its
gelling properties will allow it to suspend particles, making it a
popular ingredient in the food industry for suspending particulates
in sauces and dressings like vinaigrette, and stabilizing fruit pulp
in beverages. Fluid gels are thixotropic and exhibit a highly
pseudoplastic flow property, i.e., the viscosity decreases with
increasing shear. This means that fluid gels are perceived as thin in
terms of mouthfeel, because the act of swallowing during eating
creates shear (Imeson. 2010). Despite being low in viscosity, the
fluid gels exhibit a high elastic modulus, which gives them their
suspending properties. At the Fat Duck restaurant in Bray, chef
Heston Blumenthal uses fluid gels to prepare one of his signature
dishes, “Hot and Cold” tea, where the same beverage is served at
two temperatures within the same cup. The fluid gels retain their
structure as long as they are kept still, thus allowing the waiter to
bring the cup to the table without the two sides mixing. However,
when you tilt the cup to drink, enough shear forces are generated
to convert the fluid gels into drinkable liquids.
However, when most chefs prepare fluid gels, they are gener-
ally referring to blended gels rather than truly “fluid” gels, and
they are working with much higher concentrations of gelling
agent. These culinary creations are prepared either by preparing
a solid gel out of gellan gum and then shearing it with a blender
to make a puree or by shearing the gel as it cools so that it never
gets the chance to form a solid gel.
Applications of Gellan Gums
A major food application of gellan gum gels is for dessert gels,
particularly for Asian desserts, confectionery, dairy desserts and
bakery filings (Imeson, 2010). Amongst chefs who use these
Hydrocolloid Usages
types of ingredients, gellan gum is mainly used to make gels
(which can be served hot), blended gels and gel coatings.
Hydrocolloids as Emulsifiers
In emulsions, one liquid (for example, melted fat, O) is dispersed
in a continuous phase made of another liquid (for example, an
aqueous solution W). Emulsions are generally only metastable,
except for nanoemulsions. The physico-​chemical principles of
O/​W emulsion stability are based on the classical colloid the-
ories of electrostatic and steric stabilization (Dickinson, 1992;
McClements, 2005 as cited in Dickinson, 2009b).
The soft matter approach to food materials is definitely
gaining ground in food science. In the different phases of the
food structuring process, a multitude of concepts from soft
matter physics can be applied. Their universality gives good
opportunity for improved understanding of the complexity
of the food materials and their structuring processes (van der
Sman, 2012).
Culinary emulsions can take two different forms: fat dispersed
into water (O/​W) and water dispersed into fat (W/​O). Common
O/​W emulsified food systems include mayonnaise, aioli, milk,
cream, hollandaise and pan sauces. W/​O systems are most com-
monly found in the form of vinaigrettes and whole butter.
The criteria for choosing emulsifiers for kitchen use are
different from those used in the food industry. For kitchen
use, preserving the organoleptic properties of the food is the
most important criterion, and technological functionality (in
particular long-​term stability) is a secondary consideration.
In other words, a cook will choose an unstable emulsion with
optimal flavor over an emulsifier that ensures stability but leaves
an aftertaste.
Whether in industrial applications or for culinary purposes, the
texture of an emulsion depends on the type of emulsifier used and
depends even more on the concentration used. For instance, in
the case of French salad dressing, i.e., an O/​W emulsified system,
gum tragacanth and propylene glycol alginate can be used to
stabilize the emulsion. Other salad dressings can be stabilized
with starches and whey protein products, which can replace the
egg yolk in salad dressings. In some applications, the use of a
mixture of a few gums is favored due to their synergistic effects
(Nussinovitch and Hirashima, 2019).
As discussed in the Introduction, hydrocolloids can act as
emulsifiers due to their roles as viscosity modifiers, but some
hydrocolloids in addition contain a surface-​active component in
their structure. These include exudate gums, starch, cellulose,
pectin and proteins and will be discussed here.
Exudate Gums
The exudate gums represent an abundant raw material. Rana
et al. (2011) have mainly studied the gums due to their sustain-
able, biodegradable and biosafe characteristics. Initially, the
word “gum” might be used to describe natural exudates oozing
from trees and shrubs in tear-​like, striated nodules or amorphous
399
lumps. In 1963, Glicksman defined a gum as any polymer that
can be dissolved or dispersed in water to give a viscous solution
or dispersion (Glicksman, 1982).
Exudate gums are polysaccharides produced by plants as a
result of stress, including physical injury and/​or fungal attack.
Gum arabic, gum tragacanth, gum karaya and gum ghatti have
been used by humans for many thousands of years in various
food and pharmaceutical applications. Generally, these gums
are structurally related to arabinogalactans, galacturonans or
glucuronomannans. They all contain a high proportion of glu-
curonic or galacturonic acid residues (Milani and Maleki, 2012).
Many plant exudates are known all over the world; never-
theless, only gum arabic (CFR No.184.1330), gum ghatti (CFR
No.184.1333), karaya gum (CFR No.184.1349) and gum traga-
canth (CFR No.184.1351) are utilized in the food industry. The
reason is that the choice of hydrocolloids also depends on their
safety (Li and Nie, 2016). A large number of these exudates have
not been granted “GRAS” status: “GRAS”, an acronym for the
phrase “Generally Recognized As Safe”, is a U.S. Food and Drug
Administration designation validating that a substance is gener-
ally recognized as having been adequately shown to be safe under
the conditions of its intended use, or unless the use of the sub-
stance is otherwise excluded from the definition of a food addi-
tive (FDA, 2020).
The use of gums to produce higher-​quality food can be exem-
plified by ice cream; homemade ice cream commonly has poor
textural qualities, like the presence of ice crystals, a sandy tex-
ture and an absence of smooth meltdown. Today’s technologic-
ally manufactured ice creams contain multiple hydrocolloids
as stabilizers and emulsifiers to eliminate these quality defects
(Nussinovitch and Hirashima, 2019).
Gum Arabic
Chemical Composition/​Structure
Among the natural emulsifiers that support the consumer-​driven
movement from animal-​based ingredients to more label-​friendly
plant-​based ones, acacia gum, also known as gum Arabic, is widely
relevant. Indeed, this gum is a dried exudate obtained from the
trunk and branches of Acacia trees. Two species, A. senegal and
A. seyal, are authorized by the Food and Agriculture Organization
(FAO, 1999) and commonly used (arabic gums, E414 EC) to
stabilize O/​W emulsions, in particular odorant compounds in
beverages. Arabic gum is a complex polysaccharide, either neu-
tral or slightly acidic, found as a mixed calcium salt of a polysac-
charide acid (arabic acid). The backbone is composed of 1,3-​linked
β-​D-​galactopyranosyl units. The side chains are composed of two
or five 1,3-​linked β-​D-​galactopyranosyl units, joined to the main
chain by 1,6-​linkages. Both the main and the side chains con-
tain units of α-​L-​arabinofuranosyl, α-​L-​rhamnopyranosyl, β-​D-​
glucuronopyranosyl and 4-​O-​methyl-​β-​ D-​glucuropyranosyl, the
last two mostly as end units (Anderson and Stoddart, 1996; Islam
et al., 1997; Verbeken et al., 2003).
It was reported by Idriss et al. (1998) that the residues of arabic
gum are galactose (39–​42%), arabinose (24–​27%), rhamnose
400
(12–​16%) and glucuronic acid (15–​16%); it includes 1.5–​2.6%
protein and 12.5–​16.0% moisture. Arabic gum is a highly
branched polyanionic polysaccharide.
Emulsifying Properties
The most commonly recognized hydrocolloid gum that acts as
an emulsifier is gum arabic. It is widely used in the soft drinks
industry for emulsifying flavor oils (like orange oil) under acidic
conditions.
The film-​ forming ability of gum arabic has long since
established that this is indeed a genuine emulsifier that confers
functionality not only by modifying the rheology of the aqueous
phase but also by leading to the formation of a stabilizing layer
around oil droplets (Dickinson, 2003). However, the surface
activity is rather low in comparison with typical food emulsifiers.
This is the reason why a high gum-​to-​oil weight ratio (around
1:1) is necessary to generate stable sub-​micron-​sized droplets
(McNamee et al., 1998).
Experiments conducted by Dickinson et al. (1988) with acacia
gums having nitrogen contents in the range 0.1–​7.5% suggest a
good correlation between the protein content and the limiting
long-​time interfacial tension, and also between emulsifying cap-
acity and the initial rate of change of tension with time. It is today
widely accepted that it is the protein-​rich high-​molecular-​weight
fraction of acacia gum, the arabinogalactan proteins (AGP) com-
plex, that mainly provides the surface properties of gum (Sanchez
et al., 2018). For more information regarding the history, struc-
ture and different physico-​chemical properties of acacia gum, the
reader is referred to the recent review by Sanchez et al. (2018).
Concerning emulsifying properties, the physical chem-
ical parameters (pH, ionic strength, type of ions, temperature,
homogenization pressure, etc.) all influence the structure of
biopolymer-​stabilized emulsions and their stability. The gum
from A. senegal is generally described as most efficient to form
and stabilize emulsions compared with the gum from A. seyal
FIGURE 57.12 Global stability index measured by Turbiscan for emulsions using the gum from A. senegal and A. seyal as emulsifier. In these experiments,
the emulsions contained respectively 15 weight % sunflower oil (a) and 10% limonene (b) (personal data).
Rachel Edwards-Stuart, Reine Barbar
because of its chemical composition (richer in protein) and
structural properties of AGP. Indeed, the emulsifying activity of
acacia gum has been attributed to the proteinaceous moiety and
to a specific AGP fraction with high molar mass (Randall et al.,
1989; Garti and Leser, 2001; Al-​Assaf et al., 2006). The emulsion
stabilization, on the other hand, is attributed mainly to the poly-
saccharide portion, contributing to viscosity, steric and electro-
static effects (Chanamai and McClements, 2001). However, it
is worth mentioning that several technical factors, such as pro-
cessing conditions of emulsions and high shear forces, as well
as the nature of the emulsion oil and the ratio of gum/​oil could
have opposite trends. The Turbiscan stability index (TSI) corres-
ponds to the signal variation at definite positions (h) throughout
various height (H) ranges of the sample between the scani and
the scani−1. TSI calculated over the entire tube height takes into
account the ensemble of destabilization phenomena occurring
during storage (creaming, flocculation/​coalescence and clari-
fication). As seen in Figure 57.12, where the increase of TSI
accounts for increasing global destabilizing kinetics, the gum
from A. seyal can induce better emulsion stability at an amount
of 15% weight with sunflower oil than the gum from A. senegal,
which works better with limonene at these specific conditions of
emulsion preparation.
Gum Ghatti
Chemical Composition/​Structure
Gum ghatti is a natural gum that is exuded from the tree species
Anogeissus latifolia, which is native to India and Sri Lanka.
The major fraction has a 1,6-​linked β-​galactose backbone with
branches at O-​ 3 and O-​ 4 positions, which can be regarded
as the “hairy region”, while the “smooth region” consists of
→2)-​Araf-​(1→4)-​GlcpA-​(1→6)-​Galp-​(1→6)-​Galp-​(1→. The
terminal side chains are arabinofuranosyl (Araf) and occa-
sionally rhamnopyranosyl (Rhap), arabinopyranosyl (Arap),
Hydrocolloid Usages
galactopyranosyl (Galp) or glucuronopyranosyl (GlcpA) residues
(Kang et al., 2011a; 2011b). Gum ghatti with 4.34% (w/​w) pro-
tein covalently linked to the polysaccharide has been shown to be
a good emulsifier (Kang et al., 2012).
Emulsifying Properties
The emulsification performance of gum ghatti is greatly influenced
by the presence of insoluble components. These insoluble
components are usually filtered and removed during the manufac-
ture of beverage emulsions. Gum ghatti, like other hydrocolloid
emulsifiers such as gum arabic and sugar beet pectin, contains a
small amount of protein linked to the polysaccharide unit. Enzyme
digestion with protease does not greatly reduce the molecular
weight. However, when the gum is subjected to processing, such
as spray drying, the effect of enzyme digestion can be clearly seen
from changes in protein conformation, which then becomes more
accessible. Further evidence of protein–​polysaccharide linkage
is supported by the interaction with Yariv’s reagent, and hence,
it can be classified as an arabinogalactan–​protein complex. The
emulsification mechanism in gum ghatti is similar to that of gum
Proteins
FIGURE 57.13 Schematic of part of gum ghatti molecule adsorbed to oil droplet.
401
Arabic, where the protein is responsible for the surface activity
by acting as a strong anchor point at the oil–​water interface with
the hydrophilic polysaccharide chains providing the protective
layer (Dickinson, 2003) (Figure 57.13).
The interfacial rheology of gum ghatti is superior to that of
gum arabic at the same concentration due to the high protein
content and possibly the conformation in solution (Al-​Assaf
et al., 2008). Gum ghatti holds a superior emulsification ability
due to great oil binding capacity, acid resistance and salt tol-
erance, which can be used in formulations that are difficult to
emulsify with gum arabic or other available commercial gums
(Kang et al., 2011c; Hobbs et al., 2012). Recently, the reasons
behind this enhanced emulsifying capacity were investigated.
More gum components were adsorbed onto oil droplets in the
emulsion of gum ghatti (30%, w/​w) than that of gum arabic
(7–​10%); on the other hand, the adsorbed components of gum
ghatti in the emulsion were distributed over the whole molecular
weight range, whereas only the high-​molecular-​weight fraction
of gum arabic was adsorbed on to the oil surface. This is why
gum ghatti is used at concentrations as low as a quarter of those
Oil
droplet
Carbohydrate part
402 Rachel Edwards-Stuart, Reine Barbar

of gum arabic. Moreover, the protein content of gum ghatti Emulsifying Properties
is higher than that of gum arabic (Kang et al., 2012; Kang A range of chemical and physical modifications and their com-
et al., 2014). binations have been done to make starch particles suitable for
Pickering emulsion formulations, which are emulsions stabilized
Other Polysaccharide Hydrocolloids by solid particles (Zhu, 2019).
For instance, starch that has been hydrophobically modified
Starch, cellulose and pectin are well known for their rheological
by reaction with octenyl succinate anhydride (OSA) has been
properties, which allow them to function as thickening and
shown to be strongly surface active (Prochaska et al., 2007)
gelling agents. However, it is their ability to stabilize emulsions
and to have excellent emulsifying and emulsion-​ stabilizing
due to their surface-​active components that is less well known
properties (Figure 57.14) (Chanamai and McClements, 2002;
and will therefore be discussed here.
Nilsson and Bergenståhl, 2007; Taherian et al., 2007). The
(Modified) Starch short octenyl succinate side chains anchor the polysaccharides
Chemical Composition/​Structure to the oil–​water interface, and the long amylopectin backbone
protects the droplets against flocculation by the mechanism of
Starch is GRAS, non-​allergenic, abundant and cheap. Starch
steric stabilization (Dickinson, 1992). These lipophilic starches
granules are made mainly (99.99%) of two alpha glucans: linear
are used in concentrated flavor emulsions for soft drinks and in
amylose and branched amylopectin (Zhu, 2019). The ratio of the
encapsulated ingredients, e.g., spray-​dried flavors and creamers.
two polysaccharides varies according to the botanical origin of
Generally, these starches provide improved oxidation resistance
the starch. The “waxy” starches contain less than 15% amylose,
and low-​ temperature emulsion stability. Lipophilic starches/​
“normal” 20–​ 35% and “high” amylose starches greater than
starch derivatives act more as emulsion stabilizers rather than
about 40% (Tester et al., 2004).
true emulsifiers (Dickinson, 2009a; 2009b).
Amylose is a relatively long, linear α-​glucan containing around
Other modification paths include size reduction for the produc-
99% (1→4)-​α-​and (1→6)-​α-​linkages and differs in size and
tion of starch nanoparticles (acid hydrolysis, high-​pressure hom-
structure depending on botanical origin. Amylose has a molecular
ogenization, ultrasonication, reactive extrusion, γ-​irradiation,
weight of approximately 1 × 105–​1 × 106 (Mua and Jackson,
steam jet cooking, nanoprecipitation, enzymatic debranching and
1997; Buléon et al., 1998; Biliaderis, 1998). Amylopectin is a
recrystallization, emulsion, polyelectrolyte complex formation,
much larger molecule of 1 × 107–​1 × 109 (Morrison and Karkalas,
electrospinning, electrospray and self-​ assembly) (Zhu, 2019;
1990; Mua and Jackson, 1997; Biliaderis, 1998; Buléon et al.,
Sun, 2018).
1998) and a heavily branched structure built from about 95%
Multiple factors affect the stability of starch-​based Pickering
(1→4)-​ α–​and 5% (1→6)-​α-​linkages.
emulsions. They include the type of starch particles, degree of

FIGURE 57.14 Micrographs of W/​O/​W Pickering emulsions stabilized by octenyl succinic anhydride (the scale bar is 10 μm).
(Matos, 2013, reprinted with permission from M. Rayner)
Hydrocolloid Usages 403

starch modification, starch source, starch concentration, type of viscous adsorbed layers at liquid–​ liquid interfaces, and their
oils, emulsion composition, pH and ionic strength (Zhu, 2019). dynamic adsorption behavior and surface rheological properties
have been the subject of several investigations (Erni et al., 2007;
Cellulose Derivatives Mezdour et al., 2007; Perez et al., 2008).
Chemical Composition/​Structure Cellulose derivatives behave similarly to any flexible or semi-​
Cellulose consists of a linear chain of β (1→4)-​linked D-​glucose flexible amphiphilic polymer, providing steric stabilization.
residues and is an important structural component in the primary Nanocellulose particles follow the behavior typically observed
cell wall of green plants and in biofilms produced by certain bac- in “Pickering” emulsions (Medronho et al., 2018). Cellulose
teria (Tavernier et al., 2016). Cellulose is abundantly available, nanocrystals (Figure 57.15) are able to adsorb at the oil–​water
biodegradable and non-​toxic. Native cellulose exists as macro- interface, suggesting surface-​ active properties. These proper-
scopic fibers or as microfibrillated cellulose (MFC) (Tingaut ties have been attributed to a surface heterogeneity induced by
et al., 2012). The amorphous domains can be removed by acid the crystalline organization (Tavernier et al., 2016). Cellulose
hydrolysis to yield microcrystalline cellulose (MCC). If MCC is nanocrystals modified with carboxymethylcellulose sodium
hydrolyzed for a longer period of time, then nanocrystalline cel- (CNCC) were prepared by homogenization technology, and it
lulose (NCC) is produced. was demonstrated that novel re-​dispersible powdered Pickering
emulsions could be prepared by cellulose nanocrystals coupling
Emulsifying Properties with water-​soluble cellulosic derivatives (Xie et al., 2019).
Good stabilization of emulsions and suspensions by adsorbed Another approach to producing colloidal particles of cellulose-​
polysaccharides can be achieved with various surface-​ based polymers is through non-​ covalent complexation with
active derivatives of celluloses such as methylcellulose, reactive natural polyphenols. Colloidal complexes formed by
hydroxyethylcellulose, hydroxypropylmethylcellulose and spontaneous interaction of methylcellulose (MC, a food-​grade
ethylhydroxyethylcellulose but with different underlying sta- cellulose derivative) and tannic acid showed excellent surface
bilization mechanisms (Wollenweber et al., 2000). Like activity and were further utilized for stabilization of foams,
hydrophobically modified starch, these macromolecules form emulsions and mixtures of both (Patel et al., 2012).

FIGURE 57.15 SEM image of cellulose nanocrystals.

(By Ahmad Hivechi, https://​creativecommons.org/​licenses/​by-​sa/​4.0/​deed.en)
404
Pectins
Chemical Composition/​Structure
Pectins are a group of widely distributed plant cell wall
polysaccharides that contain residues of galacturonic acid linked
at both the 1 and 4 positions. Many structurally different regions
were identified in the composition of pectins, and of these, three
classes of pectic polysaccharides have been extensively studied in
primary cell walls: homogalacturonan (HG), rhamnogalacturonan
I (RG-​I), and rhamnogalacturonan II (RG-​II) (Christiaens et al.,
2016). A particularity in the structure of pectic substances is the
esterification by methyl groups (at C-​6) and/​or acetyl groups (at
O-​2 and/​or O-​3) of galacturonic acid (GalA) residues on the con-
tinuous poly-​(GalA) chain of HG (Yapo and Koffi, 2013). The
overall degree of substitution is known as degree of methyla-
tion (DM) and acetylation (DAc), respectively. The DM, based
on which pectins are divided into two categories: high-​methoxyl
pectins (HMP, DM > 50%) and low-​methoxyl pectins (LMP, DM
< 50%), is an important parameter in choosing the most suitable
pectin application (Łekawska-​Andrinopoulou et al., 2013).
Emulsifying Properties
In the food industry, the main use for extracted pectin (labeled
in the European Union as E440) is as a gelling agent in the
production of fruit-​based products and fillings for bakery and
confectionary products. Pectin is also applied in food produc-
tion as a thickening, stabilizing and emulsifying agent. From
the point of view of these applications, the most frequently
exploited structure–​function relationship is the one between the
methoxylation of the galacturonic acids in the pectin backbone
and pectin gelation (Dranca and Oroian, 2018).
Among pectins, sugar beet pectin (SBP) has shown excellent
emulsifying properties due to many factors, such as the highly
branched polysaccharide structures, high content of hydrophobic
acetyl groups on the GalA chain, and protein moiety, which plays
an important role (Dranca and Oroian, 2018). SBP is not able
to form gels due to the low molecular mass and high degree of
acetylation (DAc). However, the acetyl groups and the protein
moiety are held to be responsible for the good emulsifying cap-
acity of SBP (Dea and Madden, 1986; Endreb and Rentschler,
1999; Leroux et al., 2003). Due to their low protein content and
number of acetyl groups, citrus and apple pectins are considered
less useful for emulsification (Schmidt et al., 2015). Karnick and
Wicker (2018) compared the stability of emulsions prepared with
protein-​rich and protein-​poor fractions of SBP by measuring the
particle size, steady stress controlled tests, and visual analysis
using dark field microscopy, and observed that the protein-​poor
fraction, with smaller particle size but higher degree of methyl
esterification (DE) and molecular weight, was a more effective
emulsifier than the pectin fraction rich in protein.
Pectin added as a stabilizer increases the viscosity of the
continuous aqueous phase of an emulsion and can influence
the droplet size of this emulsion in several ways. First, droplet
movement is reduced. This suppresses creaming (or sedimenta-
tion), which directly leads to an increased long-​term stability or
shelf life of emulsions but does not influence the droplet size.
Less droplet movement also leads to decreased droplet collisions.
Rachel Edwards-Stuart, Reine Barbar
This reduces droplet coalescence, which in turn influences long-​
term and also short-​term stability of emulsions (Sjöblom, 2006).
Protein Hydrocolloids and Their Interaction with
Polysaccharides
Physico-​Chemical Parameters of Interaction
The physical and chemical properties that govern protein func-
tionality include size, shape, amino acid composition and
sequence, net charge and distribution of charges, hydrophobicity/​
hydrophilicity ratio, secondary, tertiary and quaternary structures,
molecular flexibility/​rigidity, and the ability to interact/​react with
the other components (Damodaran, 1996). Proteins, similarly to
polysaccharides, play a key role in the structure and stabiliza-
tion of food systems through gelling, thickening and surface-​
stabilizing properties (Tolstoguzov, 1991). The rapidity with
which a protein can adsorb to air–​water or oil–​water interfaces
depends on the distribution pattern of hydrophobic and hydro-
philic patches on the surface.
Many polymers admixed with another had better functional
properties than the sum of them measured separately (Whistler
and BeMiller, 1977; Glicksman, 1982; Kulicke et al., 1996). The
blends bring about new functional properties and/​or alter the
rheology of food products. According to the physico-​chemical
conditions of protein–​ polysaccharide interactions (pH, ratio,
total concentrations of biopolymers and ionic strength), assem-
blies ranging from soluble nanocomplexes to coacervates induce
different functionalities (Figure 57.16). These functionalities
include emulsifying behaviors (Girard et al., 2002), foaming
properties (Liu et al., 2010), interfacial properties (Ducel et al.,
2005), rheological properties (Ru et al., 2012), and so on.
The pH effects on the physical state of the polymeric
species after performing acid or base titration are usually
monitored by light scattering intensity, hydrodynamic radius
and turbidity measurements (Mekhloufi et al., 2005). There
are three critical values of pH (pHc, pHφ and pHopt) that
describe the coacervation phenomena between oppositely
charged polysaccharides and proteins (Figure 57.17). pHc
(pH corresponding to the first change in the slope) (Weinbreck
et al., 2004; Liu et al., 2015) is related to the onset of soluble
complex formation. Above pHc, the biopolymers are like-​
charged and thus do not considerably interact with each other;
i.e., no complexes are formed. The other important pH value
is the value where insoluble coacervate particles are formed
(pHφ1), involving charge neutralization (estimated at the point
when the turbidity abruptly increases). Afterwards, the optical
density will reach a maximum at a certain pH value (pHopt),
where charge neutrality is reached between the biopolymers;
i.e., maximum coacervation is observed, and this is usually the
pH value utilized to produce coacervates in most of the studies
(Moschakis and Biliaderis, 2017).
Electrostatic soluble nanocomplexes can be formed due to
the attractive electrostatic interactions between the oppositely
charged molecules/​aggregates of proteins and polysaccharides
(Semenova, 2017).
Hydrocolloid Usages 405

Total pH,
Ion
biopolymer stre ic
n ngt
concentratio h

Mixing
Stirring, ratio
T°,
Pressure

Coacervates

FIGURE 57.16 Parameters predominantly affecting coacervate formation and their structural morphological characteristics.

FIGURE 57.17 Diagram of pH-​induced transitions upon performing acid or base titration in biopolymer systems (charge modification) that form coacervates.
Arrows indicate the characteristic pH values for conformational transitions and intermolecular associations/​dissociations.
(Adapted from Moschakis and Biliaderis, 2017)

Complex coacervation is another way of using the protein–​
polysaccharide associative interactions in order to produce
biopolymer-​based particles (Joye and McClements, 2016). This
is the association of the initially soluble protein–​polysaccharide
complexes into larger particles (“coacervates”), generally by
reducing pH closer to the pI of the protein (Semenova, 2017).
Although coacervation research is not novel, the majority
of work has been focused on studying interactions involving
polysaccharides with animal proteins (β-​lactoglobulin,
ovalbumin and bovine serum albumin). Plant proteins have
received a tremendous amount of attention over the past decade
from both consumers and industry due to the growing attention
towards the safety of animal-​ source food products, dietary
practices based on religious and cultural dietary matters, and
the lower cost relative to animal proteins (Warnakulasuriya
and Nickerson, 2018). Examples in the literature involve the
use of pea, lentil, canola, faba bean, napin, soybean, chia seed,
flaxseed, etc.
406
Functional Properties
Functional properties arise from a synergetic combination of the
functional features of both the protein (generally hydrophobic
and/​or hydrophilic and globular) and the polysaccharide (gener-
ally hydrophilic and branched) (Schmitt et al., 2009). Protein–​
polysaccharide complexes, either soluble or insoluble, are used to
stabilize emulsions for food applications, and the overall stability
is governed by their individual functionality and the nature of
interactions (Dickinson, 2008). The increased adsorption of the
complexes at the oil/​water interface compared with that of the
protein alone could be related to the higher hydrophobicity of the
complexes. In that case, hydrophobic residues of the protein could
be exposed toward the oil phase, with the hydrophilic groups of
polysaccharides remaining masked inside the complexes (Ducel
et al., 2005).
REFERENCES
Al-​Assaf S, Phillips GO, Williams PA. 2006. Controlling the
molecular structure of food hydrocolloids. Food Hydrocolloids,
20(2–​3), 369–​377.
Al-​Assaf S, Amar V, Phillips GO. 2008. Characterization of gum
ghatti and comparison with gum arabic. In Williams PA,
Phillips GO (eds.), Gums and Stabilisers for the Food Industry
14, Royal Society of Chemistry, Wrexham, 280–​290.
Anderson DMW, Stoddart JF. 1996. Studies on uronic acid materials:
Part XV. The use of molecular-​sieve chromatography in studies
on acacia senegal gum (gum arabic), Carbohydrate Research,
2, 104–​114.
Armisén R, Galatas F, Hispanagar S, Spain A. 2009. Agar. In Phillips
GO, Williams PA (eds.), Handbook of Hydrocolloids, 2nd ed.,
Woodhead, Cambridge, UK, 82–​107.
Banerjee S, Ravi R, Bhattacharya. 2013. Textural characterisation of
gellan and agar based fabricated gels with carrot juice. Food
Science and Technology, 53, 255–​261.
Biliaderis CG. 1998. Structures and phase transitions of starch
polymers. In Walter RH (ed.), Polysaccharide Association
Structures in Foods, Marcel Dekker, New York, 57–​168.
Branen AL, Davidson PM, Salminen S, Thorngate JH. 2002. Food
Additives, 2nd ed., Marcel Dekker, New York.
Brownlee IA. 2011. The physiological roles of dietary fibre. Food
Hydrocolloids, 25, 238–​250.
Buléon A, Colonna P, Planchot V, Ball S. 1998. Starch granules:
structure and biosynthesis. International Journal of Biological
Macromolecules, 23, 85–​112.
Chanamai R, McClements DJ. 2001. Depletion flocculation of bev-
erage emulsions by gum arabic and modified starch. Journal of
Food Science, 66(3), 457–​463.
Chanamai R, McClements DJ. 2002. Comparison of gum arabic,
modified starch, and whey protein isolate as emulsifiers: influ-
ence of pH, CaCl2 and temperature. Journal of Food Science,
67, 120–​125.
Chawla R, Patil GR. 2010. Soluble dietary fiber. Comprehensive
Reviews in Food Science and Food Safety, 9, 178–​196.
Christiaens S, Van Buggenhout S, Houben K, Kermani ZJ, Moelants
KRN, Ngouémazong ED, Van Loey A, Hendrickx MEG. 2016.
Process-​structure-​function relations of pectin in food. Critical
Reviews in Food Science and Nutrition, 8398 (February
2015), 37–​41.
Coultate T. 2009. Food: The Chemistry of its Components. RSC
Publishing, London, UK.
Rachel Edwards-Stuart, Reine Barbar
Damodaran S. 1996. Amino acids, peptides and proteins. In Fennema
OR (ed.), Food Chemistry, Marcel Dekker Inc., New York,
321–​429.
Day L, Golding M. 2016. Food structure, rheology and texture.
Encyclopedia of Food Chemistry, 125–​129.
Dea ICM, Madden JK. 1986. Acetylated pectic polysaccharides of
sugar beet. Food Hydrocolloids, 1, 71–​88.
Dickinson E. 1992. An Introduction to Food Colloids, Oxford
University Press, Oxford, UK.
Dickinson E. 2003. Hydrocolloids at interfaces and the influence
on the properties of dispersed systems. Food Hydrocolloids,
17, 25–​39.
Dickinson E. 2004. Effect of hydrocolloids on emulsion stability. In
Williams PA, Phillips GO (eds.), Gums and Stabilisers for the
Food Industry, Royal Society of Chemistry, Cambridge, UK,
394–​404.
Dickinson E. 2008. Interfacial structure and stability of food
emulsions as affected by protein–​polysaccharide interactions.
Soft Matter, 4, 932–​942.
Dickinson E. 2009a. Hydrocolloids and emulsion stability. In
Williams PA, Phillips GO (eds.), Handbook of Hydrocolloids
(2nd ed.), Woodhead, Cambridge, UK, 23–​49.
Dickinson E. 2009b. Hydrocolloids as emulsifiers and emulsion
stabilizers. Food Hydrocolloids, 23, 1473–​1482.
Dickinson, E. 2017. Biopolymer-​based particles as stabilizing agents
for emulsions and foams. Food Hydrocolloids, 68, 219–​231.
Dickinson E, Murray BS, Stainsby G, Anderson DJ. 1988. Surface
activity and emulsifying behavior of some acacia gums. Food
Hydrocolloids, 2, 477–​490.
Dranca F, Oroian M. 2018. Extraction, purification and charac-
terization of pectin from alternative sources with potential
technological applications. Food Research International, 113,
327–​350.
Ducel V, Richar, J, Popineau Y, Boury F. 2005. Rheological inter-
facial properties of plant protein-​arabic gum coacervates at the
oil-​water interface. Biomacromolecules, 6, 790–​796.
Endreß HU, Rentschler C. 1999. Chances and limit for the use of
pectin as emulsifier-​part 1. The European Food and Drink
Review, Summer, 49–​53.
Erni P, Windhab EJ, Gunde R, Graber M, Pfister B, Parker A, Fischer
P. 2007. Interfacial rheology of surface-​active biopolymers:
Acacia senegal versus hydrophobically modified starch.
Biomacromolecules, 8, 3458–​3466.
European Union. 2006. European Parliament and Council Directive
No 95/​2/​EC|of 20 February 1995| on food additives other than
colours and sweeteners (PDF). EUR-​Lex | Access to European
Union law. 15 August 2006.
FAO. 1999. Gum Arabic. Food and nutrition paper. In N52, addendum
7 (Rome).
FDA. 2020. www.fda.gov/​Food/​IngredientsPackagingLabeling/​
GRAS/​, last access 14 December 2020.
Gao Z, Fang Y, Cao Y, Liao H, Nishinari K, Phillips, GO,
2017. Hydrocolloid-​ food component interactions. Food
Hydrocolloids, 68, 149–​156.
Garti N, Leser M. 2001. Emulsification properties of hydrocolloids.
Polymers for Advanced Technologies, 12(1–​2), 123–​135.
Girard M, Turgeon SL, Paquin P. 2002. Emulsifying properties of
whey protein-​carboxymethylcellulose complexes. Journal of
Food Science, 67, 113–​119.
Glicksman M. 1982. Food Hydrocolloids, vol. 1., Glicksman M (ed.),
CRC Press, Boca Raton, Florida.
Hitchens HP, Leikind MC. 1939. The Introduction of agar-​agar into
bacteriology, Journal of Bacteriology, 37(5), 485–​493.
Hydrocolloid Usages
Hobbs CA, Swartz C, Maronpot R, Davis J, Recio L, Hayashi SM.
2012. Evaluation of the genotoxicity of the food additive, gum
ghatti. Food and Chemical Toxicology, 50, 854–​860.
Idris OMH, Williams PA, Phillips GO. 1998. Characterisation of
gum from Acacia senegal trees of different age and location
using multidetection gel permeation chromatography. Food
Hydrocolloids, 12, 379–​388.
Islam AM, Phillips GO, Sljivo A, Snowden MJ, Williams PA. 1997.
A review of recent developments on the regulatory, structural
and functional aspects of gum arabic. Food Hydrocolloids, 11,
493–​505.
Imeson A (ed.). 2010. Food Stabilizers, Thickeners and Gelling
Agents, Blackwell Publishing Ltd.
IUPAC. 2019. Compendium of Chemical Terminology, 2nd ed. (the
“Gold Book”), McNaught AD, Wilkinson A (eds.), Blackwell
Scientific Publications, Oxford (1997). https://​doi.org/​
10.1351/​goldbook.
Joye IJ, McClements DJ. 2016. Biopolymer-​based delivery systems:
challenges and opportunities. Current Topics in Medicinal
Chemistry, 16, 1026–​1039.
Kang J, Cui SW, Guo Q, Chen J, Wang Q, Phillips GO, Nikiforuk
J. 2012. Structural investigation of a glycoprotein from gum
ghatti. Carbohydrate Polymers, 89, 749–​758.
Kang J, Cui CW, Phillips GO, Chen J, Guo Q, Wang Q. 2011a. New
studies on gum ghatti (Anogeissus latifolia) Part II. Structure
characterization of an arabinogalactan from the gum by
1D, 2D NMR spectroscopy and methylation analysis. Food
Hydrocolloids, 25, 1991–​1998.
Kang J, Cui CW, Phillips GO, Chen J, Guo Q, Wang Q. 2011b. New
studies on gum ghatti (Anogeissus latifolia) Part III: struc-
ture characterization of a globular polysaccharide fraction by
1D, 2D NMR spectroscopy and methylation analysis. Food
Hydrocolloids, 25, 1999–​2007.
Kang J, Cui SW, Chen J, Phillips GO, Wu Y, Wang Q 2011c.
New studies on gum ghatti (Anogeissus latifolia) part
I. Fractionation, chemical and physical characterization of the
gum. Food Hydrocolloids, 25, 1984–​1990.
Kang J, Guo Q, Phillips GO, Cui SW. 2014. Understanding the
structure-​emulsification relationship of gum ghatti –​a review
of recent advances. Food Hydrocolloids, 42, 187–​195.
Karnik D, Wicker L. 2018. Emulsion stability of sugar beet pectin
fractions obtained by isopropanol fractionation. Food
Hydrocolloids, 74, 249–​254.
Kulicke WM, Eidam D, Kath F, Kix M, Kull AH. 1996. Hydrocolloids
and rheology: regulation of visco-​ elastic characteristics of
waxy rice starch in mixtures with galactomannans. Starch, 48,
105–​114.
Lee P, Rogers MA. 2012. Effect of calcium source and exposure
time on basic caviar spherification using sodium alginate.
International Journal of Gastronomy and Food Science, 1,
96–​100.
Łekawska-​ Andrinopoulou L, Vasiliou EG, Georgakopoulos DG,
Yialouris CP, Georgiou CA. 2013. Rapid enzymatic method
for pectin methyl esters determination. Journal of Analytical
Methods in Chemistry, 2013, 854763.
Leroux J, Langendorff V, Schick G, Vaishnav V, Mazoyer J. 2003.
Emulsion stabilizing properties of pectin. Food Hydrocolloids,
17, 455–​462.
Li JM, Nie SP. 2016. The functional and nutritional aspects of
hydrocolloids in foods. Food Hydrocolloids, 53, 46–​61.
Liu S, Elmer C, Low NH, Nickerson MT. 2010. Effect of pH on
the functional behaviour of pea protein isolate gum Arabic
complexes. Food Research International, 43, 489–​495.
407
Liu J, Shim YY, Wang Y, Reaney MJT. 2015. Intermolecular inter-
action and complex coacervation between bovine serum
albumin and gum from whole flaxseed (Linum usitatissimum
L.). Food Hydrocolloids, 49, 95–​103.
McClements DJ. 2005. Food Emulsions, 2nd ed., CRC Press, Boca
Raton, FL.
McGee H. 2004. On Food and Cooking. An Encyclopedia of Kitchen
Science, History and Culture. Hodder & Stoughton, UK.
McNamee BF, O’Riordan ED, O’Sullivan M. 1998. Emulsification
and encapsulation properties of gum arabic. Journal of
Agricultural and Food Chemistry, 46, 4551–​4555.
Matos M, Timgren A, Sjöö M, Dejmek P, Rayner M. 2013.
Preparation and encapsulation properties of double Pickering
emulsions stabilized by quinoa starch granules. Colloids and
Surfaces A: Physicochemical and Engineering Aspects, 423,
147–​153.
Medronho B, Filipe A, Costa C, Romano A, Lindman B, Edlund H,
Norgren M. 2018. Microrheology of novel cellulose stabilized
oil-​in-​water emulsions. Journal of Colloid and Interface
Science, 531, 225–​232.
Mekhloufi G, Sanchez C, Renard D, Guillemin S, Hardy J. 2005.
pH-​ induced structural transitions during complexation and
coacervation of β-​lactoglobulin and acacia gum. Langmuir,
21, 386–​394.
Mezdour S, Cuvelier G, Cash MJ, Michon C. 2007. Surface rheo-
logical properties of hydroxypropyl cellulose at air–​ water
interface. Food Hydrocolloids, 21, 776–​781.
Milani J, Maleki G. 2012. Hydrocolloids in Food Industry,
Food Industrial Processes –​Methods and Equipment,
Valdez B (ed.). Available from: www.intechopen.com/​
books/​food-​industrial-​processes-​methods-​a nd-​e quipment/​
hydrocolloids-​in-​foodindustry.
Morell P, Fiszman SM, Varela P, Hernando I. 2014. Hydrocolloids
for enhancing satiety: relating oral digestion to rheology, struc-
ture and sensory perception. Food Hydrocolloids, 41, 343–​353.
Morris ER, Nishinari K, Rinaudo M. 2012. Gelation of gellan –​a
review. Food Hydrocolloids, 28, 373–​411.
Morrison WR, Karkalas J. 1990. Starch. In Dey PM (ed.), Methods in
Plant Biochemistry, vol. 2, Academic Press, London, 323–​352.
Moschakis T, Biliaderis CG. 2017. Biopolymer-​based coacervates:
structures, functionality and applications in food products.
Current Opinion in Colloid & Interface Science, 28, 96–​109.
Myrhvold N, Young C, Bilet X. 2011. Modernist Cuisine: The Art
and Science of Cooking, The Cooking Lab.
Mua JP, Jackson DS. 1997. Fine structure of corn amylose and
amylopectin fractions with various molecular weights. Journal
of Agricultural and Food Chemistry, 45, 3840–​3847.
Nilsson L, Bergenståhl B. 2007. Emulsification and adsorption prop-
erties of hydrophobically modified potato and barley starch.
Journal of Agricultural and Food Chemistry, 55, 1469–​1474.
Noda S, Funami T, Nakauma M, Asai I, Takahashi R, Al-​Assaf S,
Ikeda S, Nishinari KO, Phillips G. 2008. Molecular structures
of gellan gum imaged with atomic force microscopy in rela-
tion to the rheological behavior in aqueous systems. 1. Gellan
gum with various acyl contents in the presence and absence of
potassium. Food Hydrocolloids, 22, 1148–​1159.
Nussinovitch A, Hirashima M. 2019. More Cooking Innovations:
Novel Hydrocolloids for Special Dishes, Taylor & Francis/​
CRC Press, Boca Raton.
Osorio FA, Bilbao E, Bustos R, Alvarez F. 2007.Effects of concen-
tration, bloom degree, and pH on gelatin melting and gelling
temperatures using small amplitude oscillatory rheology,
International Journal of Food Properties, 10(4), 841–​851.
408
Ouwerx C, Velings N, Mestdagh MM, Axelos MAV. 1998. Physico-​
chemical properties and rheology of alginate gel beads formed
with various divalent cations. Polymer Gels and Networks, 6,
393–​408.
Patel AR, Drost E, Blijdenstein TBJ, Velikov KP. 2012. Stable and
temperature-​responsive surfactant-​free foamulsions with high
oil-​volume fraction. ChemPhysChem, 13, 3777–​3781.
Perez OE, Carrera Sanchez C, Pilosof AMR, Rodriguez Patino
JM. 2008. Dynamics of adsorption of hydroxypropyl
methylcellulose at the air–​water interface. Food Hydrocolloids,
22, 387–​402.
Prochaska K, Kedziora P, Le Thanh J, Lewandowicz G. 2007. Surface
activity of commercial food grade modified starches. Colloids
and Surfaces B, 60, 187–​194.
Rana V, Rai P, Tiwary AK, Singh RS, Kennedy JF, Knill CJ. 2011.
Modified gums: approaches and applications in drug delivery.
Carbohydrate Polymers, 83, 1031–​1047.
Randall RC, Phillips GO, Williams PA. 1989. Fractionation and char-
acterization of gum from Acacia senegal. Food Hydrocolloids,
3(1), 65–​75.
Ru Q, Wang Y, Lee J, Ding Y, Huang Q. 2012. Turbidity and rheo-
logical properties of bovine serum albumin/​pectin coacervates:
effect of salt concentration and initial protein/​polysaccharide
ratio. Carbohydrate Polymers, 88, 838–​846.
Sanchez C, Nigen M, Tamayo MV, Doco T, Williams P, Amine
C, Renard C. 2018. Acacia gum: history of the future. Food
Hydrocolloids, 78, 140–​160.
Schmidt US, Schmidt K, Kurz T, Endreß HU, Schuchmann JP. 2015.
Pectins of different origin and their performance in forming
and stabilizing oil-​in-​water-​emulsions. Food Hydrocolloids,
46, 59–​66.
Schmitt C, Aberkane L, Sanchez C. 2009. Protein-​polysaccharide
complexes and coacervates. In Phillips GO, Williams PA
(eds.), Handbook of hydrocolloids, CRC Press, New York,
420–​462.
Science and cooking. 2016. www.scienceofcooking.com/​chemical_​
physical_​properties_​agar.htm, last access 10 January 2020.
Semenova M. 2017. Protein–​polysaccharide associative interactions
in the design of tailor-​ made colloidal particles. Current
Opinion in Colloid & Interface Science, 28, 15–​21.
Sjöblom J. 2006. In Sjöblom J (ed.), Emulsions and Emulsion
Stability (2nd ed.), Taylor & Francis Group, Boca Raton, 688.
Stokke BT, Drager KI, Yuguchi Y, Urakawa H, Kajiwara K. 1997.
Small-​angle X-​ray scattering and rheological characterization
of alginate gels, Macromolecular Symposia, 120, 91–​101.
Sun Q. 2018. Starch nanoparticles. In Sjöö M, Nilsson L (Eds.),
Starch in Food: Structure, Function and Applications, 2nd ed.,
Woodhead Publishing, Duxford, 691–​745.
Sworn G. 2009. Gellan gum. In Phillips GO, Williams PA (Eds.),
Handbook of Hydrocolloids, 2nd ed., Woodhead, Cambridge,
UK, 204–​227.
Taherian AR, Fustier P, Ramaswamy HS. 2007. Effects of added
weighting agent and xanthan gum on stability and rheological
properties of beverage cloud emulsions formulated using
modified starch. Journal of Food Process Engineering, 30,
204–​224.
Tavernier I, Wijaya W, Van der Meeren P, Dewettinck K, Patel AR.
2016. Food-​grade particles for emulsion stabilization. Trends
in Food Science & Technology, 50, 159–​174.
Tester RF, Karkalas J, Qi X. 2004. Starch composition, fine structure
and architecture. Journal of Cereal Science, 39, 151–​165.
Rachel Edwards-Stuart, Reine Barbar
Tingaut P, Zimmermann T, Sebe G. 2012. Cellulose nanocrystals
and microfibrillated cellulose as building blocks for the design
of hierarchical functional materials. Journal of Materials
Chemistry, 22, 20105–​20111.
Tolstoguzov B. 1991. Functional properties of food proteins and role
of protein-​polysaccharide interaction. Food Hydrocolloids, 4,
429–​468.
Tsai FU, Kitamura Y, Kokawa M. 2017. Liquid-​ core alginate
hydrogel beads loaded with functional compounds of radish
by-​products by reverse spherification: optimization by
response surface methodology. International Journal of
Biological Macromolecules, 96, 600–​610.
Valdez B. 2012. Food Industrial Processes –​Methods and Equipment.
InTech, Croatia.
Van der Sman RGM. 2012. Soft matter approaches to food structuring.
Advances in Colloid and Interface Science, 176–​177, 18–​30.
Vega C, Ubbink J, van der Linden E (eds.). 2012. The Kitchen as a
Laboratory –​Reflections on the Science of Food and Cooking,
Columbia University Press, New York.
Verbeken D, Dierckx S, Dewettinck K. 2003. Exudate gums:
occurrence, production, and applications. Applied Microbiology
and Biotechnology, 63, 10–​21.
Vilgis T. 2015. Gels; model systems for soft matter food physics.
Current Opinion in Food Science, 3, 71–​84.
Warnakulasuriya SN, Nickerson MT. 2018. Review on plant protein–​
polysaccharide complex coacervation, and the functionality
and applicability of formed complexes. Journal of the Science
of Food and Agriculture, 98, 5559–​5571.
Weinbreck F, Nieuwenhuijse H, Robijn GW, de Kruif CG. 2004.
Complexation of whey proteins with carrageenan. Journal of
Agricultural and Food Chemistry, 52, 3550–​3555.
Whistler RL, BeMiller JN. 1977. Carbohydrate Chemistry for Food
Scientists, Eagan Press, St. Paul, Minnesota, 63–​217.
Wijaya W, Patel AR, Setiowati AD, Van der Meeren P. 2017.
Functional colloids from proteins and polysaccharides for food
applications. Trends in Food Science & Technology, 68, 56–​69.
Williams PA, Phillips GO. 2009. Introduction to food hydrocolloids.
In Phillips GO, Williams PA (eds.), Handbook of Hydrocolloids,
CRC Press, New York, 1–​22.
Wollenweber C, Makievski AV, Miller R, Daniels R. 2000. Adsorption
of hydroxypropyl methylcellulose at the liquid/​liquid interface
and the effect on emulsion stability. Colloids and Surfaces A,
172, 91–​101.
Xie J, Luo Y, Chen Y, Liu Y, Ma Y, Zheng Q, Yue P, Yang M.
2019. Redispersible Pickering emulsion powder stabilized
by nanocrystalline cellulose combining with cellulosic
derivatives. Carbohydrate Polymers, 213, 128–​137.
Yapo, B, Koffi K. 2013. Extraction and characterization of highly
gelling low methoxy pectin from cashew apple pomace. Foods,
3, 1–​12.
Zhu F. 2019. Starch based Pickering emulsions: fabrication, proper-
ties, and applications. Trends in Food Science & Technology,
85, 129–​137.
Zion. 2018. Food Hydrocolloids Market by Type (Gelatin, Xanthan,
Carrageenan, Alginate, Agar, Pectin, Guar, Locust Bean, Gum
Arabic, and Carboxymethyl Cellulose), by Application (Bakery
& Confectionery, Meat & Poultry, Sauces & Dressings,
Beverages, and Dairy Products) –​Global Industry Perspective,
Comprehensive Analysis and Forecast, 2017–​ 2024. www.
zionmarketresearch.com/​report/​food-​hydrocolloids-​market.
Imaging Foodstuffs and Products of Culinary
Transformations
Mathias Porsmose Clausen1, Morten Christensen2 and Ole G. Mouritsen3
1 Department of Green Technology, SDU Biotechnology, University of Southern Denmark, 55 Campusvej, DK-​5230 Odense M,
Denmark
2 Taste for Life, University College Lillebælt, c/​o Kold College, 55 Landbrugsvej, DK-​5260 Odense S, Denmark
3 Department of Food Science, Taste for Life, Design and Consumer Behavior, University of Copenhagen, 26 Rolighedsvej,
DK-​1958 Frederiksberg C, Denmark
Imaging and visualization of foodstuffs and the products of
culinary preparations can serve as a means to understand how
texture and mouthfeel may arise from the structure of food and
food products on various length and time scales. As examples, we
will show results of advanced microscopy applied to emulsions
(such as mayonnaise, butter, and whipped cream), cheese, dried
vegetables, and cooked squid.
The old proverb about an image saying more than a thousand
words can also be applied to gastronomic reflections concerning
the relationship between food structure, on the one side, and the
sensory perception (specifically texture and mouthfeel) on the
other side. Images can be the first visual verification as to whether
a hypothesis about a microscopic food structure is likely to be
true, but conversely, images can also be seductive and deceiving
if they are incorrectly interpreted or assumed to be more typical
for the situation than is actually the case. Still, there is a lot to
learn from imaging of food for both the observant scientist and
the curious chef.
Due to their biological origin, raw food ingredients display
a hierarchical structure on many time and length scales (Vilgis,
2015; Mouritsen and Styrbæk, 2017a). This structure can become
dramatically modified under culinary transformations, because
the different macromolecules and their internal organization
and mutual interactions are affected by conditions of, e.g., tem-
perature, pH, ions, and the activity of water in addition to a host
of physico-​ chemical processes and reactions (Barham et al.,
2010). Formulated foods like sauces, bread, pastry, chocolate,
and cheese, whether they are prepared by traditional cooking
techniques or by note-​by-​note cooking (This, 2014), also have
a rich internal structure and dynamics on many scales caused by
self-​assembly processes, although less complicated than for most
raw ingredients directly derived from plants, animals, or algae.
The imaging techniques of food proceed traditionally by various
kinds of simple transmission light or fluorescence microscopies,
although mechanical imaging techniques like ultrasound (Chemat
et al., 2011), atomic force microscopy (Gunning and Morris,
2018), and indirect, non-​invasive, and non-​destructive techniques
using spectroscopic methods such as nuclear magnetic reson-
ance (NMR) imaging (Ebrahimnejad et al., 2018) have also made
progress, particularly regarding imaging on the molecular scale
and with regard to water structure and dynamics. X-​ray imaging
(Nielsen et al., 2016) combined with modern data processing and
tomography capabilities has also been demonstrated to image
food, and electron microscopy (Dudkiewicz et al., 2011), typ-
ically of solvent-​ exchanged and critical-​ point dried samples,
continues to be important for imaging foodstuff on small length
scales, often with molecular resolution.
Optical microscopy with and without extrinsic labeling has
witnessed dramatic development in recent years, not least due to
the advent of super-​resolution methods and non-​linear, laser-​based
optical techniques with high penetration power and sectioning
capabilities. Many of these techniques are promising candidates
for investigating food matter. Super-​resolution microscopy offers
the benefits of optical microscopy, but with a ~10-​fold better
spatial resolution compared with classical diffraction-​ limited
microscopy. However, this requires fluorescent labeling of the
structures of interest. In contrast to this are so-​called CARS-​
microscopy (CARS: coherent anti-​Stokes Raman scattering) and
so-​called SHG (second-​harmonic generation) microscopy. Both
types of microscopy are limited in resolution by the diffraction of
light to about half the wavelength of the light used. The advan-
tage of CARS and SHG microscopy is that they need no probes
or staining of the material, and it is possible for CARS to dis-
cern regions of, e.g., fat, protein, carbohydrates, and water in the
sample (Han et al., 2018; Brüggemann et al., 2010). SHG can
be used on foodstuffs with a certain molecular ordering without
center symmetry, which will be the case for, e.g., collagen.
We shall show below how various kinds of advanced micros-
copy can be used to image food and products of culinary trans-
formations, with a focus on how one can possibly relate structure
to texture in specific cases (Bourne, 2002). We believe that
imaging holds potential for gastronomic development, in addition
409
410
to being a useful tool in culinary science and science education
and for enhancing the public understanding of the interaction
between science and gastronomy (Sörensen and Mouritsen,
2019). Furthermore, it is our experience that the availability of
a simple light microscope in the gastronomic kitchen will also
stimulate chef apprentices who are interested in learning more
about the fundamentals of food structure and culinary techniques,
with the underlying aim of enabling them to make more qualified
decisions regarding preparations in the kitchen (Christensen and
Edwards-​Stuart, 2019).
Imaging of Emulsions: Mayonnaise, Butter, and
Whipped Cream
Emulsified systems are widespread in processed foodstuffs, like
sauces, ice cream, butter, cheese, chocolate, foams, fudge, etc.
(Barham et al., 2010; Mouritsen and Styrbæk, 2017a). They are
all fairly complex systems involving several different thermo-
dynamic phases, be they gas, liquid, or solid. In general, these
systems are dispersions of different materials that do not easily
mix because of conflicts between hydrophobic and hydrophilic
components, e.g., oil/​air and aqueous phases. The stability of such
colloidal systems is ensured by interfacially active molecules,
such as certain amphiphilic lipids and proteins (Mouritsen and
Bagatolli, 2016), e.g., lecithin from egg yolk in the case of an oil–​
water emulsion or globulins and ovotransferrin from egg white in
the case of an air–​water emulsion (a foam). The structure of the
emulsions on scales from a few micrometers and greater is essen-
tial for the texture, i.e., the perceived mouthfeel, of the foodstuff.
As an example, we show in Figure 58.1 the structure of a may-
onnaise, which is an oil-​in-​water emulsion (Christensen et al.,
2015). The image demonstrates that this structure is a disper-
sion of spherical droplets of oil in a continuous aqueous phase.
FIGURE 58.1 Imaging mayonnaise, an oil-​in-​water emulsion. The spheres
are droplets of oil, and the blue areas in between them make up a continuous
phase of water. (a) CARS microscopy image, scale bar 10 µm. (b) Illustration
of the mode of action of the emulsifier (egg lecithin) at the surfaces of the oil
droplets. (The mayonnaise was prepared by mixing 5 g of pasteurized egg
yolk, 0.5 mL white wine vinegar, and 0.2 g table salt, followed by slowly
adding 40 mL refined sunflower oil while whisking.)
Mathias Porsmose Clausen et al.
It is the small sizes of the oil droplets (1–​5 µm), the fact that
the droplets can move among each other easily and unhindered
upon squeezing (e.g., between the tongue and the palate), and
the shear-​thinning properties that provide the mechanism that
explains the creaminess of the emulsion (Liu et al., 2007).
Systematic imaging of such emulsion formations under
different circumstances, e.g., by varying oil/​water composition,
pH, temperature, whipping intensity, etc., holds promise for
optimizing the texture and mouthfeel of mayonnaise and other
similar oil-​in-​water emulsions. Reversal of the structure into a
water-​in-​oil emulsion results in the structure of foodstuffs like
butter and margarine. An example of the imaging of butter is
shown in Figure 58.2a.
Some foods that are very rich in fats can form a type of
foam without the help of emulsifiers. In whipped cream, the
solid milk fat globules are partially broken into pieces and
clumped together in a network that contributes stability and
a certain degree of stiffness (Mouritsen and Styrbæk, 2017a).
The fat particles stabilize the air bubbles, which lie far apart
FIGURE 58.2 Imaging dairy products. (a) Microstructure of butter: the
yellow solid mass is fat, and the blue droplets are water. Left: image obtained
by CARS microscopy, scale bar 10 μm. Right: schematic illustration of butter
structure; the fat is partially crystalline (indicated by elongated diamonds)
and partially semi-​solid. (b) Microstructure of whipped cream: the large dark
bubbles are air, and the walls between them are made up primarily of fats
interspersed with water. Left: image obtained by CARS microscopy, scale bar
200 µm. Right: schematic illustration of whipped cream. (c) Microstructure
of a cheese curd formed when the casein micelles in milk coagulate into a
network in the presence of rennet. Left: image obtained by STED super-​
resolution microscopy (top left insert: confocal microscopy, scale bar 4 µm).
Right: schematic illustration of this network. (The cheese curd sample was
prepared by staining skim milk with Atto488, then adding rennet, and leaving
the mixture for 1 hour before imaging.)
Imaging Foodstuffs and Products
from one another; this is the opposite of what happens in a
real foam with thin walls. Whipped cream consists of a solidly
packed network of air bubbles held together by small spheres
of fat that attach themselves to the surface of the air bubbles.
At least 30% fat, and preferably more, is required to make a
stable foam. The larger fat particles in the cream must be broken
up into pieces that are sufficiently small to be captured by the
air bubbles. This process releases the milk proteins from their
original fat spheres, which in turn, renders them more unstable
and increases their tendency to form larger fat particles. That is
why the cream needs to be beaten, in fact whipped, to the stage
where the fat particles are so small that they do not coalesce
before the stiff network of air bubbles is formed. This is also
why temperature is important; when the cream is cold, the milk
fat globules will aggregate, but it is much harder for the solid fat
to coalesce (McGee, 2014).
Figure 58.2b shows a microscopic image of whipped cream
along with a schematic illustration of the structure. These
images support the description of the whipped cream given
here and provide an understanding of the creamy mouthfeel of
whipped cream.
Imaging of Cheese
Hard cheeses basically owe their solid structure to a network of
protein particles called casein micelles that contain from 10,000
to 100,000 protein (casein) molecules and have a diameter of
0.01–​0.3 μm (McGee, 2014; Mouritsen and Styrbæk, 2017a).
In a commonly accepted model of the casein micelle surface,
the kappa-​casein molecules look a bit like hairs on the casein
micelles (De Kruif and Holt, 2003). The native casein micelles
have a negative electrical charge that causes them to repel one
another and prevents them from aggregating. If the milk is acid-
ified (soured), the electrostatic repulsion decreases, and the milk
coagulates into cheese curds when the casein micelles form a
network. This may also capture any milk fat globules present
in the milk. For most cheese types, the process of coagulation
is initiated by the aid of a particular enzyme called chymosin
(originally found in the extract called rennet). Rennet specific-
ally cleaves off the hydrophilic hair-​like structure on the casein
micelles, and the now slightly hydrophilic casein micelles aggre-
gate to form a network (Walstra et al., 2005).
The considerable differences in texture of different cheeses
depend on the forces that bind the casein proteins within the
cheese curd and the change in structure upon maturation. The
proteins provide structural integrity and keep the fats and the water
content from going their separate ways. How well the proteins are
bound to one another is a function of the calcium content and the
associated calcium ions (Gaucheron, 2005). In turn, the ability of
the calcium ions to do the job depends on other factors, especially
the acidity of the cheese. Moreover, as the cheese ages, some
of the protein is broken down, which leads to changes in tex-
ture and the release of amino acids, including glutamate, which
enhances the umami-​taste (taste of monosodium glutamate).
Figure 58.2c shows a STED (stimulated emission depletion)
super-​resolution microscopic image of the microstructure of a
411
cheese mass formed when the casein micelles in milk coagulate
into a network in the presence of rennet, along with a sche-
matic illustration of this network (right). It is seen that the
micelle network is much better resolved in the STED image in
comparison with the confocal microscopy image (insert in top
left corner). A highly resolved image is essential for analyzing,
e.g., the specific position of the casein micelles in relation to
understanding the aggregation mechanism when the structure
is formed.
Imaging of Dried Vegetables: Crunchy Japanese
Tsukemono
Classical Japanese cuisine uses a number of pickling
techniques, so-​ called tsukemono (Mouritsen and Styrbæk,
2021a; Mouritsen, 2018), that involve different salts, sugar, vin-
egar, alcohol, enzymatic and bacterial fermentation, as well as
various pickling beds of miso, soy sauce, and sake lees (see
chapter on tsukemono by Mouritsen in this book). The pick-
ling aids in conservation as well as enhancement of nutritional
value. It also has significant effects on taste, aroma, and par-
ticularly mouthfeel. Often, the umami taste is increased, but the
most striking feature of most tsukemono is its unique mouthfeel
of crunchiness (Mouritsen, 2018). An understanding of the sci-
entific mechanisms behind this texture involves the effect of
ions, in particular divalent ions such as calcium and magne-
sium ions (from sea salt), pH, and enzymes. In many cases, the
crunchy mouthfeel is enhanced by first drying the vegetables
before marinating them (unpublished data).
Dehydration is a rough process for the cells of vegetables; it
changes their shape, and the cell walls can be broken, altering the
texture of the vegetable. The sugar content of the dried vegetable
also affects its texture; those with more sugar will be chewier and
more flexible. Sugar binds the water and will also cause water to
be absorbed from the surroundings. Vegetables with a great deal
of soluble dietary fiber have a tendency to turn out more crunchy
than those with a high sugar content.
Typically, the temperature in a vegetable dehydrator is kept in
the range 40−60 °C in order to preserve color as well as taste and
aroma. The partial dehydration of the vegetables diminishes the
growth of bacteria and enzymatic activity on their surfaces, thus
lessening the risk of deterioration. Nevertheless, drying in itself
is not sufficient to preserve the vegetables; this will require the
addition of salt, which increases the osmotic pressure across cell
membranes, causing the rupture of cell walls.
Figure 58.3 shows fluorescence wide-​field microscopy images
of the cellular structure of a root of a radish (Raphanus sativus)
in the fresh state, after drying (dehydration), and after rehydra-
tion (in a marinade). During the dehydration, the cells shrink, and
the stiff and regular network of cell walls becomes crumpled and
more flexible. During subsequent rehydration (in marinade), the
cells take up water again, but the network remains irregular and
flexible. This implies that the perceived texture in the mouth is one
where the vegetable is pliable and deformable, but when the teeth
cut through the vegetable it fractures with a crunchy mouthfeel
and a noise or pleasant auditory sensation (Bourne, 2002).
412
FIGURE 58.3 Fluorescence widefield microscopy images of the cell struc-
ture in a radish. From left to right: while fresh, after dehydration, and after
rehydration (marinating). Scale bar is 200 µm. (Plant cell walls were stained
using CalcoFluor white.)
Imaging Squid
Cephalopods, i.e., squid, octopuses, and cuttlefish, are widely
used as marine food by humans across the world and in different
food cultures. In addition to being consumed raw, various world
cuisines have prepared cephalopods by many different culinary
techniques, such as boiling, frying, grilling, marinating, smoking,
drying, and fermenting. Squid, octopuses, and cuttlefish are
common foods in Southeast Asia and some parts of Europe, but
they seldom enter regional cuisines in North America and nor-
thern Europe, although the local waters there often have abun-
dant sources of edible species. There is increasing interest among
chefs, gastroscientists, and environmentalists in sourcing from
local waters in a more diverse and sustainable fashion, including
novel uses of cephalopods to counterbalance dwindling fish-
eries and to find new protein sources to replace meat from land
animals (Mouritsen and Styrbæk, 2021b). The focus of the chefs
and gastroscientists is on the texture and flavor properties of the
different species subjected to different culinary transformations.
Combining this gastronomic development with the observa-
tion that the global populations of cephalopods are on the rise
holds an interesting promise for a future for cephalopod cooking
(Mouritsen and Styrbæk, 2018).
Squid, cuttlefish, and in particular octopuses are notoriously
known to be challenging to prepare due to their high content of
collagen, typically three to four times as much as fish and land
animals (Mouritsen and Styrbæk, 2021b). In many cases, the col-
lagen network of the muscles needs to be tenderized in order to
provide a desirable mouthfeel. We have undertaken a systematic
study of the texture of squid (Loligo forbesii) subject to a wide
range of different preparation techniques, e.g., sous vide cooking
(Faxholm et al., 2018), and we have studied the relationship
between macroscopic texture measurements and microscopic
imaging of the structure of the collagen network. Such studies,
Mathias Porsmose Clausen et al.
FIGURE 58.4 Second-​ harmonic microscopy images of the collagen
structure in raw squid (a) and after sous vide treatment (b) of squid (Loligo
forbesii). The scale bar is 50 μm.
combined with sensory analysis, can aid in designing optimized
culinary precisions for preparing squid (Schmidt et al., 2020). As
an example of the results of this kind of approach, we show, in
Figure 58.4, SHG microscopy images of the collagen structure of
the mantle of the squid Loligo forbesii before and after sous vide
cooking. The images clearly show how the heating of the muscle
leads to a breakdown of the connectedness of the collagen net-
work, rendering the tissue more tender.
Conclusion
As demonstrated in this chapter, by applying novel light micros-
copy tools, we can directly observe structural changes inside food
materials that have undergone culinary processes. We believe that
the use of microscopy tools and other imaging techniques will be an
important step in further understanding food texture at the macro-
scopic level and thereby serve to facilitate future food innovation.
Acknowledgments
This work was supported in part by a grant to Smag for Livet
(Taste for Life) from Nordea-​fonden and by a Villum YIP grant
(ID: 00025414).
REFERENCES
Barham P, Skibsted LH, Bredie WLP, Frøst MB, Møller P, Risbo
J, Snitkjaer P, Mortensen LM. 2010. Molecular gastronomy:
A new emerging scientific discipline. Chem. Rev., 110,
2313–​2365.
Bourne M, 2002. Food Texture and Viscosity: Concept and
Measurement, 2nd ed., Academic Press, San Diego.
Brüggemann DA, Brewer J, Risbo J, Bagatolli L. 2010. Second har-
monic generation microscopy: A tool for spatially and tem-
porally resolved studies of heat induced structural changes in
meat. Food Biophys., 5, 1–​8.
Chemat F, Zill-​ e-​Huma KMK. 2011. Applications of ultrasound
in food technology: Processing, preservation and extraction.
Ultrasonics Sonochem., 18, 813–​835.
Christensen M, Clausen MP, Mouritsen OG. 2015. Se ind i maden.
Akt. Nat., 6, 22–​27.
Imaging Foodstuffs and Products
Christensen M, Edwards-​Stuart R. 2019. Teaching science to chefs:
The benefits, challenges and opportunities. J. Gast. Food Sci.,
16, 100133.
De Kruif CG, Holt C. 2003. Casein micelle structure, functions and
interactions. In Advanced Dairy Chemistry. Vol. 1. Proteins
(Fox F, McSweeney PLH, eds.) Springer Science+Business
Media, New York, pp. 233–​276.
Dudkiewicz A, Tiede K, Loeschner K, Jensen LHS, Jensen E,
Wierzbicki R, Boxall ABA, Molhave K. 2011. Characterization
of nanomaterials in food by electron microscopy. Trends. Anal.
Chem., 30, 28–​43.
Ebrahimnejad H, Ebrahimnejad H, Salajegheh A, Barghi H. 2018.
Use of magnetic resonance imaging in food quality control: A
review. J. Biomed. Phys. Eng., 8, 127–​132.
Faxholm PL, Schmidt CV, Brønnum LB, Sun YT, Clausen MP,
Flore R, Olsen K, Mouritsen OG. 2018. Squids of the North:
Gastronomy and gastrophysics of Danish squid. Int. J. Gast.
Food Sci., 14, 66–​76.
Gaucheron F. 2005. The minerals of milk. Reprod. Nutr. Dev., 45,
473–​483.
Gunning AP, Morris VJ. 2018. Getting the feel of food structure with
atomic force microscopy. Food Hydrocoll., 78, 62–​76.
Han M, Clausen MP, Christensen M, Vossen E, Van Hecked T, Bertram
HC. 2018. Enhancing the health potential of processed meat:
The effect of chitosan or carboxymethyl cellulose enrichment
on inherent microstructure, water mobility and oxidation in a
meat-​based food matrix. Food Funct., 7, 4017–​4027.
Liu H, Xu XM, Guob SD. 2007. Rheological, texture and sensory
properties of low-​fat mayonnaise with different fat mimetics.
LWT –​Food Sci. Technol., 40, 946–​954.
McGee H. 2014. On Food and Cooking: The Science and Lore of the
Kitchen. Scribner, New York.
413
Mouritsen OG. 2018. Tsukemono –​crunchy pickled foods from
Japan: A case study of food design by gastrophysics and
nature. Int. J. Food Design, 3, 103–​124.
Mouritsen OG, Bagatolli LA. 2016. Life –​as a Matter of Fat.
Springer, New York.
Mouritsen OG, Styrbæk K. 2017a. Mouthfeel: How Texture Makes
Taste. Columbia University Press, New York.
Mouritsen OG, Styrbæk K. 2021b. Tsukemono: Decoding the Art
and Science of Japanese Pickling. Springer, New York.
Mouritsen OG, Styrbæk K. 2018. Cephalopod gastronomy –​a promise
for the future. Front. Comm. Sci. Environ. Comm., 3, 38.
Mouritsen OG, Styrbæk K. 2021b. Cephalopods –​Octopuses, Squid
& Cuttlefish: Seafood for Today and for the Future. Springer,
New York.
Nielsen MS, Munk MB, Diaz A, Pedersen EBL, Holler M, Bruns S,
Risbo J, Mortensen K, Feidenhans’l RK. 2016. Ptychographic
X-​ray computed tomography of extended colloidal networks in
food emulsions. Food Struct., 7, 21–​28.
Schmidt CV, Plankensteiner L, Faxholm PL, Olsen K, Mouritsen
OG, Frøst MB. 2021. Physicochemical characterisation of
sous vide cooked squid (Loligo forbesii and Loligo vulgaris)
and the relationship to selected sensory properties and hedonic
response. Int. J. Gast. Food Sci., 23:100298.
Sörensen PM, Mouritsen OG. 2019. Science education and public
understanding of science via food, cooking, and flavour. Int.
J. Gast. Food. Sci., 15, 36–​47.
This H. 2014. Note-​by-​Note Cooking. Columbia University Press,
New York.
Walstra P, Walstra P, Wouters JTM, Geurts TJ. 2005. Dairy Science
and Technology. 2nd ed. CRC Press, Boca Raton.
Vilgis TA. 2015. Soft matter food physics –​the physics of food and
cooking. Rep. Prog. Phys., 78, 124602.
Meat: Meat Tenderness and the Impact of Cooking

Jean-​François Hocquette1 and Alain Kondjoyan2

1 Clermont University, INRAE, VetAgro Sup, UMR 1213 Herbivores, Theix, 63122 Saint-​Genès Champanelle, France
2 INRAE, Unité Qualité des Produits Animaux, Theix, 63122 Saint-​Genès Champanelle, France

Since consumer satisfaction when eating meat (especially beef) Muscle Structure and Transformation of Muscle
is mainly driven by tenderness, a great deal of research has been into Meat during Ageing
done on muscle biochemistry and meat cooking. Skeletal muscles
Skeletal muscle is made of different tissues, mainly muscle
are made up of several cell types and an extracellular matrix, the
fibres, but also connective and adipose tissues (often named by
characteristics of which differ across and within muscles. These
consumers as “tendon” and “fat”, respectively), as well as nerves
differences induce variability in eating quality of meat either dir-
and blood vessels. Fibres and connective tissues are mainly
ectly or indirectly by affecting meat ageing. The latter is well
comprised of proteins. Muscle fibres, which are elongated and
known to butchers as a major way of increasing tenderness, but it
multinucleated cells, represent about 90% of the muscle tissue,
is a complex biochemical post-​mortem process.
and their size increases with animal age, thus explaining post-
Cooking is also a major factor affecting tenderness. Indeed,
natal muscle growth. The intracellular volume of muscle fibres is
heat during cooking induces contraction of the muscle struc-
mainly occupied by myofibrils, which are themselves composed
ture and a progressive denaturation of proteins and the extracel-
of myofilaments. The contractile functional unit of myofibrils is
lular matrix, to an extent that depends on cooking temperature.
the sarcomere, which consists of at least 30 different proteins, the
Grading schemes to predict eating quality, such as that of Meat
most abundant being myosin and actin.
Standards Australia, have started to include many of these factors
Thick and thin filaments of sarcomeres are mainly made of
from the production side to the consumer end. A complementary
myosin and actin, respectively. Myosin catalyses the breakdown
approach is to optimize cooking equipment and procedures.
of ATP into ADP, providing chemical energy used for contraction
Eating quality of meat, and therefore satisfaction of consumers,
is mainly driven by tenderness and flavour. Consumers know by
experience that problems of tenderness are mainly associated with
Farm animals (breed, sex, age, etc.)
red meat, and that meat obtained from older animals is gener-
ally tougher than meat from younger animals (Geay et al., 2001).
They are also aware that meat cuts differ in their tenderness and
therefore implicitly in biochemical characteristics (Listrat et al., Meat biochemical/structural characteristics Rearing
2016), which explains some differences in price between them. (connective tissue, intramuscular fat, muscle fibres) practices

Consumers are also concerned by the juiciness of white meat and

often find poultry meat, chosen as being lean meat, too dry.
Slaughtering conditions
An abnormal presence of juice in the packaging, excessive
cooking losses, or tougher meat may be interpreted as being the
Meat tenderization during ageing (proteolytic Storage
result of intensive genetic selection, poor management, stress enzymes, pH evolution) conditions
during slaughter, or even the effect of hormone injections into the
animal (this is strictly forbidden in Europe). The direct associ- Cooking method/equipment
ation by consumers of tough meat with problems connected with
animal husbandry or slaughter is often due to their poor know-
Meat sensory traits
ledge of the effect of meat ageing and cooking on meat tender- (tenderness, flavour, juiciness,
ness. In fact, tenderness of cooked meat is the result of many overall liking)

factors involved during animal production, slaughtering, meat

chilling, storage, and cooking (Figure 59.1). This chapter will
explore what is scientifically known about the evolution of meat Consumers

tenderness through all these different steps.

FIGURE 59.1 Main factors affecting meat sensory traits for consumers.

415
416
of sarcomeres. Muscle fibres are characterized by their contractile
(slow or fast) and metabolic (oxidative or glycolytic) properties.
Three main fibre types can be distinguished: slow oxidative (type
I), fast oxido-​glycolytic (type IIa), and fast glycolytic (types IIb
and IIx).
Connective tissue has three levels of scale: the endomysium,
which surrounds each muscle fibre; the perimysium, which
groups muscle fibres in fibre bundles; and finally the epimy-
sium, which is the external envelope of the whole muscle. The
extracellular matrix is made of a complex network of collagen
fibres in which other cells are embedded. The characteristics of
collagen, including its solubility, are the main determinants of
toughness. Briefly, muscles with less collagen or younger col-
lagen are more tender. Indeed, collagen fibres are increasingly
stabilized with intramolecular or intermolecular bonds, known as
cross-​links, with increasing age, resulting in greater toughness as
animals get older (Lepetit and Culioli, 1994). Mechanical prop-
erties of the muscle are also affected by the ratio of the volume
occupied by the fibres relative to the volume of the perimysium.
Early researchers reported that meat toughness, especially in beef
meat, was influenced by the dimensions of the main bundles that
are separated by the perimysium and by the thickness of this peri-
mysium (Purslow, 2005). This “meat grain” is still often taken
into account by chefs or butchers when they try to visually assess
the tenderness of meat.
Among the other cell types present in the extracellular matrix,
intramuscular adipocytes determine the intramuscular fat content.
The beef industry prefers working on “marbling”, which refers
to visible white flecks or streaks of intramuscular fat between
bundles of muscle fibres. Intramuscular fat or marbling has been
the subject of active research (Hocquette et al., 2010) due to its
positive influence on beef eating quality (flavour, juiciness, and
tenderness). The genetic determinism of intramuscular fat seems
specific compared with other fat deposits (Bernard et al., 2009).
The more intramuscular fat, the better the beef sensory quality.
Connective tissue is relatively stable after an animal’s death.
Thus, post-​mortem tenderization of meat, called ageing or mat-
uration, results mainly from a softening of the myofibrillar struc-
ture after rigor mortis by protein degradation due to a variety
of endogenous proteases and peptidases acting at different pH
values (Ouali et al., 2006). Protein degradation is also associated
with glycogen degradation; muscle glycogen is converted into
lactate, leading to a progressive acidification of the muscle tissue
(for a review on ageing tenderization, see Listrat et al., 2016).
Any perturbation in this process may hamper tenderness.
For instance, in animals under stress during transportation to
the abattoir, glycogen is depleted before slaughter, and due to
a lack of lactate produced from glycogen, the resulting pH in
the meat is too high, causing deficits in meat quality. The vari-
ability of ageing effects on meat quality is also largely affected
by the type of animal, the type of muscle, and chilling and storage
conditions. The speed of contraction of muscle fibres determines
the ageing rate after the animal’s death, with ageing of fast fibres
being more rapid, which explains why ageing is faster in pigs and
poultry compared with cattle (the maturation difference between
“red” and “white” meats known from experience by meat
experts). In beef, relationships between fibre characteristics differ
Jean-François Hocquette, Alain Kondjoyan
according to animal sex, age, and breed and also according to
the cut (Listrat et al., 2016). Abnormal ageing development can
be connected to genetics, as for pale soft exudative (PSE) pork
meat, or to slaughtering stress, as for dark firm and dry (DFD)
young bull meat. Ageing and meat tenderization are favoured
by slow chilling and higher temperatures; rapid chilling of the
carcass can stop the normal maturation and lead to very tough
beef meat due to a non-​reversible contraction of the myofibres
named “cold shortening”. Abnormal development of biochemical
reactions is generally associated with variations in the decrease of
meat pH, while cold shortening is characterized by shorter sarco-
mere length (Lepetit et al., 2000).
Evolution of Meat Structure during Cooking
Meat cuts can be cooked in many different ways: immersion in
liquid (boiling, deep-​frying, etc.), heating by contact (grilling,
pan-​frying, searing, etc.), and/​or convection, radiation (infrared
or microwaving), phase changes (steaming, etc.), or induction
(ohmic heating). Differences in the cooking technique, combined
with variations in the shape and size of the meat cut, lead to
differences in the local variations of temperature and water con-
centration in the cuts.
For example, the following experimental results were obtained
under laboratory-​ controlled heating conditions. Myofibrillar
proteins (myosin and actin) and proteins of intramuscular con-
nective tissues are the major compounds involved in the variation
of meat tenderness during cooking. Mechanical measurements
mimicking meat cutting by consumer teeth (Warner–​ Bratzler
tests) have proven that meat toughness increases with cooking,
with a first phase of increase between 40 and 50 °C, followed
by either a plateau or a decrease between 50 and 60 °C, and then
by a second phase of increase above 65 °C (Purslow, 2005). An
intermediate plateau of the shearing stress is commonly observed
between 50 and 60 °C for beef muscle, while a sharp decrease
in the stress has been observed for rabbit muscle (Combes
et al., 2003).
Heat contraction of the proteins leads to a three-​dimensional
contraction of the whole muscle structure, which can be observed
by magnetic resonance imaging (MRI) (Bouhrara et al., 2011).
Muscle contractions begin at 38 °C but become very important
between 54 and 70 °C (Bouhrara et al., 2012). Shearing stress of
muscle myofibres generally increases up to 90 °C, this increase
being related at low temperatures to the denaturation of myosin
and then to the denaturation of actin, which is more thermally
stable (Purslow, 2005). Direct measurements on perimysium
isolated from beef meat show that its strength increases in meat
cooked up to 50 °C and then decreases (Lewis and Purslow,
1990), leading to a major contribution to toughness by the con-
nective tissues in the range 20–​50 °C and by the myofibrillar
proteins above 60 °C (Purslow, 2005). Connective tissue still has
a predominant effect on muscle contraction between 55 °C and
65 °C (Lepetit et al., 2000).
In fact, the exact role of the denaturation of the different
proteins in muscle contraction remains under debate, because
the intramuscular connective tissue, which is made up of a
Meat: Meat Tenderness
continuous perimysium and endomysium network, differs
between muscles and within meat cuts, the myofibres being
connected to this network. The thermal contraction of the con-
nective tissue is also limited by the state and the variation of the
diameter of myofibres. This complex interaction between the
shortening of the perimysium collagen fibres and the variation in
diameter of the myofibre bundle was calculated by Lepetit et al.
(2000) from the sarcomere length of the raw meat, the number of
collagen cross-​links, and the drip and cooking losses. This theor-
etical approach has been validated using small meat cuts cooked
in water bath conditions.
Meat contraction is associated with water migration from
the intracellular to the extracellular area and then the forma-
tion of channels of various sizes along the bundle boundaries,
through which the juice is expelled outside the meat cut under the
constraints exerted by the contracted myofibres and connective
tissues. The quantity of juice that remains in the meat and the
equilibrium water content depend on the final meat temperature
and not on the meat cut size. The shape of the curve that provides
the equilibrium water content as a function of the final meat tem-
perature is the same for beef, pork, sheep, and horse meat, except
for some specific muscles such as the masseter (Oillic et al.,
2011; Bombrun et al., 2015). However, equilibrium values can be
different from one type of muscle to another and between animal
species (Oillic et al., 2011).
Is it Possible to Totally Control the Tenderness of
Cooked Meat?
Consumers still experience problems of meat tenderness, mainly
for “red meats”, or juiciness, mainly for “white meats”. This is
also true for trained cooks or butchers, who know how to select
meat from different animals, breeds, and muscles. In practice,
even top chefs can have some difficulties in obtaining the same
cooked meat quality when they change their heating equipment
or try to adapt a recipe or develop a new one.
The first reason for this problem is connected with the vari-
ability of the mechanical properties of the raw meat, because of
the biological and biochemical differences due to animal genetics
or bad control of the ageing process.
The second reason is connected with the fact that most heating
equipment leads to important gradients of temperature and water
concentration in the meat cuts. Cooked meat quality is judged,
after all, averagely, and the perception of its tenderness is largely
influenced by its other sensory qualities (juiciness, odour, and
taste). This mainly explains why changes of equipment or of
recipe are often difficult. A traditional way of avoiding heating
problems for collagen-​rich muscles is to apply slow cooking
techniques to help to solubilize the collagen, thus reducing meat
toughness. Another way is to apply low-​temperature cooking
methods to muscles, as traditionally used for meat roasting.
However, this does not address the whole problem, since it is
important not to consider tenderness alone, but also safety and the
nutritional traits of meat. The design of new cooking equipment
based on the theoretical knowledge obtained from the predic-
tion of the different quality traits of meat during cooking is a
417
promising solution (Kondjoyan et al., 2013; Kondjoyan et al.,
2014; Kondjoyan et al., 2018).
It is important to control all these steps to increase consumer
satisfaction, since meat tenderness is the result of many factors
that take place from animal selection, production, and slaughter
to meat maturation and cooking (Figure 59.1). Attempts have
been made to grade beef carcasses by including some indicators
associated with eating quality, such as ultimate pH of meat and
marbling score, or others such as carcass fatness, which is posi-
tively correlated with marbling. The most advanced grading
scheme is the Meat Standards Australia grading scheme, which
includes, among others, indicators of growth rate, animal sex, and
physiological maturity, which all affect the muscle characteristics
mentioned earlier (Thompson, 2002). The MSA system goes fur-
ther by including ageing time and the average effects of some
cooking methods. This system has been proved to detect vari-
ability in quality across and within muscles, depending on their
biochemical characteristics. It works to predict eating quality not
only in Australia but also in many other countries (reviewed by
Hocquette et al., 2014).
Conclusion
Meat tenderness, as it is perceived by consumers, results from
a combination of complex mechanisms associated with animal
species, muscle structure, slaughtering procedures, chilling, meat
ageing, and cooking processes. How to control all the parameters
through these steps remains a challenge. This often results in con-
sumer dissatisfaction, especially for beef consumption. Grading
schemes like that of Meat Standards Australia are aimed at inte-
grating all factors, from animal production to cooking methods,
in order to predict the eating quality of lamb and beef for each
combination of cut and cooking method, and offering a solution
for improving consumer satisfaction, not only with meat ten-
derness but also with flavour and juiciness. A complementary
strategy is to improve the design of the equipment and methods
used to cook these different combinations.
REFERENCES
Bernard C, Cassar-​Malek I, Renand G, Hocquette JF. 2009. Changes
in muscle gene expression related to metabolism according
to growth potential in young bulls. Meat Science, 82(2),
205–​212.
Bombrun L, Gatellier P, Portanguen S, Kondjoyan A. 2015.
Analysis of the juice and water losses in salted and unsalted
pork samples heated in water bath. Consequences for the pre-
diction of weight loss by transfer models. Meat Science, 99,
113–​122.
Bouhrara M, Clerjon S, Damez JL, Chevarin C, Portanguen S,
Kondjoyan A, Bonny JM. 2011. Dynamic MRI and thermal
simulation to interpret deformation and water transfer in meat
during heating. Journal of Agricultural and Food Chemistry,
59(4), 1229–​1235.
Bouhrara M, Clerjon S, Damez JL, Kondjoyan A, Bonny JM. 2012.
In-​
situ imaging highlights local structural changes during
heating: The case of meat. Journal of Agricultural and Food
Chemistry, 60(18), 4678–​4687.
418
Combes S, Lepetit J, Darche B, Lebas F. 2003. Effect of cooking
temperature and cooking time on Warner–​ Bratzler tender-
ness measurement and collagen content in rabbit meat. Meat
Science, 66, 91–​96.
Geay Y, Bauchart D, Hocquette JF, Culioli J. 2001. Effect of nutri-
tional factors on biochemical, structural and metabolic
characteristics of muscles in ruminants; consequences on
dietetic value and sensorial qualities of meat. Reproduction
Nutrition Development, 41, 1−26 (Erratum, 41, 377).
Hocquette JF, Gondret F, Baeza E, Medale F, Jurie C, Pethick DW.
2010. Intramuscular fat content in meat-​producing animals:
Development, genetic and nutritional control, and identifica-
tion of putative markers. Animal, 4(2), 303–​319.
Hocquette JF, Van Wezemael L, Chriki S, Legrand I, Verbeke W,
Farmer L, Scollan ND, Polkinghorne R, Rødbotten R, Allen
P, Pethick DW. 2014. Modelling of beef sensory quality for a
better prediction of palatability. Meat Science, 97, 316–​322.
Kondjoyan A, Oillic S, Portanguen S, Gros JB. 2013. Combined heat
transfer and kinetic models to predict cooking loss during heat
treatment of beef meat. Meat Science, 95(2), 336–​344.
Kondjoyan A, Kohler A, Realini CE, Portanguen S, Kowalski R,
Clerjon S, Gatellier P, Chevolleau S, Bonny JM, Debrauwer L.
2014. Towards models for the prediction of beef meat quality
during cooking. Meat Science, 97, 323–​331.
Kondjoyan A, Portanguen S, Duchene C, Mirade PS, Gandemer
G. 2018. Predicting the loss of vitamins B3 (niacin) and B6
(pyridoxamine) in beef during cooking. Journal of Food
Engineering, 238, 44–​53.
Jean-François Hocquette, Alain Kondjoyan
Lepetit J, Culioli J. 1994. Mechanical properties of meat. Meat
Science, 36, 203–​237.
Lepetit J, Grajales A, Favier R. 2000. Modelling the effect of
sarcomere length on collagen thermal shortening in cooked
meat: Consequence on meat toughness. Meat Science, 54,
239–​250.
Lewis GJ, Purslow PP. 1990. Connective-​tissue differences in the
strength of cooked meat across the muscle-​fibre direction due
to test specimen size. Meat Science, 28, 183–​194.
Listrat A, Lebret B, Louveau I, Astruc T, Bonnet M, Lefaucheur L,
Picard B, Bugeon J. 2016. How muscle structure and com-
position influence meat and flesh quality. The Scientific World
Journal, Article ID 3182746.
Oillic S, Lemoine E, Gros JB, Kondjoyan A. 2011. Kinetic analysis
of cooking losses from beef and other animal muscles heated in
a water bath –​effect of sample dimensions and prior freezing
and ageing. Meat Science, 88(3), 338–​346.
Ouali A, Herrera-​Mendez CH, Coulis G, Becila S, Boudjellal A,
Aubry L, Sentandreu MA. 2006. Revisiting the conversion
of muscle into meat and the underlying mechanisms. Meat
Science, 74(1), 44–​58.
Purslow PP. 2005. Intramuscular connective tissue and its role in
meat quality. A review. Meat Science, 70, 435–​447.
Thompson JM. 2002. Managing meat tenderness. Meat Science, 60,
365–​369.
Meat: Heat Transfer in Meat
Douglas Baldwin
Breville Pty Ltd 19400, South Western Ave, Torrance, California 90501, United States
Meat is a complex structure comprising roughly 75% water,
20% protein, and 5% fat and other substances. This poses some
challenges modeling heat transfer from frozen to thawed meat,
from raw to cooked meat, and for searing the surface.
Cooking
Cooking raw meat can be thought of in terms of denaturing
proteins. There are several ways that proteins are denatured in
the kitchen, including mechanical agitation when whipping egg
whites, using acids like vinegar and lemon juice, curing with inor-
ganic salts, alcoholic marinades, and cleaning with detergents,
but cooking mainly involves heat. It is convenient to group meat
proteins into muscle fibers (mostly myosin and actin), soluble
proteins (mostly enzymes and myoglobin), and connective tissue
(mostly collagen). When heating, the muscle fibers can shrink,
the water-​soluble proteins can aggregate and gel, and connective
tissue can shrink and solubilize depending on the temperature
profile.
As it heats, first, the muscle fibers start to shrink around 35
to 40 °C, and this shrinking increases up to about 80 °C and
squeezes water out of the meat. The soluble proteins start aggre-
gating and gelling around 40 °C, and this is complete around
60 °C. This gelling firms the meat, so medium and medium-​rare
meat is less chewy than rare meat. The connective tissue starts
shrinking around 60 °C but shrinks more intensely over about
65 °C. Most of this connective tissue is collagen, and further
heating can convert it into gelatin. For more on the chemistry of
meat, see Tornberg (2005).
This denaturation of proteins corresponds to the cook’s con-
cept of ‘doneness’; meat is rare around 50 °C, medium-​rare
around 55 °C, medium around 60 °C, and well-​done over about
70 °C. The muscle fibers start to shrink for rare meat, and some
soluble proteins denature. Between medium-​rare and medium,
more soluble proteins denature, but the connective tissue has
not started to shrink. For well-​done, the soluble proteins have
mostly denatured, changing the meat’s color, and the muscle
fibers have squeezed significant amounts of water out of the
meat. For meat with significant fat marbling and connective
tissue, holding the meat at 70 °C and above will hydrolyze the
collagen and release the melted fat to give it succulence. The
temperatures corresponding to rare, medium, and well-​done are
about 5 °C higher in poultry and about 10 to 15 °C lower in fish
compared with red meat. This temperature difference roughly
corresponds to the difference in the animal’s body temperature,
possibly because the biochemistry in it is naturally selected to
this temperature.
Let us consider the following partial differential equation to
model heat transfer in meat:
∂T
= ∇ ⋅ ( α∇T ) (60.1)
∂t
where T is temperature (°C) and α is the thermal diffusivity (m²/​s).
If we know the initial temperature distribution, we can estimate
how α changes with temperature, time, and location, and we can
describe how the surface temperature changes, then mathematic-
ally we know that this model has a unique solution for all future
times at all locations in the meat.
Since the structure of meat is so complex, we will need to make
many simplifications and assumptions in relation to equation (60.1)
to solve it in practice. First, we will assume that α does not depend
on position, i.e., that we can average locally over muscle fibers,
fat, cartilage, bone, and so on, even though they have different
thermal diffusivities. Next, we will assume that α does not depend
on time, so we are assuming that mass transfer and protein denatur-
ation cause negligible effects on how heat is conducted through the
meat. For example, salt diffusivity in meat at room temperature
is 5–​10 × 10⁻10 m²/​s, compared with meat’s thermal diffusivity of
1–​2 × 10⁻⁸ m²/​s; this suggests that over shorter times, we can neg-
lect mass transfer, but that it can be important over longer times.
Indeed, mass transfer is essential near the surface to desiccate the
meat’s surface so that it can be heated to over 130 °C, whereupon
the Maillard reaction can brown the surface and create roasted and
savory volatile flavor compounds.
Even assuming that α only depends on temperature, equation
(60.1) is challenging to solve for arbitrary shapes and boundary
conditions. If we are mainly concerned with the meat’s surface
and core temperatures, we can approximate equation (60.1)
with a one-​dimensional heat equation using a geometric factor,
E, that relates to the meat’s effective dimensionality. So, using
( )
α ≡ k / ρC p , where k (T ) is thermal conductivity (W/​mK), ρ is
419
420 Douglas Baldwin

FIGURE 60.1 A boneless beef rib roast on a rack in a roasting pan was placed in a domestic oven set to 175 °C. Four thermocouples were used to collect
temperature data: a needle-​type probe with the tip near the meat’s core, another just below the meat’s surface, another in the air (the dry-​bulb), and the last
in the top of a sponge with the bottom of the sponge in water so that wicking kept it wetted to approximate an ideal wet-​bulb measurement. When the oven
door is opened, the air temperature rapidly drops and then recovers after about 10 min. The meat’s surface temperature increases quickly and then slows as it
approaches the wet-​bulb temperature. The evaporating juices cool the meat’s surface and also increase the oven’s relative humidity, which slightly raises the
wet-​bulb temperature. Around 1 hour, the surface has desiccated enough that it can exceed the wet-​bulb temperature. When the meat’s core reached 52 °C,
the roast was removed from the oven; residual heat continued to diffuse and raise the core temperature to 60 °C despite the surface rapidly cooling to 40 °C.

conditions, and E = 1, 2, and 3 exactly correspond to the heat

density (kg/​m³), and C p (T ) is the specific heat (kJ/​kg.K), and a equation in cartesian, cylindrical, and spherical coordinates with
geometric factor, E, gives: radial symmetry.
The boundary condition, equation (60.4), can be used to under-
∂T  ∂2T E − 1 ∂T 
ρC p (T ) = k (T )  2 +  (60.2) stand many counter-​intuitive experiences. For example, let us
∂t  ∂r r ∂r  consider a barbecue stall, where the meat’s temperature goes up
and then is maintained around 65 to 75 °C, often for hours, before
∂T climbing up to 85 to 90 °C. At these temperatures, k (T ) is essen-
T (r , 0 ) = T0 , (0, t ) = 0 (60.3) tially constant, so we just need to look at the right hand side (RHS)
∂r 
of equation (60.4). Initially, the RHS is positive, so the meat’s
surface temperature increases; this increase is usually dominated
k (T )
∂T
∂r
(
( R, t ) = h T fluid − T ( R, t ) ) ( )
by the first term, h T fluid − T ( R, t ) , unless there is condensing
(60.4)
( )
steam, which can be very efficient at heating. Then, evaporative
+ K m L Pv, fluid − aw Pv,surface
cooling starts to slow the temperature increase, until the RHS
(
is close to zero. The second term, K m L Pv, fluid − aw Pv,surface , )
where 0 ≤ r ≤ R, t ≥ 0, T0 is the initial temperature, Tfluid is the fluid is negative because Pv, fluid < aw Pv,surface and balances the first
(air, steam, water, or oil) temperature the meat is being cooked in, term, slowing the cook at around 65 to 75 °C. This ‘stalling’
h is the heat transfer coefficient (W/m².K), Km is the mass transfer temperature is lowered by decreasing the air’s relative humidity
coefficient (kg/m².s.Pa), aw is water activity at the meat’s surface, (decreasing Pv, fluid), increasing the airflow (increasing K m),
P(v,fluid) is partial water vapor pressure away from the food and aw mopping the meat (increasing aw Pv,surface ), or lowering the air
P(v,surface) at the food’s surface (Pa), and L is the latent heat of evap- temperature T fluid . The stall can be shortened by increasing the
oration for water (J/kg). air temperature, increasing the airflow, or reducing the air’s rela-
For a slice of ham or a steak, the shortest path for heat is from tive humidity. The RHS becomes positive after the stall because
the top and bottom, so the effective dimension is about E ≈ 1 the surface desiccates (decreasing aw ), and so the surface can
for a steak; similarly, for a tenderloin, the shortest path to the approach the air temperature, but below the desiccated layer is
core is in two dimensions, and so E ≈ 2 for a tenderloin; for a another wet region that is below the boiling point.
meat ball, the shortest path is in all three dimensions, so E ≈ 3. If we assume that the water activity at the meat’s surface is
Practically, 1 ≤ E = t slab /t ≤ 3, where t is the heating time and t slab aw > 0.99, so it has not started to desiccate, then we can further
is the heating time for an infinite slab under the same, constant simplify equation (60.4) to get
Meat: Heat Transfer in Meat
∂T
k (T )
∂r
(
( R, t ) = h Twet −bulb − T ( R, t ) ) meat heats is complicated and is best done with already tested
where Twet − bulb is the wet-​bulb temperature of the fluid (air, steam,
or water) and h is the effective heat transfer coefficient. When
poaching, steaming, boiling, or sous-​vide cooking, the wet-​and
dry-​bulb temperatures, Twet − bulb = T fluid , are the same. In air that
is not saturated with water, such as in ovens and smokers, the
wet-​bulb temperature is often much lower than the dry-​bulb tem-
perature; the wet-​bulb in a convection oven might be 80 to 90 °C
when the dry-​bulb is between 125 and 200 °C. The lower the air’s
relative humidity, the larger the difference between the wet-​and
dry-​bulb temperatures. Indeed, the difference between wet-​and
dry-​bulb temperatures is a common method for measuring rela-
tive humidity.
Modern convection steam ovens allow the injection of steam
to increase the wet-​bulb temperature to near the dry-​bulb tem-
perature, between about 65 and 90 °C, and so increase the
rate of cooking. Likewise, increasing the difference between
the dry-​and wet-​bulb temperatures increases evaporation and
leads to the surface desiccating. The magnitude of h can also
give insights into how the meat will cook; in ovens, with nat-
ural and forced convection, the magnitude is relatively low, and
variations between ovens can result in significantly different
heating times. However, the magnitude of h is relatively high
in sous-​vide baths, convection steam ovens near 100% relative
humidity, deep-​fat fryers, and so on, so heating times are con-
sistent and predictable.
Now that we understand the boundary condition, let us return
to the heat equation (60.2). While the meat’s thermal conduct-
ivity, k (T ), and specific heat, C p (T ), depend on temperature
and have nonlinearities associated with freezing and desicca-
tion, they are essentially constant for thawed meat. So, we can
 2  glucose, also increases the reaction rate in meat because reducing
simplify equation (60.2) to ∂T = α  ∂ T + E − 1 ∂T  . Recall that
∂t  ∂r
2 r ∂r 
E = t slab / t allows us to estimate the time for other shapes based
on the time required for a slab. Thus, let us consider the heat
equation:
∂T ∂2T
=α 2
∂t ∂r
From calculus, we can show that doubling the thickness
quadruples the time, so cooking time is proportional to R 2 .
This means that the meat’s characteristic thickness, R, has a
huge effect on the cooking time. This can also be seen from
the meat’s thermal diffusivity, which is between 0.12 and
0.16 mm²/​s (Nicolaï and Baerdemaeker, 1996); the estimated
minimum cooking time for a 30 mm steak, heating from both
top and bottom, is from
(15 mm )2 = 23.4
(0.16 mm2 /s)(60 s/min)
(15 mm )2 = 31.3 min.
(0.12 mm2 /s)(60 s/min)
421
Using equations (60.2) through (60.4) to compute how the
and validated computer code.
Thawing and Freezing
Unlike pure water, not all the water in meat freezes at 0 °C. As
ice forms, the remaining solution becomes more concentrated with
salts, sugars, and other solutes. As the meat cools further, some
of the solutes crystallize out of solution to give a mix of ice, solu-
tion, and solute. As the meat is cooled still further, the solution
reaches a maximal freeze concentration and is transformed into a
rubber state; on cooling further, the rubber state transforms to a
glassy state. Frozen meat and fish are very stable in the glassy state,
and this is used to store some high-​value products. Differences in
fat and other substances between species and muscles affect these
transition temperatures as well as other thermophysical properties.
Generally, product quality is maximized by freezing and thawing
quickly and storing high-​value items in the glassy state, e.g., using
liquid nitrogen or dry ice to make ice creams with small ice crystals
or storing tuna for sushi in the glassy state below about –55 °C. For
a review on predicting freezing, see Delgado and Sun (2001).
Browning
Once the surface sufficiently desiccates, the water activity, aw ,
decreases, and the temperature exceeds 130 °C, then the Maillard
reactions between amino acids and reducing sugars start browning
the surface and developing roast and savory flavors. The Maillard
reaction starts noticeably around 130 °C, and good browning
occurs above 150 °C (Baldwin, 2012), with the rate increasing
with temperature. Increasing the amount of reducing sugar, like
sugars are relatively less available than amino acids. This is why
meat with a sweet glaze browns so quickly. The browning rate
also increases as the water activity, aw , decreases.
There are many ways to brown the meat’s surface: searing
in a dry pan, shallow-​and deep-​fat frying, using a gas or char-
coal grill, using a salamander or broiler, roasting in an oven, and
so on. Returning to equation (60.4), we see that many of these
methods use a high cooking temperature (a high T fluid ) and low
partial water vapor pressure (low Pv, fluid) to increase evaporation
and make a desiccated surface layer. When frying with fat, for
example, the partial water vapor is essentially zero except for the
water coming out of the meat’s surface. This is also why basting
with fat increases the wet-​bulb temperature, as the fat coats the
surface and thus increases the amount of evaporation, which
leads to the surface desiccating.
to
REFERENCES
Baldwin DE. 2012. Sous vide cooking: A review. International
Journal of Gastronomy and Food Science, 1(1), 15–​30.
Delgado AE, Sun DW. 2001. Heat and mass transfer models for
predicting freezing processes –​a review. Journal of Food
Engineering, 47, 157–​174.
422 Douglas Baldwin

Nicolaï BM, De Baerdemaeker J. 1996. Sensitivity analysis with Tornberg E. 2005. Effect of heat on meat proteins –​implications
respect to the surface heat transfer coefficient as applied to on structure and quality of meat products. Meat Science, 70,
thermal process calculations. Journal of Food Engineering, 493–​508.
28, 21–​33.
Meat: Reduction of Nitrate and Nitrite Salts in Meat Products –​
What Are the Consequences and Possible Solutions?
Régine Talon and Sabine Leroy
Université Clermont Auvergne, INRAE, MEDIS, Clermont-​Ferrand, France
The use of nitrite and nitrate salts as additives in the manufacture
of cured meat products is regulated by European directives. The
nitrite ion can be toxic, and nitrate becomes so upon reduction
to nitrite. However, nitrites have multiple functions, as they con-
tribute to sensorial qualities (colour, flavour) and to the inhib-
ition of some microorganisms. The European Commission is still
pressing for further reductions in the levels of these additives.
Although currently there are no compounds that replace nitrites in
all their functions, some maintain the sensorial or hygienic qual-
ities of cured meat products and, together with a reduced level
of nitrites, could be an alternative. The use of nitrate/​nitrite salts
is still controversial, however, as some studies have highlighted
their health benefits.
Introduction
Numerous foodstuffs contain nitrates and, to a lesser extent,
nitrites. The average dietary intake of nitrates in France is 141 mg/​
day/​person, with 75% provided by vegetables and fruits, 14% by
water and 6% by animal products. The average dietary intake of
nitrites is 2 mg/​day/​person, 40% of which comes from vegetables
and fruits, 39% from processed animal products, 15% from other
foodstuffs and 5% from water (EFSA, 2008a).
Sodium and potassium salts of nitrates (E251, E252) and
nitrites (E249, E250) are commonly used as additives in the
manufacture of meat products, such as salt-​cured meats. Nitrites
that are added or that derive from the reduction of nitrates are
the active curing agent. They have a very broad spectrum of
action and contribute to sensorial properties (colour, flavour) and
the microbiological safety of cured products, notably regarding
spore-​forming bacteria.
The use of these additives is regulated by European direct-
ives 95/​2/​EC, 2006/​52/​EC and 1129 (11 November 2011), which
authorize the addition of 100 to 150 mg of nitrites (E249, E250)
per kg for heat-​treated meat products, and 150 mg of nitrites with
150 mg of nitrates (E251, E252) per kg for non-​heat-​treated meat
products. These directives were drawn up on the basis of the tox-
icity of nitrites, while the toxicity of nitrates is estimated from its
reduction to nitrites. As a comparison, the mean nitrate content
of lettuce is 4000 mg/​kg. The acceptable daily intake (ADI) is
3.7 mg/​kg bodyweight/​day for nitrates and 0.07 mg/​kg bw/​day
for nitrites (EFSA 2010, 2017).
In 2017, EFSA experts estimated that the exposure to nitrates
solely from their use as a food additive was less than 5% of the
overall exposure to nitrates in food and did not exceed the ADI, and
thus, that exposure to nitrites as food additives was within safe levels.
The toxicity of nitrites is due to their potential to form carcinogenic
substances called nitrosamines (R2N2O), both in foodstuffs and in
the body (Hammes, 2012). The residual nitrite and/​or nitrate con-
centration in sausages and cooked ham was below 20 mg/​kg, so
the probability of forming nitrosamines in meat products was low
(Honikel, 2008). Nitrosamines could, however, be generated when
meat products were heat-​treated at temperatures above 130 °C, such
as during the grilling of bacon (Honikel, 2008).
Given the potential risk associated with nitrites, the European
Commission is considering reducing levels of nitrites or nitrates or
both in some foodstuffs, including meat products. The main risks
associated with reducing or eliminating nitrites in the manufac-
ture of cured meats are the growth of unwanted microorganisms,
a reduction of about one-​third in shelf-life, a greyish colour
instead of a stable pink or red colour, and a loss of character-
istic taste of these products. Numerous questions on the future
of these products, therefore, remain unanswered. Is, for instance,
this reduction compatible with the concomitant decrease in salt
that food professionals must provide for consumer health? How
can these reductions be achieved while ensuring the safety and
sensory qualities of these foods?
To our knowledge, there are presently no compounds that
replace nitrites in all their functions, but some more or less effi-
cient alternatives are used or are being tested (Table 61.1).
Do Nitrate-​Rich Plant Extracts Offer a Viable
Alternative?
To avoid adding nitrates yet retain the sensorial and microbio-
logical properties of meat products, nitrate-​rich plant extracts
have been tested. Sebranek and Bacus (2007) proposed that
spices, spinach or celery juice were viable alternatives; they
reported that the nitrate concentration in a powder from celery
juice concentrate was high (2.75%) and that its addition to a meat
423
424
TABLE 61.1
Role of Nitrate/​Nitrite in Cured Meats and Examples of Alternatives
NO3 (nitrate)/​NO2 (nitrite)
Colour
NO3 → NO2 via nitrate reductase of staphylococci
NO2 →NO +myoglobin = nitrosomyoglobin red colour stable
Flavour
Interaction with meat components = flavourful derivatives of
nitroso-​compounds
Inhibition of lipid oxidation = avoid rancidity
Microbiological qualities
No effect on growth for lactic acid bacteria
Altered survival of staphyloccoci and physiology (nitrosative stress)
Inhibition of spoilage bacteria Enterobacteria and Brochothrix
thermosphacta
Inhibition of pathogenic bacteria Clostridium botulinum, Listeria
monocytogenes, Salmonella
Source: Adapted from: Honikel, 2008; Ras et al., 2017; Sindelar and Milkowski, 2011; Christieans et al., 2017; Vermassen et al., 2014; Weiss et al., 2010.
batter in the manufacture of sausages resulted in a nitrite concen-
tration of 100 mg/​kg. Products made in this way are categorized
as “clean label” food. Nitrates added in this “natural” way must
be reduced to nitrites, which is possible via the action of starters
with nitrate reductase activity, such as the species Staphylococcus
carnosus and S. xylosus. The addition of “natural” nitrates to
cooked meat products also called for the addition of starters that
reduced nitrates to nitrites for the development of colour and
thus led to manufacturing changes; in particular, the temperature
should be kept at 42 °C for 90 min before the usual manufac-
turing procedure of these products (Sebranek and Bacus, 2007).
The addition of plant extracts with a Staphylococcus starter cul-
ture to curing brines yielded cooked ham with a colour and low
level of lipid oxidation similar to that treated with nitrites, while
ensuring lower residual levels (Krause et al., 2011).
The Standing Committee on the Food Chain and Animal Health
considers that when a starter and nitrate-​rich plant extracts are added,
the nitrites obtained via the enzymatic activity of this starter should
be considered as an additive and must be subject to authorization.
Natural cherry extracts, which contain high levels of ascorbic acid,
can be added instead of the acid itself to facilitate the conversion of
nitrites into nitric oxide (NO) (Pegg and Honikel, 2015).
The use of nitrate-​rich plant extracts rather than nitrates as an
additive may therefore seem like an alternative, but it is in fact
only a “false” solution, because the risk associated with the nitrite
residue is the same.
Role of Nitrates/​Nitrites in Colour
Colour Formation
Colour is formed through the reduction of nitrates to nitrites via
the nitrate reductase activity of the two staphylococci used as
Régine Talon, Sabine Leroy
Alternatives
Nitrate-​rich plant extracts: celery, spinach ... = addition of nitrate
Natural colourants: tomato extract, paprika ... + decreased level of NO2
from 150 to 100 mg/​kg
Addition of bacteria: staphyloccoci with a nitric oxide synthase activity
No alternatives
Antioxidant compounds: vitamin E, ascorbic acid, citrate, polyphenols
Extracts of aromatic plants: rosemary, oregano, sage … + decreased level
of NO2 from 150 to 80 or 60 mg/​kg
Plant extracts: essential oil of savory, nutmeg, sage or clove + decrease
level of NO2 (10 mg/​kg)
Acidifying lactic acid starter (pH < 5.0)
starters, and also by numerous chemical reactions of nitrites with
meat pigments (Talon et al., 1999; Honikel, 2008). These enzym-
atic and chemical reactions depend on pH, the pigment concen-
tration in the raw material, the redox potential and temperature.
Nitrate reductase activity differed greatly between the two
species commonly used to initiate colour formation: S. carnosus
and S. xylosus (Gøtterup et al., 2008; Sanchez et al., 2015; Leroy
et al., 2017). Most S. carnosus strains reduced nitrates effect-
ively, whereas fewer than 50% of S. xylosus strains did so.
Various chemical reactions convert nitrites into derivatives such
as NO, which binds to the iron of the haem group of myoglobin
to form nitrosomyoglobin, a stable, red pigment that accounts for
the typical colour of salt-​cured meats (Honikel, 2008).
Various studies have shown that nitrite concentrations between
25 and 50 mg/​kg are, in general, sufficient to produce a satisfactory
colour in cured meats (Sindelar and Milkowski, 2011). However,
higher concentrations are needed to develop and maintain accept-
able cured meat colour of products with a long shelf-​life.
Alternatives
Addition of Natural Colourants
The addition to meat batters of tomato extract (2.5–​3.0%) or pap-
rika as natural colourants at reduced nitrite levels (100 mg/​kg
instead of 150 mg/​kg in the control) yielded a colour close to that
of the control, confirming that the nitrite level could be reduced
without affecting colour (Bazán-​Lugo et al., 2012). Deda et al.
(2007) showed that colour could be preserved in frankfurters
by the addition of 12% tomato paste when the nitrite level was
100 mg/​kg.
Addition of Bacteria
The lactic acid bacteria Lactobacillus plantarum, L. fermentum
and Pediococcus acidilactici, the staphylococci S. xylosus and
Meat: Reduction of Nitrate and Nitrite Salts
S. carnosus when used as starters, and four other Staphylococcus
species produced nitrosylmyoglobin in a laboratory culture
supplemented with metmyoglobin (Morita et al., 1998; Gündoğdu
et al., 2006; Gøtterup et al., 2007; Ras et al., 2017). The forma-
tion of nitrosylmyoglobin by S. xylosus was also noted in meat
batter (Li et al., 2013). The bacteria could produce NO from
arginine via nitric oxide synthase (NOS) (Crane et al., 2010).
The gene encoding NOS has not been found in the genomes of
lactic acid bacteria, but it has been identified in various species of
coagulase-​negative staphylococci, including two used as starters,
S. carnosus and S. xylosus (Jansens et al., 2014; Ras et al., 2017).
NO production via NOS has been demonstrated in a strain of
S. xylosus (Ras et al., 2018).
Role of Nitrates/​Nitrites in Flavour
Flavour Development, Antioxidant Role
Various sensory studies have shown that consumer panels are able
to distinguish between food products made with different quan-
tities of nitrites (Sindelar and Milkowski, 2011). In the absence of
nitrites, products have a taste described as “meaty”, whereas they
have a taste typical of cured meat when made in the presence of
nitrites. Concentrations of about 50 mg/​kg seemed sufficient for
differences to be perceived by a panel (Gray et al., 1981). Pegg and
Shahidi (2000) reported that 40 to 70 mg/​kg nitrites was enough to
ensure the development of the aroma characteristic of cured meats.
Despite this reported sensorial perception, the molecules respon-
sible for this characteristic flavour and the mechanisms involved
are largely unknown. Flavourful compounds are formed by the
interaction of nitrite and/​or NO with different ingredients of meat,
resulting in nitroso-​compounds. The nitrite can react with haem
and non-​haem proteins, low-​molecular-​weight peptides or amino
acids, and lipids (Honikel, 2008). Nitrite and NO also have an
antioxidant role (Sindelar and Milkowski, 2011); NO links to iron
in myoglobin and thus prevents its oxidation and transformation
into a potent oxidant able to peroxidize lipids. NO also chelates
free radicals, and nitroso-​compounds have antioxidant properties
(Sebranek and Bacus, 2007; Sindelar and Milkowski, 2011).
There were more organoleptic compounds in ham cooked
with nitrites than without: 49 versus 37 out of a total of 53, and
lipid oxidation products were found at lower levels in nitrite-​
containing ham (Guillard, 1998). Likewise, the concentration of
volatile compounds associated with lipid oxidation was lower
in sausages made with a nitrate/​nitrite mixture than in control
sausages without nitrate/​nitrite (Hospital et al., 2012). This anti-
oxidant effect was noted at nitrate and nitrite concentrations above
75 mg/​kg in fermented dry sausages. It should also be noted that
the antioxidant power of a nitrate/​nitrite mixture was well above
that of nitrate alone (Christieans et al., 2017). In another study,
20 mg/​kg nitrite was sufficient to inhibit lipid oxidation in fish,
chicken, pork and beef (Sindelar and Milkowski, 2011).
Alternatives to Nitrates and Nitrites
There is an abundant literature on antioxidants tested in meat
products (Weiss et al., 2010). Some, like vitamin E, lycopene,
425
ascorbic acid, and extracts of rosemary, oregano and sage, trap or
stabilize free radicals that cause oxidation reactions, while others,
such as citrates and polyphenols, chelate metals that initiate oxi-
dation. For example, rosemary extract added to liver pâté enabled
a reduction in nitrites from 150 mg/​kg to 80 mg/​kg (Doolaege
et al., 2012). A combination of rosemary (E392) and 60 mg/​kg
nitrite preserved the colour and flavour of cooked ham.
Role of Nitrates/​Nitrites in Microbiological
Qualities
Nitrates serve as a reservoir of nitrites, which are the active
molecules involved in inhibition of various bacterial populations.
The mechanisms responsible for the inhibition are poorly under-
stood and depend on the bacterial species. Furthermore, inhib-
ition depends on factors like pH, salt concentration and the
presence of reducing agents like ascorbate and iron (Sindelar
and Milkowski, 2011). The nitrite ion is a stronger inhibitor at
acid pH (as in meat), doubtless because of the various chemical
reactions that yield NO and nitrous and nitric acids (Honikel,
2008). Inhibition can also be influenced by iron, which binds to
and inactivates nitrite. Thus, the presence of liver in certain cured
meats can inhibit nitrites because it contains high levels of iron
(EFSA, 2003).
We shall now review the effects of nitrate/​nitrite on different
bacterial populations, i.e., starters, spoilage bacteria and patho-
genic bacteria.
Growth, Survival and Activity of Starters
Bacterial starters composed of a mixture of lactic acid bac-
teria and staphylococci are added during the manufacture of
dry sausages. The microbial ecology of sausages depends on
numerous endogenous factors, such as the composition of the
raw material, its initial quality (pH, water activity and microbial
load), salt, saccharides, spices, additives (nitrates and nitrites),
added starters, casings, exogenous factors (temperature and
humidity) and microbial interactions (Hammes, 2012).
In general, the growth and survival in sausages of lactic
acid bacteria added as starters are not affected by the different
concentrations of nitrate and/​or nitrite used. Thus, populations
of Lactobacillus sakei alone or in the presence of Pediococcus
pentosaceus increased more than 2.5 log-​fold and then remained
stable in dry sausages made with different concentrations of
nitrate (150–​300 mg/​kg) or nitrite (150 mg/​kg) or both (80/​
80 mg/​kg, 120/​120 mg/​kg) (Marco et al., 2006; Christieans et al.,
2017). L. plantarum grew and survived in the absence or presence
of nitrate/​nitrite mixtures (150/​150; 112.5/​112.5; 75/​75 mg/​kg) in
dry sausages (Hospital et al., 2012; Hospital et al., 2014).
The survival of two staphylococci used as starters (S. xylosus,
S. carnosus) was modified by the presence of nitrate and/​or
nitrite. It was greater in a sausage made without nitrate/​nitrite or
with 75/​75 mg/​kg nitrate/​nitrite than in the presence of greater
quantities (Hospital et al., 2012; Hospital et al., 2014).
Very few studies have examined the impact of nitrate/​nitrite
on the physiology of starters. The absence or very low levels of
426 Régine Talon, Sabine Leroy

these additives favoured the formation of volatile compounds and cooked ham) prepared with nitrites (0, 75 or 120 mg/​kg)
via degradation of saccharides or amino acids (Hospital et al., survived. Type B botulinum toxin was produced during storage at
2012), a finding linked to results showing that these additives 8 °C in the products without nitrite, whereas no toxin was present
affected the survival of staphylococci (Hospital et al., 2014). in the products prepared with nitrite during 5 weeks of storage
Inoculation of S. xylosus into a meat matrix in the presence of (Keto-​Timonen et al., 2012). When ground pork was inoculated
nitrate/​nitrite resulted in the modulation of expression of 24% with group I C. botulinum and subjected to short (80 °C, 7 min)
of its genes, reflecting a large change in its physiology in this and long (80 °C, 1 h) heat treatments in the presence of 100,
condition (Vermassen et al., 2014). Notably, nitroso-​compounds 200 or 300 mg/​kg nitrite, the probability of toxin production was
generated by nitrate/​ nitrite exerted nitrosative stress, and the reduced from 96% to 35% and from 86% to 23% for the short and
bacterial strain responded by overexpression of genes involved long treatments, respectively (EFSA, 2003). In the presence of
in iron homeostasis and those encoding antioxidant enzymes ascorbate, the probability of toxin production was decreased from
(Vermassen et al., 2014). 26% to 1% and from 8% to 0% for the short and long treatment,
respectively (EFSA, 2003).
Growth and Survival of Spoilage Bacteria Various studies have suggested that 50–​100 mg/​kg nitrite is
necessary to inhibit the growth of C. botulinum and hence the
Enterobacteriaceae in dry sausages containing 150/​150 or 112.5/​
production of toxin in heat-​treated meat products (EFSA, 2003;
112.5 mg/​kg nitrate/​nitrite decreased in number during fermen-
Cui et al., 2010; Keto-​Timonen et al., 2012). In non-​heat-​treated
tation and were completely eliminated during drying in the
meat products, however, the addition of 150 mg/​kg nitrite is
presence of the highest concentrations (Hospital et al., 2012).
considered necessary to inhibit C. botulinum.
In control (without nitrate/​nitrite) or in sausages made with 75/​
75 mg/​kg nitrate/​nitrite, Enterobacteriaceae multiplied during Salmonella
fermentation and then decreased faster in samples with nitrate/​ In France, the prevalence of Salmonella on the surface of pig
nitrite than without (Hospital et al., 2012). Similarly, the growth carcasses was about 17.6% in 2006/​ 2007 (EFSA, 2008b).
of the spoilage bacterium Brochothrix thermosphacta in minced This degree of contamination is an issue when the cuts from
pork stored at 10 °C under vacuum for 12 days was slowed by contaminated carcasses are used for the manufacture of dry cured
the presence of nitrate/​nitrite compared with a control (Hospital meats consumed as such. Even if products like dry sausages are,
et al., 2014). during processing, subject to technological stages (fermentation,
drying) that in theory have an unfavourable effect on bacterial
Inhibition of Pathogenic Bacteria growth and survival, the presence of Salmonella at the end of
Numerous studies have shown that nitrite is valuable in control- drying is today a reality. Since 2007, the prevalence of Salmonella
ling spore-​forming bacteria like Clostridium botulinum, and also in French finished products has increased, as seen from batch
Listeria monocytogenes and Salmonella. withdrawals and two cases of food poisoning following the con-
sumption of dry sausage in June 2010 and December 2011. This
Clostridium botulinum emergence is often ascribed to upstream conditions (method of
Botulinum poisoning results from the consumption of food feeding animals) and to changes introduced in recent years by
containing preformed botulinum toxin, produced by Clostridium industrial manufacturers, both technologically and in terms of
botulinum during the growth phase of vegetative cells following formulations (reduction of fat, salt and nitrite levels). France is not
the germination of spores. Toxin poisoning often occurs after the only country affected, and cases of salmonellosis associated
consumption of poorly sterilized preserves or of “homemade” with pork consumption have been reported elsewhere in Europe.
ham and cured meat. Nitrites are used to control the growth of In studies on sausages inoculated with Salmonella enterica
C. botulinum, thereby avoiding toxin production in cured meat serovar Typhimurium and made with different concentrations of
products. The growth of C. botulinum is inhibited because nitroso-​ nitrate or a nitrate/​nitrite mixture, when these compounds were
compounds interact with the iron-​sulfur clusters of respiratory absent or when only nitrate was added (150, 200 or 205 mg/​kg)
chain enzymes of this strictly anaerobic bacterium, reduce the a 2–​2 .5 log growth was observed during the fermentation phase,
level of intracellular ATP, and reduce the efficiency of active followed by a decrease during maturation and storage, albeit
transport systems by blocking enzyme pathways (Milkowski insufficient to eliminate these bacteria (Hospital et al., 2014;
et al., 2010). Christieans et al., 2017). In the presence of a nitrate/​nitrite mix-
There are two groups of C. botulinum: group I-​proteolytic ture, the growth of Salmonella was low (1 log) or nonexistent, and
C. botulinum (toxins A, B, F), the spores of which are heat-​ a population decrease was observed during the maturation phase;
resistant and can germinate and grow at a water activity >0.94 an absence of Salmonella was only noted at the end of storage
(10% NaCl) and pH > 4.6, and group II-​nonproteolytic C. botu- (90 days). In these two studies, Salmonella was not eliminated
linum (toxins B, E, F), the spores of which are heat-​sensitive and by the time the sausages could potentially be marketed (Hospital
can germinate and grow at a water activity >0.97 (5% NaCl) and et al., 2014; Christieans et al., 2017).
pH > 5.0, but with growth possible from 3 °C, which is not the
case with the proteolytic strains (EFSA 2003). Keto-​Timonen Listeria
et al. (2012) have shown that group II C. botulinum inoculated Listeria in sausages grew during fermentation and then
into heat-​treated meat products (Wiener or Bologna sausages decreased slightly during maturation in the absence of nitrate/​
Meat: Reduction of Nitrate and Nitrite Salts
nitrite or in the presence of nitrate only (Hospital et al., 2012;
Christieans et al., 2017). In the presence of a nitrate/​nitrite
mixture, the Listeria population remained stable or decreased
during fermentation, and a 2.5 log decrease was noted at the
end of maturation in the presence of 150/​150 mg/​kg nitrate/​
nitrite, whereas 120/​120 to 75/​75 mg/​kg nitrate/​nitrite only
caused a 1–​1.5 log reduction. At the end of storage (90 days),
Listeria was no longer detected only in the sausages made with
at least 120/​120 mg/​kg nitrate/​nitrite. These results prompted
the recommendation by the code of practice in France (Code
des usages, 2020) that 120 mg/​kg nitrate/​nitrite be used instead
of 150 mg/​kg (Christieans et al., 2017).
Alternatives
Plant Extracts
Numerous studies have envisaged controlling health risks by par-
tially replacing nitrites with plant extracts that have antimicro-
bial activity. As an example, cranberry powder (3%) together
with 0.4% celery powder when used instead of nitrates prevented
the growth of Listeria monocytogenes in frankfurters (Xi et al.,
2012). Clostridium perfringens was inhibited in mortadella
sausages prepared with essential oil of savory and 100 mg/​kg
nitrite (Coutinho de Oliveira et al., 2011). Likewise, a synergistic
effect was noted in the control of C. botulinum in a meat matrix
by combination of 10 mg/​kg nitrite with nutmeg, sage or clove
(Cui et al., 2010).
Lactic Acid Bacteria
Salmonella persisted during the manufacturing of dry sausage
in the presence of nitrites at weakly acid pH (>5) and water
activity of 0.90 or 0.85 at the end of drying (Hospital et al.,
2012; Christieans et al., 2017). In dry fermented sausages made
with lactic acid bacteria that yielded a pH of 5.1–​5.2 at the end
of fermentation and of 5.4–​5.5 at the end of drying, or a pH of
4.8–​4.9 at the end of fermentation and of 4.8–​5.0 at the end of
drying, approximately 3.5 log growth of Salmonella occurred
during the fermentation phase at moderate pH, whereas only
2 log growth was recorded at the more acidic pH (Christieans
et al., 2017).
Thus, it appears that new hurdle strategies can be envisaged
by acting on two levels: (i) processing, notably by optimizing
the fermentation stage, and (ii) using substitutes that compensate
for reductions in salt and nitrites while preserving the traditional
characteristics of the products.
Conclusion
The use of nitrates/​nitrites in the manufacture of cured meats
remains controversial. Should it be banned, given the toxicity
of nitrites? Or should nitrates/​nitrites be authorized in more
limited amounts, given that they enhance the safety and sen-
sorial qualities of cured meats? What about the consumption of
leafy plants, which contain large amounts of nitrate dispropor-
tionate to those added to meat products? These questions remain
unanswered, especially as several studies (Milkowski et al.,
427
2010; Hammes, 2012; Parthasarathy and Bryan, 2012) tend
to highlight the human health benefits of the consumption of
moderate amounts of nitrate/​nitrite, which are responsible for
the formation of NO, deficiency in which can lead to several
diseases. Lundberg et al. (2011) even go so far as to suggest
that “we may have to reconsider our current thinking and realize
that inorganic nitrate may not necessarily be a threat to human
health. Instead, in some years we might even consider it as an
essential nutrient.”
REFERENCES
Bazan-​Lugo E, Garcia-​Martinez I, Alfaro-​Rodriguez RH, Totosaus
A. 2012. Color compensation in nitrite-​reduced meat batters
incorporating paprika or tomato paste. Journal of the Science
of Food and Agriculture, 92, 1627–​1632.
Christieans S, Picgirard L, Parafita E, Lebert A, Gregori T. 2017.
Impact of reducing nitrate/​nitrite levels on the behavior of
Salmonella Typhimurium and Listeria monocytogenes in
French dry fermented sausages. Meat Science, 137, 160–​167.
Code des usages. 2020. www.code-​des-​usages-​charcuterie.fr/​”www.
code-​des-​usages-​charcuterie.fr, last accessed 26 March 2020.
Crane BR, Sudhamsu J, Patel BA. 2010. Bacterial nitric oxide
synthases. Annual Review of Biochemistry, 79, 445–​470.
Cui H, Gabriel AA, Nakano H. 2010. Antimicrobial efficacies of
plant extracts and sodium nitrite against Clostridium botu-
linum. Food Control, 21, 1030–​1036. www.sciencedirect.com/​
science/​article/​pii/​S0956713509003491.
Coutinho de Oliveira TL, Araújo Soares R, Mendes Ramos E,
Cardoso M, Alves E, Hilsdorf Piccoli R. 2011. Antimicrobial
activity of Satureja montana L. essential oil against
Clostridium perfringens type A inoculated in mortadella-​type
sausages formulated with different levels of sodium nitrite.
International Journal of Food Microbiology, 144, 546–​555.
Deda JG, Bloukas JG, Fista GA. 2007. Effect of tomato paste
and nitrite level on processing and quality characteristics of
frankfurters. Meat Science, 76, 501–​508.
Doolaege EHA, Vossen E, Raes K, De Meulenaer B, Verhé R, Paelinck
H, De Smet S. 2012. Effect of rosemary extract dose on lipid
oxidation, colour stability and antioxidant concentrations, in
reduced nitrite liver pâtés. Meat Science, 90, 925–​931.
EFSA. 2003. The effects of nitrites/​nitrates on the microbiological
safety of meat products. EFSA Journal, 14, 1–​31.
EFSA. 2008a. Nitrates in vegetables. EFSA Journal, 689, 1–​79.
EFSA. 2008b. Report on the task force on zoonoses data collection
on the analysis of the base line survey on the prevalence of
Salmonella in slaughter pigs. EFSA Journal, 135, 1–​111.
EFSA. 2010. Statement on nitrites in meat products. EFSA Journal,
1538, 1–​12.
EFSA. 2017. Re-​evaluation of sodium nitrate (E251) and potassium
nitrate (E252) as food additives. Scientific opinion. EFSA
Journal, 15 (6), 4787.
Gøtterup J, Olsen K, Knöchel S, Tjener K, Stahnke LH, Møller JKS.
2007. Relationship between nitrate/​nitrite reductase activities
in meat associated staphylococci and nitrosylmyoglobin for-
mation in a cured meat model system. International Journal of
Food Microbiology, 120, 303–​310.
Gøtterup J, Olsen K, Knøchel S, Tjener K, Stahnke LH, Møller
JKS. 2008. Colour formation in fermented sausages by meat-​
associated staphylococci with different nitrite-​and nitrate-​
reductase activities. Meat Science, 78, 492–​501.
Gray JI, MacDonald B, Pearson AM, Morton ID. 1981. Role of nitrite
in cured meat flavor: a review. Journal of Food Protection, 44,
302–​312.
428
Guillard AS. 1998. Role of sodium nitrite on phospholipid com-
position of cooked cured ham. Relation to its flavor. In Food
Flavors: Formation Analysis and Packaging Influences,
Elsevier Science, 245–​253.
Gündoğdu A, Karahan A, Çakmakç M. 2006. Production of nitric
oxide (NO) by lactic acid bacteria isolated from fermented
products. European Food Research and Technology,
223, 35–​38.
Hammes WP. 2012. Metabolism of nitrate in fermented meats: the
characteristic feature of a specific group of fermented foods.
Food Microbiology, 29, 151–​156.
Honikel KO. 2008. The use and control of nitrate and nitrite for the
processing of meat products. Meat Science, 78, 68–​76.
Hospital XF, Hierro E, Fernandez M. 2012. Survival of Listeria
innocua in dry fermented sausages and changes in the typ-
ical microbiota and volatile profile as affected by the concen-
tration of nitrate and nitrite. International Journal of Food
Microbiology, 153, 395–​401.
Hospital XF, Hierro E, Fernandez M. 2014. Effect of reducing nitrate
and nitrite added to dry fermented sausages on the survival of
Salmonella Typhimurium. Food Research International, 62,
410–​415.
Jansens M, Van der Mijnsbrugge A, Sanchez Mainar M, Balzarini M,
De Vuyst L, Leroy F. 2014. The use of nucleotides and arginine
as alternative energy sources by coagulase negative staphylo-
cocci in view of meat fermentation. Food Microbiology,
39, 53–​60.
Keto-​Timonen R, Lindström M, Puolanne E, Niemistö M, Korkeala
H. 2012. Inhibition of toxigenesis of group II (nonproteolytic)
Clostridium botulinum type B in meat products by using
a reduced level of nitrite. Journal of Food Protection, 75,
1346–​1349.
Krause BL, Sebranek JG, Rust RE, Mendonca A. 2011. Incubation of
curing brines for the production of ready-​to-​eat, uncured, no-​
nitrite-​or-​nitrate-​added, ground, cooked and sliced ham. Meat
Science, 89, 507–​513.
Leroy S, Vermassen A, Ras G, Talon R. 2017. Insight into the genome
of Staphylococcus xylosus, a ubiquitous species well adapted
to meat products. Microorganisms, 52, 1–​18.
Li P, Kong B, Chen Q, Zheng D, Liu N. 2013. Formation and identifi-
cation of nitrosylmyoglobin by Staphylococcus xylosus in raw
meat batters: a potential solution for nitrite substitution in meat
products. Meat Science, 93, 67–​72.
Lundberg JO, Carlström M, Larsen FJ, Weizberg E. 2011. Roles of
dietary inorganic nitrate in cardiovascular health and disease.
Cardiovascular Research, 89, 525–​532.
Marco A, Navarro JL, Flores M. 2006. The influence of nitrite and
nitrate on microbial, chemical and sensory parameters of slow
dry fermented sausage. Meat Science, 73, 660–​673.
Milkowski A, Garg HK, Coughlin JR, Bryan NS. 2010. Nutritional
epidemiology in the context of nitric oxide biology: a risk-​
benefit evaluation for dietary nitrite and nitrate. Nitric Oxide,
22, 110–​119.
Régine Talon, Sabine Leroy
Morita H, Sakata R, Nagata Y. 1998. Nitric oxide complex of iron(II)
myoglobin converted from metmyoglobin by Staphylococcus
xylosus. Journal of Food Science, 63, 352–​355.
Morita H, Yoshikawa H, Sakata R, Nagata Y, Tanaka H. 1997.
Synthesis of nitric oxide from the two equivalent guanidino
nitrogens of L-​arginine by Lactobacillus fermentum. Journal
of Bacteriology, 179, 7812–​7815.
Parthasarathy DK, Bryan NS. 2012. Sodium nitrite: The “cure” for
nitric oxide insufficiency. Meat Science, 92, 274–​279.
Pegg RB, Honikel KO. 2015. Principles of curing. In Handbook of
Fermented Meat and Poultry, ed. Toldra F, John Wiley and
Sons Ltd, 19–​30.
Pegg RB, Shahidi F. 2000. Nitrite Curing of Meat –​The N-​Nitrosamine
Problem and Nitrite Alternatives. Food & Nutrition Press, 268.
Ras G, Bailly X, Chacornac JP, Zuliani V, Derkx P, Seibert TM,
Talon R, Leroy S. 2018. Contribution of nitric oxide synthase
from coagulase-​ negative staphylococci to the development
of red myoglobin derivatives. International Journal of Food
Microbiology (in press).
Ras G, Zuliani V, Derkx P, Seibert TM, Leroy S, Talon R. 2017.
Evidence for nitric oxide synthase activity in Staphylococcus
xylosus mediating nitrosoheme formation. Frontiers in
Microbiology, 8, 598.
Sanchez Mainar M, Leroy F. 2015. Process driven bacterial com-
munity dynamics are key to cured meat colour formation
by coagulase negative staphylococci via nitrate reductase or
nitric oxide synthase activities. International Journal of Food
Microbiology, 212, 60–​66.
Sebranek JG, Bacus JN. 2007. Cured meat products without direct
addition of nitrate or nitrite: what are the issues? Meat Science,
77, 136–​147.
Sindelar JJ, Milkowski LA. 2011. Sodium nitrite in processed meat
and poultry meats: a review of curing and examining the risk
and benefit of its use. American Meat Science Association
White Paper Series, 3, 1–​14.
Talon R, Walter D, Chartier S, Barrière C, Montel MC. 1999. Effect
of nitrate and incubation conditions on the production of cata-
lase and nitrate reductase by staphylococci. International
Journal of Food Microbiology, 52, 47–​56.
Vermassen A, de La Foye A, Loux V, Talon R, Leroy S. 2014.
Transcriptomic analysis of Staphylococcus xylosus in the
presence of nitrate and nitrite in meat reveals its response to
nitrosative stress. Frontiers in Microbiology, 5, 1–​15.
Weiss J, Gibis M, Schuh V, Salminen H. 2010. Advances in ingre-
dient and processing systems for meat and meat products.
Meat Science, 86,196–​213.
Xi Y, Sullivan GA, Jackson AL, Zhou GH, Sebranek JG. 2012.
Effects of natural antimicrobials on inhibition of Listeria
monocytogenes and on chemical, physical and sensory
attributes of naturally-​cured frankfurters. Meat Science, 90,
130–​138.
Microwave Heating and Modern Cuisine

Alan L. Kelly1 and Hervé This vo Kientza2,3

1 School of Food and Nutritional Sciences, University College Cork, Ireland
2 INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
3 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France

Introduction for microwave–​matter interactions, which can be generalized as

dielectric losses, conductive losses, magnetic losses, etc. These
Since it was discovered after the Second World War that micro-
mechanisms are responsible for the commonly known forms
wave generators used at that time in radar systems could also heat
of heating such as dielectric, Joule and induction heating, all
food, microwave ovens have become a commonplace feature of
of which depend on the electromagnetic field characteristics
many kitchens, domestic and commercial (restaurant).
and the material’s properties (Huang and Richert, 2008). As the
The idea of turning radars into ovens is often attributed to
electromagnetic field has both an electric field component and a
Percy Spencer, who was working on magnetron generators at the
magnetic field component, they will interact with materials by
Raytheon company: it is said that one day, working with a high-​
different mechanisms.
intensity magnetron microwave generator, he noticed a strange
The electric field of microwaves is responsible for the dielec-
feeling as he walked past a magnetron and he recognized that
tric heating, which is effected via two primary mechanisms:
a chocolate bar in his pocket had melted; this would have given
dipolar polarization and ionic conduction. In the polarization
him the idea to turn the magnetron into an oven (Parker and
mechanism, a dipole is sensitive to electric fields in which it is
Vollmer, 2004). However, this story can be debated, because if
placed and will have a tendency to rotate in order to align itself
fat (a bad absorbent of microwaves) had been heated, he would
with the field. However, under a high-​frequency electric field, the
first have felt a heating sensation in his body, as the human body
dipoles do not have sufficient time to respond to the oscillating
is abundant with water which absorbs more microwaves. Another
field; as a result of this phase lag, they collide with each other
story was told by the late Nicholas Kurti, who was involved in
when they attempt to follow the field, and power is dissipated to
radar development during the Second World War (This, 1997):
generate heat in the material (Figure 62.1a). This is the case for
the physicists working for this development had observed that
water and other polar materials. The dipolar polarization mech-
pigeons fell cooked in front of radars. Whatever the truth, the
anism is the primary principle of microwave dielectric heating,
compact devices created for kitchens are now ubiquitous, because
which involves the heating of electrically insulating materials by
they allow rapid and energy-​efficient heating. This important fea-
dielectric loss. In the conduction mechanism, any mobile charge
ture is a consequence of operating on different heating principles
carriers (electrons, ions, etc.) move back and forth through the
than other heating systems, leading to new opportunities for
material under the influence of the microwave electric field, cre-
innovation and creativity.
ating an electric current (Figure 62.1b). These induced currents
will cause heating in the sample due to any electrical resistance
caused by the collisions of charged species with neighbouring
Principle of Operation of a Microwave Oven molecules or atoms.
Whereas traditional culinary processes primarily transmit energy For domestic microwave ovens, the target is water, which
inside food from the outside by conduction, microwave ovens explains the typical values of wavelength and frequency used
are unique, because energy is deposited directly in the mass of in domestic microwaves being around 122 mm and 2.45 GHz,
the food ingredients. This is not so strange as some think; if one respectively, so that part of the microwaves is not absorbed
puts a finger over a red LED, it can be observed that the light readily by the surface but can propagate inside the food (Datta and
can go through tissues, part of it being absorbed and thus giving Davidson, 2000). Materials can be characterized by their dielec-
energy to the absorbing medium (the finger). More generally, tric properties, which indicate their ability to convert microwave
when electromagnetic radiation encounters a medium, it can be energy to heat (Chandrasekaran et al., 2013). Dielectric proper-
reflected, absorbed, transmitted or any combination of these three ties are mainly affected by the operating temperature and the fre-
interactions (Sun et al., 2016). Many mechanisms are responsible quency of microwaves used, and materials can be classified as:

429
430 Alan L. Kelly, Hervé This vo Kientza

(a) strong absorbers of microwaves (high dielectric loss released from the food being cooked. Heating of the water in the
materials); food is rapid, but food should be allowed to stand for a period
(b) low dielectric constant materials, which are essen- of time after exposure to microwaves to allow heat to transfer to
tially transparent to microwaves; water-​poor regions through conduction.
(c) opaque or conducting materials, which reflect In a typical microwave oven, the core of the equipment is a
microwaves. magnetron, which converts electrical current to microwaves,
which are then channeled into the cooking chamber using a wave-
Lipid molecules can also heat in a microwave oven, but to a guide. The oven walls are constructed of metal, and the box shape
lesser extent (remember that the oxygen atoms in triglycerides focusses the microwaves on the food; the microwaves are further
have nonbinding doublets of electrons), as well as proteins and dispersed and focussed and reflected using a fan, while uniform
saccharides, but many glass or ceramic containers, as well as the heating is typically facilitated through the rotation of the food
air surrounding the food, are unaffected, except through the heat being heated on a central turntable. The door of the microwave is
usually made of glass to allow the cooking to be monitored, but
crucially, the glass is coated in a metal mesh, the dimensions of
(a) which are such that microwaves are entrapped and cannot pene-
trate the door. A schematic diagram of the operation of a micro-
wave oven is presented in Figure 62.2.
+ + +
The principal control variables in a microwave oven concern
– – – the power applied, which is typically chosen as a percentage of the
maximum power (in watts, and actually reflecting the percentage
O– H+ H+ O– H+ H+ O– H+ H+ of time for which microwaves are being emitted, such that an oven
H+ H+ O– H+ H+ O– H+ H+ O– set to 50% power only is emitting microwaves from the magnetron
50% of the time) and the duration of heating. A demonstration of
the alternating on and off emissions can be observed by placing
a light bulb in the microwave oven; the filament will act as an
(b)
antenna, and it will brighten when microwaves are emitted (mind
that the filament will melt if stimulated for too long).
+ + +

– – –
Microwave Heating of Food
The lack of an external source of heat during microwave heating
Na+ Cl– Na+ Cl– Na+ Cl– Na+ Cl– Na+ Cl– Na+ Cl– generally leads to a lack of browning, because the surface of food
does not heat as much as in a traditional oven or during radiative
heating in a grill. Microwaving can indeed dry materials without
FIGURE 62.1 The principle of microwave heating: (a) heating water overheating and burning; where dehydration is not desirable,
molecules and (b) ionic polarization. food should be covered by a lid or thin film before heating.

G F

C A
E

FIGURE 62.2 The principle of operation of a microwave oven: food in a suitable container (A) is placed on a rotating platform in a metal box with a trans-
parent but microwave-​impermeable door (C). Microwaves (F) are then generated by a magnetron (D) at a power level controlled by the user (E), passed through
a waveguide (G) and distributed by a fan (H) to focus on the food, including by bouncing off the interior walls of the oven (I).
Microwaves and Modern Cuisine
Browning can be achieved in some cases by cooking the
food in contact with a material that absorbs microwaves and
heats (typically metal sheets called susceptors), with the food in
direct contact with the hot surface then heating by conduction.
Browning may also occur for some food products due to caramel-
ization of sugars. There have been studies of the reactions of food
components in simple systems such as mixtures of sugars and
amino acids, and a wide range of reactions have been noted, some
of which indicate a certain extent of glycation reactions occurring
(Yaylayan et al., 1994), but not concentrated at the surface as
for conventional cooking methods. Glycation reactions have
also been identified in microwave-​heated milk (Meissner and
Erbersdobler, 1996). Amino acids with alkyl side chains, sulphur-​
containing amino acids and basic amino acids were associated
with formation of caramel, meaty and baked/​nutty notes, respect-
ively (Yaylayan et al., 1994). Valero et al. (1999) reported that the
volatile profiles of microwave-​and conventionally heated milk
were similar if the heating was uniform and the temperature con-
trol satisfactory.
There are many unique aspects to the cooking of food in micro-
wave ovens, such as the counter-​intuitive observation that the size
of the food material matters, with food of very small size being
‘missed’ by the microwaves, while food of large size is harder to
heat, as the penetrative depth of the microwaves (typically only
2–​3 cm) is insufficient to transmit heat deeply into the material
(Puligundla et al., 2013). Thus, evenly sized food pieces of
optimal size will cook best in this type of oven. There may also be
safety concerns where ‘cold spots’ persist in unheated regions of
the food, another reason why allowing to stand for heat to equili-
brate is advised (typically for a recommended time of 2 min,
although this should depend on the size and composition of the
food portion being heated; Parker and Vollmer, 2004). The distri-
bution of heat during microwave heating of milk was reviewed
by Sieber et al. (1996). Nutrient retention following microwave
heating does not seem to be inferior to any other heating method;
for example, several studies have reported no negative effects of
microwave heating on the nutritional properties of milk (Sieber
et al., 1996; Sierra and Vidal-​Valverde, 2000; Wu et al., 2018).
Strange effects can often be observed when using microwave
ovens for cooking; for example, when a dough prepared from
flour and water is made into a spherical shape (diameter 10 cm)
and heated in this way, the outside may rapidly cool while the
interior remains very hot, due to the heat being unable to escape
through the surface, which has essentially case-​hardened.
Microwave ovens are useful for multiple applications in a
kitchen beyond cooking, including reheating pre-​cooked food,
melting butter or chocolate, making caramel, and softening
skins of some fruits; they are also popular for heating popcorn in
domestic situations. Foods with a skin, such as sausages, should
be pricked before cooking to avoid bursting due to released steam
building up inside; for this reason, shell eggs should not be heated
in a microwave (and there are reports in the literature of ocular
injuries due to eggs exploding under such conditions).
Deshelled egg can be safely cooked, but controlling the nature
of the cooking can be difficult due to the nature of the heat transfer
involved, and so scrambled eggs are perhaps the most common
application. To boil an egg, the yolk and white should be pierced
431
with a knife after removing from the shell to prevent explosion
and placed in a microwave-​safe vessel lightly coated in cooking
oil, to which a little salt can be added (which heats due to ionic
polarization). For a simple demonstration of the nature of the heat
transfer in a microwave oven, consider heating a tube containing
egg white, with the tube sealed with a bung at the end of the tube
where the egg is located. A first cycle of microwave heating will
result in the egg gelling, while the water remains liquid. If the
cork sealing the tube is then removed and the tube heated again
in the microwave oven, the egg gel will be rapidly expelled from
the open end, because the steam from the heating water cannot go
through the gel. This effect explains the explosion of eggs.
Defrosting of food can be carried out using a microwave, but
only at low power, as water molecules in the rigid lattices of ice
crystals do not vibrate and heat as well as those in the liquid state
(Agmon, 1996). If defrosting is not undertaken correctly, runaway
heating can occur, whereby some regions of the food heat rapidly
while others remain frozen. In a dramatic demonstration of this, a
‘bowl’ made of ice can be prepared by freezing water in a suitably
shaped container with a weight creating an inner hollow space.
If this is then filled with liquid water and heated in a microwave
oven, the liquid water will boil even while the ice remains solid.
An early example of the application of microwave heating
in molecular cooking was the ‘Frozen Florida’, which was the
inverse of the idea of a baked Alaska, a traditional dessert in
which ice cream was prevented from melting in an oven by the
presence of an insulating layer of meringue, giving a product
with counter-​intuitive proximity of hot and frozen regions. In the
Frozen Florida, first described by Nicolas Kurti at a lecture to
the Royal Society entitled ‘The physicist in the kitchen’ in 1969,
a liqueur that does not freeze is placed in a shell of meringue
(which freezes) and frozen; when heated, the liquid heats, but the
outside remains cold (Kurti, 1969; Kurti and This, 1994).
In another concept, called the ‘vauquelin’ (This, 2008), egg
white whipped into a large volume of foam, with the addition of
a liquid and sugar (to increase viscosity), is then ‘set’ into a solid
form by rapid heating in a microwave oven. Ferran Adria at his
restaurant El Bulli in Spain used microwave cooking to rapidly
cook a ‘lighter than air’ (clearly a promotional name, because this
is not true) chocolate sponge cake, and this has since become a
common use of microwave cooking. In such a recipe, eggs, salt
and flour are mixed, melted butter is added, and the mixture is put
into a nitrous oxide siphon and shaken to generate a foam. It is
then extruded into a suitable container, such as a cup with vents
cut to allow steam to escape, and microwaved; on removal from
the cup, a highly porous ‘fluffy’ structure results.
The English chef Heston Blumenthal has recommended using
microwave heating for cooking fennel, and other chefs have
adopted its unique effects for specific culinary purposes. Meat may
also be cooked in innovative ways using microwaves, such as when
pieces of duck that have been grilled or fried to begin browning
are injected with the orange liqueur Cointreau and then heated in a
microwave, which results in the meat boiling from the inside with
an orange medium, giving a modernist version of duck a l’orange
called roast duck Pravez-​Cointreau (Kurti and This, 1994).
And finally, a ‘Gibbs’ system can be made by whipping oil in
an egg white or in an aqueous protein solution and then heating
432
the emulsion in the microwave for some tens of seconds: the
emulsion is then trapped in a gel because of the coagulation of
the proteins. This gel of an emulsion (dispersed system formula-
tion (DSF) formula (O/​W)xS) can then be dried in a traditional
oven in order to create an oleogel (formula OxS) (This, 2012; Le
Ball, 2015).
Today, combined microwave ovens are available, in which
convection cooking, or even rotisserie cooking, is combined with
microwave heating. In addition, some large-​scale continuous-​
throughput systems are becoming available, particularly for
treatment of fluids such as milk; it is perhaps surprising that
such a commonplace technology has not been widely scaled
up, as heat-​ exchangers based on microwave heating should
have advantages of even heating without hot surfaces at which
burning could occur (even at temperatures up to those used for
ultra-​high-​temperature (UHT) treatment), but it is only in recent
years that even prototypes of such systems have become available
(Baxendale et al., 2007).
Overall, although commonplace since the 1970s, and hence
perhaps seen as a relatively traditional and ‘unglamorous’ tech-
nology, there is likely still untapped potential for the use of
microwaves in culinary innovation, due to the fact that the prin-
ciple of heating involved is so different from that in other cooking
methods.
REFERENCES
Agmon N. 1996. Tetrahedral displacement: the molecular mech-
anism behind the Debye relaxation in water. Journal of
Physical Chemistry, 100, 1072–​1080.
Baxendale IR, Hayward JJ, Ley SV. 2007. Microwave reactions
under continuous flow conditions. Combinatorial Chemistry
and High Throughput Screening, 10, 802–​836.
Chandrasekaran S, Ramanathan S, Basak T. 2013. Microwave
food processing –​a review. Food Research International, 52,
243–​261.
Datta AK, Davidson PM. 2000. Microwave and radio frequency pro-
cessing. Journal of Food Science, 65, 32–​41.
Huang W, Richert R. 2008. The physics of heating by time dependent
fields: microwaves and water revisited. Journal of Physical
Chemistry B, 112, 9909–​9913.
Alan L. Kelly, Hervé This vo Kientza
Kurti N. 1969. The physicist in the kitchen. Proceedings of the Royal
Institution of London, 42, 199, 451–​467.
Kurti N, This H. 1994. Chemistry and physics in the kitchen.
Scientific American, 4, 44–​50.
Le Ball A. 2015. Hervé This and the future of food. The New York
Times Style Magazine, www.nytimes.com/​2015/​09/​17/​t-​
magazine/​herve-​this-​nbn-​future-​food.html
Meissner K, Erbersdobler HF. 1996. Maillard reaction in microwave
cooking: Comparison of early Maillard products in conven-
tionally and microwave-​heated milk. Journal of the Science of
Food and Agriculture, 70, 307–​310.
Parker K. and Vollmer M. 2004. Bad food and good physics:
The development of modern microwave cookery. Physics
Education, 39, 82–​90.
Puligundla P, Abdullah SA, Choi W, Jun S, Oh SE, Ko S. 2013.
Potentials of microwave heating technology for select food
processing applications –​A brief overview and update. Journal
of Food Processing and Technology, 4, 1–​9.
Sieber R, Eberhard P, Gallmann PU. 1996. Heat treatment of milk
in domestic microwave ovens. International Dairy Journal, 6,
231–​246.
Sierra I, Vidal-​Valverde C. 2000. Influence of heating conditions in
continuous-​flow microwave or tubular heat exchange systems
on the vitamin B1 and B2 content of milk. Lait, 80, 601–​608.
Sun J, Wang W, Ye Q. 2016. Review on microwave–​matter inter-
action funamentals and efficient microwave-​associated heating
strategies. Materials, 9, 231.
This H. 2008. Le vauquelin, www.pierregagnaire.com/​ pierre_​
gagnaire/​travaux_​detail/​23
This H. 2012. Cours de gastronomie moléculaire N°1, Quae/​
Belin, Paris.
Valero E, Sanz J, Martínez-​Castro I. 1999. Volatile components in
microwave-​and conventionally-​heated milk. Food Chemistry,
66, 333–​338.
Wu Y, Fan D, Hang F, Yan B, Zhao J, Zhang H, Chen W. 2018. Effect
of calcium on absorption properties and thermal stability of
milk during microwave heating. International Journal of
Molecular Science, 19, 1–​11.
Yaylayan VA, Forage NG, Mandeville S. 1994. Microwave and
thermally-​ induced Maillard reactions. In Parliament TH,
Morello MJ, McGorrin RJ (eds.) Thermally Generated
Flavors: Maillard, Microwave and Extrusion Processes,
American Chemical Society, Washington, DC, 449–​456.
Mineral Ions and Cooking

Christian Salles
CSGA (Centre des Sciences du Goût et de l’Alimentation), AgroSup Dijon, CNRS, INRAE,
Université de Bourgogne Franche-​Comté, F-​21000 Dijon, France

Minerals are essential nutrients that our body needs in small on taste perception of different food systems, on the colour of
amounts to work and keep us in good health (Gharibzahedi foods, and on the texture and structure of foods.
and Jafari, 2017). They are necessary in particular for building
strong bones and teeth, controlling body fluids inside and out-
side cells, and turning food into energy. Minerals are provided
Involvement of Mineral Ions in Food in Taste
mainly by foods and drinks such as meat, cereals, fish, milk and
dairy foods, fruit and vegetables, and nuts, but no food contains Perception and Appreciation
all the minerals and can provide all mineral species (Table 59.1). Dairy Products
A healthy balanced diet should provide all the minerals to avoid
Mineral ions in cheeses and changes during cheese making,
diet deficiencies. Macrominerals are essential mineral nutrients
including the maturation process, have been already extensively
such as sodium, potassium, calcium, magnesium, chloride and
studied and reported. We will only focus on the taste influence of
sulphur. Other minerals such as iron, manganese, copper, iodine,
minerals in model dairy products.
zinc, cobalt, fluoride and selenium are considered as trace elem-
The water-​ soluble extract of cheese mainly contains non-​
ents, because they are needed in smaller amounts. They are found
volatile compounds. It is a complex mixture of numerous poten-
in small amounts in a large variety of foods.
tially taste-​active compounds, such as mineral salts, organic acids,
The quality of a food product is driven by the combination of
sugars, amino acids, nucleotides, biogenic amines and peptides.
several characteristics, such as appearance, consistency, flavour
In cheeses, sodium chloride, mainly added during the salting
and nutritional value. In addition to their nutritional properties,
step of cheese processing, is usually fully responsible for percep-
minerals have a crucial and complex role in flavour perception
tion of cheese saltiness (Engel et al., 2001; Guinee, 2004; Guinee
when eating foods, and influence the texture and colour of foods.
and O’Kennedy, 2007). However, salts of potassium, calcium and
Thus, mineral ions are an essential ingredient category for food
magnesium also have a salty taste dimension.
quality.
As an example, the role of each mineral ion (sodium, potas-
This multifunctionality of some mineral ions in food can be
sium, calcium/​magnesium chlorides and phosphates) in taste
easily illustrated by the different approaches developed to take
characteristics in a model water-​soluble extract of goat cheese
into account the recommendations in term of salt content reduc-
was investigated through a method of systematically omitting
tion in food formulation. Changes in the concentration of sodium
each in turn (Engel et al., 2002), as shown in Figure 63.1.
or substitution of sodium chloride by other mineral or organic
Although sodium chloride appeared to have the biggest impact
compounds affect many properties of the food such as taste,
on saltiness perception, the decrease of the mean saltiness devi-
odour, texture and structure, and relevant strategies should be
ation due to the omission of sodium chloride was greater than
combined and adapted to the type of food in order to keep the
with other minerals taken one by one but weaker than when all
reformulated food acceptable to consumers (Salles et al., 2017).
of them were omitted (Engel et al., 2000a; Engel et al., 2000b).
Thus, these food characteristics play an important role in the
This indicated that each of the four mineral salts contributes
main quality criteria allowing the consumer to judge the overall
positively to the salty note in an additive way to explain the
quality of food, contributing to the rejection or the acceptance of
saltiness of the whole water-​soluble extract. The contribution of
the product.
phosphates to saltiness could be explained by their presence as
The objective of this chapter is not to give a complete over-
sodium salts.
view of the role of mineral ions in food properties, as this
Phosphates showed a decreasing effect on sourness, while
would require an entire book, but only to present the main food
sodium chloride had an enhancing influence. This effect of phos-
characteristics impacted by both the nature and the content of
phate ions can mainly be related to their role in pH, although this
mineral ions through a few selected examples. The interested
taste characteristic has never been directly correlated with pH,
reader will find information on the involvement of mineral ions
which means that other compounds are probably involved in that

433
434 Christian Salles

TABLE 63.1
Examples of Main Mineral Sources in Foods
Foods Ca Fe I Mg K Na P S Cu
Dairy products x x X X x x x
Meat products x X x x x
Cured meat products X x x
Fish x X x x
Raw cereals x X
Seafoods x X x x
Fruits, potatoes x X
Green vegetables, salads x x X X
Fruits and vegetables x X x x
Legumes x X x x x
Seeds, nuts x X X x x x
Drinking waters x X x x

Source: Gharibzahedi and Jafari (2017).

c
d b
c SALTINESS
b
b
b

NaCl/PO4
a
Minerals
b
NaCl b
KCl b
BITTERNESS
CaCl2/MgCl2
c
PO4
b

b
a
d
c SOURNESS
c
a

–70 –60 –50 –40 –30 –20 –10 0 10 20 30 40 50

Mean taste intensity deviation from the model mixture (/100)

FIGURE 63.1 Impact of omitting minerals from a model water-​soluble extract of goat cheese on flavour perception.
(Reprinted with permission from (Engel et al., 2000a, b). Copyright (2020) American Chemical Society.)

perception. Concerning the role of sodium chloride, it is admitted
that the perception of saltiness enhances that of sourness. Thus,
overall sourness appeared to be due to an enhancing effect of
sodium chloride on compounds acting on the pH level. In this
model food system, the significant increase of the sourness due to
phosphate omission was partially compensated by the omission
of sodium chloride, which tended to decrease the sour note
because of its enhancing contribution.
For each attribute, the means with the same letter (a, b, c, d) are
not significantly different at the 5% level according to Newman–​
Keuls tests. The mean deviation is drawn for each mean (from
six replicates).
The omission of both calcium and magnesium chlorides was
responsible for a significant decrease in the bitter taste, which
means that they might be at least partially responsible for
bitterness of the water-​soluble extract. Such an effect of diva-
lent cations has already been reported, as the taste of cheeses
salted with calcium or magnesium chloride was extremely bitter
after 16 weeks of ripening (McSweeney, 1997). In addition,
the simultaneous omission of phosphate (sodium form) and
sodium chloride significantly increased the bitterness perceived,
whereas the omission of phosphate or sodium chloride alone
had no significant effect (Engel et al., 2000b). This effect was
explained by a larger amount of sodium present in the model
Mineral Ions and Cooking
solution originating from both sodium chloride and phosphates,
leading to a significant masking effect of sodium chloride on the
perceived bitterness of the entire model solution. This confirmed
that it is impossible to consider only the individual taste prop-
erties of a compound to explain its impact on the water-​soluble
extract mixture context. Although it had no significant impact,
the omission of sodium seems to have a tendency to increase the
bitterness.
Stepwise multiple linear regressions processed on the omission
test data confirmed these findings (Engel et al., 2000b). Sodium,
calcium, magnesium and potassium chlorides and phosphates
were, respectively, responsible for 75%, 13.8%, 5.8% and 4.4%,
respectively, of the salty taste of the water-​soluble solution of
cheese, and bitterness was mainly due to the antagonistic effects
between calcium/​ magnesium chlorides (41.3%) for the posi-
tive contribution and both sodium chloride (29.2%) and sodium
phosphates (9.4%) for the negative one.
This study highlights that, in a complex mixture context, the
individual taste properties of a compound are often not suffi-
cient to explain its impact on the overall taste but interactions at
different levels should be taken into account.
Variations in salt content also modify the rheological proper-
ties and microstructure of model cheeses, due mainly to sodium
ion binding to caseins and its mobility in the food matrix, which
could influence saltiness perception. A multimodal approach
was used to understand the effects of changes to the compos-
ition of model cheeses (lipid/​protein ratio and sodium chloride
content) on sodium ion mobility (using 23Na nuclear magnetic
resonance (NMR)), in-​mouth sodium release and saltiness per-
ception (Boisard et al., 2013; Boisard et al., 2014). An increase
in the salt content was shown to decrease cheese firmness and
perceived hardness and to increase sodium ion mobility, in vivo
sodium release and both saltiness and aroma perception. It was
also reported that the mobility of sodium ions in the food matrix
decreased when the protein concentration increased. Moreover,
the mobility of sodium ions depended on the type of protein
and seemed to be related to its content of negative sites (Mosca
et al., 2015).
Meat Products
Meat is recognized as an important source of protein in the human
diet, but it also provides a large variety of minerals, which are
involved in many physiological functions (Beisenov et al., 2017;
Stasiak et al., 2017; Tomovic et al., 2015) and are thus essential
for good health.
Sodium chloride is an essential mineral in processed meat
products, which contributes to water retention, colour, fat binding
and flavour. This mineral is fully involved in taste characteristics
in several meat products. After fractionation of a water-​soluble
extract of dry sausages, the fractions containing sodium chloride,
free amino acids and peptides were described as having the
highest taste intensity, which was described as bouillon, salty,
bitter and sour (Henriksen and Stahnke, 1997). Sodium chloride
is not only responsible for salty taste but is also suspected to
enhance the taste of co-​eluted amino acids. Similar observations
were reported for cooked ham, in which sodium chloride is
435
suspected to enhance the umami taste of peptides and glutamate
(Valentin et al., 1998).
Another important cation provided in the diet by meat products
is iron, which plays many biological functions in the human body
(Czerwonka and Tokarz, 2017). The large majority of iron in the
body is in the form of haem molecules consisting of four pyr-
role rings joined together by methionine bridges. In the haem or
non-​haem form, iron is critical to the structure of various proteins
associated with different biological functions. The haem ion ratio
varies according to the type of meat (Pretorius et al., 2016). The
average bioavailability of haem iron of the meat is 23% and that
of non-​haem iron 3% (Schoenfeldt and Hall, 2011). However,
raw meat is rarely consumed as such, but mostly submitted to
a cooking process at high temperature. Because of the cooking
process, a thermal leakage is observed, depending on the type and
intensity of the thermal process. This leads to a reduction of the
share of haem by about 4–​25% (Czerwonka and Szterk, 2015).
Nitrite, mainly used in sodium (or potassium) nitrite salt form,
is an essential ingredient currently used in the curing of different
meat products (Pegg and Shahidi, 1997). It is responsible not
only for characteristic colour but also for flavour, extended shelf-​
life and microbiological stability. In combination with sodium
chloride, it has a significant bacteriostatic action, particularly in
inhibiting the growth of Clostridium botulinum, the bacterium
responsible for life-​threatening botulism.
However, the food industry is recommended to reduce or phase
out the usage of nitrite in cured meat because it may contribute to
the formation of carcinogenic nitrosamines. The development of
substitutes has been explored for several years (Pegg and Shahidi,
1997). Thus, EFSA (European Food Safety Agency) has recently
re-​assessed the safety of sodium and potassium salts of nitrite
(E 249–​250) and proposed an admissible daily intake (ADI) of
0.07 mg/​kg bw/​day (EFSA, 2017). In cooked cured ham, sodium
nitrite was found to impact volatile compound composition
(Guillard et al., 1998). Those authors reported a significantly
lower amount of phospholipids during its processing, in par-
ticular for phosphatidyl choline and phosphatidyl ethanolamine,
the two major phospholipids of pork meat. For the major fatty
acids contained in these two phospholipids, the strongest effect of
nitrite was observed on arachidonic acid for both phospholipids,
and on oleic acid only for phosphatidyl choline, as the content of
these fatty acids was reduced in the presence of sodium nitrite.
For the other fatty acids, the effect was, rather, due to the cooking
process. In consequence, volatile compounds such as hexanal,
heptanal and oct-​1-​en-​3-​ol were found in lower amounts than in
cooked cured meat untreated with sodium nitrite.
Tomatoes
The presence of mineral salts in tomato has a positive influence
on its taste characteristics (Petro-​Turza, 1986–​1987; Petro Turza
and Teleky-​Vamosky, 1989). Those authors reported in particular
that a model juice made of mineral salts, amino acids, sugars
and organic acids presents overall taste characteristics of tomato
juice. In addition, minerals can have an indirect action on the
taste, though the saltiness characteristic was not the dominant
taste in this vegetable, for which the overall taste was driven
436 Christian Salles

by sweetness and sourness. Potassium, which is the major min- sodium bicarbonate profiles and low pH (Platikanov et al., 2017).
eral ion in tomatoes, can influence the free acid content, while Moreover, minor constituents such as magnesium, potassium and
phosphate can interact by its buffering capacity (Petro-​Turza, silicon ions also appeared to be relevant for the taste evaluation
1986–​1987). The association of organic acids and potassium ions of water samples, probably due to their spurious correlation with
was determinant for the saltiness characteristic of tomato (Salles the major concentrations of calcium and sodium (Platikanov
et al., 2003). Indeed, through omission tests performed on a com- et al., 2017).
plex tomato juice model mixture, these authors reported that the The role of nitrate and sulphate anions in the taste of water is
main part of saltiness perception arose from the anionic form of not entirely clear yet (Lopez et al., 2017). Sulphate at low and
citric and malic acids, with potassium as the main counterion. moderate concentrations tended to improve the overall taste of
Hence, the acid form of the organic acids was responsible for water. However, at high concentration, it was perceived nega-
sourness and the dissociated potassium form was implied in salti- tively, with saltiness and bitterness tastes. Nitrate was not found
ness. This shows that changes in ionic ratio and equilibrium could to have a relevant role in the taste of water (Lopez et al., 2017).
strongly influence the taste characteristics of food. The mineral content of water may impact its extraction cap-
acity when cooking. The higher the mineral concentration in
Drinking Water the water, the lower the extraction yields of aluminium, organic
carbon and polyphenols in infusions (Mossion et al., 2008). This
Common dissolved minerals such as calcium, magnesium, potas- is explained by the calcium in the mineral water, which could
sium, sodium, bicarbonate, chloride, nitrate and sulphate affect complex with pectins of the cell walls of tea leaves and decrease
the taste of water. In the absence of off-​flavours due to organic component extraction.
contaminants responsible for odours, mineral content is a major The sensory quality of tea infusions is influenced by water
determinant of both taste and consumer acceptability of drinking containing different concentrations of mineral ions. The taste
water (Bruvold, 1970; Platikanov et al., 2013). Total dissolved and the visual appearance of water are less appreciated when the
solid reflects the overall mineral content of water and is the most calcium ion concentration is over 40 mg/​L, which promotes the
important determinant of consumer liking. formation of tea sediments (Xu et al., 2013). Bitter, umami and
A study on the taste of waters with different sensory meth- sweet taste are weakened by a high calcium concentration, whilst
odologies reported a bitter and metallic taste for low-​mineral-​ the astringent perception is strengthened for green tea infusion
content waters, a neutral taste and a sensation of freshness for (Yin et al., 2014). The type of brewing water influences the taste
medium-​mineral-​content waters and finally, a more salty taste for quality, catechin concentration and antioxidant capacity of tea
waters with the highest mineral content (Teillet et al., 2010a). infusions (Xu et al., 2017). Those authors reported that calcium,
A consumer study performed on waters with different mineral- magnesium and sodium ions in the tea infusion were mainly
ization levels showed that the most likely preferred type of water derived from the water used, while the potassium originated
was that with medium mineralization, with total dissolved solids mainly from the tea leaves. Generally, the concentration of metal
about 300–​350 mg/​L, perceived as tasteless and cooler (Teillet ions in the tea infusion is significantly influenced by the concen-
et al., 2010b). tration of metal ions in the water.
A Japanese study reported that evaluation of potable water For example, the higher the calcium concentration in water, the
might be possible using organic matter, total hardness (calcium), lower the calcium extraction yield from tea, while a high concen-
iron and bicarbonate concentrations, as they were well correlated tration of sodium in water improves the sodium extraction yield
with sensory appreciation (Sasaki et al., 1996). in the infusion tea. Concerning anions, it was found that chloride,
In natural waters, synergistic and antagonistic effects can nitrate and sulphate in tea infusions originate mostly from the
take place between the different ions. Thus, the influence of water used, while in contrast, fluoride ions were mostly extracted
each mineral ion on taste perception and preference of waters from the tea leaves (Xu et al., 2017). As for cations, the con-
has been detailed. The preferred water samples were associated centration in brewing water influences the extraction yield in tea
with a moderate content of total dissolved solids and with rela- infusion. The colour of the tea is also influenced by the quality of
tively high concentration of bicarbonates, sulphate, calcium and the water. The colour of tea infusions darkened with increasing
magnesium ions, as well as relatively high pH values. In con- conductivity and pH of the water, probably due to the oxidation
trast, high concentrations of sodium, potassium and chlorides of the extracted catechins in the infusion. A high conductivity
were scored low by the consumers, while residual chlorine did decreases the extraction yield of catechins and caffeine, and a
not affect the ratings but was perceived and allowed discrimin- high pH lowers the stability of catechins. Thus, reducing the pH
ation of waters (Platikanov et al., 2013). The role of magnesium of water can partially improve the taste quality and acceptability,
cations was unclear and strongly dependent on the nature of the and increases the concentration of catechins, in tea infusions (Xu
associated anion. Magnesium bicarbonate and sulphate probably et al., 2017).
have a positive or a neutral effect on waters, while magnesium
chloride has a negative bitter effect.
It has also been reported that the majority of consumers liked
waters rich in calcium sulphate and calcium bicarbonate/cal- Mineral Ions May Affect the Colour of Food
cium sulphate, with high alkalinity and pH; however, a small Along with aroma, taste and texture perception, colour is one of
group of consumers preferred waters with sodium chloride and the key factors for food appreciation by consumers. It provides
Mineral Ions and Cooking
symbolic and associative information about the food. Moreover,
it has been reported that light colours modulate consumers’ will-
ingness to eat and their hedonic impressions of foods (Yang et al.,
2016). For example, bread making can be optimized by taking
into account the crust colour of maximum acceptability (Castro
et al., 2017). Similar observations have been reported for beef
products (Holman et al., 2016).
Crude Food Products
Iron, zinc and iodine are important micronutrients, which are
commonly deficient worldwide. Among these, iron is a highly
reactive mineral ion and may lead to a dull colour and off-​taste
when present in excess in foods due to its addition as a micronu-
trient. As an example, for fruit and vegetable purées, iron fortifi-
cation may lead to colour and taste modification (Hurrell, 1997).
Moreover, a consequence can be a partial or total loss of product
identity, where consumers use colour as an important key indi-
cator of quality. This colour change is mainly due to the forma-
tion of iron–​polyphenol complexes, and different strategies have
been developed to limit the formation of such complexes and to
stabilize the colour, such as increasing pH or using complexing
or competitive agents (Habeych et al., 2016).
On the other hand, cations such as calcium can promote
attractive colours. For example, calcium ions are considered to
be a promoter of anthocyanin accumulation, as they are involved
in signal transduction regarding anthocyanin biosynthesis
(Vitrac et al., 2000). In the case of grapes, calcium chloride can
be associated with ethephon to improve skin colour (Scavroni
et al., 2018). It has also been reported that spraying of calcium
or potassium salt solutions, alone or in combination, signifi-
cantly increased fruit weight, firmness and anthocyanin concen-
tration (Solhjoo et al., 2017). Another example for vegetables
is the effect of calcium chloride treatment during storage on the
colour properties of fresh-​cut green beans (Kasim and Kasim,
2015). By comparing the effect of solutions of different calcium
chloride concentrations with a control without calcium chloride,
it was reported that calcium chloride treatment of fresh-​cut
green beans was effective in the retention of their colour for
a short period of storage. This phenomenon was related to the
reduction of polyphenol oxidase activity of the samples in the
presence of a high level of calcium chloride.
Another example worth mentioning is the colour of mollusc
shells (Lertvachirapaiboon et al., 2015), which are made of
aragonite calcium carbonate structured in different stratified
layers varying in thickness. The shells with multiple aragonite
thicknesses expressed a multicolour nature, as the upper ara-
gonite layers reflected colour from the lower layers. Thus, the
expressed colour was directly associated with the structural
architecture of the shell (thickness and number of layers), and
colour is known to influence our appreciation (Kim et al., 2011).
During Cooking
The Maillard reaction plays an important role in the appearance
and flavour of food subjected to cooking or roasting (Martins et al.,
2001). This complex reaction is initiated by an amine–​carbonyl
437
reaction of free or bound amino acids and reducing sugars, and
leads to numerous active intermediates through a viable electro-
chemical system (Rizzi, 2003). Maillard browning is accelerated
by the dihydrogen phosphate ion acting as a base for catalysing
Amadori rearrangement (Potman and Vanwijk, 1989), leading to
acceleration of the reaction (Davidek et al., 2002).
In a study of the effect of phosphate and carboxylate ions on
browning, browning for a series of reducing sugars was measured
with and without the presence of beta-​alanine (Rizzi, 2004).
Significant browning was observed for sugar alone, suggesting
that polyatomic anions can form reactive Maillard reaction
intermediates directly from sugars. These chemically reducing
species formed during phosphate-​ion-​catalysed degradation of
reducing sugar can be directly quantified by electrochemical
methods (Rizzi et al., 2010). As a mechanism, it is suggested
that polyatomic anions (XO2− form) facilitate the aldose/​ketose
equilibration of reducing sugars and aldose sugar dehydration to
form 3-​deoxyosones. This leads to the formation of additional
alpha-​dicarbonyls, which are key intermediates leading to col-
oured melanoidins (Rizzi, 2004).
Sodium tetraborate decahydrate (Borax), employed as a pre-
servative agent in food, was found to enhance Maillard reaction
browning to a degree uniformly greater than that produced by
phosphate anions on an equimolar basis (Rizzi, 2007). As for
phosphate, borate anions operate during the early stage of the
Maillard reaction but by completely different mechanisms.
Indeed, borax accelerates most browning for aldose and ketoses
with trans-​ configured hydroxyl groups in the vicinal C2/​ C3
positions of their hemiacetal forms. The complexation occurs,
therefore, in the open-​chain form, leading to a higher concentra-
tion of carbonyl moieties and thereby enhancing Maillard reac-
tion browning (Rizzi, 2007).
As well as bidentate anions like phosphates, carboxylates and
tetraborate leading to enhanced browning, it has been reported that
the addition of sodium, potassium, magnesium and calcium ions
and choline chloride also leads to increased browning (Rizzi, 2010).
Among cations, lithium, sodium, potassium, rubidium and caesium
were found to produce only a small increase in browning, while
a greater effect was observed for calcium and magnesium (Rizzi,
2008). It is noteworthy that browning was reduced or suppressed by
the addition of an ammonium salt to the sugar–​amino acid reaction
(Rizzi, 2010) and by triethylammonium ion, but was not affected by
the salt of a stronger base such as guanine (Rizzi, 2008).
Mineral Ions Affect the Texture and Structure
of Food
Mineral ions are well known to influence the structure of most
food products, such as polymeric gels, dairy, meat, bakery
products and sauces. We will limit this chapter to the presentation
of examples for vegetables and emulsions.
Vegetables
Calcium interacts with pectin to form calcium pectate, which
maintains the integrity of cell walls of vegetables and assures
438
firmness by cross-​linking with cell walls and pectin. Thus,
treatment of fruits and vegetables with calcium salts is cur-
rently used to retain firmness during storage (Rico et al.,
2007b; Martin-​Diana et al., 2007). Moreover, this treatment
is effective in reducing chlorophyll and protein losses, and
inhibiting plant tissue senescence (Rico et al., 2007a). As
examples, calcium lactate has been used to reduce respiration,
increase firmness retention and reduce the impact of physio-
logical disorders in apple, and more generally for delicate
fruits. It has also been used specifically as a firming agent for
several varieties of fruits. Compared with calcium chloride,
calcium lactate avoids bitterness and off-​ flavour defaults.
For pears, it has been shown that calcium lactate reinforces
the structure of the fruit, maintaining the fibrillary packing
in the cell and the intercellular contacts. This reinforcement
counteracted the increase of enzyme activities related to the
degradation of cell wall components (pectinmethylesterase and
polygalacturonase), which could be activated by calcium add-
ition (Alandes et al., 2009). Thus, this leads to an improvement
of the structural integrity of the fresh-​cut fruit.
The combined effect of calcium chloride solutions and
heat treatment on fresh-​ cut melons positively maintained
or improved firmness (Luna-​ Guzman et al., 1999). In the
same way, the combination of heat-​shock and calcium lac-
tate treatments was shown to be very efficient in maintaining
firmness in carrots during storage (Rico et al., 2007b). Sensory
analysis experiments showed that carrots washed with calcium
lactate at 50 °C showed higher scores for texture (firmness)
and visual appearance than with standard treatments, such
as calcium lactate at 25 °C and chlorine, at 10 days’ storage.
These carrot samples were also rated better in terms of sale-
ability than the others. This is associated with the control of
browning-​related enzymes, which are inhibited by heat-​shock.
Cryo-​ scanning electron microscopy showed that combined
heat-​shock and calcium lactate treatments were more effective
in maintaining the turgor of cortex tissue cells and reduced
the extent of lignification at the peeler-​wounded areas (Rico
et al., 2007b). The authors also reported better diffusion and
solubility of calcium in the tissues at 50 °C, and explained
the beneficial effect on texture of temperature treatments and
calcium solutions by the stimulation of pectin methyl esterase
by temperature. This enzyme is able to cleave the methoxyl
groups from methylated pectin substances, which generates
free carboxyl groups able to bind endogenous and added cal-
cium, leading to firmer tissues. Similar observations were
reported for fresh-​cut Galia melon (Cecilia Silveira et al.,
2011). Moreover, those authors observed an important loss of
flavour in all treatments except with calcium chloride, lactate
and ascorbate, which was the only treatment found acceptable
by consumers.
Emulsions
Emulsions are destabilized by different minerals at elevated
temperature and low pH (Yuan et al., 2011). The addition of
minerals can destabilize emulsions in these conditions. Low
pH and elevated temperature decrease the surface potential of
Christian Salles
oil microdroplets and enhance their gathering. The addition of
minerals contributes to the decrease of surface potential and
thus contributes to flocculation and coalescence, which are the
two main steps of emulsion destabilization. The key role of the
interfacial charging was also demonstrated for double emulsions,
such as oil-​in-​water-​in-​oil (O/​W/​O) emulsions, as their formation
is enhanced at low pH and low salt concentration (Pradhan and
Rousseau, 2012). Concerning water-​in-​oil-​in-​water (W/​O/​W)
emulsion, osmotic pressure differences between the internal and
external aqueous phases influences the stability of the emulsions,
the release rate of components from the internal phase, and the
engulfment of the internal droplets by the flow of water from
the external continuous aqueous phase to the internal phase
(Benichou et al., 2004). These authors observed that, if no salt
is present in the inner phase but the external phase contains salt,
the inner phase disappears completely after a few minutes, indi-
cating a water flow from the inner to the external aqueous phase.
When salt is present in the inner phase, water migration progres-
sively decreases and stops when an equilibrium between the two
aqueous phases is reached. Thus, the presence of an electrolyte
such as salt in the inner phase is necessary for the stability of the
double emulsion achieved by the osmotic pressure equalization.
Using glucose to balance or unbalance osmotic pressure, it was
shown that the release of salt is driven by the chemical poten-
tial difference between the two aqueous phases rather than the
unbalanced osmotic pressures (Pawlik et al., 2010). Gelation of
the inner phase using gelatin (Sapei et al., 2012) or whey pro-
tein isolate (Perez-​Moral et al., 2014) strongly increased the sta-
bility of the double emulsion and slowed down the diffusion of
electrolytes.
Conclusion
Minerals are present in most foods, constitute a major ingre-
dient family, and are considered as essential nutrients. Although
minerals exist in a large variety, no food contains all minerals,
and quantitative differences between foods may be very high.
They are associated with several important functionalities in
food related to sensory perception, such as bitterness, salti-
ness, metallic perception, colour, general appearance, texture
and acceptability by the consumer. Concerning taste perception,
beside their direct involvement in saltiness (sodium chloride) and
bitterness (calcium/​magnesium chlorides), they may act as taste
enhancers or inhibitors at the taste receptor level. At too high
levels in food, they can be responsible for strong off-​flavours,
leading to food rejection. Despite their high involvement in the
final quality of most foods, studies on minerals have been rather
neglected compared with studies on organic compounds. All the
complex mechanisms directly or indirectly involving mineral ions
in perception need to be unravelled for a better understanding of
the mechanisms leading to perception.
REFERENCES
Alandes L, Perez-​Munuera I, Llorca E, Quile A, Hernando I. 2009.
Use of calcium lactate to improve structure of “Flor de Invierno”
fresh-​cut pears. Postharv. Biol. Technol., 53, 145–​151.
Mineral Ions and Cooking
Beisenov AK, Amanzholov KZ, Mirzakulov SM, Micinski B,
Pogorzelska J, Micinski J. 2017. Fattening, slaughter features
and meat mineral composition of 3 beef cattle breeds. J.
Elementol., 22, 1141–​1154.
Benichou A, Aserin A, Garti N. 2004. Double emulsions stabilized
with hybrids of natural polymers for entrapment and slow
release of active matters. Adv. Colloid Interface Sci.,
108, 29–​41.
Boisard L, Andriot I, Arnould C, Achilleos C, Salles C, Guichard, E.
2013. Structure and composition of model cheeses influence
sodium NMR mobility, kinetics of sodium release and sodium
partition coefficients. Food Chem., 136, 1070–​1077.
Boisard L, Andriot I, Martin C, Septier C, Boissard V, Salles C,
Guichard E. 2014. The salt and lipid composition of model
cheeses modifies in-​ mouth flavour release and perception
related to the free sodium ion content. Food Chem., 145,
437–​444.
Bruvold WH. 1970. Mineral taste and potability of domestic water.
Water Res., 4, 331–​340.
Castro W, Oblitas J, Chuquizuta T, Avila-​George H. 2017. Application
of image analysis to optimization of the bread-​making process
based on the acceptability of the crust color. J. Cereal Sci., 74,
194–​199.
Cecilia Silveira A, Aguayo E, Chisari M, Artes F. 2011. Calcium salts
and heat treatment for quality retention of fresh-​cut “Galia”
melon. Postharv. Biol. Technol., 62, 77–​84.
Czerwonka M, Szterk A. 2015. The effect of meat cuts and thermal
processing on selected mineral concentration in beef from
Holstein-​Friesian bulls. Meat Sci., 105, 75–​80.
Czerwonka M, Tokarz A. 2017. Iron in red meat –​friend or foe. Meat
Sci., 123, 157–​165.
Davidek T, Clety N, Aubin S, Blank I. 2002. Degradation of the
Amadori compound N-​(1-​deoxy-​D-​fructos-​1-​yl)glycine in
aqueous model systems. J. Agric. Food Chem., 50, 5472–​5479.
EFSA. 2017. Nitrites and nitrates added to food [in French] (www.
efsa.europa.eu/​fr/​corporate/​pub/​nitritesandnitrates170614)
[Online].
Engel E, Nicklaus S, Garem A, Septier C, Salles C, Le Quéré JL.
2000a. Taste active compounds in a goat cheese water-​soluble
extract. 1. Development and sensory validation of a model
water-​soluble extract. J. Agric. Food Chem., 48, 4252–​4259.
Engel E, Nicklaus S, Salles C, Le Quéré JL. 2002. Relevance of
omission tests to determine flavour-​active compounds in food:
Application to cheese taste. Food Qual. Prefer., 13, 505–​513.
Engel E, Nicklaus S, Septier C, Salles C, Le Quéré JL. 2000b. Taste
active compounds in a goat cheese water-​ soluble extract.
2. Determination of the relative impact of water-​soluble extract
components on its taste using omission tests. J. Agric. Food
Chem., 48, 4260–​4267.
Engel E, Septier C, Leconte N, Salles C, Le Quéré JL. 2001.
Determination of taste-​active compounds of a bitter Camembert
cheese by omission tests. J. Dairy Res., 68, 675–​688.
Gharibzahedi SMT, Jafari SM. 2017. The importance of minerals in
human nutrition: Bioavailability, food fortification, processing
effects and nanoencapsulation. Trends Food Sci. Technol., 62,
119–​132.
Guillard AS, Goubet I, Salles C, Le Quéré JL, Vendeuvre JL.
1998. Role of sodium nitrite on phospholipid composition of
cooked cured ham. Relation to its flavor. In: Contis ET, Ho
CT, Mussinan CJ, Parliment TH, Shahidi F, Spanier AM (eds.)
Food Flavors: Formation, Analysis and Packaging Influences.
Elsevier, Amsterdam, 245–​253.
Guinee TP. 2004. Salting and the role of salt in cheese. Int. J. Dairy
Technol., 57, 99–​109.
439
Guinee TP, O’Kennedy BT. 2007. Reducing salt in cheese and
dairy spreads. In: Kilcast D, Angus F. (eds.) Reducing Salt in
Foods: Practical Strategies. Woodhead Publishing Limited,
Cambridge, UK, 316–​357.
Habeych E, Van Kogelenberg V, Sagalowicz L, Michel M, Galaffu N.
2016. Strategies to limit colour changes when fortifying food
products with iron. Food Res. Int., 88, 122–​128.
Henriksen AP, Stahnke LH. 1997. Sensory and chromatographic
evaluations of water soluble fractions from dried sausages. J.
Agric. Food Chem., 45, 2679–​2684.
Holman BWB, Mao Y, Coombs CEO, Van De Ven RJ, Hopkins DL.
2016. Relationship between colorimetric (instrumental) evalu-
ation and consumer-​defined beef colour acceptability. Meat
Sci., 121, 104–​106.
Hurrell RF. 1997. Preventing iron deficiency through food fortifica-
tion. Nutr. Rev., 55, 210–​222.
Kasim R, Kasim MU. 2015. Biochemical changes and color prop-
erties of fresh-​ cut green bean (Phaseolus vulgaris L. cv.
gina) treated with calcium chloride during storage. Food Sci.
Technol., 35, 266–​272.
Kim A, Kim HS, Park SO. 2011. Measuring of the perceptibility and
acceptability in various color quality measures. J. Opt. Soc.
Korea, 15, 310–​317.
Lertvachirapaiboon C, Parnklang T, Pienpinijtham P, Wongravee K,
Thammacharoen C, Ekgasit S. 2015. Selective colors reflection
from stratified aragonite calcium carbonate plates of mollusk
shells. J. Struct. Biol., 191, 184–​189.
Lopez P, Perez-​Rodriguez I, Estrany F, Devesa R. 2017. Effects of
sulfate and nitrate on the taste of water: A study with a trained
panel. J. Water Supply Res. T, 66, 598–​605.
Luna-​ Guzman I, Cantwell M, Barrett DM. 1999. Fresh-​ cut
cantaloupe: Effects of CaCl2 dips and heat treatments on
firmness and metabolic activity. Postharv. Biol. Technol.,
17, 201–​213.
Martin-​Diana AB, Rico D, Frias JM, Barat JM, Henehan GTM,
Barry-​Ryan C. 2007. Calcium for extending the shelf life of
fresh whole and minimally processed fruits and vegetables: A
review. Trends Food Sci. Technol., 18, 210–​218.
Martins SIFS, Jongen WMF, Van Boekel MAJS. 2001. A review of
Maillard reaction in food and implications to kinetic model-
ling. Trends Food Sci. Technol., 11, 364–​373.
Mcsweeney PLH. 1997. The flavour of milk and dairy products: III.
Cheese: taste. Int. J. Dairy Technol., 50, 123–​128.
Mosca AC, Andriot I, Guichard E, Salles C. 2015. Binding of Na+ ions
to proteins: Effect on taste perception. Food Hydrocolloids,
51, 33–​40.
Mossion A, Potin-​Gautier M, Delerue S, Le Hecho I, Behra P. 2008.
Effect of water composition on aluminium, calcium and
organic carbon extraction in tea infusions. Food Chem., 106,
1467–​1475.
Pawlik A, Cox PW, Norton IT. 2010. Food grade duplex emulsions
designed and stabilised with different osmotic pressures. J.
Colloid Interface Sci., 352, 59–​67.
Pegg RB, Shahidi F. 1997. Chemistry and processing aspects of
nitrite-​free cured meats. In: Spanier AM, Tamura M, Okai H,
Mills O. (eds.) Chemistry of Novel Foods. Allured Publishing
Corporation, Carol Stream, IL, 273–​306.
Perez-​Moral N, Watt S, Wilde P. 2014. Comparative study of the
stability of multiple emulsions containing a gelled or aqueous
internal phase. Food Hydrocolloids, 42, 215–​222.
Petro-​Turza M. 1986–​1987. Flavor of tomato and tomato products.
Food Rev. Int., 2, 309–​351.
Petro Turza M, Teleky-​Vamosky G. 1989. Study on taste substances
of tomato. III. Sensory evaluations. Nahrung, 33, 387–​394.
440
Platikanov S, Garcia V, Fonseca I, Rullan E, Devesa R, Tauler R.
2013. Influence of minerals on the taste of bottled and tap
water: A chemometric approach. Water Res., 47, 693–​704.
Platikanov S, Hernandez A, Gonzalez S, Luis Cortina J, Tauler R,
Devesa R. 2017. Predicting consumer preferences for mineral
composition of bottled and tap water. Talanta, 162, 1–​9.
Potman RP, Vanwijk TA. 1989. Mechanistic studies of the Maillard
reaction with emphasis on phosphate-​mediated catalysis. ACS
Symposium Series, 409, 182–​195.
Pradhan M, Rousseau D. 2012. A one-​step process for oil-​in-​water-​
in-​oil double emulsion formation using a single surfactant. J.
Colloid Interface Sci., 386, 398–​404.
Pretorius B, Schoenfeldt HC, Hall N. 2016. Total and haem iron
content lean meat cuts and the contribution to the diet. Food
Chem., 193, 97–​101.
Rico D, Martin-​ Diana AB, Barat JM, Barry-​ Ryan C. 2007a.
Extending and measuring the quality of fresh-​cut fruit and
vegetables: A review. Trends Food Sci. Technol., 18, 373–​386.
Rico D, Martin-​Diana AB, Frias JM, Barat JM, Henehan GTM,
Barry-​ Ryan C. 2007b. Improvement in texture using cal-
cium lactate and heat-​shock treatments for stored ready-​to-​eat
carrots. J. Food Eng., 79, 1196–​1206.
Rizzi GP. 2003. Electrochemical study of the Maillard reaction. J.
Agric. Food Chem., 51, 1728–​1731.
Rizzi GP. 2004. Role of phosphate and carboxylate ions in Maillard
browning. J. Agric. Food Chem., 52, 953–​957.
Rizzi GP. 2007. On the effect of tetraborate ions in the generation of
colored products in thermally processed glycine-​carbohydrate
solutions. J. Agric. Food Chem., 55, 2016–​2019.
Rizzi GP. 2008. Effects of cationic species on visual color formation
in model Maillard reactions of pentose sugars and amino acids.
J. Agric. Food Chem, 56, 7160–​7164.
Rizzi GP. 2010. Control of color formation by ionic species in non-​
enzymic browning reactions. In: Mottram DS, Taylor AJ (eds.)
Controlling Maillard Pathways to Generate Flavors. ACS
Symposium series, American Chemical Society, Washington,
DC, 121–​128.
Rizzi GP, Amba EE, Heineman WR. 2010. Quantification of
chemically reducing species in the phosphate ion catalyzed
degradation of reducing sugars. J. Agric. Food Chem., 58,
9739–​9743.
Salles C, Kerjean JR, Veiseth-​Kent E, Stieger M, Wilde P, Cotillon
C. 2017. The TeRiFiQ project: Combining technologies to
achieve significant binary reductions in sodium, fat and sugar
content in everyday foods whilst optimising their nutritional
quality. Nutr. Bull., 42, 361–​368.
Salles C, Nicklaus S, Septier C. 2003. Determination and gustatory
properties of taste-​active compounds in tomato juice. Food
Chem., 81, 395–​402.
Sapei L, Naqvi MA, Rousseau D. 2012. Stability and release
properties of double emulsions for food applications. Food
Hydrocolloids, 27, 316–​323.
Christian Salles
Sasaki K, Iwanaga C, Watanabe M, Suzuki K, Hamaoka T, Kondo
S. 1996. Correlation between sensory good taste and chem-
ical compositions in natural and mineral water. Nippon Nogeik.
Kaishi, 70, 1103–​1116.
Scavroni J, Ferreira LC, Ferrarese MDLL, Ono EO, Rodrigues JD.
2018. Ethephon and calcium chloride, a combination that
improves skin color of “rubi” table grape. Rev. Bras. Frutic.,
40, e-​777–​777.
Schoenfeldt HC, Hall NG. 2011. Determining iron bio-​availability
with a constant heme iron value. J. Food Comp. Anal., 24,
738–​740.
Solhjoo S, Gharaghani A, Fallahi E. 2017. Calcium and potassium
foliar sprays affect fruit skin color, quality attributes, and min-
eral nutrient concentrations of ‘Red Delicious’ apples. Int.
J. Fruit Sci., 17, 358–​373.
Stasiak K, Roslewska A, Stanek M, Cygan-​Szczegielniak D, Janicki
B. 2017. The content of selected minerals determined in the
liver, kidney and meat of pigs. J. Elementol., 22, 1475–​1483.
Teillet E, Schlich P, Urbano C, Cordelle S, Guichard E. 2010a.
Sensory methodologies and the taste of water. Food Qual.
Prefer., 21, 967–​976.
Teillet E, Urbano C, Cordelle S, Schlich P. 2010b. Consumer percep-
tion and preference of bottled and tap water. J. Sensory Stud.,
25, 463–​480.
Tomovic V, Jokanovic M, Sojic B, Skaljac S, Tasic T, Ikonic P. 2015.
Minerals in pork meat and edible offal. Procedia Food Sci., 5,
293–​295.
Valentin J, Guillard AS, Septier C, Salles C, Le Quéré JL. 1998.
Identification of tasty compounds of cooked cured ham:
Physico-​chemical and sensory approaches. In: Contis ET, Ho
CT, Mussinan CJ, Parliment TH, Shahidi F, Spanier AM. (eds.)
Food Flavors: Formation, Analysis and Packaging Influences.
Elsevier, Amsterdam, 195–​205.
Vitrac X, Larronde F, Krisa S, Decendit A, Deffieux G, Merillon JM.
2000. Sugar sensing and Ca2+-​calmodulin requirement in Vitis
vinifera cells producing anthocyanins. Phytochemistry, 53,
659–​665.
Xu YQ, Zhong XY, Yin JF, Yuan HB, Tang P, Du QZ. 2013. The
impact of Ca2+ combination with organic acids on green tea
infusions. Food Chem., 139, 944–​948.
Xu YQ, Zou C, Gao Y, Chen JX, Wang F, Chen GS, Yin JF. 2017.
Effect of the type of brewing water on the chemical compos-
ition, sensory quality and antioxidant capacity of Chinese teas.
Food Chem., 236, 142–​151.
Yang FL, Cho S, Seo HS. 2016. Effects of light color on consumers’
acceptability and willingness to eat apples and bell peppers. J.
Sensory Stud., 31, 3–​11.
Yin JF, Zhang YN, Du QZ, Chen JX, Yuan HB, Xu YQ. 2014. Effect
of Ca2+ concentration on the tastes from the main chemicals in
green tea infusions. Food Res. Int., 62, 941–​946.
Yuan S, Tong M, Wu G. 2011. Destabilization of emulsions by nat-
ural minerals. J. Hazard. Mater., 192, 1882–​1885.
Osmosis in the Kitchen
Hervé This vo Kientza1,2
1 INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​
AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
In culinary circles, the phenomenon of osmosis is sometimes
given as an explanation for water transfers during culinary
processes, but the mechanisms given are not always correct.
As discussed here, cooking is not the sole field in which wrong
descriptions of osmosis are given, so it is probably helpful to give
some information on what osmosis really is and how it works.
Here, we consider practical cases, and we conclude by discussing
the possible mechanisms, as well as incorrect descriptions.
Osmosis and Semipermeable Membranes
In some textbooks, osmosis is explained using a U tube, with
the two halves separated by a “semipermeable membrane”, i.e.,
a membrane that is said to be crossed by water but not by other
molecules or ions, such as sodium or chloride ions (Figure 64.1).
It is often said that “water moves from the compartment with
solutes and water toward the other compartment” (Goubet, 2017)
until an equilibrium is reached.
One early observer of osmosis in this kind of circumstances
was the French clergyman and physicist Jean-​Antoine Nollet
(1700–​ 1770): in 1748, using an animal bladder to separate
chambers containing water and wine, he noted that the volume
FIGURE 64.1 In this classical description from elementary textbooks, it is
said that water moves from the side where the solute concentration is lower
towards the side where the solute concentration is higher, through a “semi-
permeable membrane”.
in the wine chamber increased and, if this chamber was closed, a
pressure developed (Nollet, 1748).
However, the term “osmosis” was introduced only in 1826,
under the two forms endosmosis and exosmosis, by the French
physician and physiologist Henri Dutrochet (1776–​ 1847):
“Indeed, when the denser of the two fluids is in the cavity, water
moves into it by the action that I called endosmosis; on the con-
trary, when the denser of the two fluids is outside the cavity, the
less dense fluid inside is pushed outside by an opposite action that
I shall call exosmosis” (Dutrochet, 1826). Dutrochet derived the
term “osmosis” from the Greek ωσμος, impulse.
Later, other scientists such as the British chemist Thomas
Graham (1805–​1869) and the German plant physiologist Wilhelm
Pfeffer (1845–​1920) created more resistant artificial membranes
(Wisniak, 2013) with which they could study the phenomenon:
Graham made them from rubber, and Pfeffer made them in the
walls of an unglazed porcelain vessel by reacting copper salts
with potassium ferrocyanide to form a copper ferrocyanide pre-
cipitate membrane on the surface of the vessel. Pfeffer used this
membrane to separate a sucrose solution inside the vessel from
water outside, and he found that water flowed from the water side
to the sucrose side. He observed that the flow was proportional to
the sucrose concentration; furthermore, a pressure applied inside
the vessel produced a filtration flow proportional to the pressure.
Pfeffer defined osmotic pressure as the hydrostatic pressure
necessary to stop osmotic flow across a barrier (e.g., a membrane)
that is impermeable to the solute.
In 1887, the Dutch chemist Jacobus Henricus van’t Hoff used
the experimental results of Pfeffer and others to postulate that,
for dilute solutions, the osmotic pressure Π is related to the solute
concentration, establishing the law:
Π = ∆ c RT (64.1)
where Δc is the solute imbalance between reservoirs, c the molar
concentration of the solute (mol.m−3), T the absolute tempera-
ture (K) and R the constant of ideal gases (8.31 J.mol−1.K−1). For
example, for seawater (3%, w/​w, sodium chloride) at 25 °C, the
osmotic pressure is about 25 bars.
In the 19th century, the American physical chemist Josiah
Willard Gibbs (Gibbs, 1875) calculated the osmotic equilibrium
441
442
using the chemical potential that he had just introduced, the
chemical potential of the solution being equal to the chemical
potential of pure water plus the potential energy of the water
raised by osmosis. It is noteworthy to add that this thermo-
dynamic treatment avoided any consideration of mechanisms; in
particular, it needed no assumption about the “diaphragm”, i.e.,
the semipermeable membrane.
Of course, the scientific study of osmosis did not stop with this
result: Marbach and Bocquet (2019) discussed recent results, in
particular using molecular dynamics, for various cases.
Where Are These Semipermeable Membranes?
When does osmosis takes place in the kitchen? Not always where
it is said to be! For example, it is often proposed that salt (more
or less pure sodium chloride) would draw the juices from grilled
meat by osmosis, but is it true? And more precisely, how much of
the mass decrease of the muscular tissue is really due to osmosis?
Experiments to study this question (see the chapter in this
volume on salt and grilled meat) showed that the quantitative
contribution to osmosis is probably much less than the effect
of the disorganization of the collagenic tissue through heating,
triggering a contraction of meat fibres, and juices being excluded.
Osmosis is also sometimes discussed for meat stock prepar-
ation, and, even more precisely, it was used as an argument for
the wrong theory of “cooking by expansion” in public French
culinary education; in stocks, water would have flowed into the
meat by osmosis, but weighing the meat (Figure 64.2) is enough
to demonstrate that, on the contrary, meat shrinks during cooking,
again because of collagenic denaturation (This, 2008).
FIGURE 64.2 In some culinary textbooks, it is claimed that meat stock
should be made by putting meat in cold water, “otherwise juices cannot escape
from the meat”, and sometimes, it is also stated that osmosis would be respon-
sible for the “expansion” of meat during the process. A simple experiment of
measuring the mass of two pieces of beef meat (Bos taurus, semitendinosus)
having the same initial mass refutes this incorrect theory: the piece of meat
put in cold water (blue crosses) lost less mass than the piece in hot water
(red crosses) until about 150 minutes, after which time the masses of the two
pieces were equal. Also, the volume of the pieces of meat was measured, and
the decrease corresponded to the mass lost. In particular, regarding the topic
of this chapter, no osmosis could be active enough against the meat shrinkage
due to the contraction of the collagenic tissue.
Hervé This vo Kientza
On the other hand, osmosis can be readily observed with the
product obtained after putting a chicken egg (with its shell) in
vinegar. First, the shell dissolves (effervescence is observed,
carbon dioxide being produced during the reaction of calcium
carbonate with acetic acid), so that an egg inside a soft membrane
is obtained (Figure 64.3).
With this system, it is easy to experiment with osmosis in the
kitchen; adding salt to the vinegar makes the egg shrink, whereas
putting the egg in pure water makes it expand (reversibly).
For these two experiments with meat or egg, the question
was primarily one of education, but osmosis can also be used
practically when making fruits in syrup (Figure 64.4). The issue
for this preparation is the shrinkage of the fruits (e.g., cherries,
mirabelles, plums, etc.) in concentrated (high sucrose concentra-
tion) syrups, and their swelling and degradation in light (low con-
centration) syrups. Indeed, here again, the skin of the fruits acts
as a semipermeable membrane separating the syrup outside and
the sweet inside. Of course, the solution is to store the fruits in a
syrup for which the osmotic pressure is zero, but how to make it?
In 1997, I proposed that fruits float at the surface of concentrated
syrups, whereas they sediment in light syrups, because of the
respective densities of the outer and the inner solution. This leads
FIGURE 64.3 When an egg is dipped in vinegar, the shell dissolves, and
the egg can be observed inside a semipermeable membrane (in particular,
the yolk can be observed, floating in the egg white, which demonstrates that,
inside the egg, the yolk is always in the upper place, whatever the orien-
tation of the egg). When the shell-​deprived egg is put in pure water, or on
the contrary, in a concentrated brine, it will expand or contract by osmosis,
respectively.
FIGURE 64.4 When fruits such as plums are in a “light syrup” (aqueous
solution with a low concentration of sucrose), they sediment; in contrast, they
float when the concentration of sucrose is greater than inside the fruits (the
effect of the kernel is weak). Storing the fruits in a syrup of the same osmotic
force as the fruits avoids both shrinkage and expansion of the fruits.
Osmosis in the Kitchen
to an obvious proposal: put the fruits in a concentrated syrup and
add water until the fruits gently begin to descend.
Other Culinary Processes with Osmosis
Osmosis occurs along with other phenomena in many culinary
processes. For example, storing zucchinis in salt is sometimes
advised, for various reasons, but mainly for avoiding bitterness:
for 30 minutes according to Montagné (1996); 1 hour according
to Verboom (1880) and Comolly (2003), the latter adding that
one-​third of the weight is lost; for more than 2 hours according to
Raymond (1887); for 24 hours according to Bonnechère (1904).
These differences in culinary precisions are important, but what
are the results?
We tested this “culinary precision” before extending the
experiment to other fruits and vegetables. First, assuming that
the soaking would take place primarily by the cut surfaces of
the plant tissue, slices of various thicknesses were prepared
(this assumption was tested later), soaked and weighed regularly
(Figure 64.5).
It can be observed that the mass decreases first rapidly, for
about 75 minutes, before stabilization. If the goal is to soak as
rapidly as possible, 1 hour and a quarter is a better choice than
1 hour.
Of course, one could ask whether the size of salt crystals is
important, but experiments with 1-​cm thick slices covered with
either fine or coarse culinary salt did not show any difference
after 50 minutes of soaking (mass measurements were performed
every 5 minutes). However, the maximum loss was reached 10
minutes earlier for fine salt.
Now, because we wanted to know whether the stabilization at
about 75 minutes was the limit, we studied this specifically by
burying slices in salt for many days. At the same time, we dipped
FIGURE 64.5 The mass of slices of zucchinis dipped in salt. Here, slices of
different thicknesses (i.e., masses) were studied. The size of the dots corres-
ponds to the standard deviation for three slices of the same mass.
443
slices of the same fruit in sugar, in vinegar or in flour. It appeared
that soaking is faster and more efficient with sugar (60% in one
week) than with salt (37–​43% depending on the thickness). With
vinegar (8% acetic acid), the loss was slightly lower than with
salt, and with flour, it was almost nil.
In some other culinary books, the reason for soaking zucchinis
in salt is not bitterness but oil absorption during frying (Mordelet,
1994). Soaking, here, is advised to last for 1 hour, but this time
was challenged again, the main result of these experiments
showing that soaking for 2 hours produced over-​salted slices.
Coming back to muscular tissues, which are also made of cells,
what about the same experiment as for zucchinis? Here, one has
to take into consideration the anisotropy of the tissue, and results
also depend on the collagenic content. In Figure 64.6, we show an
example of the results for the muscular tissue obliquus internus,
for which the intact fibres were parallel to the surface of the piece.
Mechanisms
Osmosis is not fully understood, and the mechanisms are still
being discussed, because various cases have to be considered.
Of course, these mechanisms depend on the particular solutions
(ionic or not) and membranes, as has been apparent since the
beginning of the scientific studies of osmosis, but other conditions
can also be considered, such as the application of electric fields,
for example.
There are many misconceptions about the mechanisms of
osmosis, as discussed by Kil (1982), Stein (1986), Borg (2003),
Kramer and Myers (2013) and Marbach and Bocquet (2019).
Some theories are incorrect because they have a teleological
flavour, such as when it is said that the water molecules move
“in order to” dissolve the more concentrated solution: water
molecules have no intention or goal –​they just move at random.
For some other theories, the defects are more subtle.
For example, Van’t Hoff suggested a physical explanation
(Van’t Hoff, 1887; Meyer, 1890) in analogy with the kinetic
gas theory, i.e., the theory considering that gases are made of
molecules that collide according to the laws of mechanics; the
pressure exerted against the walls of the container equals the total
momentum transferred by the molecules colliding per unit time.
Van’t Hoff proposed that, in a solution, the water molecules would
move through the membrane, whereas the solute molecules would
transfer their momentum to the membrane; only the impermeable
solute molecules would account for the osmotic pressure (“bom-
bardment hypothesis”) (Kil, 1982). However, it appeared that this
theory had shortcomings, as the osmotic flow between solutions
of identical hydrostatic pressure does not deform a membrane,
whereas a similar flow induced by a hydrostatic pressure drop
across the membrane causes bulging. The osmotic flow is there-
fore generated by a flow of water through the membrane, and not
by the impact of solute molecules.
Another incorrect theory is that the volume occupied by the
solute, at the aperture of any pore of the membrane, displaces
some water molecules and thereby decreases the number of
water molecules per unit volume; as a result, fewer water
molecules arrive per unit volume at the pore from the side with
444
FIGURE 64.6 The mass decrease of beef meat (Bos taurus, obliquus internus) soaked in salt. After a “fast decrease” (30% of the mass lost in 10 days), the
mass decrease is slower when a “crust” is observed to have formed.
higher solute concentration, and this should result in a net flux
of water through the pore from lower to higher solute concen-
tration. However, this was refuted by Paganelli and Stein, 1957;
Stein, 1986), who measured the diffusive entry of tritiated water
into red blood cells.
Borg (2003) discussed other incorrect theories and concluded
that any mechanical description of osmosis has to go back to the
very concept of pressure and its explanation in terms of the virial
theorem; the solute-​retardation mechanism could be used when
looking for a simple but true explanation of osmosis. Manach
and Bocquet (2020) reviewed recent developments on osmotic
theories, and it would be both useless and impossible to sum-
marize everything here; it can merely be observed that molecular
simulations today have now become an efficient tool to explore
osmosis. Direct simulations can be performed using two explicit
reservoirs with difference of solute concentration. For example,
such implementation was used by Kalra et al. (2003) in the study
of osmosis across carbon nanotubes. Also, osmosis has been
rationalized in more simple terms by simplifying the explicit
membrane description to reduce it to a confining potential acting
on the solute only (Yoshida et al., 2017).
Applications
Osmosis is widely used for various applications, but one of the
most important is seawater desalination. Here, the particular
osmosis performed is “reverse osmosis”, which relies on the
very simple principle of applying an external large hydrostatic
pressure to counterbalance the osmotic pressure difference and
induce a flow of water towards the low-​concentration side; the
pressures involved are typically in the range of 30–​50 bars in
order to exceed the osmotic pressure.
In a different approach, forward osmosis (combined with
thermal methods) makes use of draw solutions to counterbalance
Hervé This vo Kientza
the salinity-​induced osmotic pressure. Generating a high osmotic
pressure, typically above the 30 bars of pressure between sea-
water and fresh water, requires draw solutes that are highly sol-
uble in water and also of a sufficiently small size (hence, low
molecular weight) (Marbach and Bocquet, 2019).
More applications are discussed in the chapter dealing with
membrane filtration, as the fractionation of plant tissues is a pro-
cess that can become important for producing the raw ingredients
for note by note cooking (see the chapter on this in this book).
Conclusion
In the kitchen, many incorrect ideas about osmosis are trans-
mitted by textbooks, and, on the other hand, there are many
circumstances in which a liquid has to be concentrated (fond,
demi-​glace, glacé or sauces), or diluted (syrups or brines), or
modified (sauces), and the full potential of osmotic techniques is
still in its infancy. There is no doubt that the advent of membrane
processes can be an asset for the future of cooking techniques.
REFERENCES
Bonnechère C. 1904. La cuisine du siècle, Brochard, Paris.
Borg F. 2003. What Is Osmosis? Explanation and Understanding of
a Physical Phenomenon, F. G. Borg, arXiv preprint physics/​
0305011.
Comolly M. 2003. Gourmande et pressée, Hachette, Paris.
Dutrochet H. 1826. L’agent immédiat du mouvement vital dévoilé
dans sa nature et dans son mode d’action, chez les végétaux et
chez les animaux, Paris.
Gibbs JW. 1875. On the equilibrium of heterogeneous substances,
Transactions of the Connecticut Academy of Arts and Sciences,
3, 108–​248.
Goubet F. 2017. Pourquoi certains plats sont-​ ils meilleurs le
lendemain?, Le Temps, 1 September 2017, www.letemps.ch/​
lifestyle/​certains-​plats-​sontils-​meilleurs-​lendemain
Osmosis in the Kitchen
Kalra A, Garde S, Hummer G. 2003. Osmotic water transport
through carbon nanotube membranes, Proceedings of the
National Academy of Sciences of the United States of America,
100, 10175–​10180.
Kil F. 1982. Mechanisms of osmosis, Kidney International, 21,
303–​308.
Kramer EM, Myers DR. 2013. Osmosis is not driven by water dilu-
tion, Trends in Plant Science, 18(4), 195–​197.
Marbach S, Bocquet L. 2019. Osmosis, from molecular insights
to large-​scale applications, Chemical Society Reviews, 48,
3102–3144.
Meyer L. 1890. Ueber das Wesen des osmotischen Druckes,
Zeitschrift für Physikalische Chemie, 5, 23–​27.
Montagné P. 1996. Larousse gastronomique, Larousse-​Bordas, Paris
(1st ed. 1938).
Mordelet A. 1994. Cuisine des palais d’Orient, Editions de l’aube,
La Tour d’Aigues, 134.
Nollet JA. 1748. Recherches sur les causes du bouillonnement des
liquides, Mémoires de Mathématique et de Physique, tirés des
Registres de l’Académie Royale des Sciences de l’année 1748,
57–​104.
445
Paganelli CV, Solomon AK. 1957. The rate of exchange of tritiated
water across the human red cell membrane, Journal of General
Physiology, 41, 259–​277.
Raymond E. 1887. Le nouveau livre de cuisine, Firmin-​Didot, Paris.
Stein WD. 1986. Transport and Diffusion across Cell Membranes,
Academic Press, 110–​111.
This H. 2008. Molecular Gastronomy, Exploring the Science of
Flavor, Columbia University Press, New York.
Van’t Hoff JH. 1887. Dir Rolle des osmotischen Druckes in der
Analogie zwischen Lösungen un Gasen, Zeitschrift fur
Physikalische Chemie, 1, 481–​507.
Verboom A. 1880. La table, guide complet de la maitresse de maison,
Moniteur des dames et des demoiselles, Paris-​Bruxelles.
Wisniak J. 2013. Thomas Graham, I. Contributions to thermodynamics,
chemistry and occlusion of gases, Educación Química, 24(3),
316–​325, DOI: 10.1016/​S0187-​893X(13)72481–​9
Yoshida H, Marbach S, Bocquet L. 2017. Osmotic and diffusion-​
osmotic flow generation at high solute concentration. II.
Molecular dynamics simulations. Journal of Chemical Physics,
146, 194702.
Pasta: Durum Wheat Proteins –​a Key Macronutrient for Pasta
Qualities
Coline Martin1, Marie-​Hélène Morel1 and Bernard Cuq2
1 INRAE, UMR IATE, 2 place Viala, 34060 Montpellier cedex, France
2 L’Institut Agro – Montpellier SupAgro, UMR IATE, 2 place Viala 34060 Montpellier cedex, France
Pasta Quality Is Driven By Both Raw Material and
Process
Durum wheat (Triticum durum) pasta is a staple food in the diet
of many countries. Its colour and firmness, its smooth texture,
and its convenient use make it really popular. Pasta is made from Pasta-​Making Process
a blend of only durum wheat semolina and water. Pasta quality
in each step of the pasta-​making process is mainly driven by the
control of structuration mechanisms and drying temperature,
which strongly impact protein transformations and the visco-
elastic texture of the cooked pasta. Durum wheat proteins (10–​
14%) are strongly modified during the pasta-​making process
and are involved in the texture and visual aspects of pasta. After
giving an overview of durum wheat composition and the pasta-​
making process, this chapter aims to explain the contribution of
durum wheat proteins to pasta quality.
Durum Wheat Components
Durum wheat semolina is obtained by milling durum wheat
grain for the purpose of recovering semolina particles (200 to
300 µm in size) free from outer grain layers. Semolina derives
from the central part of the grain and includes starch (80%),
proteins (10–​14%) and smaller quantities of lipids, non-​starch
polysaccharides and minerals. Starch occurs as semi-​crystalline
granules composed of assemblies of linear and branched
polymers of glucose, referred to as amylose and amylopectin,
respectively. Most of the semolina protein (85%) coincides with
grain storage proteins, also called gluten. The gluten proteins
are insoluble in water and, once hydrated, form a cohesive and
viscoelastic mass stabilized by numerous weak interactions
(hydrogen bonds, hydrophobic and van der Waals interactions).
Gluten includes two types of proteins: gliadins and glutenins.
Gliadins are a group of more than 20 different polypeptides ran-
ging from 20,000 to 75,000 g/​mol, sharing the same solubility in
ethanol/​water solvent (70%, v/​v) and having no free thiol groups.
Glutenins are in the form of disulfide-​linked polypeptides ran-
ging from 100,000 to millions of grams per mole. While gliadin
contributes to gluten viscosity, glutenins provide mechanical
resistance and elasticity. The texture of cooked pasta relies on
the viscoelastic properties of gluten.
The pasta-​making process can be described in four major steps
(Figure 65.1).
(i) The first step consists in hydrating and mixing semo-
lina with a sufficient amount of water (32% w/​w) to
ensure protein hydration, but not enough to allow
dough development. Low-​ speed mixing is used to
seamlessly distribute water on semolina particles and
is continued until each particle core becomes hydrated
by means of water diffusion. The diffusion time
depends on particle diameter and mixing temperature.
In modern industrial processes, semolina particles
and water droplets are sprayed together to force their
direct contact, thus improving the global homogeneity
of hydration.
(ii) Hydrated semolina is shaped into a malleable pasta
through extrusion, which requires a high input of
mechanical energy. Structuration mechanisms are
induced by mechanical stresses (shearing and com-
pression) applied to the dough conveyed between the
sheath and the extrusion screw. At the end of the screw,
the structured dough is forced through a die to give it
the final shape. Curved (e.g., short macaroni) or spiral
(e.g., torti or fusilli) pasta shapes are generated by a
differential flow speed of the dough in the die. Pasta is
cut to the desired length by a knife blade scraping the
die’s external surface.
(iii) Extruded pasta is dried under a hot air flow in a
temperature-​and humidity-​ controlled atmosphere
until it reaches a target water content of 12%. Drying
decreases water activity to 0.5 and ensures the micro-
biological and physico-​ chemical stability of pasta
throughout its shelf-​life. Drying also leads to a large
447
448 Coline Martin et al.

decrease in pasta weight. High temperatures (>70 °C) cooking increases its digestibility and leads to the spe-
speed up the drying step and reinforce the gluten pro- cific firm and smooth texture of pasta.
tein network by promoting disulfide bonding between
gliadin and glutenin polymers, thus improving the tex-
ture qualities of cooked pasta (Figure 65.2).
Pasta Qualities
(iv) Dry pasta is cooked before intake by immersion in a
large volume of salted boiling water for a necessary Colour –​ The act of purchase is substantially driven by
and sufficient time span (5 to 12 min depending on dry pasta colour. Dry pasta has a light amber-​yellow
colour, which turns to a characteristic light yellow after
pasta shape). As simple as it seems, cooking is a crit-
cooking. Pasta colour can be described using three
ical step, which determines pasta’s organoleptic and
indices in the CIE Lab frame of reference: lightness L*
nutritional qualities. Cooking leads to an increase in
(55–​65); yellowness b* (20–​30); redness a* (4–​7).
water content (200–​250% wet basis) at the optimal
Surface roughness –​A smooth and shiny surface after cooking
cooking time. Starch gelatinization occurring during relates to a firm non-​sticky pasta, while a rough and
degraded surface corresponds to a soft and sticky pasta.
Stickiness –​Cooked pasta stickiness correlates well with
Durum wheat its surface aspect. A smooth and shiny surface attests
semolina Water that pasta is not too sticky. Cooked pasta sticki-
ness correlates with a weak gluten network, which is
distorted by starch granule swelling and leads to high
(i)
Mixing quantities of soluble amylose in the cooking water.
Texture –​When eating pasta, consumers are looking for
Hydrated semolina a specific firm and melt in the mouth texture. Cooked
agglomerates pasta texture qualities can be described using sensory
or instrumental methods.
Extrusion Pasta nutritional value –​Durum wheat pasta is rich in
(ii) easily digestible energy coming from the high propor-
tion of gelatinized starch. Pasta also provides significant
Fresh pasta
amounts of proteins, minerals (iron) and some vitamins
(thiamine, riboflavin and nicotinic acid). Nevertheless,
Drying as pasta is especially poor in lysine, its consumption
(iii)
alone will not provide all the essential amino acids.
Ready for sale pasta

Cooking Gluten Proteins Control Pasta Qualities

Protein Contribution to Quality throughout
(iv) Ready to eat pasta
Pasta-​Making Process
(i) Semolina particle wetting is critical for reaching an
FIGURE 65.1 The principal steps of the pasta-​making process. ideal rubbery state of gluten. Hydration lowers the

FIGURE 65.2 Schematic representation of cross-​linking reactions of durum wheat proteins.

(From Martin et al., 2019)
Pasta: Durum Wheat Proteins
glass transition temperature of gluten from more
than 100 °C to ambient temperature. Gluten proteins
are then ready to interact together, and enzymes are
activated. During extrusion, mechanical energy input
forces interactions between durum wheat components
and allows the creation of efficient interactions
between protein chains. Extrusion leads to the creation
of a sufficient number of weak interchain interactions
to form a continuous protein network around starch
granules that ensures dough cohesion and continuity.
The extrusion temperature must be lower than 50 °C
to prevent early gelatinization, which would trigger
starch granule swelling. Low extrusion temperature
also prevents early thermal reticulation of native
proteins, which would limit their ability to form a
strong gluten network during drying.
(ii) Applying high temperatures (>70 °C) at the end of
drying allows cross-​linking of the gluten protein net-
work by generation of additional interchain disul-
fide bonds between gliadin and glutenin polymers.
This reticulation mechanism participates in the
strengthening of pasta texture. Thermal reticulation
of the gluten network built during extrusion must
occur before starch gelatinization. Indeed, the elastic
resistance of the gluten network limits starch granule
swelling, which otherwise would disrupt pasta cohe-
siveness and induce excessive matter loss during
cooking. At an early stage of the drying cycle, pasta
water content is still high, and starch is highly recep-
tive to thermal changes. As a consequence, applying
high drying temperatures at the beginning of drying is
avoided to prevent starch granule gelatinization before
gluten network setting. Applying high temperatures at
the end of drying may induce a modification of pasta
colour. Excessive temperature at low water content
triggers a Maillard reaction between amino acids
and reducing sugars, increasing the red hue of pasta,
a typical feature of high-​ temperature-​dried pasta
(Lamacchia et al., 2007).
(iii) Cooking pasta in boiling water (100 °C) continues
gluten cross-​linking mechanisms, and this phenom-
enon is competitive with starch granule gelatinization.
The gluten protein network in high-​temperature-​dried
pasta is almost completely cross-​linked after drying
stage, but some new disulfide bonds can still be
created. Starch granule swelling is therefore restrained
by the strong gluten network. As the gluten network is
acting as an elastic net around starch granules, it limits
amylose chain diffusion and thereby pasta stickiness.
In contrast, the gluten network from low-​temperature-​
dried pasta is less reticulated and more deformable.
Starch granules are free to absorb water, swell and gel-
atinize, which leads to strong amylose chains leaking
out of starch granules and to sticky pasta (Shewry and
Tatham, 1997).
449
Protein and Pasta Qualities
Colour and enzymes –​Pasta colour can be altered by enzym-
atic activity of some durum wheat semolina proteins
(peroxidases, polyphenol oxidases and lipoxygenases).
Being able to control the pasta-​making process can
minimize the adverse impacts of enzymatic reactions on
pasta lightness (L*) and yellowness (b*). The semolina
hydration step is enough to increase water activity and
to activate enzymes. These enzymes remain efficient as
long as water activity and temperature conditions are
favourable. Two technological tools might be used to
minimize the impact of enzymatic reactions and pre-
serve pasta yellowness and lightness. Firstly, mixing can
be conducted under vacuum to remove oxygen, which
would otherwise be involved in oxidation reactions.
Secondly, increasing the temperature above 70 °C at the
beginning of the drying step leads to enzyme activation
(Icard-​Vernière and Feillet, 1999).
Surface aspect, texture and gluten network –​The control
of cooked pasta stickiness and texture depends on sev-
eral parameters. The quality and quantity of semolina
gluten proteins play a critical role through their ability
to embed starch granules into a continuous viscoelastic
network. The continuous gluten network is developed
during extrusion and cross-​ linked during the drying
step. During cooking, protein network cross-​ linking
continues, constraining starch granules from swelling
during gelatinization and restricting amylose chain
leakage. Cooking is the final step contributing to pasta
surface quality (Del Nobile et al., 2005).
Gluten network and glycemic index –​The slow digestion
rate of starch is associated with the low glycaemic index
of pasta. The compact texture of pasta and the water
insolubility of gluten are key factors in reducing starch
digestion rates. The low glycaemic index of pasta can
be attributed to the strong, reticulated gluten network
(Fardet et al., 1998). The continuous gluten network
causes an encapsulation of starch granules that limits
their swelling during cooking. The compact and dense
structure of pasta limits the surface area over which
digestive enzymes are able to gain access to available
starch, thereby limiting rates of digestion.
Protein digestibility –​High drying temperatures increase
the occurrence of Maillard-​ type reactions, resulting
in higher protein cross-​linking and decreasing lysine
availability, which affects both protein digestibility and
antigenicity (Colonna et al., 1990).
Conclusion
Gluten proteins act as a key macronutrient for the control of
durum wheat pasta quality. In recent decades, a varietal selec-
tion has been made in favour of durum wheat varieties containing
high levels of qualitative gluten proteins. Throughout the pasta-​
making process, gluten proteins are activated during semo-
lina hydration and structured in a continuous network during
450
extrusion. High-​ temperature treatment during drying and
cooking strengthens the gluten network through the coupling of
gliadin to glutenin through disulfide bonds. A good knowledge
of starch and protein reactivity towards hydration, mechanical or
thermal stresses allows the pasta-​making process to be designed
to achieve the optimal visual aspects and textural quality.
REFERENCES
Colonna P, Barry JL, Cloarec D, Bornet F, Gouilloud S, Galmiche JP.
1990. Enzymic susceptibility of starch from pasta, Journal of
Cereal Science, 11(1), 59–​70.
Del Nobile MA, Baianoa A, Contea A, Moccic G. 2005. Influence of
protein content on spaghetti cooking quality, Journal of Cereal
Science, 41, 347–​356.
Fardet A, Hoebler C, Baldwin PM, Bouchet B, Gallant DJ, Barry
JL. 1998. Involvement of the protein network in the in vitro
Coline Martin et al.
degradation of starch from spaghetti and lasagna: a micro-
scopic and enzymic study, Journal of Cereal Science, 27(2),
133–​145.
Icard-​Vernière C, Feillet P. 1999. Effects of mixing conditions on
pasta dough development and biochemical changes, Cereal
Chemistry, 76(4), 558–​565.
Lamacchia C, Di Luccia A, Baiano A, Gambacorta G, la Gatta B,
Pati S, La Notte E. 2007. Changes in pasta proteins induced
by drying cycles and their relationship to cooking behaviour,
Journal of Cereal Science, 46(1), 58–​63.
Martin C, Morel MH, Reau A, Cuq B. 2019. Kinetics of gluten
protein-​insolubilisation during pasta processing: decoupling
between time-​and temperature-​dependent effects, Journal of
Cereal Science, 88, 103–​109.
Shewry PR, Tatham AS. 1997. Disulphide bond in wheat gluten
proteins, Journal of Cereal Science, 4(2), 97–​106.
Pasteurization in the Kitchen
Gabriela Precup and Dan-​Cristian Vodnar
Faculty of Food Science and Technology, Institute of Life Sciences, University of Agricultural Sciences and
Veterinary Medicine Cluj-​Napoca, Calea Mă nă ştur 3–​5, 400372 Cluj-​Napoca, Romania
Introduction
Thermal processing methods used in the kitchen, like pasteur-
ization, sterilization and blanching, are of crucial importance for
food safety and the sensory quality of food (Szponar et al., 2017).
These methods have been extensively used in the food
industry to control the spoilage of food by bacteria, fungi and
other microorganisms, as well as undesirable enzymes in foods.
As a result of applying thermal processing to foods and dishes,
physical, chemical and biochemical reactions take place, leading
to changes in nutritional value and flavour. Moreover, long
heating at high temperatures can also cause quality degradation
and render the food product unwholesome (Fellows, 2009). For
instance, colour changes, development of off-​odours or off-​taste,
and loss of freshness or nutrients occur frequently during and
after thermal processing (Liaotrakoon, 2013).
The scientific understanding of the cooking processes,
including the transformation of raw products during thermal pro-
cessing, is the objective of molecular gastronomy. This “science
exploring cooking” offers scientific explanations of the chemical
and physical processes related to the preparation and consump-
tion of dishes (This, 1995; 2003; 2013).
Pasteurization is a form of thermal processing that is widely
used worldwide in order to reduce spoilage organisms and elim-
inate vegetative bacteria but not bacterial spores (Fellows, 2017).
The method is named after the French chemist and microbiolo-
gist Louis Pasteur, who accidentally discovered, during a summer
holiday in Arbois in 1864, that microorganisms could cause food
spoilage. He had conducted an experiment in order to correct the
frequent acidity of the local aged wines and discovered that it
was sufficient to heat a young wine to only about 50–​60 °C for
a short time to kill the “microbes”. He had also noticed that the
wine could subsequently be aged and still retain its quality. In
honour of Pasteur, the process became known as “pasteurization”
(Wilbey et al., 2007).
Later on, the practice of pasteurization became popular in
order to improve the quality and shelf-life of foods and ensure
consumer safety. It was originally used as a way of preventing
wine and beer from souring (Carlisle, 2004). Today, pasteur-
ization has been widely accepted as an effective preservation
method for reducing the unwanted vegetative cells of patho-
genic and spoilage microorganisms in foods with minimal loss
of desired food quality. The goals of pasteurization, besides food
decontamination, are also to extend the food shelf-life, promote
food safety, and allow the reduction and elimination of chemical
preservatives added to foods. In order to satisfy these goals, new
techniques have been developed in recent years. The appearance
of these emerging techniques also requires a broadening of the
definition of pasteurization. However, there is no universally
accepted definition for pasteurization. In the United States,
the National Advisory Committee on Microbiological Criteria
for Foods (NACMCF) defined pasteurization as any process,
treatment or combination thereof that is applied to food to reduce
the most resistant microorganism(s) of public health significance
to a level that is not likely to present a public health risk under
normal conditions of distribution and storage.
A great variety of techniques are applied to different foods for
pasteurization, including thermal (steam and hot water heating,
ohmic heating, microwave heating, infrared processing, etc.) and
non-thermal techniques (high-​ pressure processing, ultraviolet
radiation, irradiation, pulsed electric field, chemical treatments,
ultrasound, filtration, high-​voltage arc discharge, etc.). Many of
these techniques (e.g., ohmic heating and pulsed electric fields)
can provide higher product quality and better productivity and
have less impact on the environment (lower energy consumption,
water savings) (Masanet, 2008; Pereira, 2010).
Today, pasteurization as a thermal treatment is applied to
many foods, including milk, meat and fish products (e.g., cured
cooked ham, hot smoked fish), liquid eggs, crab, fruit juices, beer,
wine, dairy products (e.g., milk for making cheese), some sauces,
pickles, sauerkraut and food ingredients. The elimination of
pathogenic bacteria in foods consumed directly is of paramount
importance, while in food products that do not present public
health threats, control of spoilage bacteria is the main goal (Silva
et al., 2014).
Several types of pasteurization are used in the food industry,
including low-​temperature, short-​time (LTST), high-​temperature,
short-​time (HTST), higher-​heat, shorter-​time (HHST) and steam
pasteurization. The higher the heat, the shorter the time needed
to pasteurize a food. Depending on the type of food matrix,
451
452
the parameters used are different, but mild heat treatments are
preferably used.
Pasteurization Processes Used in the
Food Industry
Low-​Temperature, Short-​Time Pasteurization
This method depends on a mechanism in which low heat and low
pressure are utilized to pasteurize foods such as dairy products.
The parameters used might be, for example, at 60–​65 °C for
20–​40 min. Myer et al. (2016) demonstrated the effectiveness of
low-​temperature short-​time pasteurization (LTST) to extend the
shelf-​life of milk, as well as keeping equal or better organoleptic
characteristics without increased energy consumption and with a
significant reduction in microbial load.
High-​Temperature, Short-​Time Pasteurization
High-​temperature short-​time (HTST) processing, or flash pas-
teurization, is a method of heat pasteurization of perishable
beverages like fruit and vegetable juices, milk, beer and wine.
Flash pasteurization is performed to kill spoilage microorganisms
prior to filling containers, in order to make the products safer and
to extend their shelf-​life compared with the unpasteurized food-
stuff (Giribaldi et al., 2016).
HTST is extensively used in the milk industry. It involves
heating milk rapidly to 72 °C, keeping it there for a few seconds
(usually 15 s) and cooling it down immediately. Compared with
other pasteurization processes, it maintains colour and flavour
better, preserving the functionality of its biologically active
components (Browne and Candy, 2001). Modern pasteurizers
have many advantages: they are capable of both small and large
scales of operation, they are energy efficient, and the process is
continuous. Pasteurization should ideally be used in conjunc-
tion with sterile fill technology (similar to aseptic processing)
to prevent post-​pasteurization contamination and ensure product
quality (Browne and Candy, 2001).
A pasteurized product has a relatively short expiration date,
because only the pathogenic microorganisms are eliminated in
order to preserve the organoleptic characteristics of the food-
stuff. It is important to note that pasteurized foods can become
contaminated, and therefore hygienic practices like handling
food properly with clean hands and keeping it at a safe tempera-
ture are crucial.
Steam Pasteurization
This technique uses heat to control or reduce harmful
microorganisms in beef. This system passes freshly slaughtered
beef carcasses that are already inspected, washed and trimmed
through a chamber that exposes the beef to pressurized steam for
approximately 6 to 8 s. The steam raises the surface temperature
of the carcasses to 88–93 °C. The carcasses are then cooled with
a cold-​water spray. This process has proven to be successful in
reducing pathogenic bacteria, such as Escherichia coli O157:H7,
Salmonella and Listeria, without the use of any chemicals. Steam
pasteurization is used on nearly 50% of U.S. beef.
Gabriela Precup, Dan-Cristian Vodnar
Sous-​Vide Cooking
“Sous-​vide cooking” or “under vacuum” is a method of cooking
food in vacuum-​ sealed plastic bags at precisely controlled
temperatures; bags are placed in a water bath or steam environ-
ment for longer than normal cooking times (usually 1 to 7 h, up
to 48 h or more in some cases). The technique is also defined as a
“low-​temperature-​long-​time” cook, or the “cook-​in-​bag” system.
In sous-​vide cooking, typical temperatures around 50–​85 °C are
used, which is why it requires longer heating times compared
with conventional cooking methods.
From a culinary viewpoint, this method is used to obtain a
uniformly cooked product, ensuring that the inside is properly
cooked without overcooking the outside, and to retain moisture.
In addition, it contributes to the development of desired organo-
leptic flavours, creating a consistent and appealing texture, and
limits off-​flavours due to oxidation (Baldwin, 2012).
Sous-​vide cooking has been extensively used and studied since
the 1990s, but only in the early 2010s did this method begin to be
utilized in restaurants and homes. American and French engineers
developed under-​pressure cooking as a solution for food preser-
vation in the mid-​1960s. They discovered that the food texture
and flavour improved when pressure was applied through vacuum
sealing. A French chef, Georges Pralus, adopted the method in
his kitchen and noticed, for example, that foie gras maintained
its appearance and presented a better texture when cooked
under pressure. He developed sous-​vide cooking and cooked at
high temperatures. Another pioneer was Bruno Goussault, who
developed the parameters of cooking times and temperatures. In
1974, he presented a study on sous-​vide cooking of beef shoulder,
which highlighted that using this cooking method would prolong
the shelf-​life to 60 days. Since then, sous-​vide has remained a
standard method for cooking beef and other meats in restaurants
worldwide.
Compared with other cooking methods, like bain-​ marie,
roasting and grilling, cooking under vacuum has many benefits.
It allows heat to be efficiently transferred from the water (or
steam) to the food; it increases the food’s shelf-​life by elimin-
ating the risk of cross-​contamination during storage; it inhibits
off-​flavours from oxidation; and it prevents evaporative loss of
odorant volatiles and moisture during cooking (Baldwin, 2012).
Sous-​vide cooking applied to vegetables requires temperatures
closer to 100 °C, whereas for meat the temperatures could be
reduced to 65–​70 °C. During cooking at a temperature between
82 and 85 °C, the structure of vegetables dissolves, which results
in softening, but the cells walls remain almost intact (Sila et al.,
2006). Compared with traditional cooking, sous-​vide cooking
of vegetables also reduces the degradation and oxidation of
pigments, such as chlorophylls and carotenoids (Chiavaro et al.,
2012). In the case of meat, sous-​vide cooking is applied at 58–​
63 °C for 10–​48 h (beef, pork or lamb), resulting in intense col-
lagen solubilization, which in turn leads to great formation of
gelatine, making the meat tender.
Moreover, overcooking can be prevented, as the temperature of
the bath is already set and cannot get higher. Therefore, very pre-
cise control of cooking and temperature can be achieved (Stringer
and Metris, 2017). In order to ensure product safety after mild
heat treatment, effective refrigerated storage (time, temperature)
Pasteurization 453

is subsequently necessary, as some bacteria could survive the heat physical, chemical, microbiological and nutritional changes of
treatment and grow at refrigeration temperatures, contaminating the final food products.
the food products (Baldwin, 2012).
Meat
During processing of meat products, heat is often applied in
Pathogens of Concern for Thermal Pasteurization different ways to enhance their flavour, to extend their shelf-​life
As already mentioned, pasteurization has the objective of inacti- and to improve their hygienic quality by the inactivation of patho-
vating pathogenic vegetative microorganisms of public health genic microorganisms. Traditional methods for cooking meat
significance. A mild heat treatment (70–​100 °C) inactivates vege- products involve heating the product by immersion in hot water
tative cells and many enzymes while preserving the nutritional or by steam cooking. In such cooking processes, heat is predom-
quality of heated vegetables (Aamir et al., 2013). inantly transferred by convection from the cooking media to the
Recently, Listeria monocytogenes has become the target micro- product surface and then by conduction from the surface to the
organism for pasteurization of meats, seafood and vegetables, geometrical centre of the products. This could lead to overheating
instead of Mycobacterium tuberculosis, which was the most of the surfaces while waiting for the interior to reach the required
feared in the past. Pasteurization temperatures are not sufficient temperature. In contrast, low-​temperature cooking, between 50 and
to inactivate the spores of Clostridium botulinum; therefore, 65 °C, can generally maintain the nutritional values of the cooked
pasteurized foods, especially vegetables with a pH higher than products and is widely used in the food industry (Yu et al., 2017).
4.6 and water activity higher than 0.92, need to be refrigerated.
Protein Modifications
However, some pathogenic bacteria can grow even at refriger-
ation temperatures, resulting in a shorter shelf-​life for the food Meat is one of the richest sources of dietary protein. Since meat
products, of about one month or less (Aamir et al., 2013). contains all the essential amino acids that the body requires daily
In order to design an optimal pasteurization process and ensure to build proteins, it is also classified as a high-​quality, complete
food safety, parameters like the pH, initial microbial load and protein. The amount of protein present in one portion of meat
resistance of the target microorganism, but also the structure and varies by type and cut, and the content of early postmortem
nature of the food product, are important. Therefore, several com- muscle is about 19 g/​100 g wet weight (Lawrie and Ledward,
binations of time and temperature are selected for microbial con- 2006). Proteins are responsible for the texture-​forming proper-
trol (Aamir et al., 2013). ties of muscle foods, including water-​holding, fibre-​swelling,
According to the European Chilled Food Federation, the extractability, gelation, emulsification and binding. The cooking
target for pasteurized and chilled foods is to have a 6 log reduc- processes can introduce a wide range of structural changes of its
tion of the most heat-​ resistant vegetative pathogen Listeria proteins, as meat is composed of approximately 75% water, 20%
monocytogenes (this treatment will control other vegetative protein, and 5% fat and other substances. The effects of pasteur-
pathogens) or Clostridium botulinum (this treatment will not ization temperature and time on different physico-​chemical and
control other spore-​forming pathogens such as Bacillus cereus) microbiological parameters of meat have been studied by many
(ECFF, 2006). Therefore, a mild pasteurization at 70 °C for researchers.
2.0 min for low-​acid food would be suitable for a shelf-​life Thermal processing causes the oxidation of lipids and proteins,
of maximally 10 days at 5 °C, with a 6 log cycle reduction in which could finally result in a decrease of meat quality and
numbers of L. monocytogenes. In addition, a severe pasteurization acceptability (Roldán et al., 2015). When meat is cooked, oxida-
process at 90 °C for 10.0 min, aiming at a 6D inactivation (heat tive changes continue during refrigerated storage, as free radicals
treatments proposed to eliminate 106 non-proteolytic spores) of from lipid oxidation may induce further oxidation of fatty acids
non-proteolytic C. botulinum, allows a product shelf-​life of up to and proteins (Christensen et al., 2013).
six weeks at 5 °C (ECFF, 2006). Heat will denature the proteins in meat by changing their native
For sous-​vide cooking, the main pathogens of interest are conformation and providing the polypeptides with kinetic energy
Salmonella species and pathogenic strains of Escherichia coli. and thus breaking the weak intramolecular forces (such as non-
There are, of course, many other food pathogens, but these two polar interactions, various kinds of electrostatic interactions, and
species are relatively heat resistant and require very few vegeta- disulfide bonds) that hold the proteins together. Temperature will
tive bacteria per gram to make an immunocompromised person influence which type of proteins will be denatured and to what
sick (Baldwin, 2012). Foods treated with moist heat (steam or hot extent. As the temperature increases, a protein starts to unfold.
water immersion) reduce populations of surface organisms that When almost all the tertiary and secondary structures are lost,
may be responsible for spoilage or cause illness and can also lead the unfolded proteins may aggregate, have their disulfide bonds
to enhanced refrigerated shelf-​life (Aamir et al., 2013). shuffled (be oxidized with non-​native disulfide bonds), undergo
side-​chain modifications, and cross-​link with other polypeptides.
Thus, heat-​denatured proteins whose hydrophobic groups have
turned outward into the surrounding water, in order to adopt a
Physical-​Chemical Modifications of Foods during lower energy state, can interact through non-polar interactions
Pasteurization (Yu et al., 2017).
The application of different temperatures in the range of 60–​ During heating, the myofibrillar proteins (mostly myosin
100 °C to different foods, for specific periods of time, results in and actin) and the connective tissue proteins (mostly collagen)
454
contract, while the sarcoplasmic proteins (soluble proteins,
enzymes and myoglobin) expand. Specifically, the muscle fibres
begin to shrink transversely at 35–​40 °C, and shrinkage increases
with temperature up to 80 °C. Above 60–​65 °C, the muscle fibres
shrink longitudinally and cause substantial water loss, and the
extent of this contraction increases with temperature. The aggre-
gation and gelation of sarcoplasmic proteins begins around 40 °C
and finishes around 60 °C. Connective tissues start shrinking
around 60 °C but contract more intensely above 65 °C (Yu
et al., 2017).
Changes in meat tenderness and juiciness during cooking
strongly depend upon the cooking temperature and time applied;
these factors determine collagen solubilization and shrinkage
of myofibrillar proteins (This, 2006). The tenderness of meat
is increased by dissolving collagen into gelatine and reducing
inter-​fibre adhesion, which is an important feature when cooking
tough meats, which take either a long time or high temperatures
to cook. The temperature of 65 °C increases tenderness, as the
sarcoplasmic proteins aggregate and gel, and it becomes easier
to fracture the meat with the teeth. Over 65 °C and up to 80 °C,
the meat is tougher, because the elastic modulus increases and
requires higher tensile stress to extend fractures, which means
that the meat converts from a viscoelastic to a more or less elastic
material (Yu et al., 2017).
Prolonged cooking has been shown to double the tenderness
of the meat by dissolving all the collagen into gelatine and redu-
cing inter-​fibre adhesion to almost nothing. Cooking beef mus-
cular tissues for 24 h at 55 °C and 60 °C increases its tenderness
due to weakening of connective tissue, inactivation of proteolytic
enzymes and decreasing myofibrillar tensile strength. Collagen
begins to dissolve into gelatine above 55 °C (Tornberg, 2005).
Furthermore, the sarcoplasmic protein enzyme collagenase
remains active below 60 °C and can significantly tenderize the
meat if held for more than 6 h (Becker et al., 2016). In the case
of tough cuts of meat, like beef chuck and pork shoulder, it takes
10–​12 h at 80 °C or 1–​2 days at 55–​60 °C to become fork-​tender.
Intermediate cuts of meat, like beef sirloin, only need 6–​8 h at 55–​
60 °C to become fork-​tender, because the tenderization from the
enzyme collagenase is sufficient (Becker et al., 2016). Chicken
and turkey breasts are moist, plump and juicy when pasteurized
between 58 and 63 °C and duck breasts are usually pasteurized
at 57 °C for different periods of time in order to kill pathogenic
bacteria (Becker et al., 2016).
Lipid Modification
Lipids play an important role in food product quality, making
them more desirable by improving the organoleptic properties of
flavour, colour and texture. In addition, they confer nutritive value
on the product, constituting a source of metabolic energy, essen-
tial fatty acids and fat-​soluble vitamins. On the other hand, the
lipid components are susceptible to attack by molecular oxygen,
resulting in lipid oxidation, with the generation of cholesterol
oxides and alteration of fatty acids.
Oxidation of lipids, also called “autoxidation”, is one of the
most important factors in the non-​microbial degradation of meat
(see also the chapter on this question in this handbook). Lipid
oxidation has been extensively investigated in meat because
Gabriela Precup, Dan-Cristian Vodnar
some products of the reaction can readily react with proteins,
leading to organoleptic modification and the loss of nutritional
value. Lipid oxidation occurs in three phases as a result of many
radical chain reactions (initiation, propagation and completion).
During the first two phases, compounds like conjugated dienes
and hydroperoxides are produced, which further decompose into
carbonyl compounds, ketones, alcohols and aldehydes (Broncano
et al., 2009; Filgueras et al., 2010).
During cooking, lipid oxidation in meat is important for the
formation of taste and odour compounds, but at the same time
it produces undesirable modifications of consistency, rancidity
level, production of toxic compounds or nutritional losses. The
major prooxidants in meat lipid oxidation are considered to
be the ferric haem pigments. Pigment and lipid oxidation are
interrelated, and ferric haems are believed to promote lipid oxi-
dation. The resulting oxidation destroys the haems. Non-​haem
iron and ascorbic acid may also function as prooxidants in meat.
Sodium chloride accelerates oxidation of the triglycerides,
although the mechanism of salt catalysis is not completely
known. Cooked meat undergoes rapid deterioration due to tissue
lipid oxidation. The meat pigment in the cured pink ferrous
form does not promote the rapid oxidation undergone by cooked
uncured meat.
However, in order to control the lipid oxidation in meat, it is
of utmost importance to use the right parameters for tempera-
ture and time of cooking. Lower cooking temperature could
reduce energy consumption, but a final internal temperature of
65–​85 °C must be reached to ensure safety. It is known that high
temperatures and long heating times produce oxidative changes
in food, which could be highly negative for quality.
The most common method to measure lipid oxidation in
meat is the thiobarbituric acid (TBA) test, which determines
malondialdehyde (MDA) content; MDA is in many instances
the most abundant individual aldehyde that results from lipid
peroxidation in foods, and its concentration in meat and fish
products can reach 300 μM or more (Reitznerová et al., 2017).
The flavour of cooked meat comes from many chemical
reactions, among them the glycation reactions, which occur
intensely around 130 °C and provide savoury, roast and boiled
flavours. Cooked flavour also arises from the thermal and oxida-
tive degradation of lipids (Belitz et al., 2004).
Fruits and Vegetables
Fruits and vegetables are a primary source of macronutrients,
including fibre and saccharides, micronutrient vitamins (folate,
vitamins A and C), minerals (potassium), phenolics, carotenoids
and glucosinolates. A range of phytochemicals have high antioxi-
dant capacity and are crucial for preventing nutritional deficien-
cies and oxidative stress and reducing the risk for various types
of cancer, heart disease, diabetes, diverticulosis, stroke, hyper-
tension, birth defects, cataracts and obesity (Barba et al., 2017).
However, fruits and vegetables are highly perishable and need
appropriate preservation techniques to prolong shelf-​life while
maintaining nutritional and sensory qualities. Thermal pro-
cessing technologies have been commonly used in order to meet
these consumer needs.
Pasteurization
During the thermal pasteurization of fruits and vegetables, one
of the goals is to maintain a fresh-​like quality besides achieving
microbial inactivation for safety reasons. The endogenous
enzymes from fruits and vegetables are inactivated during pas-
teurization, as they could have a negative effect during pro-
cessing and storage. While fruits and vegetables are a rich source
of vitamins and minerals, boiled or steamed vegetables lose
nutrients to their cooking water.
Several articles have reported the effects of thermal pasteur-
ization on the quality of fruits and vegetables, including colour,
texture, carotenoids, phenolics, antioxidant activity, vitamins
and other nutritional attributes (Barba et al., 2012). The pasteur-
ization of whole fruits and vegetables affects the matrix and its
structure; thus, it has an influence on the bioaccessibility of the
bioactive compounds therein. It helps release the compounds
from the food matrix by hydrolysing the pectins in cell walls,
making bioactive compounds more accessible to absorption
(Barba et al., 2017). Recent research has now established that
food processing operations have positive effects that improve the
quality and health benefits of foods (Augustin et al., 2016).
Saccharide Modification during Pasteurization
During cooking, the cell walls of fruits and vegetables undergo
textural changes and become softer due to chemical and struc-
tural transformations of some of their saccharides (mono-​and
disaccharides). Plant cell walls are complex polysaccharide
matrices with diverse structural and physiological roles; the
saccharides or dietary fibres found in fruit and vegetable cell
walls comprise pectins, cellulose and hemicelluloses (Peng
et al., 2017).
Pasteurization of fruits and vegetables causes the hydrolysis
of pectins, gelatinization of starches, solubilization of
hemicelluloses, and loss of cell turgor. Increased temperatures
during pasteurization lead to the breakage of weak bonds
between polysaccharide chains. Also, glycosidic linkages in the
dietary fibre polysaccharides may be broken. These changes
are important from analytical, functional and nutritional points
of view.
A decreased association between fibre molecules and/​ or
a depolymerization of the fibre results in solubilization. If the
depolymerization is extensive, alcohol-​soluble fragments can be
formed, resulting in a decreased content of dietary fibre. Moderate
depolymerization and/​ or decreased association between fibre
molecules may have only a minor influence on the dietary fibre
content, but functional (e.g., viscosity and hydration) and physio-
logical properties of the fibre will be changed. Other reactions
during pasteurization that may affect the dietary fibre content and
its properties are leakage into the processing water, formation of
glycation products, thus adding to the lignin content, and forma-
tion of resistant starch fractions. Structural alterations in the cell
wall architecture are also important to follow during pasteuriza-
tion, as these are highly correlated with sensory and nutritional
characteristics (Houška and da Silva, 2017).
Colour and Texture
Colour plays a vital role in consumer choice and acceptance
(Pathare et al., 2013). Such visual appeal comes mainly from
455
pigments such as chlorophylls, anthocyanins and carotenoids. The
pasteurization process leads to changes in food colour, influenced
by various mechanisms, such as degradation of pigments, oxi-
dation of ascorbic acid, enzymatic browning and non-​enzymatic
browning (glycation reactions) (Ling et al., 2015).
Colour degradation in fruits and vegetables by thermal pas-
teurization depends mainly on the heat intensity, duration,
media, compounds responsible for colour, and storage time. For
example, Koskiniemi et al. (2013) pasteurized three vegetables
(broccoli, red bell pepper and sweet potato) and found that the
green colour of broccoli florets changed the most, while the sweet
potato colour was stable over the course of processing.
Texture is another important characteristic for consumer
acceptance and involves a series of physical characteristics that
arise from the composition and structure of the food. The texture
of the food matrix is modified through pasteurization, and the
mechanisms that contribute to texture loss include turgor loss due
to the breakdown of cellular membranes and cell wall degradation
due to solubilization of pectic material. For example, the texture of
starchy vegetables (beans, peas and lentils) is changed at 80 °C by
the gelatinization of the starch granules in their cells, making them
more digestible. Sous-​vide cooked vegetables have been shown to
be preferred to boiled vegetables, as the cell walls remain mostly
intact and the vegetables become more tender because some of the
cementing material that holds the cells together is dissolved, while
the nutritive value is also retained (Baldwin, 2012).
Antioxidant Activity
Antioxidant capacity depends on the levels of some bioactive
compounds in foods, such as phenolics, vitamin C and lyco-
pene. Thermal processing (10 min at 80 °C) of red-​grape waste
and red-​beet waste increased total antioxidant activities when
compared with fresh samples due to the increase in levels of
betacyanins and other polyphenols (Vodnar et al., 2017). In add-
ition, the same paper highlighted that the thermally processed
samples exhibited stronger inhibitory activity against Salmonella
typhimurium when compared with fresh samples. Similarly,
processed tomato and sweet corn exhibited higher antioxidant
activities than fresh samples due to the increased release of
bound phenolic compounds in the food matrices (Dewanto et al.,
2002). However, other researchers did not find differences in the
antioxidant activity between fresh and pasteurized tomato juice,
and explained this by the formation of novel compounds, such
as products from the glycation reactions, which have antioxidant
activity (Odriozola-​Serrano et al., 2008).
Phenolic Compounds
Phenolics are important phytochemicals that act as bioactive
compounds and exert antioxidant activities. Effects of thermal
pasteurization on the total phenolics in fruits and vegetables
depend on the matrix, package and storage conditions. For
instance, canning of raspberries (100 °C, 28 min) and blue-
berries (100 °C, 22 min) increased the phenolic content and anti-
oxidant activity by 50% and 53%, respectively (Sablani et al.,
2010; Syamaladevi et al., 2012). Strawberry juices pasteurized
at 90 °C for 60 s showed a 10% increase in anthocyanin content
(Odriozola-​Serrano et al., 2008).
456
In addition, the pasteurization of grape juices resulted in an
increase of procyanidins and a decrease of catechins (Fuleki and
Ricardo-​Da-​Silva, 2003). However, a decrease in the total phen-
olic content of pasteurized pumpkin and carrot juice (85 °C for
5 min) was observed (Zhou et al., 2014), while another paper
reported no significant differences in the total phenolic content
between pasteurized and fresh tomato juice (90 °C for 30/​60 s)
(Odriozola-​Serrano et al., 2008).
Carotenoids
Carotenoids are among the predominant organic pigments and
micronutrients present in yellow, orange and dark green leafy
vegetables, and include α-​ and β-​carotenes (yellow/​orange), lyco-
pene (red/​orange), xanthophyll (yellow), lutein and zeaxanthin
(green/​yellow). Carotenoids are classified on the basis of chem-
ical structure as oxycarotenoids or xanthophylls. The primary
carotenoids required by plants for photosynthesis are β-​carotene,
violaxanthin and neoxanthin. Other carotenoids localized in fruits
and flowers include α-​carotene, β-​cryptoxanthin, zeaxanthin,
antheraxanthin, capsanthin and capsorubin (Lichtenthaler, 1987;
Lichtenthaler and Buschmann, 2001). These are important bio-
active compounds, and α-​ and β-​carotenes found in carrots and
sweet potato have antioxidant roles, being important for vision
and reducing the risk of degenerative diseases. Lycopene,
another important bioactive compound, which gives the red
colour to tomatoes, is considered a potential antioxidant and
cancer-​preventing agent.
However, carotenoids are not stable to processes like thermal
pasteurization and storage, depending on the heat intensity and
properties of the products. For example, lycopene and α-​ and β-​
carotenes may undergo isomerization, oxidation and other chem-
ical changes during thermal processing and storage due to their
highly unsaturated structure (Rodriguez-​ Amaya and Kimura,
2004). Total carotenoids found in vegetables are relatively stable
to mild pasteurization, unless oxygen or ultraviolet light is pre-
sent (Peng et al., 2017).
In general, thermal processing results in the decrease of total
carotenoid content and an increase of carotenoid bioaccessibility,
as the food matrix is changed through cell wall softening
(Benlloch-​Tinoco et al., 2015).
Bioaccessibility refers to the fraction of a nutrient that is
released from its food matrix during digestion and made access-
ible for absorption into the intestinal mucosa. However, there
is not yet a consensus on the effects of thermal processing on
carotenoid bioaccessibility, as diverse papers have reported
various effects on bioaccessibility, from no changes to significant
increase or decrease.
An increase in total carotenoid content was observed for lyco-
pene and β-​carotene from tomato juice at a temperature of 90 °C
for 30 s or 60 s due to the heat treatment conditions. However, the
concentrations of α-​ or β-​carotene from carrots did not notably
change at temperatures of mild (70 °C, 2 min) or severe pasteur-
ization (90 °C, 10 min), due to the protective role of the matrix
(Vervoort et al., 2012). The carotenoid content of vegetables
was reported to decrease after pasteurization in some studies:
for carrot juice at 100 °C for 10 min (Rayman et al., 2011) or
for red sweet pepper after pasteurization at 70 °C for 10 min
Gabriela Precup, Dan-Cristian Vodnar
(Hernández-​Carrión et al., 2014). In the case of lycopene, pas-
teurization of tomato juice at 90–​100 °C for 7 min resulted in a
1.1–​1.7% decrease in lycopene, but at a higher temperature, at
130 °C for 7 min, the loss was even greater (17.1%). The nature
and extent of lycopene degradation depend upon temperature and
time of heating (Aamir et al., 2013).
Vitamins
Fruits and vegetables are great sources of various essential
vitamins, in particular vitamin C (ascorbic acid), which is a
thermolabile vitamin. In the presence of oxygen and light, it is
rapidly degraded. High pasteurization temperatures accelerate
the degradation process of ascorbic acid, and high loss is recorded
(Plaza et al., 2006; Torregrosa et al., 2006; Elez-​Martínez and
Martín-​Belloso, 2007; Koo et al., 2008). For instance, thermal
pasteurization of gazpacho (a cold vegetable soup) at 90 °C for
1 min reduced the vitamin C level to 79.2% of its initial value
(Elez-​
Martínez and Martín-​ Belloso, 2007). The content of
vitamin C also decreased in thermally treated pumpkin (85 °C
for 5 min) by 19.2%. Other vitamins, such as vitamins E and D,
have also been reported to decrease in vegetable beverages after
pasteurization (Barba et al., 2012).
Novel Pasteurization Methods
At present, novel processing technologies such as high hydro-
static pressure (HHP), pascalization or high-​pressure processing
(HPP), high-​intensity pulsed electric fields (HIPEF), ultrasound
(US), ultraviolet light (UV), irradiation and cold plasma, among
others, are being applied or explored to process foods at low
temperatures, avoiding the negative changes induced by heat.
Microwave volumetric heating (MVH) is the newest available
pasteurization technology. It uses microwaves to heat liquids,
suspensions or semi-​ solids in a continuous flow. During the
last few years, different studies have demonstrated that these
non-thermal treatments may represent gentle food preservation
techniques capable of inactivating pathogenic microorganisms
and deteriorative microorganisms and enzymes, providing safe
and fresh-​like products with minimum changes to their nutri-
tional, physico-​chemical and sensorial properties (Houška and da
Silva 2017; Johnson et al., 2017).
REFERENCES
Aamir M, Ovissipour M, Sablani SS, Rasco B. 2013. Predicting
the quality of pasteurized vegetables using kinetic models: a
review, International Journal of Food Science, https://​doi.org/​
10.1155/​2013/​271271.
Augustin MA, Riley M, Stockmann R, Bennett L, Kahl A, Lockett
T, Osmond M, Sanguansri P, Stonehouse W, Zajac I, Cobiac L.
2016. Role of food processing in food and nutrition security,
Trends in Food Science & Technology, 56, 115–​125.
Baldwin DE. 2012. Sous-​ vide cooking: a review, International
Journal of Gastronomy and Food Science, 1(1), 15–​30.
Barba FJ, Esteve MJ, Frigola A. 2012. Impact of high-​pressure pro-
cessing on vitamin E (α-​, γ-​, and δ-​tocopherol), vitamin D
(cholecalciferol and ergocalciferol) and fatty acid profiles in
liquid foods, Journal of Agricultural and Food Chemistry, 60,
3763–​3768.
Pasteurization
Barba FJ, Mariutti LR, Bragagnolo N, Mercadante AZ, Barbosa-​
Cánovas GV, Orlien V. 2017. Bioaccessibility of bioactive
compounds from fruits and vegetables after thermal and
nonthermal processing, Trends in Food Science & Technology,
67, 195–​206.
Becker A, Boulaaba A, Pingen S. Krischek C, Klein G. 2016. Low
temperature cooking of pork meat –​physicochemical and sen-
sory aspects, Meat Science, 118, 82–​88.
Belitz HD, Grosch W, Schieberle P. 2004. Food Chemistry, Springer,
Berlin-​Heidelberg, 8–​91.
Benlloch-​ Tinoco M, Kaulmann A, Corte-​ Real J, Rodrigo D,
Martinez-​ Navarrete N, Bohn T. 2015. Chlorophylls and
carotenoids of kiwifruit puree are affected similarly or less by
microwave than by conventional heat processing and storage,
Food Chemistry, 187, 254–​262.
Broncano JM, Petrón MJ, Parra V, Timón ML. 2009. Effect of
different cooking methods on lipid oxidation and formation of
free cholesterol oxidation products (COPs) in Latissimus dorsi
muscle of Iberian pigs, Meat Science, 83(3), 431–​437.
Browne J, Candy E.2001. Excellence in Packaging of Beverages,
Binsted Group, Hook, Hampshire, U.K., 178.
Carlisle R. 2004. Scientific American Inventions and Discoveries,
John Wiley & Sons, Inc., New Jersey, 357.
Chiavaro E, Mazzeo T, Visconti A, Manzi C, Fogliano V, Pellegrini
N. (2012). Nutritional quality of sous vide cooked carrots and
Brussels sprouts, Journal of Agricultural and Food Chemistry,
60(23), 6019–​6025.
Christensen L, Ertbjerg P, Løje H, Risbo J, van den Berg FW,
Christensen M. 2013. Relationship between meat toughness
and properties of connective tissue from cows and young bulls
heat treated at low temperatures for prolonged times, Meat
Science, 93(4), 787–​795.
Dewanto V, Wu XZ, Adom KK, Liu RH. 2002. Thermal processing
enhances the nutritional value of tomatoes by increasing
total antioxidant activity, Journal of Agricultural and Food
Chemistry, 50, 3010–​3014.
Doyle MP, Erickson MC. 2006. Reducing the carriage of foodborne
pathogens in livestock and poultry, Poultry Science, 85(6),
960–​973.
Elez-​Martínez P, Martín-​Belloso O. 2007. Effects of high intensity
pulsed electric field processing conditions on vitamin C and
antioxidant capacity of orange juice and gazpacho, a cold
vegetable soup, Food Chemistry, 102, 201–​209.
European Chilled Food Federation (ECFF). 2006. Recommendations
for the production of prepacked chilled food (2nd ed), www.ecff.
net/​wp-​content/​uploads/​2018/​10/​ECFF_​Recommendations_​
2nd_​ed_​18_​12_​06.pdf
Fellows PJ. 2009. Food Processing Technology: Principles and
Practice (3rd ed.), CRC Press, Boca Raton.
Fellows PJ. 2017. Food Processing Technology Principles and
Practice. Woodhead Publishing Series in Food Science,
Technology and Nutrition, 563–​578.
Filgueras RS, Gatelier P, Aubry L, Thomas A, Bauchart D, Durand
D, Zambiazi RC, Sante-​ Lhoutellier V. 2010. Colour, lipid
and protein stability of Rhea americana meat during air-​and
vacuum-​packaged storage: influence of muscle on oxidative
processes, Meat Science, 86, 665–​673.
Fuleki T, Ricardo-​Da-​Silva JM. 2003. Effects of cultivar and pro-
cessing method on the contents of catechins and procyanidins
in grape juice, Journal of Agricultural and Food Chemistry,
51, 640–​646.
Giribaldi M, Coscia A, Peila C, Antoniazzi S, Lamberti C, Ortoffi
M, Moro GE, Bertino E, Civera T, Cavallarin L. 2016.
Pasteurization of human milk by a benchtop High-​Temperature
457
Short-​Time device, Innovative Food Science & Emerging
Technologies, 36, 228–​233.
Hernández-​Carrión M, Hernando I, Quiles A. 2014. High hydro-
static pressure treatment as an alternative to pasteurization to
maintain bioactive compound content and texture in red sweet
pepper, Innovative Food Science & Emerging Technologies,
26, 76–​85.
Houška M, da Silva FVM. 2017. High Pressure Processing of Fruit
and Vegetable Products, CRC Press, Boca Raton.
Johnson PS, Lee SJ, Altalhi ASD. 2017. Novel Food Product and
Method of Use. U.S. Patent Application 15/​507,197.
Koo KM, Kim HW, Lee DS, Lyu ES, Paik HD. 2008. Quality changes
during storage of cook-​chilled soybean sprouts, Food Science
and Biotechnology, 17, 540–​546.
Koskiniemi CB, Truong VD, McFeeters RF, Simunovic J. 2013.
Quality evaluation of packaged acidified vegetables subjected
to continuous microwave pasteurization, LWT –​Food Science
and Technology, 54(1), 157–​164.
Lawrie RA, Ledward DA. 2006. The storage and preservation of
meat. I. Temperature control. In Lawrie’s Meat Science (7th
ed.), CRC Press, Boca Raton, 192–​202.
Liaotrakoon W, Clercq N, Hoed V, Van de Walle D, Lewille B,
Dewettinck K. 2013. Impact of thermal teatment on physico-
chemical, antioxidative and rheological properties of white-​
flesh and red-​flesh dragon fruit (Hylocereus spp.) purees, Food
and Bioprocess Technology, 6(2), 416–​430.
Lichtenthaler HK. 1987. Chlorophylls and carotenoids: pigments of
photosynthetic biomembranes, Methods in Enzymology, 148,
350–​382.
Lichtenthaler HK, Buschmann C. 2001. Chlorophylls and
carotenoids: measurement and characterization by UV-​ vis
spectroscopy, Current Protocols in Food Analytical Chemistry,
F4.3.1–​F.4.3.8.
Ling B, Tang J, Kong F, Mitcham EJ, Wang S. 2015. Kinetics of food
quality changes during thermal processing: a review, Food and
Bioprocess Technology, 8(2), 343–​358.
Masanet E, Worrell E, Graus W, Galitsky C. 2008. Energy efficiency
improvement and cost saving opportunities for the fruit and
vegetable processing industry, an energy star guide for energy
and plant managers. Technical report of the Environmental
Energy Technologies Division.
Myer MR, Parker KR, Kanach AT, Zhu T, Morgan MT, Applegate
BM. 2016. The effect of a novel low temperature-​ short
time (LTST) process to extend the shelf-​life of fluid milk,
SpringerPlus, 5(1), 660.
Odriozola-​ Serrano I, Soliva-​ Fortuny R, Martin-​ Belloso O. 2008.
Phenolic acids, flavonoids, vitamin C and antioxidant capacity
of strawberry juices processed by high-​intensity pulsed electric
fields or heat treatments, European Food Research Technology,
228, 239–​248.
Pathare PB, Opara UL, Al-​Said FAJ. 2013. Colour measurement
and analysis in fresh and processed foods: a review, Food and
Bioprocess Technology, 6(1), 36–​60.
Peng J, Tang J, Barrett DM, Sablani SS, Anderson N, Powers JR.
2017. Thermal pasteurization of ready-​ to-​
eat foods and
vegetables: critical factors for process design and effects
on quality, Critical Reviews in Food Science and Nutrition,
57(14), 2970–​2995.
Pereira RN, Vicente AA. 2010. Environmental impact of novel
thermal and non-​ thermal technologies in food processing,
Food Research International, 43(7), 1936–​1943.
Plaza L, Sánchez-​Moreno C, Elez-​Martínez P, de Ancos B, Martín-​
Belloso O, Cano MP. 2006. Effect of refrigerated storage on
vitamin C and antioxidant activity of orange juice processes
458
by high-​pressure or pulsed electric fields with regard to low
pasteurization, European Food Research and Technology, 223,
487–​493.
Rayman A, Baysal T, Demirdöven A. 2011. Optimisation of
electroplasmolysis application for increased juice yield in
carrot juice production, International Journal of Food Science
& Technology, 46(4), 781–​786.
Reitznerová A, Šuleková M, Nagy J, Marcinčák S, Semjon B, Čertík
M, Klempová T. 2017. Lipid peroxidation process in meat
and meat products: a comparison study of malondialdehyde
determination between modified 2-​Thiobarbituric acid spec-
trophotometric method and reverse-​phase high-​performance
liquid chromatography, Molecules, 22(11), 1988.
Rodriguez-​Amaya DB, Kimura M. 2004. Harvest Plus Handbook
for Carotenoid Analysis (Vol. 2), International Food Policy
Research Institute (IFPRI), Washington, DC.
Roldán M, Antequera T, Hernández A, Ruiz J. 2015. Physicochemical
and microbiological changes during the refrigerated storage of
lamb loins sous-​vide cooked at different combinations of time
and temperature, Food Science and Technology International,
21(7), 512–​522.
Sablani SS, Andrews PK, Davies NM, Walters T, Saez H, Syamaladevi
RM, Mohekar PR. 2010. Effect of thermal treatments on
phytochemicals in conventionally and organically grown
berries, Journal of the Science of Food and Agriculture, 90,
769–​778.
Sila DN, Doungla E, Smout C, Van Loey A, Hendrickx M. 2006.
Pectin fraction interconversions: insight into understanding
texture evolution of thermally processed carrots, Journal of
Agricultural and Food Chemistry, 54(22), 8471–​8479.
Silva FVM, Gibbs PA, Nunez H, Almonacid S, Simpson R. 2014.
Thermal processes: pasteurization. In Batt CA, Tortorello ML
(Eds.) Encyclopedia of Food Microbiology (2nd ed.), Elsevier,
Amsterdam, 577–​595.
Soliva-​
Fortuny R, Balasa A, Knorr D, Martín-​ Belloso O. 2009.
Effects of pulsed electric fields on bioactive compounds in
foods: a review, Trends in Food Science and Technology, 20,
544–​556.
Stringer SC, Metris A. 2017. Predicting bacterial behaviour in sous
vide food, International Journal of Gastronomy and Food
Science, 13, 117–​118.
Syamaladevi RM, Andrews PK, Davies NM, Walters T, Sablani
S. 2012. Storage effects on anthocyanins, phenolics and
Gabriela Precup, Dan-Cristian Vodnar
antioxidant activity of thermally processed conventional
and organic blueberries. Journal of the Science of Food and
Agriculture, 92, 916–​924.
Szponar B, Marzec A, Skrzypek M. 2017. Modern gastronomic tech-
nologies in optimization of pro-​health food potential, Polish
Journal of Public Health, 127(3), 134–​139.
This H. 1995. La gastronomie moléculaire et physique. PhD thesis.
University Paris VI, Paris, France.
This H. 2003. La gastronomie moléculaire, Sciences des aliments,
23(2), 187–​198.
This H. 2006. Molecular Gastronomy: Exploring the Science of
Flavor, Columbia University Press, New York.
This H. 2013. Molecular gastronomy is a scientific discipline, and
note by note cuisine is the next culinary trend, Flavour, 2, 1.
Tornberg E. 2005. Effect of heat on meat proteins –​implications
on structure and quality of meat products, Meat Science, 70,
493–​508.
Torregrosa F, Esteve MJ, Frígola A, Cortés C. 2006. Ascorbic acid
stability during refrigerated storage of orange-​ carrot juice
treated by high pulsed electric field and comparison with
pasteurized juice, Journal of Food Engineering, 73, 339–​345.
Vervoort L, Van der Plancken I, Grauwet T, Verlinde P, Matser
A, Hendrickx M, Van Loey A. 2012. Thermal versus high
pressure processing of carrots: a comparative pilot-​scale study
on equivalent basis, Innovative Food Science & Emerging
Technologies, 15, 1–​13.
Vodnar DC, Călinoiu LF, Dulf EV, Ştefănescu BE, Crişan G, Socaciu
C. 2017. Identification of the bioactive compounds and anti-
oxidant, antimutagenic and antimicrobial activities of ther-
mally processed agro-​industrial waste, Food Chemistry, 231,
131–​140.
Wilbey RA. 2007. Advances in Thermal and Non-​Thermal Food
Preservation, edited by Tewari G, Juneja V, Blackwell
Publishing, John Wiley & Sons, Hoboken, New Jersey, 96–​116.
Yu TY, Morton JD, Clerens S, Dyer JM. 2017. Cooking-​induced pro-
tein modifications in meat, Comprehensive Reviews in Food
Science and Food Safety, 16(1), 141–​159.
Zhou CL, Liu W, Zhao J, Yuan C, Song Y, Chen D, Ni YY, Li QH.
2014. The effect of high hydrostatic pressure on the micro-
biological quality and physical–​ chemical characteristics of
Pumpkin (Cucurbita maxima Duch.) during refrigerated
storage, Innovative Food Science & Emerging Technologies,
21, 24–​34.
Plating: The Science of Plating
Charles Spence
Head of the Crossmodal Research Laboratory, Department of Experimental Psychology, Anna Watts Building,
University of Oxford, OX2 6GG, United Kingdom
Internet-​based testing techniques are increasingly allowing chefs
to optimize the visual presentation of their food online. This
scientific approach to plating is part of the emerging field of
gastrophysics, namely, the application of psychophysical testing
techniques to the design of enhanced food experiences. Here,
three rules of thumb concerning the preferred orientation of food
on the plate that have been identified by the research are outlined,
namely: (1) angular elements on the plate are preferred when they
point away from the diner/​viewer; (2) people typically exhibit a
preference for those elements on the plate that ascend to the right
(rather than to the left); (3) people also exhibit a preference when
linear/​rectangular food presentations are aligned along the hori-
zontal/​vertical axis, rather than when they are oriented away from
these principal axes.
Introduction
In recent decades, many chefs have started to become increasingly
interested in the way in which their dishes are presented (Deroy
et al., 2014). Until recently, chefs would tend to rely on intuition
when determining the optimum orientation in which to present
the dishes served to their guests. However, given the growing
realization that people really do eat first with their eyes (Spence
et al., 2016), together with the explosion of interest in sharing
beautifully (or surprisingly) plated dishes of food on social media
sites such as Instagram’s Art of Plating (www.instagram.com/​_​
artofplating_​/)​, it is becoming increasingly important to assess
the plating of food scientifically (Spence et al., 2014). The emer-
ging gastrophysics research in this area shows that people exhibit
clear preferences for certain orientations over others as far as
plating of food is concerned (Spence, 2017b).
Assessing Orientation Preferences for the Plating
of Food
One of the most exciting developments in the emerging field of
gastrophysics research (Spence, 2017a) has been the emergence
of online testing as a legitimate, rapid, and cheap means of
assessing people’s impressions/​ associations/​expectations with
visually presented stimuli, such as, for example, plates of food
(Woods et al., 2015). Working together with those chefs interested
in optimizing the eye appeal of their dishes, we have conducted
a number of studies in which the orientation preferences of
large numbers of individuals are assessed online. While such an
approach can be merely exploratory in nature (i.e., involving the
hypothesis-​free assessment of people’s plating preferences), there
are a number of aesthetic ‘rules’ that have emerged from decades
of research on painting that, it turns out, can be applied to the
world of plating too (Spence, 2017b). In the following sections,
three such rules of thumb that the research demonstrated help
predict the preferred orientation of plates of food are highlighted.
Pointing Angularity Away from the Diner
There is a literature out there showing that people’s brains typic-
ally exhibit a short-​lasting fear response (e.g., in the amygdala)
when angular stimuli are seen pointing towards them (Larson
et al., 2009). This has been shown to translate into a slight prefer-
ence for angular stimuli when they are seen pointing away from
the viewer (rather than towards them). We conducted an online
study with the signature dish served by chef Alberto Landgraf
(from the restaurant Epice in Sao Paulo) showing a similar prefer-
ence as far as the plating of food is concerned, too (Figure 67.1).
The dish in question consists of red onions, tapioca, sugar cane
vinegar, peanut, and fermented cream. A picture of the dish was
uploaded onto the internet and people were invited to select the
preferred orientation for the dish (being asked the question ‘If
you were to be served this dish, how would you like it plated?’).
The 1667 participants, who were paid a few cents each to take
part in the study, saw the plate rotating (to avoid any bias attrib-
utable to end anchoring should a fixed initial position have been
shown). As expected, the results highlighted a clear preference
for the three onions arranged so as to point away from the viewer
(shown by the arrow and the bulge in the line surrounding the
plate).
459
460
Ascending to the Right: Preferential Orientation for
Linear Elements
When there is a dominant linear element on the plate, our research
highlights a preference when that element ascends to the right
FIGURE 67.1 Circular data plot showing the results of a study in which
1667 participants indicated their preferred orientation for one of chef Alberto
Landgraf’s dishes. Each dot represents the preferred orientation indicated by
one of the online participants. Hence, the greater the number of dots, the more
pronounced the preference for a particular orientation. The line surrounding
the plate provides an estimate of the preferred orientation (indicated by the
bulge of the line). The food has been added to the figure and oriented by 3.20°
clockwise (the bias-​corrected, mean orientation in which the food was placed
by participants). The arrow indicates the mean angle at which participants
placed the food (beneath which is a blue shaded wedge indicating the line’s
95% confidence interval). In this case, the chef’s decision to place the dish
pointing at 12 o’clock was closely in line with the preferences demonstrated
by the group.
(With permission from Michel et al., 2015)
FIGURE 67.2 Circular data plots and purple rose diagrams of the preferred plate orientations for each dish selected by participants in Youssef et al. (2015;
Experiment 3). The results clearly highlight a preference when the linear element ascends to the right (see left image). For clarity and ease of interpretation, the
food has been added to the figure and oriented by the mean orientation in which the food was placed by participants.
(Courtesy of Youssef et al., 2015)
Charles Spence
(rather than to the left). So, for example, Youssef et al. (2015)
presented viewers with two versions of the same dish created by
chef Jozef Youssef, one more round/​centred and the other with
a distinctive linear arrangement (Figure 67.2). These two dishes
were uploaded onto the internet, and people (N = 521) were
invited to rotate each of the plates into their preferred orienta-
tion. While the results revealed no clear preferred orientation for
the round/​centred presentation of the dish, there was a clear (and
significant) preference for the ascending to the right orientation
for the linear plating arrangement, and hence, that was the orien-
tation in which the dish was served to diners at the chef’s pop-​up
dining events.
But why, one might ask, should the ascending to the right orien-
tation be preferred? According to Arnheim (1974), the bottom-​left
to top-​right diagonal appears to be ascending, while the top-​left
to bottom-​right diagonal appears to be descending. Meanwhile,
marketing research shows that product logos ascending to the
right are associated with notions of activity in the mind of the
consumer (Schlosser et al., 2016). However, when it comes to
food, it is worth noting that our brains simulate the act of eating
a plate of food even if it is seen on the internet (or on the front
of product packaging). Importantly, anything that can be done
to make it easier for the viewer to simulate the act of eating the
plate of food tends to translate into increased liking (Elder and
Krishna, 2012). Hence, the ascending to the right preference in
plating might also be explained in terms of this presentation of
the food being easier to simulate eating than when the same food
is shown ascending to the left instead.
Orientation Preference for the Horizontal/​Vertical
In the world of painting, a preference for paintings to be oriented
along the horizontal/​vertical axis has sometimes been reported.
One might ask whether the same preference would also be
observed as far as the plating of food is concerned. Evidence of
a preference for food to be oriented along the main axes comes
from another study in which one of chef Jozef Youssef’s dishes
Plating: The Science of Plating 461

diner’s willingness to pay for the food, others may wish to plate it
as creatively as possible. While there is no space to cover it here,
it is also worth noting that much the same approach to assessing
orientation preferences for food can be used when assessing the
orientation of those foods shown on product packaging (Velasco
et al., 2015).

REFERENCES
Arnheim R. 1974. Art and visual perception: A psychology of the cre-
ative eye. Berkeley, CA: University of California Press.
Deroy O, Michel C, Piqueras-​Fiszman B, Spence C. 2014. The plating
manifesto (I): From decoration to creation. Flavour, 3, 6.
Elder RS, Krishna A. 2012. The ‘visual depiction effect’ in adver-
tising: Facilitating embodied mental simulation through
product orientation. Journal of Consumer Research, 38,
988–​1003.
Larson CL, Aronoff J, Sarinopoulos IC, Zhu DC. 2009. Recognizing
FIGURE 67.3 Circular data plot highlighting the results of a study showing
threat: A simple geometric shape activates neural circuitry
the preferred orientation of one of chef Jozef Youssef’s dishes. For clarity for threat detection. Journal of Cognitive Neuroscience, 21,
and ease of interpretation, a composite of all 401 plate orientations has been 1523–​1535.
added to the figure (colour filters have been applied to the composite to make Michel C, Woods AT, Neuhäuser M, Landgraf A, Spence C. 2015.
it more legible). The original dish is shown in the bottom right of the figure. Rotating plates: Online study demonstrates the importance of
orientation in the plating of food. Food Quality & Preference,
44, 194–​202.
was uploaded onto the internet for people (N = 401) to rotate into Schlosser AE, Rikhi RR, Dagogo-​Jack SW. 2016. The ups and
their preferred orientation. In this case, the lobster shown in the downs of visual orientation: The effects of diagonal orienta-
bowl was preferred either when oriented along the horizontal axis tion on product judgment. Journal of Consumer Psychology,
or else when oriented (pointing away from the viewer) along the 26, 496–​509.
vertical axis (Figure 67.3). Such a preference may perhaps be Spence C. 2017a. Gastrophysics: The new science of eating. London,
explained in terms of a preference for balance (Spence, 2017b). UK: Viking Penguin.
Spence C. 2017b. The art and science of plating. In Levent N,
Mihalache ID (Eds.), Food and museums. London, UK:
Bloomsbury Academic, 237–​253.
Conclusions Spence C, Okajima K, Cheok AD, Petit O, Michel C. 2016. Eating
with our eyes: From visual hunger to digital satiation. Brain &
Optimizing the visual presentation of a dish is becoming more Cognition, 110, 53–​63.
important than ever before. While the decision concerning how Spence C, Piqueras-​Fiszman B, Michel C, Deroy O. 2014. Plating
to plate a dish was traditionally left up to the intuitions of the manifesto (II): The art and science of plating. Flavour, 3, 4.
chef, an emerging branch of gastrophysics research has started Velasco C, Woods AT, Spence C. 2015. Evaluating the orientation of
to provide techniques that enable viewers to select/​rate the orien- design elements in product packaging using an online orienta-
tion task. Food Quality & Preference, 46, 151–​159.
tation of a dish and hence, provide rapid, cheap feedback to the
Woods AT, Velasco C, Levitan CA, Wan X, Spence C. 2015.
chef (Spence, 2017a; 2017b). Interestingly, while the intuitions Conducting perception research over the internet: A tutorial
of the successful chef often turn out to be preferred by the popu- review. PeerJ, 3: e1058; DOI 10.7717/​peerj.1058.
lation at large, that is not always the case. While some chefs/​ Youssef J, Juravle G, Youssef L, Woods A, Spence C. 2015. On the
restaurateurs may wish to orient the plate so as to maximize the art and science of naming and plating food. Flavour, 4, 27.
Proteins and Proteases
Linda A. Luck1 and Alan L. Kelly2
1 Professor of Chemistry-​Emeritus, State University of New York at Plattsburgh, Plattsburgh, 12901 New York, United States
2 School of Food and Nutritional Sciences, University College Cork, Ireland
Introduction
Proteins are among the most versatile and active macromolecules
in our living system. They are the work horses for critical
functions in most biological processes, and are key in affording
mechanical support, generating motion, controlling the processes
in metabolism, providing defenses (as in our immune system),
managing genetic information, and acting as hormones. To
this end, proteins interact with one another and with each of
the other three biomolecular classes: nucleic acids, lipids, and
carbohydrates. Most importantly, proteins that are enzymes cata-
lyze an enormous number of highly regulated and integrated
chemical reactions in life. Proteins are built when needed and
broken down when not needed. This process continues con-
stantly in our bodies, where we maintain a “dynamic steady
state” of these biological molecules. Proteins, therefore, are key
components of our diet, providing much-​needed nutritive value.
Amino Acids
Protein are polypeptides built of amino acids, which are linked by
covalent peptide bonds, to form long unbranched polymers. The
general structure of an amino acid is depicted in Figure 68.1 and
consists of an amino group (NH2), a carboxyl group (COOH),
a hydrogen atom, and a distinct R group, all bonded to a single
carbon atom, called the alpha-​carbon.
There are 20 amino acids in living systems, and each has its
own unique chemical and structural characteristics defined by
FIGURE 68.1 General structure of an amino acid. The R group refers to the
side groups of amino acids.
the R group. The R groups, also known as the amino acid side
chains, can be classified into four main sub-​families according to
their chemical character: charged amino acids, hydrophilic amino
acids, and hydrophobic and aromatic amino acids (Figure 68.2).
Hydrophilic R groups are soluble in water, while hydrophobic
groups are not soluble in water. Charged amino acid side groups
that are acidic are aspartic acid (Asp) and glutamic acid (Glu).
These have a negative charge at neutral pH. Charged amino acids
with a basic R group (lysine (Lys), arginine (Arg), and histidine
(His)) have a positive charge at neutral pH.
Especially interesting amino acids for molecular gastronomy
are glycine (Gly), proline (Pro), and cysteine (Cys). Glycine
is the smallest amino acid, with a single H for the R group,
and is classified in a hydrophobic sub-​group; it is a prominent
amino acid in collagen and is responsible for the tight packing
of the coils of that molecule. Besides contributing to the struc-
ture of proteins, glycine is necessary for our bodies to produce
heme, which is necessary for the function of our iron-​carrying
molecules, hemoglobin, and myoglobin.
Proline exhibits a structure with a pyrrolidine ring that
incorporates the amino nitrogen. When incorporated into a poly-
peptide chain, the nitrogen of the ring lacks a hydrogen atom and
therefore cannot participate in hydrogen bonding. The ring also
reduces flexibility, which is the reason why the structural form of
collagen is a stretched-​out left-​handed helix.
Cysteine contains an ionizable sulfhydryl group and is the
most highly reactive amino acid side group. When two cysteine
residues on a protein are located near each other in an oxidizing
environment, they form a disulfide bond or disulfide bridge, also
called a cystine. The bond can be broken by a reducing agent and
extremely high heat (Figure 68.3). This amino acid contributes
to the coagulation phenomena in eggs, discussed in several other
chapters of this book, and many other food systems, including
whey proteins.
Amino acids in their own right are also important, as they
serve as precursors for a number of metabolites used in many
life processes, including hormones such as epinephrine. The food
industry has utilized a methyl ester dipeptide of aspartic acid
and phenylalanine to synthesize an artificial, non-​carbohydrate
sweetener named aspartame.
463
464 Linda A. Luck, Alan L. Kelly

FIGURE 68.2 Structures of the 20 different amino acids with their R groups shown in red. The R groups are named by their three letter codes.

OH
NH2
O
H H O H H O
H
N C C N C C Oxidation S + 2H+ + 2e-
H OH + H OH
CH2 CH2 Reduction S
H
SH SH
O
H2N
OH

FIGURE 68.3 Illustration of the cystine bond between two cysteine (Cys) R groups.

than the theoretically predicted number, because not all combin-

Protein Structure ations of amino acids have useful functions.
Primary structure: The precise sequence of the amino acid Peptide bonds are formed by a condensation reaction between
residues in the protein chain is responsible for the final three-​ the backbone carboxylic acid group and amino group of two amino
dimensional structure of that protein, which, in turn, dictates the acids, leading to the release of a water molecule (Figure 68.4). By
function of the protein. DNA contains the informational code for convention, the amino acids in a protein are specified from the
the sequence of the amino acid residues in a protein. This infor- amino end, which is designated amino acid 1, to the carboxyl
mation is transcribed into mRNA, and then the translation of this end. The backbone sequence of atoms is (-​N-​Ca-​C—​N-​Ca-​C-​).
code occurs during protein synthesis. Most natural proteins are This peptide bond is extremely stable, and degradation of this
50–​2000 amino acid residues in length. Proteins are very diverse bond, which on average has a half-​life of 10 years in water, can
in nature, and the different combinations of these 20 amino acids only be accomplished by biological catalysts called proteases.
allow trillions of different entities to be formed. However, the Proteases are inherent in our digestive system, where they act
actual number of biologically relevant proteins is much smaller on dietary proteins, reducing them to amino acids to be recycled
Proteins and Proteases 465

H H in subsequent protein synthesis. The significance of proteases in

H O H O gastronomy is discussed later in this article.
N C C + N C C It is imperative that we take in dietary protein, as our bodies
H H
OH OH can only synthesize 11 of the 20 amino acids necessary for life.
R R’
Those not included are called essential amino acids: histidine
amino acid 1 amino acid 2 (His), isoleucine (Ile), leucine (Leu), lysine (Lys), methionine
(Met), phenylalanine (Phe), threonine (Thr), tryptophan (Trp),
H
and valine (Val).
H O Secondary structure: The polypeptide chains can fold into
R’
N C C OH regular formations with stabilizing hydrogen bonds between
H N C C + HOH backbone peptide NH and CO groups of amino acids, often near
R H one another on the chain. These secondary elements are largely
O
H independent of primary amino acid sequence. The first type
of structural unit is the alpha helix, which is a coiled structure
FIGURE 68.4 Depiction of formation of peptide bonds with the release stabilized by intra-​strand hydrogen bonds between NH and CO
of a water molecule. As illustrated, the first amino acid in a protein chain is groups on the backbone four residues apart (Figure 68.5).
located on the amino terminus and the subsequent amino acids are added to Most amino acids are compatible with this secondary struc-
the carboxyl terminus in the growing protein chain.
ture except for proline (vide supra), which usually ends the
helical structure. The prominent feature of this helix is that the
R groups are on the outside of the helix. This allows the helix
faces to adopt the chemical character of the specific R groups on
that side. Helices have been known to be amphipathic, in that

FIGURE 68.5 Depiction of the alpha helix. The ribbon form of the helix is shown in A. The ball and stick form is shown with the ribbon in B. As
illustrated, the R groups of the protein are on the outside of the coiled structure. The backbone atoms including NH and CO involved in hydrogen bonding
are on the inside of the helix. The atomic form of the alpha helix in ball and stick form is shown in C. The three-dimensional alpha helix figure was
generated using amino acids 44-59 of the glucose and galactose binding protein using the protein data base file 2GBP.pdb. This figure was generated using
amino acids 444-59 Dassault Systèmes BIOVIA, Discovery Studio Modeling Environment, Release 2016, San Diego.
466
FIGURE 68.6 Beta sheet depiction. (a) Extended polypeptides with
interstrand hydrogen bonding. (b) Parallel and antiparallel strands in ribbon
form. The arrow heads are the carboxyl ends of the protein chains, and the
blunt ends are the amino ends.
FIGURE 68.7 The three-​dimensional ribbon diagram of myoglobin using
the protein data base file 1MBO.PDB (Phillips, 1980) is shown in brown. The
illustration shows the alpha helical structures and the loops connecting them.
The heme group, which binds iron and oxygen, is displayed in ball and stick
mode. This figure is produced using Discovery Studio Visualizer (Dassault
Systèmes BIOVIA, 2015).
Linda A. Luck, Alan L. Kelly
one side of the helix has charged R groups –​hydrophilic –​and
the other side of the helix has hydrophobic groups. This allows
the helix to face inwards or outwards depending on the cellular
environment.
Beta sheets are stabilized by inter-​strand hydrogen bonding
between polypeptide strands and are markedly different from
the alpha helix. The beta sheet consists of two or more strands
of polypeptides that are more fully extended rather than tightly
coiled as in the alpha helix. The beta sheet is formed by linking
two or more stands lying next to one another through hydrogen
bonds. Adjacent strands can be running in the same direction
(parallel) or the opposite direction (antiparallel) (Figure 68.6).
Antiparallel sheets are more stable, as the hydrogen bonds are in
a straight line; this is illustrated in Figure 68.6 (top).
The third type of secondary structure in proteins is beta
turns and loops. Since most proteins are globular in shape,
containing both beta sheets and alpha helices, there must be
polypeptide strands that connect the structural units. Beta turns
are common for connecting two beta strands in an antiparallel
beta sheet. The turns contain four amino acids, two of which
are hydrogen bonded. In beta loops, there are six to ten amino
acids, and they are located mainly on the surfaces of proteins.
The loops are more flexible because they are held together by
multiple weak interactions within the structures. In Figure 68.7,
loop structures are shown as thin licorice structures in the myo-
globin molecule.
Tertiary Structure
Another level of protein structure, namely tertiary structure,
refers to the spatial arrangement of all atoms in the protein –​the
three-​dimensional view of the protein, for example, myoglobin,
shown in Figure 68.7. On this level, we observe the arrangement
of the secondary structures, the interaction of the R groups, and
the role of disulfide bonds. The overall structure of any protein is
the most stable arrangement of elements that generally minimizes
solvent contact and overall free energy. The tertiary structure
is held together by a combination of four non-​covalent weak
interactions, which play a role in stabilization of the overall pro-
tein structure, and in some cases disulfide bonds and/​or metal ion
interactions. The following are descriptions of these interactions.
(1) Weak interactions between oppositely charged atoms
or groups are called ionic interactions and are a
type of electrostatic interaction. An example of this
would be the attraction of a positively charged amino
group NH3+ and a negatively charged carboxyl group
COO−. The strength of the interaction depends on
the environment of the ions and the distance between
them. These interactions are often referred to as salt
bridges.
(2) Hydrogen bonds are not just reserved for water
molecules. These bonds can form between a hydrogen
atom and an electron donor group such as oxygen
or nitrogen and sometimes sulfur. The strength of
the bond is dependent on the distance and the angle
between the atoms. Although these hydrogen bonds
are the main stabilizing factors in secondary structure,
Proteins and Proteases
they play an equally important role in tertiary and qua-
ternary structures.
(3) Hydrophobic interactions occur when non-​ polar
molecules tend to cluster together in water, not because
they are attracted to one another but because, when
they associate, they release the water clustered around
them. The water released is then less ordered and the
entropy of the system is higher. This is an entropy-​
driven association, and it forms spontaneously. In the
overall scheme of protein folding, this entropy effect
is the driving force for hydrophobic amino acids to be
buried in the center of a globular protein molecule.
(4) van der Waals interactions, the weakest of the four
discussed, depend on asymmetry in electrical charge
around atoms in the R group. Many molecules are
neither charged nor polar but can interact with each
other electrostatically due to asymmetry of the elec-
tron cloud distribution about an atom or group causing
a temporary dipole moment. At any instant, the cloud
can have transient regions of positive and negative
FIGURE 68.8 The three-​ dimensional space-​ filling (CPK) structure of
hemoglobin using the protein data base file 1A3N.PDB illustrates that the
protein has four subunits, two alpha and two beta units, shown in yellow
and green (Tame and Vallone, 2000). The heme units that binds the iron
and oxygen molecules are shown in red. This protein is only active when all
subunits are bound together. This figure is produced using Discovery Studio
Visualizer (Dassault Systèmes BIOVIA, 2015).
FIGURE 68.9 Three-​dimensional structure of collagen produced by Discovery Studio Visualizer (DSV) (Dassault Systèmes BIOVIA, 2015) using the file
1CAG.PDB in CPK mode (Bella et al., 1994). Three protein chains, A, B, and C, are colored in green, violet, and light brown. These chains are extended and
held together by weak interactions described earlier.
467
charges. These charged regions can induce a comple-
mentary asymmetry on a neighboring atom, which
in turn does the same to another atom nearby. The
overall results within the biomolecule can be additive
and provide a stabilizing force in the tertiary structure.
(5) A covalent interaction that plays a major role in ter-
tiary structure is disulfide bond formation between
two cysteine residues in the protein. This interaction
is also very important in the quaternary structure, to
be discussed later.
(6) Coordinated metal ions such as iron and zinc are
another way to stabilize tertiary structure through
cysteine and histidine residues. A common protein
using this configuration is the zinc finger protein
which binds DNA, while coordinating metals are also
found in lactoferrin.
Quaternary Structure
Proteins that consist of more than one polypeptide chain display
quaternary structure, with each individual protein chain being
called a subunit. Quaternary structure can be as simple as two
identical subunits or as complex as dozens of different subunits.
Figure 68.8 shows the subunits in the molecule hemoglobin. In
most cases, and as in hemoglobin, the subunits are held together
by the four weak forces described earlier, and in many other
cases, such as in insulin, disulfide bonds at the interface of the
subunits. The shared surfaces are chemically and physically com-
plementary to each other, providing a stabilizing environment for
their interaction. The multi-​subunit array in quaternary structured
proteins provides increased functionality to the proteins not
found in the individual units. This may be in terms of structural
properties not present in the individual units, or it may be a means
for regulation of protein function. This is evident in the four-​
subunit hemoglobin protein that uses oxygen binding to one sub-
unit to increase binding action to the other three. This cooperative
binding is observed only with quaternary structure. Multiprotein
complexes also increase the efficiency of biochemical processes
by bringing linked chemical reactions into close proximity.
Not all proteins have quaternary structure. For example, myo-
globin has tertiary structure but not quaternary structure, because
it consists of only one subunit. On the other hand, collagen and
keratin have quaternary structure, as they are both multi-​subunit
proteins. The structure of collagen is shown in Figure 68.9.
468
FIGURE 68.10 A simple illustration of the unfolding process. The globular proteins are folded under native conditions with the water molecules displayed in
stick mode, colored by element. When exposed to heat, change in pH, or different buffer or salt conditions, the proteins unfold from globular form to a straight
or almost straight chain. Under these non-​native conditions, the water molecules have access to amino acid residues in the interior of the proteins (Roy, 2020).
FIGURE 68.11 The unfolding curve of a protein during a denaturing pro-
cess. The midpoint of the curve is the Tm, which is defined as the point where
half of the proteins are unfolded.
Protein Denaturation
Since the three-​dimensional structure of proteins is held together
by weak forces, they are quite susceptible to denaturation or
unfolding. The weak forces can be disrupted by a number of
methods that interfere with the above-​mentioned non-​covalent
interactions (without peptide bond cleavage) holding the molecule
in its native form. The unfolded form is usually an outstretched
chain with all the internal amino acid residues exposed to the
solvent. When the protein loses its native form, it becomes
denatured or unfolded, and if it is impossible for it to regain the
native structure, then it is considered to be irreversibly denatured.
Protein unfolding usually results in a loss of solubility due to the
exposure of the hydrophobic core regions in the center of the pro-
tein. Exposed hydrophobic regions of one protein are likely to
associate with hydrophobic regions of other unfolded proteins,
causing aggregation and subsequent precipitation or coagulation.
Linda A. Luck, Alan L. Kelly
Denaturation can be induced by heating, changes in salt
concentrations, pH changes, mechanical treatments such as
shearing (as in whipping, kneading, or rolling), exposure to
organic solvents, or exclusion or chelation of metals needed for
structural stability. Heating causes denaturation mainly by pro-
viding heat energy, which breaks hydrogen bonds and van der
Waals interactions. pH changes will affect the charged R groups,
either adding or removing H+ ions and disrupting the necessary
ionic interactions to keep the protein in the native form. Shearing
forces disrupt alpha helices and cause loss of the tertiary struc-
ture. Organic solvents or hydrophobic substances can change the
solvation pattern of the protein. The outside of the protein is usu-
ally populated by hydrophilic groups, and if the solvent is organic
in nature, those hydrophilic groups will want to be buried inside
the protein, while the hydrophobic groups will be more stable on
the outside of the protein immersed in the hydrophobic solvent.
Basically, the organic solvent will turn the protein “inside out”.
A simple illustration of the denaturation process is shown in
Figure 68.10.
Each protein has a different chemical make-​up, and thus
different weak forces are unique to each individual protein.
An example of a denaturation curve is shown in Figure 68.11,
in which the relative amount of unfolded protein increases as
the denaturing condition increases. The conditions could be an
increase or decrease in pH or salt or an increase in temperature.
The midpoint of the curve is called the Tm, and it depends on
the protein structure and the conditions in which the experi-
ment is conducted. Thus, different proteins will unfold at
different Tms.
A practical example of this is the denaturation of an egg by
heat. The egg white will denature at a lower temperature than
the yolk, because the white part of the egg has different proteins
than the yolk. The egg white proteins have lower Tms than the
yolk proteins. In another example, whey proteins from milk
denature, unfold, and aggregate when heated, and, depending on
the specific heating applied, plus environmental conditions such
as protein concentration, pH, and ionic environment, can form
aggregates or gels.
The denaturation of an enzyme results in loss of activity because
it is not in its native form. This is very useful for destroying the
Proteins and Proteases
action of enzyme proteins with catalytic activity. The actions of
these enzymes can be halted by denaturation by heat, pH, or other
means depending on the cooking application. For example, the
protease activity of bromelain from pineapple will break down
gelatin, which is why pineapple is heated first before it is used in
Jello (jelly) preparation. Also, enzymes called phenolases cause
chemical color changes in a food, such as the browning of an
apple, but pH changes, for example by adding lemon juice, will
slow the enzymes in fruits so they do not brown.
Another application of this phenomenon is in reducing food
spoilage by bacteria. Heating, pH change, or irradiation will
denature bacterial proteins, rendering them inactive and killing
the bacterial cells.
Proteases and Proteins
As discussed, the structure of many food products and dishes is
greatly influenced by the properties of proteins, and changes on
cooking or preparing food often result from changes such as heat-​
induced denaturation or changes in structure due to changing
acidity or salt concentrations. Examples such as the gel structure
of cheese and yoghurt, transformations on cooking or marinating
meat, and the range of ways in which the texture of a raw egg
can change on cooking in different ways all represent different
aspects of the behavior of food proteins.
For this reason, it is not surprising that reactions which result
in the breakdown of proteins have significant impacts on the
culinary properties of food. The key biological process through
which such breakdown can occur is proteolysis, whereby enzymes
from a range of sources, either naturally present in the food (indi-
genous) or added to the food (exogenous), result in desirable (or,
occasionally, undesirable) changes in the food.
There are four main types of protease, depending on their
mode of catalytic action: serine, aspartic, metallo, and cysteine.
These differ in their requirements for optimal activity and so may
be more or less active in different food systems depending on
factors such as temperature, pH, ionic environment, and presence
of inhibitory substances. The activity of other enzymes is rou-
tinely controlled by manipulation of such factors (e.g., the inhib-
ition of enzymatic browning of fruit by oxidases by reducing pH
through the use of acids such as lemon juice), while the inactiva-
tion of most proteases depends on heat treatment.
For example, fresh pineapple, or pineapple waste, can be
extracted to yield a potent protease mixture called bromelain, but
in canned pineapple, which has been subjected to significant heat
treatment (retorting), the enzyme activity has been inactivated.
This becomes relevant when either fresh or tinned pineapple is
added to another product, such as gelatin-​based desserts, which the
former can rapidly soften due to the residual activity present. As
the fresh fruit ripens, increasing enzyme availability can give pine-
apple a very sharp taste and even give rise to a tingling sensation as
the enzyme actually attacks the eater’s mouth during consumption.
Proteases are also used in the preparation of many foods and
beverages, for example to reduce haze in beer and wine or to
clarify blackcurrant juice (Mamo and Assefa, 2018), and in baking
to control dough structure and firmness. In baking, proteases can
469
be targeted to hydrolyze gluten and increase dough extensibility,
improve gas-​holding capacity, and enhance swelling of starch
granules (Hamada et al., 2013; Bender and Schönlecher, 2020).
Enzymes and Cheese
To take a beneficial example of very limited proteolysis, cheese
has been made for millennia using an extract from calf stomach
(rennet), which contains an enzyme called chymosin, which very
specifically cleaves one bond within aggregated protein structures
in milk called casein micelles, destabilizing them in such a way
that they form three-​dimensional networks that transform the milk
from a liquid to a solid. This solid matrix is then dehydrated by
cutting, stirring, and cooking to expel whey, and pressed into a
curd, which is stored for periods from weeks to years to allow other
enzymes, including proteases, to slowly break down the proteins
and other milk constituents to give textures and flavors charac-
teristic of the variety in question (Uniacke-​Lowe and Fox, 2017).
Fruit Enzymes
As mentioned earlier, many fruit products are potent sources of
proteases. Bromelain extracted from pineapple (Ananas comosus
L.) is a crude mixture containing several proteases as well as other
enzymes, including phosphatases, glucosidases, and cellulases.
Bromelain can be extracted from various parts of the pineapple,
including stem, peel, core, and crown, and methods for extracting
and purifying bromelain were reviewed by Vasiljevic (2020). As
well as bromelain, another potent cysteine protease is papain,
which can be extracted from the latex of the papaya tree, while
fig tree latex contains a protease called ficin. Papain is quite heat-​
stable and can act at temperatures of 60–​70 °C, which means that
cooking under mild conditions will be insufficient to inactivate
the enzyme, and stored cooked meat may lose texture due to con-
tinuing enzyme action.
Marinades and Meat
Proteases have substantial effects on the properties of meat.
Many of the post mortem changes in meat texture and tender-
ness, for example, arise due to the action of proteases within the
tissue breaking down proteins such as collagen and myofibrillar
proteins; different meat types and characteristics depend on the
exact handling of the carcasses and treatments at this point.
There have been many studies about increasing the tender-
ness or palatability of meat using fruit-​derived enzymes such
as papain and bromelain (Bekhit et al., 2014; Botinstean et al.,
2018). Enzymes may be introduced to meat through a number of
means. One common approach is to include them in a marinade
(either as a powder or as the fruit or other source), while they may
also be sprinkled onto meat that has been pierced, e.g., with a
fork, to allow the enzyme access to the interior of the meat.
One of the most notable references to the culinary use of
proteases was in the book But the crackling is superb (Kurti and
Kurti, 1988), in which the use of a hypodermic syringe to inject
fresh pineapple juice (in which, critically, the enzyme bromelain
remains active) into cuts of meat (e.g., pork chops) with a view to
tenderizing them is described; the injected meat is left to stand for
470
two hours, to allow the enzyme to act, before being cooked. For
a larger piece of meat, such as a gammon joint, longer standing
periods were recommended to allow percolation and action.
Many herbs and culinary ingredients also contain proteases,
and their application in certain dishes may actually be favored
for this reason; for example, ginger contains proteases, which
can help to tenderize meat. In addition, culinary preparations
containing papain and bromelain are available (e.g., in powder
form), which can be used as meat tenderizers.
Conclusions
Biochemical processes are central to food applications, and their
causes and effects are ubiquitous in food and cooking systems.
Many transformations during food preparation and cooking are
dependent on the properties of proteins, and phenomena such as
denaturation, while the controlled breakdown of proteins using
proteases is likewise a key aspect of many culinary processes.
A good understanding of proteins with respect to their structure
and function is critical for the future of foods, agriculture, pro-
cessing, and design.
REFERENCES
Bekhit AA, Hopkins DL, Geesink G, Beknit AA, Franks P. 2014.
Exogenous enzymes for meat tenderization. Critical Reviews
in Food Science and Nutrition, 54, 1012–​1031.
Bella J, Eaton M, Brodsky B, Berman HM. 1994. Crystal and
molecular structure of a collagen-​like peptide at 1.9 A reso-
lution. Science, 266, 75–​81.
Linda A. Luck, Alan L. Kelly
Bender D, Schönlecher R. 2020. Innovative approaches towards
improved gluten-​ free bread properties. Journal of Cereal
Science, 91, 102904.
Botinstean C, Gomez C, Nian YQ, Auty MAE, Kerry JP, Hammill
RM. 2018. Possibilities for developing texture-​modified beef
steaks suitable for older consumers using fruit-​derived proteo-
lytic enzymes. Journal of Texture Studies, 49, 256–​261.
Dassault Systèmes BIOVIA. 2015. Discovery Studio Modeling
Environment, San Diego, CA, Dassault Systèmes.
Hamada S, Suzuki K, Aoki N, Suzuki Y. 2013. Improvements in the
qualities of gluten-​free bread after using a protease obtained
from Aspergillus oryzae. Journal of Cereal Science, 57, 91–​97.
Kurti N, Kurti G (eds.). 1988. But the crackling is superb, an
anthology on food and drink by fellows and foreign members
of the Royal Society, Philadelphia, Adam Hilger.
Mamo J, Assefa F. 2018. The role of microbial aspartic protease
enzyme in food and beverage industries. Journal of Food
Quality, Article ID 7957269.
Phillips SE. 1980. Structure and refinement of oxymyoglobin at
1.6 A resolution, Journal of Molecular Biology, 142, 531–​535.
Roy U. 2020. Artistic rendering of the protein unfolding, Clarkson
University, Potsdam, NY.
Tame JR, Vallone B. 2000. The structures of deoxy human haemo-
globin and the mutant Hb Tyr alpha42His at 120 K. Acta
Crystallographia Section D, 56, 805–​811.
Uniacke-​ Lowe T, Fox PF. 2017. Chymosin, pepsins and other
aspartyl proteinases: structures, functions, catalytic mech-
anism and milk-​clotting properties. In McSweeney PLH, Fox
PF, Cotter PD, Everett DW (eds.) Cheese, 4th edn, San Diego,
Academic Press, 69–​113.
Vasiljevic T. 2020. Pineapple. In Galanakis CM (ed.) Valorization
of fruit processing by-​ products, London, Academic Press,
203–​226.
Puddings: The Secret of the Rice Pudding
Martin Lersch
https://khymos.org
Rice pudding is a staple food all over Scandinavia. In Norway,
it is known under the name risengrynsgrøt, which translates into
“rice grain porridge”. It is eaten all year round, but peaks in popu-
larity around Christmas, when it is prepared in copious amounts
and served for lunch, leaving plenty of leftovers that can be used
to prepare a popular dessert known as riskrem (rice cream) by
mixing with sweetened whipped heavy cream.
The rice pudding is prepared by letting short-​grained rice
(Oryza sativa subsp. japonica) slowly simmer in milk, thereby
allowing the dominant amylopectin starch to thicken the milk.
Some people find that dairy foods can cause bloating and that
reducing their intake of dairy products, or switching to lactose-​
reduced products, may alleviate the symptoms. With this in mind,
I therefore opted for lactose-​reduced milk when preparing rice
pudding. Since the amylopectin starch only needs a temperature
of 60–​70 °C to swell and absorb the milk, I mixed the cooked
rice and the milk and brought them up to the boiling point before
covering the pot with a lid and leaving it in an oven dialed in
at 100 °C for an hour or two. In this way, the risk of burning
the milk is reduced to a minimum. The surprise came when
I removed the lid and discovered that the entire rice pudding had
turned light brown (Figure 69.1)!
It turned out that the lactose-​reduced milk was to blame for
the unexpected color (Figure 69.2). The milk sugar lactose is a
FIGURE 69.1 Rice pudding made with lactose-reduced milk turns light
brown.
disaccharide, consisting of two simple sugar residues: glucose
and galactose. People with lactose intolerance lack the enzyme
lactase in the small intestine, which cleaves lactose into these two
sugar molecules. What happens then is that lactose is transported
unchanged to the large intestine, where bacteria from the micro-
biota feed on the lactose, resulting in production of gas. The
solution to the problem is simple: you can add the enzyme lac-
tase directly to the milk. This means that lactose-​reduced milk
contains glucose and galactose instead of lactose, and both these
sugars are sweeter than lactose. In fact the sweetness increases
approximately by a factor of four. The sweet taste is noticeable,
and this is actually utilized when preparing a fit-​for-​all chocolate
milk served in schools in Norway, which is lactose free and also
contains significantly less added sugar than normal sweetened
chocolate milk.
Having explained the presence of the monosaccharides glu-
cose and galactose, we are moving closer to an understanding
of the light brown rice pudding. Glucose and galactose can react
with proteins in the Maillard reaction. And what is more, glu-
cose and galactose react 10–​20 times faster than lactose (Naranjo,
2013). However, since the temperature was just below the boiling
point, the browning proceeded relatively slowly. This is why the
color was clearly visible, but luckily without the smell or taste
being affected.
FIGURE 69.2 The milk sugar lactose reacts slowly with proteins in the
Maillard reaction. When it is enzymatically split up into glucose and gal-
actose, the browning reaction proceeds significantly faster.
471
472
Rice Pudding
300 g short-​grained rice
8 dL water
2 L full fat milk
salt to taste
Bring the water to a rolling boil, add the rice, and let it simmer
on low heat for 20 min. In a separate pot, heat the milk to close
to the boiling point and add it to the pot with rice. Bring to boil
while constantly scraping the bottom to prevent the milk from
burning. Cover the pot with a lid and transfer it to an oven set to
90–​100 °C. After 30 min, it is recommended to stir up the rice,
which may have sunk to the bottom. The pudding will continue
to thicken as the rice slowly absorbs more of the liquid. The
Martin Lersch
consistency is normally right after one or two hours. Depending
on the type of rice used, or if left for longer than two hours, the
consistency may become too thick. In this case, adjust consist-
ency with milk prior to serving. Add salt to taste. Serve with
sugar and cinnamon. A dollop of butter, which will melt in con-
tact with the warm pudding, is also nice.
REFERENCE
Naranjo GB, Gonzales ASP, Leiva GE, Malex LS. 2013. The kin-
etics of Maillard reaction in lactose-​hydrolysed milk powder
and related systems containing carbohydrate mixtures, Food
Chemistry, 141, 3790–​3795.
Roasting

Laura Febvay1 and Hervé This vo Kientza2,3

1 Aerial, 250 Rue Laurent Fries, Parc d’innovation, 67412 Illkirch, France
2 INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
3 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,

F-​75005, Paris, France

In the kitchen, many food ingredients are “roasted”, i.e., ther- by conduction, radiation and convection, but depending on the
mally processed without addition of a liquid. The most common design of the roasting equipment, the proportion of heat received
are coffee, chocolate, nuts, almonds, flour and spices. The through each transfer mechanism can vary (Eggers and Pietsch,
various chemical changes due to this process are discussed in this 2001). For instance, in fluidized bed roasters, coffee seeds are
chapter. As the same tissues from various plants often have the exposed to fast-​moving hot air that simultaneously heats and
same overall chemical composition, one example (the seeds of tumbles individual seeds in the roast chamber; convective heating
Coffea) is analysed more thoroughly. is the main mode of heat transfer.
In the kitchen, the roasting process is defined as “the action In drum roasters, conduction tends to dominate (Baggenstoss
of cooking something in an oven or over an open fire” (Oxford et al., 2008).
Dictionaries, 2018). This is what has been done for the seeds of Based on the phenomena triggered by heat, the roasting pro-
Coffea and also for the beans of the cocoa tree Theobroma cacao cess can be divided into five steps: drying, yellowing, first crack,
(“chocolate beans”), various seeds such hazelnut (from Corylus development and second crack (Figure 70.1). Then the roasted
avellana), sesame (Sesum indicum), and also wheat flour (the beans are cooled (Hoffmann, 2014).
ground seed of plants of the genus Triticum). However for all These various steps can be observed by colour changes of
these culinary ingredients, roasting is performed in more ways seeds. Figure 70.2 shows these changes (Vosloo, 2017).
than are described by the dictionary: sometimes it is performed in
an oven, but it can also be in a frying pan, and the “open fire” is 1. Drying: when the green seeds are introduced into the
today replaced by modern heating systems (gas, electric heating preheated roaster, the temperature of the vessel drops.
plates, induction systems, electric oven, etc.). But because more energy is provided, the system tem-
perature stops dropping and increases again after a
“turning point”. When the temperature inside the seeds
The Particular Example of Coffee
For coffee, the green seeds (also called “green beans”) of Coffea
are harvested in producing countries and then sent to roasters,
who operate a thermal treatment during which the coffee seeds
get new organoleptic properties: a brown colour and a specific fla-
vour. This process increases the value of the coffee seeds by 100–​
300% compared with the green seeds (Yeretzian et al., 2002). For
a long time, coffee was simply roasted in a frying pan (Kurti and
Kurti, 1997), but the need to treat all parts of the surface of the
seeds led to the introduction of small rotating drum roasters, in
particular for large-​scale production.
In the coffee industry, roasting is performed by putting the
seeds in a heated vessel (“roaster”) at temperatures between
160 °C and 250 °C for a time between 8 and 20 min, depending
on the final desired organoleptic properties (Fabbri et al., 2011; FIGURE 70.1 The temperature variation around Coffea seeds during
Münchow, 2016). In the roaster, the coffee seeds are heated roasting.

473
474
FIGURE 70.2 Changes in colour of coffee seeds during roasting.
(Vosloo, 2017)
reaches the temperature of water vaporization, the water
inside cells becomes gaseous: this is the drying step,
during which the water content is reduced from 12% to
2% (Fadai et al., 2017), making the seeds expand. This
is the longest step in roasting, as it takes some time for
the seeds to absorb enough energy before evaporation.
2. Yellowing: when the seeds absorb more energy, volatiles
are released, and the colour of the seeds turns to yellow
(Franca et al., 2005); the seeds then have an odour
similar to that of bread (Franca et al., 2009; Putranto
and Chen, 2012). In this step, as in the first one, the
walls of the seeds remain firm, and little steam escapes
from the seeds, so that the pressure increases; the seeds
expand and lose their “chaff” (Hoffmann, 2014; Vosloo,
2017). During the roasting process, these coffee chaffs
are often recovered using an air flux inside the roaster in
order to avoid the formation of flames.
The two first steps are important for roasting, as if an insufficient
quantity of water is released, the external part will roast correctly,
but the inside of the seeds will remain “raw”, with an undesirable
bitter flavour (Hoffmann, 2014).
3. First crack: then some chemical reactions take place,
releasing a large quantity of gases (5–​12 L/​kg, mainly
carbon dioxide (CO2)) (Münchow, 2016). The tem-
perature in the seeds increases rapidly, and the gas
Laura Febvay, Hervé This vo Kientza
production increases the volume of seeds even more,
until the breaking point (Schwartzberg, 2002; Wang
and Lim, 2014). This is the first crack, associated with
a soft noise, as for popcorn. This occurs generally at a
temperature between 175 °C and 185 °C (Gloess et al.,
2014), with about 2 min elapsing from the time when
the sound appears for the first seeds until all seeds have
cracked. At this step, the volume of the seeds is twice
the initial volume.
4. Development: after the first crack, odorant compounds
are generated by many chemical reactions. This step
lasts for 20–​25% of the total roasting time (Münchow,
2016), and roasters base their decision on roasting
time on the colour associated with these processes.
Indeed, technicians decide the development time by
choosing a balance between acidity and bitterness of
the final product; more precisely, during development,
the acidity decreases, whereas bitterness increases
(Hoffmann, 2014).
5. Second crack: after the colour turns browner and
odorant compounds accumulate, a second crack can
occur because of CO2 accumulation, but this second
sound is softer (like snapping fingers) than the first
crack, and it occurs during a shorter time than for
the first. This happens when the temperature reaches
200 °C, where the seeds become darker (Gloess et al.,
2014; Hoffmann, 2014). Industrial roasting processes
Roasting
generally end between the first and the second crack.
Before the first crack, the odour and flavour of coffee
are not yet fully developed, and after the second crack,
the beans are considered burnt; favourable odorant and
taste compounds are lost, and unwanted odours are
generated (Münchow, 2016).
6. Cooling: when the final roasting stage is reached, the
seeds are transferred into another drum for cooling.
Often, liquid water is added in order to stop the roasting,
but some roasters also use cold air (Franca et al., 2005).
As a whole, roasting remains empirical, because the various
times depend on a number of parameters, such as the physical
and chemical characteristics of the seeds (density and water con-
tent), size, volume of batches, room temperature, etc. The goal
of the roaster is to choose the seeds and decide on the roasting
parameters in order to “maximize the flavour” (Illy and Viani,
2005): a strange expression, because the appreciation is either
personal or cultural and is hard to measure. They follow the pro-
cess using thermocouples inserted inside the roaster at the level
of the seeds. Most roasters use the temperature development
curve in order to observe the kinetics of the process.
The chemical changes involved have been studied, as will be
discussed later.
Some Modifications of Coffee Seeds
during Roasting
Colour changes are the most obvious change in seeds during
roasting. In large-​ scale industrial applications, the roasting
degree is even evaluated through the colour changes (Münchow,
2016), but this is not a proper choice, because different chemical
[1]‌  Trigonelline [2]‌  Galactose
[4]‌  Arabinose
475
compositions of roasted seeds can be associated with the same
colour (Febvay, 2019). Roasting triggers a multitude of chem-
ical reactions (Clarke, 1987; Clarke and Vitzthum, 2017),
generating new compounds, with sensorial, nutritional and toxi-
cological effects (Buffo and Cardelli-​Freire, 2004; Baggenstoss
et al., 2008).
Of course, the chemical reactions are many, and include in par-
ticular pyrolysis, glycation reactions (including Schiff, Fischer,
Maillard and Strecker reactions), hexose dehydration, oxidations,
and degradation of polysaccharides, chlorogenic acids, proteins
and trigonelline [1]‌(Franca et al., 2009b; Putranto and Chen,
2012; Sunarharum et al., 2014).
As a result, the main changes are:
• loss of water
• decrease in polysaccharides, loss of oligomers and
monosaccharides
• changes in quantities of aliphatic acids
• loss of chlorogenic acids
• decrease in proteins and amino acids
• disappearance of trigonelline
• formation of melanoidins
Some of these will now be considered.
1. Saccharides: polysaccharides undergo degradation,
depolymerization and structural modification (Redgwell
et al., 2002). Monosaccharides such as galactose [2]‌,
mannose [3], arabinose [4] and ribose [5] are released
(Moreira et al., 2012; Moreira et al., 2015; Oosterveld
et al., 2003; Nunes et al., 2012). The oligosaccharides
and monosaccharides produced are then transformed
into degradation products (Oosterveld et al., 2003).
[3]‌  Mannose
[5]‌  Ribose
476
Arabinogalactans and galactomannans undergo many chem-
ical modifications, including a reduction of the polymerization
and branching degree, which can lead to a reduced solubility
in water (Moreira et al., 2012; Nunes and Coimbra, 2001).
Arabinogalactans are particularly prone to thermal processing
compared with other polysaccharides of coffee. They are
depolymerized after a mild roasting through galactan backbone
fission and a loss of arabinose [4]‌by lateral chains.
Only a fraction of the oligomers or monomers in green seeds is
found in the roasted seeds, showing that they are rapidly modified
(Oosterveld et al., 2003). Modelling (Ginz et al., 2000) shows
that the sucrose [6]‌content decreases, becoming too low to be
[6]‌  Sucrose [7]‌  D-​fructose
[9]‌  Citric acid [10] Malic acid
[12] Glutaric acid [13] Itaconic acid
Laura Febvay, Hervé This vo Kientza
detected after strong roasting (at temperatures more than 280 °C).
When sucrose is added to seeds during roasting, fructose [7] and
glucose [8] are formed, showing that these sugars are produced
by degradation of sucrose.
2. Aliphatic acids: citric [9]‌and malic [10] acids are
destroyed, producing citraconic acid [11], glutaric
acid [12], itaconic acid [13] and mesaconic acid [14]
for the former, and succinic acid [15] and fumaric
[16] and maleic [17] acids for the latter, respectively
(Balzer, 2001).
[8]‌  D-​glucose
[11] Citraconic acid
[14] Mesaconic acid
Roasting
FIGURE 70.3 Dehydration of chlorogenic acids produces a lactone.
(Farah et al., 2005)
The acidity appearing during roasting is primarily due to the
formation of four aliphatic acids: formic [18], acetic [19], gly-
colic [20] and lactic [21] acids (Verardo et al., 2002). Their main
precursor is sucrose [6]‌, which is at a concentration between 3%
and 8% in green seeds. Arabinose [4], erythrose [22] and 1,6-​
anhydroglucose [23] are also produced by the thermal degrad-
ation of sucrose [6] and are precursors for acid formation.
3. Chlorogenic acids: during roasting, the chlorogenic
acids undergo many transformations, the first being a
dehydration generating a lactone (Figure 70.3).
When the roasting process continues, the chlorogenic acids are
degraded and react with sugars. Typically, 60% of chlorogenic
acids are modified, being transformed into 30 lactones through
477
the loss of water and quinic acid [24] and the formation of an
intramolecular ester (Bennat et al., 1994; Farah et al., 2005).
Of these lactones, seven were identified in roasted coffee
(Farah et al., 2005): 3-​caffeoylquinic-​1,5-​lactone (3-​CQL, the
most abundant) [25], 4-​caffeoylquinic-​ 1,5-​lactone (4-​ CQL)
[26], 3-​coumaroylquinic-​1,5-​lactone (3-​pCoQL) [27], 4-​
coumaroylquinic-​1,5-​lactone (4-​pCoQL) [28], 3-​feruloylquinic-​
1,5-​lactone (3-​FQL) [29], 4-​feruloylquinic-​1,5-​lactone (4-​FQL)
[30], and 3,4-​dicaffeoylquinic-​1,5-​lactone (3,4-​diCQL) [31].
With a prolonged roasting, the chlorogenic acids are degraded
into phenolic compounds through the hydrolysis of caffeinic
[32] and quinic [24] acids, forming new phenolic compounds
such as pyrogallol [33] or catechol [34], and also caffeic acid
[35] and hydroquinone [36] (Clifford, 1985; Müller et al.,
2006).
478 Laura Febvay, Hervé This vo Kientza

[15] Succinic acid [16] Fumaric acid [17] Maleic acid

[18] Formic acid [19] Acetic acid [20] Glycolic acid

[21] Lactic acid [22] Erythrose [23] 1,6-​anhydroglucose

[24] Quinic acid [25] 3-​caffeoylquinic-​1,5-​lactone (3-​CQL)

[26] 4-​caffeoylquinic-​1,5-​lactone (4-​CQL)

Roasting 479

[27] 3-​coumaroylquinic-​1,5-​lactone (3-​pCoQL) [28] 4-​coumaroylquinic-​1,5-​lactone (4-​pCoQL)

[29] 3-​feruloylquinic-​1,5-​lactone (3-​FQL) [30] 4-​feruloylquinic-​1,5-​ lactone (4-​FQL)

480 Laura Febvay, Hervé This vo Kientza

[31] 3,4-​dicaffeoylquinic-​1,5-​lactone (3,4-​diCQL) [32] Caffeinic acid

[33] Pyrogallol [34] Catechol [35] Caffeic acid

[36] Hydroquinone [37] Arginine

Roasting
4. Formation of melanoidins: melanoidins are polymers
that are present in many transformed food products.
In roasted coffee, they are nitrogen-​ containing
compounds of large size, with a molar mass between
3,000 and 22,000 (Ledl and Schleicher 1990; Nunes
et al., 2012). Melanoidins include residues of sugars,
proteins and phenolics (Nunes and Coimbra, 2001;
Borrelli et al., 2002; Bekedam et al., 2006; Nunes and
Coimbra, 2007; Bekedam et al., 2008; Gniechwitz
et al., 2008; Nunes and Coimbra, 2010; Silván et al.,
2010; Moreira et al., 2012; Nunes et al., 2012; Coelho
et al., 2014; Moreira et al., 2015). Because of the poor
characterization of melanoidins to date, their quantity
is difficult to assess and their formation is difficult
to follow. It has been reported that polymerization of
phenols and glycation processes contribute to the for-
mation of melanoidins in coffee (Borrelli et al., 2002;
Montavon et al., 2003).
5. Degradation of proteins: roasting denatures proteins but
also makes them insoluble. Depending on the intensity
of roasting, the loss of amino acid residues can reach
20–​40%. Some amino acids, such as arginine [37], cyst-
eine [38], lysine [39], methionine [40], threonine [41]
and serine [42], are lost partially or entirely (Moreira
et al., 2012a; Parliment, 2000; Clarke, 1987; Spiller,
1997). Amino acids with hydroxyl (-​OH), thiol (-​SH)
and amino (-​NH2) groups participate in odour formation.
6. Disappearance of nitrogenous compounds: caffeine
[43] does not appear to be modified during roasting
[38] Cysteine [39] Lysine
[41] Threonine [42] Serine
481
(Farah et al., 2005), but other xanthines, such as
adenine [44], guanine [45] and hypoxanthine [46], are
destroyed. Trigonelline [11] is changed into nicotinic
acid [47], pyridine [48] and aromatic substances (such
as furanes, pyrazines or pyrroles) (Ky, 2001). In par-
ticular, N-​methylpyridinium [49] is produced (Taguchi
et al., 1985; Stadler et al., 2002).
7. Volatiles: volatile compounds are absent in green
seeds, but they make up about 0.1% of the dry matter
of roasted seeds. Their appearance during roasting is
largely due to sucrose [6]‌degradation through cara-
melization (Ginz et al., 2000; De Maria et al., 1996).
Volatile compounds such as aldehydes, ketones, thiols,
furans, pyrroles, pyrazines and guaiacols derive from
degradation or fragmentation of sucrose, amino acids,
trigonelline and chlorogenic acids, as well as from pyr-
olysis of arabinogalactans.
Roasted coffee contains more than 600 volatile compounds;
only about 10% of these contribute to odour.
Furfurylic alcohol [50] plays a particular role, because green
seeds contain more of it (418 µg/​g) compared with other roasted
seeds in the same conditions (up to 132 µg/​g). Its production
resembles that of other compounds such as hydroxymethylfurfural
(HMF) [51] or acrylamide [52]. However, the quantity of
furfurylic acid [53] in roasted coffee does not correspond to the
quantity produced during roasting.
[40] Methionine
[43] Caffeine
482 Laura Febvay, Hervé This vo Kientza

[44] Adenine [45] Guanine [46] Hypoxanthine

[47] Nicotinic acid [48] Pyridine [49] N-​methylpyridinium

[50] Furfurylic alcohol [51] Hydroxymethylfurfural (HMF) [52] Acrylamide [53] Furfurylic acid

8. Lipids are not much transformed during roasting; the that the medium roasted process (6.5 min to the onset of the first
lipid fraction tends to be stable and survive the roastingcrack and 1.0 min to the onset of the second crack) resulted in
process with only minor changes. Linoleic [54] and “good balance” of taste and odour. However, the “sweated pro-
palmitic [55] acids are the predominant fatty acid cess” (4.5 min to the first crack and 6.5 min to the second crack)
residues in coffee. The most important modification resulted in non-​uniform bean colour, and the coffee was “sour,
is the reduction of diterpenes in dehydrocafestol [56] grassy, and underdeveloped”. Reducing the heating rate further
and dehydrokahweol [57]. Cafestol [58] and kahweol by using the “baked process” (11 min to the first crack and 18 min
[59] are diterpenes that are degraded by the roasting to the second crack) produced coffee that was “flat, woody with
process. low brightness and acidity” (Lyman et al., 2003). In another
study, Schenker et al. (2002) reported that processes for which
As a whole, for culinary purposes, it is worth observing that temperature increases from low to high (150 to 240 °C for 270 s;
not only are the temperature of roasting and the duration of 240 °C for 55 s) resulted in the formation of the highest quan-
the thermal process important, but also the different time–​ tities of odoorant volatiles, while the long-​time low-​temperature
temperature profiles during roasting develop different flavours approach (isothermal heating at 220 °C for 600 s) generated the
(Buffo and Cardelli-​Freire, 2004; Febvay et al., 2019). The lowest odorant volatiles. Moreover, the distribution of the 13
effects of time–​temperature profile on coffee odorant proper- volatile compounds monitored differed considerably depending
ties have been reported by Lyman et al. (2003), who observed on the roasting profiles used.
Roasting 483

[54] Linoleic acid

[55] Palmitic acid

[56] Dehydrocafestol [57] Dehydrokahweol

484
[58] Cafestol
Other Seeds
Roasting is also performed for other food ingredients than coffee
beans. The process has been less well studied, but many results
have been obtained. Some are given here, providing complemen-
tary information that can sometimes hold as well for coffee.
1. Cocoa: One other important (in terms of mass treated)
product that is roasted is cocoa, for which roasting also
develops unique flavour (Zzaman et al., 2017). Roasting
temperatures are often between 150 °C and 250 °C,
and roasting times are in the interval 10–​50 min. As
for coffee, the concentration of total reducing sugars
is also reduced to 60–​ 85% with increased roasting
temperature. The hydrophobic amino acids or amino
acid residues are reduced by as much as 30–​49% with
increased roasting temperature after 50 min processing.
A number of pyrazines, esters, aldehydes, alcohols,
ketones, carboxylic acids and hydrocarbons appear in all
the samples at different concentration ranges. The most
flavour-​active compounds, pyrazines, were formed at the
highest concentration (2.96 mg/​kg) at 200 °C for 10 min.
2. Carob: Other plant materials give analogous results.
For example, carob pod powder (Ceratonia siliqua
L.) roasted at different roasting temperatures become
darker, and the average moisture content, water activity,
oil content and sweetness values decrease at higher
temperatures (Boublenza et al., 2017). Here, the total
content of phenolics (and the antioxidant activity)
increases with increasing roasted temperature. Oleic
acid [60], linoleic acid [54] and palmitic acid [55] are
the main fatty acid residues present in carob oil. Results
showed that the roasted carob pod powders are sweeter
and have a more caramel-​like taste and a more cacao-​
like aroma at lower roasting temperatures but have a
more astringent taste, coffee-​like aroma and roasted
aroma at higher roasting temperatures.
Laura Febvay, Hervé This vo Kientza
[59] Kahweol
3. Walnut: For walnut flour (Santos et al., 2017), the
changes in lipids were thoroughly followed in one
study. The goal was to evaluate the composition and
the antioxidant properties of walnut flours submitted
to different roasting protocols (50, 100 and 150 °C for
30, 60 and 120 min). All walnut flours contained about
42% of protein and a significant amount of dietary fibre
(17%), which was affected by the roasting process.
Nonetheless, the fat content increased around 50% in
walnut flours subjected to longer and higher roasting
temperatures (150 °C). The lipid fraction showed
good nutritional quality with a high vitamin E content
(mainly γ-​tocopherol [61]) and a fatty acid profile rich
in linoleic [54] and linolenic acids [62]. The high phen-
olic content also provides great antioxidant capacity to
the flours. Mild roasting of the walnuts did not affect
the quality of the flours, which could be used as a func-
tional ingredient in the food industry.
4. Sesame seeds: For sesame, Kahyaoglu and Kaya (2006)
modelled moisture, colour and texture changes in sesame
seeds during roasting. Experiments were performed at
120, 150 and 180 °C for 120 min. As for other roasting
processes, seeds became browner, but the interest of this
study was to simulate the changes in the moisture con-
tent, colour values (L*, a* and b*) and textural properties
(hardness and fracturability) using exponential, kinetic
and polynomial models. The variations in the colour
parameters of sesame seeds during conventional roasting
were adequately simulated by cubic polynomials. The
zero-​order kinetic model satisfactorily described the
reduction in the hardness and fracturability of sesame
seeds. The temperature dependence of moisture and
colour models was described by a linear function rela-
tionship, but an Arrhenius-​type relationship was used for
textural models. The activation energy for sesame seeds
was estimated to be 6.9 and 9.8 kJ/​mol for hardness and
Roasting
fracturability changes, respectively, over the temperature
range of the study.
  The effect of roasting treatment on the chemical com-
position of sesame oil was also studied (Ji et al., 2019).
They reported that with roasting, the peroxide value and
colour development of oils were obviously elevated,
while the total tocopherols and sesamolin decreased
[60] Oleic acid
[61] γ-​tocopherol
[62] Linolenic acid
485
steadily. The acid values in the current experiment were
expected to grow as the roasting time increased at the
same temperature, and the acid value decreased in the
first 30 min of roasting at 160 °C. Increased roasting
temperature or time facilitates sesamol formation in
sesame oil. The fatty acid profiles are almost inde-
pendent of roasting conditions.
486
5. Wheat: Of course, the case of wheat flour and bread is
important, as baking bread is related to roasting, with
the same range of temperature and time of thermal pro-
cessing. Bread making leads to the generation of a lot of
odorant compounds, but these can be attributed to either
fermentation or roasting (Gassenmeier and Schieberle,
1995). The most important odorants in the crumbs of
wheat breads (French-​type) prepared according to two
different dough recipes using pre-​fermentation (crumb
I: liquid pre-​ ferment, containing 0.25% yeast and
1.5% yeast in the final dough; crumb II: soft dough
pre-​ferment containing 15% yeast and 4.6% yeast in
the final dough) were evaluated on the basis of aroma
extract dilution analyses. In crumb I, exhibiting the more
typical flavour, comparatively higher flavour dilution
(FD) factors were found especially for 2-​phenylethanol
[63] 2-​phenylethanol (2-​PE) [64] 3-​methyl-​butanol (3-​MB)
[66] 1-​octen-​3-​one [67] 4-​hydroxy-​2,5-​dimethyl-​3(2H)-​furanone
[69] 2-​methylbutanoic acid [70] 3-​methylbutanoic acid
Laura Febvay, Hervé This vo Kientza
(2-​PE) [63] and 3-​methyl-​butanol (3-​MB) [64], while in
crumb II, the FD factors of methional [65], 1-​octen-​3-​
one [66], 4-​hydroxy-​2,5-​dimethyl-​3(2H)-​furanone [67],
butanoic acid [68] and 2-​methylbutanoic acid [69] and
3-​methylbutanoic acid [70] were higher than in crumb
I. Quantitative studies (stable isotope dilution assays)
of 2-​PE and 3-​MB formation in the liquid pre-​ferments
containing low yeast concentrations (0.25%) revealed
that anaerobic conditions and a fermentation tempera-
ture of 35 °C favoured the production of both odorants.
Model studies, in which either the 3-​MB precursors
L-​leucine [71] and 3-​methylbutanal [72] or the 2-​PE
precursors L-​phenylalanine [73] and phenylacetaldehyde
[74] had been added to the pre-​ferments, indicated that
baker’s yeast significantly (15–​ 55%) converted these
precursors into the respective odorants.
[65] Methional
[68] Butanoic acid
[71] L-​leucine
Roasting
[72] 3-​methylbutanal [73] L-​phenylalanine
[76] (E)-​2-​nonenal [77] 3-​methylbutyric acid
Toasting gives a different perspective on the same kind of pro-
cess with the same kind of raw material (Rychlik and Groschf,
1996): slices of wheat bread were toasted until a distinct inten-
sity of brown colour was attained. Potent odourants formed were
evaluated by odorant extract dilution analysis and gas chromatog-
raphy/​olfactometry of headspace samples. Compounds showing
high dilution factors were quantified, and their odorant activity
values (OAV, ratio of concentration to odour threshold) were
calculated on the basis of their odour thresholds in starch. The
roasty smell of 2-​ acetyl-​1-​
pyrroline [75] showed the highest
OAV, followed by (E)-​2-​nonenal [76], 3-​methylbutyric acid [77],
4-​hydroxy-​2,5-​dimethyl-​3 (2H)-​furanone [66], methional and
2,3-​butanedione [78]. Formation of nine odorants ([73] to [76],
among others) was measured in relation to the intensity of the
brown colour. It was found that at the beginning of toasting, [72]
increased more rapidly in the bread slices than [77], which was
mainly produced at medium browning.
Conclusions
Plant materials share a common basic composition, with water,
polysaccharides, amino acids, proteins, etc. Seeds in particular
include lipids because of the biological significance of these
compounds for young plant development (Bergfeld et al., 1978).
It is not surprising that the thermal processing of these products
is accompanied by the same kind of chemical reaction, with the
487
[74] Phenylacetaldehyde [75] 2-​acetyl-​1-​pyrroline
[78] 2,3-​butanedione
same colour changes, and the same products being formed in the
same range of time and temperature. Of course, particular plant
tissues with a particular raw composition (a lot of starch for flour,
caffeine in coffee or chocolate, etc.) can have particular reactions,
alone or in combination, but this is not to say that all these plant
tissues will have the same flavour in the end, because it is well
known that changes in the particular composition of odorants can
have drastic consequences for the odour, and hence the flavour,
of food products.
REFERENCES
Baggenstoss J, Poisson L, Kaegi R, Perren R, Escher E. 2008.
Coffee roasting and aroma formation: Application of different
time−temperature conditions. Journal of Agricultural and
Food Chemistry, 56 (14), 5836–​5846.
Balzer HH. 2001. Acids in coffee. In Clarke RJ, Vitzthum OG (eds)
Coffee: recent developments, Blackwell Science Ltd.
Bekedam E, Henk K, Schols A, Martinus A, van Boekel AJS, Smit
G. 2006. High molecular weight melanoidins from coffee
brew. Journal of Agricultural and Food Chemistry, 54 (20),
7658–​7666.
Bekedam EK, Loots MJ, Schols HK, Van Boekel MAJS, Smit G.
2008. Roasting effects on formation mechanisms of coffee
brew melanoidins. Journal of Agricultural and Food Chemistry,
56 (16), 7138–​7145.
Bennat C, Engelhardt UH, Kiehne A, Wirries FM, Maier HG. 1994.
HPLC analysis of chlorogenic acid lactones in roasted coffee.
Zeitschrift Für Lebensmittel-​Untersuchung Und Forschung,
199 (1), 17–​21.
488
Bergfeld R, Hong YN, Kfihnl T, Schopfer P. 1978. Formation of
oleosomes (storage lipid bodies) during embryogenesis and
their breakdown during seedling development in cotyledons of
Sinapis alba L. Planta, 143, 297–​307.
Borrelli RC, Visconti A, Mennella C, Anese M, Fogliano V. 2002.
Chemical characterization and antioxidant properties of coffee
melanoidins. Journal of Agricultural and Food Chemistry, 50
(22), 6527–​6533.
Boublenza I, Lazouni HA, Ghaffari L, Ruiz K, Fabiano-​Tixier AS,
Chemat F. 2017. Influence of roasting on sensory, antioxidant,
aromas, and physicochemical properties of carob pod powder
(Ceratonia siliqua L.) quality. Journal of Food Quality, https://​
doi.org/​10.1155/​2017/​4193672
Buffo RA, Cardelli-​Freire C. 2004. Coffee flavour: An overview.
Flavour and Fragrance Journal, 19 (2), 99–​104.
Clarke RJ (ed.).1987. Coffee: Volume 2: Technology, Springer
Netherlands. www.springer.com/​la/​book/​9789401080286
Clarke RG, Vitzthum OG (eds.). 2017. Coffee: Recent Developments.
Wiley. www.wiley.com/​WileyCDA/​WileyTitle/​productCd-​
0632055537.html.
Clifford MN. 1985. Chemical and physical aspects of green coffee
and coffee products. In Clifford MN, Wilson KC (eds.) Coffee:
Botany, Biochemistry and Production of Beans and Beverage,
Boston, MA: Springer US. 305–​374.
Coelho C, Ribeiro M, Cruz ACS, Domingues MRM, Coimbra MA,
Bunzel M, Nunes FM. 2014. Nature of phenolic compounds
in coffee melanoidins. Journal of Agricultural and Food
Chemistry, 62, 7843–​7853.
De Maria CABL, Trugo C, Aquino Neto, FR, Moreira, RFA, Alviano
CS. 1996. Composition of green coffee water-​soluble fractions
and identification of volatiles formed during roasting. Food
Chemistry, 55 (3), 203–​207.
Eggers R, Pietsch A. 2001. Technology I: Roasting. In Clarke RJ,
Vitzhthum OG (eds.) Coffee, Blackwell Science Ltd, 90–​107.
Fabbri A, Cevoli C, Alessandrini L, Romani S. 2011. Numerical
modeling of heat and mass transfer during coffee roasting pro-
cess. Journal of Food Engineering, 105 (2), 264–​269.
Fadai NT, Melrose C, Please CP, Schulman A, Van Gorder RA.
2017. A heat and mass transfer study of coffee bean roasting.
International Journal of Heat and Mass Transfer, 104
(Supplement C), 787–​799.
Farah A, de Paulis T, Trugo LC, Martin PR. 2005. Effect of roasting
on the formation of chlorogenic acid lactones in coffee. Journal
of Agricultural and Food Chemistry, 53 (5), 1505–​1513.
Febvay L, Hamon E, Werner D, This D. 2019. Identification of
markers of thermal processing (“roasting”) in aqueous extracts
of Coffea arabica L. seeds through NMR fingerprinting and
chemometrics. Magnetic Resonance in Chemistry, 57, 589–​602.
Franca AS, Oliveira LS, Mendonça JCF, Silva XA. 2005. Physical
and chemical attributes of defective crude and roasted coffee
beans. Food Chemistry, 90 (1), 89–​94.
Franca AS, Oliveira LS, Oliveira RCS, Mancha Agresti PC, Rodinei
A. 2009. A preliminary evaluation of the effect of processing
temperature on coffee roasting degree assessment. Journal of
Food Engineering, 92 (3), 345–​352.
Gassenmeier K, Schieberle P. 1995. Potent aromatic compounds in
the crumb of wheat bread (French-​type) –​influence of pre-​
ferments and studies on the formation of key odorants during
dough processing. Zeitschrift für Lebensmittel-​Untersuchung
und Forschung, 201 (3), 241–​248.
Ginz M, Balzer HH, Bradbury AGW, Maier HG. 2000. Formation
of aliphatic acids by carbohydrate degradation during roasting
of coffee. European Food Research and Technology, 211 (6),
404–​410.
Laura Febvay, Hervé This vo Kientza
Gloess AN, Vietri A, Wieland F, Smrke S, Schönbächler S, Sánchez
López JA, Petrozzi S, Bongers S, Koziorowski T, Yeretzian S.
2014. Evidence of different flavour formation dynamics by
roasting coffee from different origins: On-​line analysis with
PTR-​ToF-​MS. International Journal of Mass Spectrometry,
Special issue: Tilmann Mäerk, 365–​366, 324–​337.
Gniechwitz D, Reichardt N, Ralph J, Blaut M, Steinhart H, Bunzel
M. 2008. Isolation and characterisation of a coffee melanoidin
fraction. Journal of the Science of Food and Agriculture, 88
(12), 2153–​2160.
Hoffmann J. 2014. The World Atlas of Coffee: From Beans to Brewing
–​Coffees Explored, Explained and Enjoyed. Hachette UK.
Illy A, Viani R. 2005. Espresso Coffee: The Science of Quality.
Academic Press.
Ji J, Liu Y, Shi L, Wang N, Wang X. 2019. Effect of roasting
treatment on the chemical composition of sesame oil, LWT –​
Food Science and Technology, 101, 191–​200.
Kahyaoglu T, Kaya S. 2006. Modeling of moisture, color and tex-
ture changes in sesame seeds during the conventional roasting.
Journal of Food Engineering, 75, 167–​177.
Kurti N, Kurti G. 1997. But the Crackling Is Superb. CRC Press,
London.
Ky CL, Louarn J, Dussert S, Guyot B, Hamon S, Noirot M. 2001.
Caffeine, trigonelline, chlorogenic acids and sucrose diver-
sity in wild Coffea arabica L. and C. canephora P. accessions.
Food Chemistry, 75 (2), 223–​230.
Ledl F, Schleicher E. 1990. New aspects of the Maillard reac-
tion in foods and in the human body. Angewandte Chemie
International Edition in English, 29 (6), 565–​594.
Lyman DJ, Benck R, Dell S, Merle S, Murray-​Wijelath J. 2003.
FTIR-​ ATR analysis of brewed coffee: Effect of roasting
conditions. Journal of Agricultural and Food Chemistry, 51
(11), 3268–​3272.
Montavon P, Mauron AF, Duruz E. 2003. Changes in green coffee
protein profiles during roasting. Journal of Agricultural and
Food Chemistry, 51 (8), 2335–​2343.
Moreira ASP, Nunes FM, Rosário Domingues M, Coimbra MA.
2012. Coffee melanoidins: Structures, mechanisms of for-
mation and potential health impacts. Food & Function, 3 (9),
903–​915.
Moreira ASP, Coimbra MA, Nunes FM, Passos CP, Santos SAO,
Silvestre AJD, Silva AMN, Rangel M, Domingues RM.
2015. Chlorogenic acid–​arabinose hybrid domains in coffee
melanoidins: Evidences from a model system. Food Chemistry,
185 (Supplement C), 135–​144.
Müller C, Lang R, Hofmann T. 2006. Quantitative precursor studies
on di-​and trihydroxybenzene formation during coffee roasting
using “in bean” model experiments and stable isotope dilu-
tion analysis. Journal of Agricultural and Food Chemistry, 54,
10086–​10091.
Münchow M. 2016. Book SCAE Roasting Foundation. https://​
scae- ​ s hop.myshopify.com/ ​ p roducts/ ​ b ook- ​ s cae- ​ r oasting-
foundation
Nunes FM, Coimbra MA. 2001. Chemical characterization of the
high molecular weight material extracted with hot water from
green and roasted Arabica coffee. Journal of Agricultural and
Food Chemistry, 49 (4), 1773–​1782.
Nunes FM, Coimbra MA. 2007. Melanoidins from coffee infusions.
Fractionation, chemical characterization, and effect of the
degree of roast. Journal of Agricultural and Food Chemistry,
55 (10), 3967–​3977.
Nunes FM, Coimbra MA. 2010. Role of hydroxycinnamates in
coffee melanoidin formation. Phytochemistry Reviews, 9 (1),
171–​185.
Roasting
Nunes FM, Cruz ACS, Coimbra MA. 2012. Insight into the mech-
anism of coffee melanoidin formation using modified “in
bean” models. Journal of Agricultural and Food Chemistry,
60 (35), 8710–​8719.
Oosterveld A, Voragen AGJ, Schols HA. 2003. Effect of roasting
on the carbohydrate composition of Coffea arabica beans.
Carbohydrate Polymers, 54 (2), 183–​192.
Oxford Dictionaries. 2018. https://​en.oxforddictionaries.com/​defin-
ition/​roasting
Parliment TH. 2000. An overview of coffee roasting. In Caffeinated
Beverages, ACS Symposium Series 754, American Chemical
Society, 188–​201.
Putranto A, Chen XD. 2012. Roasting of barley and coffee modeled
using the Lumped-​Reaction Engineering Approach (L-​REA).
Drying Technology, 30 (5), 475–​483.
Redgwell RJ, Trovato V, Curti D, Fischer M. 2002. Effect of roasting
on degradation and structural features of polysaccharides
in Arabica coffee beans. Carbohydrate Research, 337 (5),
421–​431.
Rychlik M, Groschf W. 1996 .Identification and quantification of
potent odorants formed by toasting of wheat bread. LWT –​
Food Science and Technology, 29 (5–​6), 515–​525.
Santos J, Alvarez-​Ortí M, Sena-​Moreno E, Rabadán A, Pardo JE,
Oliveira MBPP. 2017. Effect of roasting conditions on the
composition and antioxidant properties of defatted walnut
flour. Journal of the Science of Food and Agriculture, 98 (5),
1813–​1820.
Santos JR, Sarraguça MC, Rangel AOSS, Lopes JA. 2012. Evaluation
of green coffee beans quality using near infrared spectroscopy:
A quantitative approach. Food Chemistry, 135 (3), 1828–​1835.
Schenker S, Heinemann C, Huber M, Pompizzi R, Perren R, Escher
R. Impact of roasting conditions on the formation of aroma
compounds in coffee beans. Food Science, 67 (1), 60–​66.
Schwartzberg M. 2002. Modeling and simulation of reactive and
separation food bed processes, in Engineering and Food for
the 21st Century, ed. Welti-​Chanes J, Mguilera J-​M, Taylor &
Francis Group.
489
Silván JM, Morales FJ, Saura-​Calixto F. 2010. Conceptual study on
Maillardized dietary fiber in coffee. Journal of Agricultural
and Food Chemistry, 58 (23), 12244-​1249.
Spiller FA. 1997. Caffeine. CRC Press.
Stadler RH, Varga N, Milo C, Schilter B, Arce Vera F, Welti DH.
2002. Alkylpyridiniums. 2. Isolation and quantification in
roasted and ground coffees. Journal of Agricultural and Food
Chemistry, 50 (5), 1200–​1206.
Sunarharum WB, Williams DJ, Smyth HE. 2014. Complexity of
coffee flavor: A compositional and sensory perspective. Food
Research International, 62, 315–​325.
Taguchi H, Sakaguchi M, Shimabayashi Y. 1985. Trigonelline content
in coffee beans and the thermal conversion of trigonelline into
nicotinic acid during the roasting of coffee beans. Agricultural
and Biological Chemistry, 49 (12), 3467–​3471.
Verardo G, Cecconi F, Geatti P, Giumanini AG. 2002. new procedures
for determination of acids in coffee extracts, and observations
on the development of acidity upon ageing. Analytical and
Bioanalytical Chemistry, 374 (5), 879–​885.
Vosloo J. 2017. Heat and Mass Transfer Model for a Coffee Roasting
Process. Thesis, North-​ West University (South Africa),
Potchefstroom Campus. https://​ repository.nwu.ac.za:443/​
handle/​10394/​26094.
Wang X, Lim LT. 2014. A kinetics and modeling study of coffee
roasting under isothermal conditions. Food and Bioprocess
Technology, 7 (3), 621–​632.
Yeretzian C, Pascual EC, Goodman BA. 2002. Effect of roasting
conditions and grinding of free radical contents of coffee beans
stored in air. Food Chemistry, 131 (3), 811–​816.
Zzaman W, Bhat R, Yang TA, Mat Easa Y. 2017. Influences of
superheated steam roasting on changes in sugar, amino acid
and flavor active components of cocoa bean (Theobroma
cacao). Journal of the Science of Food and Agriculture, 97
(13), 4429–​4437.
Salt: When Should Salt Be Added to Meat Being Grilled?
Hervé This vo Kientza1,2, Marie-​Paule Pardo2 and Rolande Ollitrault2
1 INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​ AgroParisTech International Centre for Molecular Gastronomy, F-​75005, Paris,
France
Many cooks add “salt” (of course, they mean sodium chloride
including impurities) to “meat” (pieces of muscular tissue from
animals) before “grilling” (thermal treatment by contact with a
solid surface at a temperature well above 100 °C, e.g., 300 °C)
because they think and say that the salt enters into the meat.
However, the loss of an aqueous solution (“juices”) by the meat,
due to collagen shrinkage during heating, is an indication of an
outward flow that can prevent salt penetration. Using scanning
electron microscopy (SEM) with X-​ray analysis, it can be shown
that only a very thin outside layer is salted (Kuehne, 1988).
However, capillarity allows some salt penetration in meat for
some meats only (where fibre bundles are loosely grouped).
This question of salting meat before, during or after grilling was
discussed at the second INRA Seminar of Molecular Gastronomy
in November 2000 in Paris (This, 2000); cooks, culinary teachers
and scientists discussed the “more appropriate” time for adding
salt to meat being grilled. This was naïve, because the question
was not technical but artistic, and terms such as “good” or “better”
are too subjective to be defined; as well, no unique answer could
be given to “more appropriate”, because the goal is not clearly
defined (This and Gagnaire, 2008). Moreover, even for such a
simple question as when to add salt to meat and why, the chefs
attending the seminar had different ideas, with no demonstration
that their ideas could work properly.
Some chefs proposed to add the salt before grilling, saying
that the salt would enter into the meat during cooking. This idea
is also given by some culinary books (Bernardi, 1853; Bocuse,
1976), but no measurement was ever made in kitchens, to the best
of our knowledge.
Other cooks recommended to add the salt during grilling,
because (they said) the colour would be improved. Again, they did
not compare the same piece of meat processed in the two ways.
Also, some cooks advised adding the salt after grilling,
because “otherwise salt draws juices out of the meat”; this third
idea can also be found in many culinary books (Bretonnel, 1896;
Robuchon, 1993), and again, no rigorous test was made.
Interestingly, no cook present at this seminar had read the sci-
entific literature about the question (articles such as Zhou and
Wang (1998) are generally unknown in culinary circles). In order
to know whether the culinary “precisions” (This, 2005) given
by cooks and culinary books about salt were right or wrong, we
first measured the mass of various meat samples covered with
salt as a function of time at room temperature. The losses were
compared with measurement of thermally processed meat before
it was shown why, in some cases, capillarity could be responsible
for some salt absorption in meat during cooking. Finally, we used
SEM with X-​ray analysis to measure the salt content in the meat
before and after grilling.
Juice Extraction by Salt
In order to understand first how much “juice” could be extracted
from meat by salt at room temperature, samples of beef muscular
tissues (Longissimus dorsi, cut perpendicularly to the muscular
fibres, and Diaphragma crus dextrum sinistrum, cut parallel to
the fibres) and chicken muscular tissues (M. pectoralis major,
intact) were purchased from a local butcher. Meat pieces were
put in cups and completely covered with sea salt (brand La
Baleine) at a number of fixed temperatures. At various times, the
samples of animal tissues were taken out from the salt, washed,
dried with absorbing paper and compressed air, and weighed
(Mettler balance, precision 0.0001 g). Then they were trans-
ferred into fresh salt.
Figure 71.1 shows the variation of the mass of meat stored in
salt as a function of time. It can be observed that two regimes
are recorded, and the second one corresponds apparently to the
formation of a crust, slowing down water transfer. However, the
time constant for the loss of “juice” is very long compared with
the time usually used for grilling meat (minutes).
Of course, these results are obtained at room temperature,
and the exudation is faster at the temperature applied during
grilling (up to 400 °C at the lower surface, about 50 °C inside the
meat). However, assuming a Fickean diffusion of salt or of juices
extracted by osmosis, this effect would not be much increased.
Indeed, if one assumes a monodimensional diffusion of salt
or water, the second Fick law for the particle flux in moles of
molecules per unit area per unit time (J) holds that:
J = – D dc(x,t)/​dx
491
492 Hervé This vo Kientza et al.

80
70
60
50
Mass (g)

40
30
20
10
0
0 20 40 60
Time (days) FIGURE 71.2 Diffusion of a marker (here methylene blue) in two gelatine
gels: 20% (left) and 5% (right) (w/​w). These pictures were taken after 16 h
(a)
and 40 min. The depth reached by the blue colour is only 44 mm for the
80
5% gel.
60
Mass (g)

salt as shown in the previous result is negligible compared with

40
20
0 heating creates slits between bundles of muscular fibres, so that
0 10000 20000 30000 40000
Time (min)
(b)
FIGURE 71.1 Meat stored in salt loses weight very slowly, but there are
two regimes: first a slow loss, and then an even slower loss (a). In the second
picture (b), the beginning of the experiment is shown in more detail, because
the time taken is similar to that for grilling.
where c(x,t) is the concentration at distance x and time t, with the
diffusion coefficient:
D = (1/​f) kB T
where kB is the Boltzman constant, T the absolute temperature
and f the frictional coefficient. At 298 K, D = 1.33 × 10–​5 cm2 s−1
for Na+, so that it would reach only 1.7 × 10−5 cm2 s−1 at 373 K
(Atkins, 1990).
This can be compared to salt diffusion into a gel, as plant
or animal tissues are indeed complex gels (see the chapter on
this); when a strong gelatine gel is made from 1 g food gelatine
dissolved in 250 g of water, coffee powder or methylene blue
used as a colour marker diffuses in the gel at a maximum vel-
ocity of about 15 mm per day at 18 °C (Figure 71.2). For a 5 min
grilling process, assuming the speed is not changed, the penetra-
tion depth of salt would be only 0.05 mm.
Of course, such crude calculations are not enough, as meat is
cooked at more than 100 °C (it should be noted, however, that in
all parts except the outside very thin layer, temperature is limited
at about 100 °C because water is present). However, in this case,
the introduction of salt should be even more difficult, because
collagen and sarcomere shrinkage during meat cooking expels
juices (Obuz and Dikeman, 2003; Kopp et al., 1977; Lepetit
et al., 2000). It appears that at room temperature, the effect of
the juice loss by collagen contraction (Figure 71.3) (This, 1998).
Moreover, Figure 71.4 shows that a much faster capillary
effect can occur during grilling; the contraction of meat under
any salted solution created by the dissolution of salt crystals in
the juices at the upper surface is very rapidly absorbed by capil-
larity; for example, for a viscous silicon oil, the capillary motion
occurs at about 1 mm/​s (Quéré, 1997; De Gennes et al., 2013).
How Much Salt Goes into Meat during Grilling?
As the various effects can have contradictory results, a direct
SEM determination of the salt in meat cooked in culinary
conditions was performed. Beef meat (Charal Inc., total mass
316 g, Longissimus dorsi) was cut perpendicularly to the axis
of muscular fibres into two slices of thickness between 1.0 and
1.2 cm each. At the opening of the bag, it was observed that some
juice was lost. The two slices were dried with absorbent paper
and kept at room temperature until thermal equilibrium. Each
slice was divided into two identical parts (left part of slice A:
sample 1; right part of slice A: sample 2; left part of slice B:
sample 3; right part of slice B: sample 4), and the raw meat was
weighed. Exactly 2 min before grilling, samples 1 and 2 were
salted on one side (1.000 g of the same salt as used previously).
Samples 3 and 4 were salted after grilling, also on one side (same
quantity of the same salt).
Grilling was performed in a non-​stick pan (diameter 25 cm),
without fat, heated by a Scholtes electric stove at full power
(1500 W), pre-​heated for 20 min.
For the two samples (1 and 3) that were salted before grilling:
salt was deposited on only one side of the meat, with a dis-
tribution as regular as possible;
2 min after salt deposition, meat was cooked for 60 s on
each side, salted side first;
after grilling, there was a 120 s rest on a plate with a previ-
ously weighed filter paper to collect the juice, and meat
and juice were weighed.
Salt: When Should Salt Be Added to Meat? 493

Stock cooking
400

350

300

250

Mass (g)
200

150

100

50

0
0 50 100 150 200 250 300
Time (min)

FIGURE 71.3 Two identical pieces of the same muscular tissue were put either in boiling water (100 °C, pink squares) or in cold water that was put to the
boil (blue solid diamonds) (the boiling temperature was reached after 51 min). It can be observed that the increase in temperature triggers the contraction of
collagenic tissue and the loss of juices by 20–​30%, depending on the meat.
(This, 1995)

80

70

60

50
Mass (g)

40
FIGURE 71.4 For some meats, the collagenic contraction creates crevices,
into which the salty juice can be sucked by capillarity. 30

For the two samples (2 and 4) that were salted after grilling: 20

grilling was for 60 s on each side;

salt was regularly deposited on one side of the meat, after
meat was put on a plate with filter paper;
meat was turned to the other side, followed by a 120 s rest,
and finally;
meat and juice were weighed.
When the meat was cooked, fragments of each sample were cut
for SEM study with X-​ray analysis. The scanning electron micro-
scope used (Model JEOL JSM-​5200) had a maximum magnifi-
cation of 100,000. The X-​ray analysis by energy dispersion was
done using a Link ISIS system by Oxford Instruments (Charlot
et al., 2007; Roland and Haschke, 2007). The mass measurements
in the second experiment (grilling) are given in Figure 71.5.
For SEM with X-​ray analysis, the observations are difficult
because of the high water content in meat. The mass proportions
(in %) are given in Table 71.1.
10
0
0 1 2 3
Samples
FIGURE 71.5 This chart shows, for samples 1 to 4, the mass of meat before
cooking, the mass of salt added, the total mass before cooking, the mass after
cooking, the mass of juices collected, the total mass after cooking, the mass
lost by meat and the mass loss in %.
The grilling process chosen for being as close as possible to
the one cooks use in their kitchens generates a meat with red
inside. When salted before cooking, a grey outside was obtained,
in contrast to the meat salted after grilling, which was darker. The
two meats dried slowly after cooking.
494 Hervé This vo Kientza et al.

Charlot F, Crisci A, Maniguet L, Robaut R, Roussel-​Dherbey F. 2007.

TABLE 71.1 Le point sur la microanalyse X, Spectra Analyse, 256, 22–​28.
Mass Proportion of Various Elements in Grilled Meat De Gennes PG, Brochard Wyart F, Quéré D. 2013. Capillarity and
Wetting Phenomena: Drops, Bubbles, Pearls, Waves, Springer,
Sample 1 Sample 2 Sample 3 Sample 4 Raw meat Heidelberg, Germany.
Oxygen 77.55 90.01 95.87 90.61 83.97 Girard JP (ed.). 1990. Technologie de la viande et des produits
Sodium 9.91 5.78 0.79 4.64 8.13 carnés, Lavoisier Tec et Doc, Paris.
Phosphorus 2.60 2.87 1.74 1.37 1.7 Kopp J, Sale P, Bonnet Y. 1977. Contractomètre pour l’étude
Sulfur 2.13 0.94 1.10 1.41 1.21
des propriétés physiques des fibres conjonctives: tension
isométrique, degré de réticulation, relaxation. Canadian
Chlorine 7.81 0.4 0.50 1.97 4.99
Institute of Food Science and Technology Journal, 10(1), 69–​72.
Kuehne D. 1988. Analysis of common salt and sodium in meat
products. Meat, 68(8), 1007–​1010.
Samples 1 and 2 had the same treatment, but they do not Laroche M. 1988. Technologie de la viande et des produits carnés
contain the same quantity of salt. It is assumed that the more (JP Girard ed.), Editions Technique et Documentation
important juice loss of sample 2 eliminated some salt. Lavoisier, Paris.
Lepetit J, Grajales A, Favier R. 2000. Modelling the effect of
Meat salted before grilling contained less salt than raw meat,
sarcomere length on collagen thermal shortening in cooked
and they seem less salted than the others (Wirth, 1991). Possibly, meat: Consequence on meat toughness. Meat Science, 54(3),
the juices they lost dissolved the salt out. 239–​250.
This shows that, in the specific conditions of our experiments, Obuz E, Dikeman ME. 2003. Effects of cooking beef muscles from
salt put on meat before grilling does not enter more than 3 mm frozen or thawed states on cooking traits and palatability. Meat
inside the meat (i.e., the natural roughness of the meat): the Science, 65(3), 993–​997.
culinary idea is thus refuted. Ollitrault R, Rivet J, Flahaut J, Daher H, Khanafer M. 1998. Journal
This result is easy to interpret, as meat is made from mus- of the Less Common Metals, 138, 241.
Quéré D. 1997. Inertial capillarity. Europhysics Letters, 39(5),
cular fibres coated with collagen-​rich tissue (Girard, 1988). Due
533–​538.
to this structure, salt penetration should be different depending Robuchon J. 1993. Les dimanches de Joël Robuchon, Le chêne,
on whether meat is cut perpendicular or parallel to the muscular Paris, 170.
fibres (Laroche, 1998). Roland P, Haschke M. 2007. Microfluorescence X sur microscope
Finally, we have to add that Table 71.1 does not include électronique à balayage, Spectra Analyse, 256(7), 44–​46.
interesting results from previous experiments showing that meat This H. 1995. La gastronomie moléculaire et physique, PhD disserta-
cooked directly in aluminium pans contains some aluminium, tion, University Paris VI, Paris, France.
This H. 2000. Compte rendu du Séminaire de Gastronomie
compared with the same meats cooked in the same aluminium
moléculaire N°2, www2.agroparistech.fr/​-​Les-​premiers-
pans but with some oil or salt added to the pan before putting seminaires-​de-​.html
in the meat. Salt and oil seem to make a protective crust against This H. 2005. Modelisation of dishes and exploration of culinary
metal migration into the meat. “precisions”: the two issues of Molecular Gastronomy, British
Journal of Nutrition, 93, Suppl. 1, S139–​S146.
REFERENCES This H, Gagnaire P. 2008. Cooking, a Quintessential Art, The
University Press of California, Berkeley, California.
Atkins PW. 1990. Physical chemistry, W. H. Freeman and Co, Wirth F. 1991. Reducing the fat and sodium content of meat
New York, 763. products. What possibilities are there? Fleischwirtschaft,
Bernardi M. 1853. Le cuisinier national de la ville et de la campagne 71(3), 294–​297.
(ex Cuisinier royal), Gustave Barbu, Paris, 101. Zhou Shi L, Wang G. 1998. Effects of cooking methods on iodine
Bocuse P. 1976. La cuisine du marché, Flammarion, Paris, 165. content in iodized salt, Wei Sheng Yan Jiu, 27(6), 412–​414.
Bretonnel S. 1896. Etude raisonnée de la cuisine, Editions Foucher,
Paris, 77.
Sauces

Hervé This vo Kientza1,2

1 INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy, F-​75005,

Paris, France

In the kitchen, sauces are important for many reasons, such as or partially gelatinized granules with a solid core
adding a liquid in order to improve swallowing, or adding fla- (S@(W/​S)).
vour to a dish (Prinz, 1997). They have been used since antiquity
(Apicius, 390), and in the past, various classifications were Finally, 23 categories were found. In order of complexity, they
proposed, such as that of Carême (1833), with “mother sauces” are:
and sauces being derived from them. Of course, the number
of possible sauces is infinite, because any ingredient can be W
chosen for imparting a particular flavour to a sauce, but, in the O
past, sauces were codified, and variations were considered with W/​S
caution. O/​W
In the decades before the advent of molecular and physical S/​W
gastronomy, chefs making sauces often referred to the text- (O+S)/​W
book published by Th. Gringoire and L. Saulnier (Gringoire (W/​S)/​W
O+(W/​S)
and Saulnier, 1914), created from the Guide culinaire (Escoffier
(G+O)/​W (Figure 72.1)
et al., 1903). However, this was based only on processes of sauce
(G+O+S)/​W
making and not intrinsically based on the physical and chemical
(O+(W/​S))/​W
constitution of the produced systems. The introduction of the dis-
(S+(W/​S))/​W
perse system formalism (DSF) (see the corresponding chapter in ((W+S)/​O)/​S
this book) made it possible to recognize some order in the long (O+S+(W/​S))/​W
list of French classical sauces (and from any other food culture), ((W/​S)+(W@S))/​W
and also to create entirely new systems. (O + (W/​S)/​W)/​S
For French sauces, the research of the DSF formula was ((O+(W/​S))/​W)/​S
published (This, 2004; 2005). The corpus used for the physico-​ (O/​W) + ((G+O)/​W)
chemical classification was compiled from classical or official (O+(W/​S)+(W@S))/​W
culinary books (Académie des gastronomes, 1991; Gringoire (S+(W/​S)+(W@S))/​W
and Saulnier, 1914; Escoffier et al., 1903). Each sauce was made (((W/​S)+(W@S))/​W)/​S
strictly following the published recipe and analysed using optical (O+S+(W/​S)+(W@S))/​W
microscopy (size cut-​off d > 10–​5 m). For establishing the DSF (O+S+((G+O)/​W))/​W
formula, assumptions were made, such as:
Surprisingly, some simple types are missing, such as “foamed
(1) thermally treated plant samples dissociate into sep- veloutés” having the formula (G+(W/​S))/​W. Such systems are
arate cells, so that cell aggregates (W/​S) can be not difficult to produce practically, and their absence from the
formed as well as individual cells (W@S); traditional repertoire leads to the question of why such sauces
(2) culinary filtration does not separate small cell were not “invented” or recognized officially by cooks in the past.
aggregates; This led to a separate study on the number of different physical
(3) grinding animal or plant tissues disintegrates a cer- kinds of sauces as a function of time (This, 2005). For this study,
tain proportion of cells; some traditional French culinary books were used (Menon, 1755;
(4) depending on the processing time, different results Marin, 1742). The increasing number of types of sauces with
can be obtained when starch granules are heated time shows that culinary empiricism has probably not yet had
in water, such as fully gelatinized granules (W/​S) enough time to develop all possible kinds of sauces.

495
496
FIGURE 72.1 Microscope picture of an emulsified foam obtained by
whipping oil in an egg white. The DSF formula is (G+O)/​W. Such systems
can also be obtained by mixing a whipped egg white and a mayonnaise. The
proportion of air bubbles (disks with dark edge) and oil droplets can be any-
where between 0 and 1.
How to Use the New Classification?
Let us now consider technological consequences of the previous
result. As the sauce called aioli is an emulsion that is obtained
by grinding garlic with oil, it was proposed in 1993 that other
emulsions of the same kind could be made using the same pro-
cess, but with other food ingredients, such as animal or plant
tissues, because they all contain water, as well as phospholipids
from membranes, and proteins (This, 1995). Some examples of
such emulsions were known (hummus, aubergine caviar), but
most possibilities had not been explored. The general name given
for these emulsions was “olis”.
Another idea was proposed in 1995, i.e., a new dish named
“Chantilly chocolate” (This, 1996; This, 2008). This one was not
based on a generalization of ingredients, but of process: it was a
generalization of whipping cream (Montagné, 1996).
Indeed, milk cream is primarily made of fat globules
dispersed in a water phase (with an appropriate reference
size, casein micelles not taken into account in its formula)
(Montagné, 1996). It is sometimes described as “oil-​in-​water
emulsion resulting from the concentration of milk”, but this
description is inaccurate, as part of the fat is solid at room
temperature (Michalski et al., 2004); hence, a formula such as
f(O,S)/​W should be preferred to O/​W, the function f(O,S) of
O and S being as yet unknown, as it is not established whether
f(O,S) is equal to S@O or to O/​S.
Anyway, whipping cream can be described by the equation:
f(O,S)/​W + G → [G + f(O,S)]/​W
Looking for formulae is an invitation to changes: O, it was said,
can be any liquid fat, W any aqueous solution and G any gas.
In particular, “Chantilly chocolate” is obtained when, starting
from a chocolate emulsion made of chocolate dispersed in
water, whipping is performed while cooling below a tempera-
ture of 34 °C (the lecithin in chocolate is the surfactant, but if
needed, other compounds can be added, e.g., gelatine). When
Hervé This vo Kientza
the proportions are well chosen, the final consistency can be as
aerated as whipped cream/​Chantilly cream.
Alternatively, Chantilly foie gras, Chantilly cheese, Chantilly
butter, Chantilly brown butter or even “Chantilly olive oil” can
be made (when cooling is sufficient, for olive oil) from foie gras,
cheese, butter, brown butter and olive oil, respectively.
In practice, making these products is easy. For example, with
chocolate:
1. first make a chocolate emulsion O/​W by heating choc-
olate into a water phase (the proportion of chocolate
and water has to be chosen so that the final fat/​water
ratio is about the same as the fat/​water ratio in ordinary
cream);
2. then whip (+G) at room temperature while the emulsion
is cooled: after some time (some minutes, depending
of the efficiency of the cooling), a “chocolate mousse”
[G + f(O,S)]/​W is obtained.
This mousse is not a ganache montée, as it contains no cream,
and it was initially considered “impossible” by some of the most
famous pastry chefs (Thuries, 1995) because water and melted
chocolate were mixed, making chefs fear graining (but it has to
be observed that water is not added to melted chocolate; rather,
chocolate is dispersed in water). It needs no eggs, contrary to
traditional chocolate mousse, which can be an advantage for
people having food allergies; given the right amount of water, the
texture can be the same as that of whipped cream. As whipped
cream is called “Chantilly cream” when sugar is present, the
name “Chantilly chocolate” was given to the new dish made from
chocolate.
Another example can show how formulae can lead to new
systems having scientific and culinary interest. For example,
the formula O/​W can be made with oil dispersed in water, using
surfactants. In particular, when one whips oil in an aqueous
solution of gelatine, the emulsion obtained gels slowly; a jelli-
fied emulsion called a “liebig”, in honour of the German chemist
Justus von Liebig (Figure 72.2) is obtained, according to the
FIGURE 72.2 A “liebig” made by chef Pierre Gagnaire in his Paris
restaurant.
Sauces
following equation (where gelation goes with partial coalescence
of the dispersed oil droplets):
D0(O)/D3(W) → [D0(O)/D3(W)]×D3(S)
Here, the gel is a physical gel, but a chemical gel can also be
obtained by using egg white instead of the aqueous solution
of gelatine, as the heated proteins can link through disulfide
bridges. The emulsion obtained before coagulation was called a
“geoffroy”, in honour of the French dynasty of chemists of this
name, and the chemically jellified system was called a “gibbs”, in
honour of the American physical chemist Josiah Willard Gibbs.
Finally, a new dish called a “faraday” (in honour of the great
British physical chemist) was introduced as a demonstration that
physical systems could be made from formulae (here [(G+O+S1)/​
W]/​S2). The number of possibilities is innumerable. It can be
easily calculated that, using four phases and four operators,
the number of formulas is 114,688, and more than 106 with six
phases: there is plenty of room for innovation!
REFERENCES
Académie des gastronomes et Académie culinaire de France. 1991.
L’art des sauces, J. T. Lanore, Paris.
Apicius. 390. De Re Coquinaria. http://​penelope.uchicago.edu/​
Thayer/​E/​Roman/​Texts/​Apicius/​10*.html.
Carême MA. 1833. L’art de la cuisine française au XIXe siècle.
Kerangue et Pollès, Paris.
497
Escoffier E, Gilbert Ph, Fetu E, Suzanne A, Reboul B, Dietrich Ch,
Caillat A. 1903. Guide culinaire. Bibliothèque professionnelle,
Paris.
Gringoire Th, Saulnier L. 1914. Répertoire de cuisine. Flammarion,
Paris.
Montagné P (ed.). 1996. Larousse Gastronomique. Larousse-​Bordas,
Paris.
Marin M. 1742. Les dons de Comus (3 vol., facsimile of the 1742
edition published in Paris). Manucius, Pau, 2001.
Menon M. 1755. Les soupers de la cour, ou l’art de travailler toutes
sortes d’aliments. Guillyn, Paris.
Michalski MC, Ollivon M, Briard V, Leconte N, Lopez C. 2004.
Native fat globules of different sizes selected from raw milk:
Thermal and structural behavior. Chemistry and Physics of
Lipids, 132(2), 247–​261.
Prinz J, Lucas PW. 1997. An optimization model for mastication and
swallowing in mammals, Proceedings of the Royal Society of
London. B, 264, 1715–​1721.
This H. 1995. Révélations gastronomiques, Belin, Paris.
This H. 1996. Le chocolat Chantilly. Pour la Science, 230(12), 6.
This H. 2004. 14 types de sauces. Pour la Science, 317, 4.
This H. 2004. Molecular Gastronomy. The World of Food Ingredients,
4–​5, 22–​35.
This H. 2005. Modelling dishes and exploring culinary “precisions”:
The two issues of Molecular Gastronomy. British Journal of
Nutrition, 93(4), S139–​S146.
This H. 2008. Dans la famille “mousses au chocolat” … Actualité
Chimique, 319, 10–​13.
Thuries Y. 1995. Press conference in Carmaux, France (21 June).
Sauces: Hollandaise Sauce
Guro Helgesdotter Rognså
Nofima Processing Technology, Måltidets Hus, Richard Johnsens gt. 4, Po box 8034, 4068 Stavanger, Norway
Hollandaise sauce is one of our most beloved and elegant butter
sauces. It has the power to lift a piece of baked fish, asparagus
or a poached egg to something luxurious. Its history is long, and
it is one of the sauces chefs across the world find most precious.
The reason may be its versatility, but undoubtedly it has also
to do with the sauce’s velvety texture. One interesting point is
that the average person normally does not make it from scratch
at home. This probably has to do with the sauce’s reputation of
being difficult to make. In addition, the sauce contains a high
amount of butter, which may make it unattractive to modern
people on a diet. However, hollandaise sauce may be the compo-
nent that provides a dish with that something extra, and because
of this we are devoting a chapter to it. Hollandaise sauce may
have “a temper”, but by understanding its nature with some sci-
entific insights, it is probable that you will succeed with your
future butter sauces.
Hollandaise Sauce History
Hollandaise sauce is an old sauce –​in fact, so old that historians
do not agree on its origin, or why or where it got its name.
Hollandaise is French and means Dutch or made “Dutch-​style”.
Some believe that the sauce was originally developed in the
Netherlands and then imported to France. The first known recipe
resembling hollandaise sauce is presented in a Dutch cookbook
from 1593 by Carel Baten (Battus, 1593). Another account from
1651, by François Pierre La Varenne, describes a sauce made
“with good fresh butter, a bit of vinegar, salt and nutmeg, and
an egg yolk to bind the sauce” (La Varenne, 1651). This sauce
was intended for cooked asparagus. There are also other early
reports of sauces resembling hollandaise sauce and being thus
named, but they also contain other ingredients than what we con-
sider typical hollandaise sauce ingredients today. So, the question
is –​what sauce composition variations are acceptable in order
to call a sauce a hollandaise sauce? Historians do not agree on
this point either, and therefore we cannot conclude on the sauce’s
origin and why it got its name, although many hypotheses have
been made. The only thing we can be sure of is that hollandaise
sauce has a long history.
Due to the sauce’s long history, both preparation techniques
and ingredient selection have somewhat changed over time,
which is the case for many types of food with a long history.
For example, many of the old recipes for hollandaise sauce con-
tain different types of spices, but spices such as nutmeg are nor-
mally not used in hollandaise sauce today. It is important to have
in mind that foods are not static products. They continuously
adapt to the ingredients they are made of, available preparation
equipment and food trends.
There are different variations of the preparation method for
hollandaise sauce. Old recipes yield sauces with denser and more
custard-​like textures compared with modern standards, because
in these recipes the yolks are heat treated just with stirring (not
whisking) together with vinegar and/​or butter from the start.
Today, chefs normally base the sauce on the warm and stiff foam
of egg yolks, called a sabayon, which gives the sauce a much
airier texture and makes the mouthfeel lighter.
Mother and Daughter Sauces
Auguste Escoffier, the famous French chef born in 1846, included
hollandaise sauce as a mother sauce (also called a primary or
leading sauce) in the French sauce system (Escoffier, 1907), a
work he continued after another famous French chef, Marie
Antonin Carême (Carême, 1854). A sauce derived from one
of these mother sauces is called a small, secondary or deriva-
tive sauce. Hollandaise sauce has many derivative sauces, and
béarnaise sauce is the best known. Béarnaise is more or less a
hollandaise sauce flavoured with chopped tarragon and chervil,
and its wine reduction also contains chopped shallots. The old
notion of mother sauces is no longer used extensively, and hol-
landaise sauce is nowadays characterized as one of two primary
sauces of the sauce class “butter sauces” (Larousse, 1993). The
other main butter sauce is beurre blanc, which does not contain
egg yolks.
A Warm Emulsion
A hollandaise sauce is, among other things, a warm egg yolk-​
based emulsion. Many of our most indulgent foods are made up
either partly or purely of emulsions, and for other products, an
emulsion has been an important step in the preparation process
499
500
(McClements, 2005). Product examples are milk, cream, butter,
mayonnaises and remoulades, dressings, vinaigrettes, soups,
cakes, cream liquors, ice cream, desserts and sauces.
We are now going to have a closer look at emulsion science,
because it is an important point in order to understand the
behavior of warm butter sauces. By making emulsions, we are
unfortunately working against “the natural order” of things, as
emulsions are “programmed” to separate, which they may do in
a number of different ways. There is unfortunately never a good
time for emulsion separation, especially if you are expecting
guests, which is why learning how to control emulsions can lead
to less stress in the kitchen.
Mixing of the Immiscible
Many emulsions, especially the warm ones, have a reputation
for being difficult both to make and to keep stable. The reason
lies hidden in the emulsions’ structure. Emulsions are mixes
of two main ingredients –​water and fat in the form of oils or
melted butter. As we know, water and oils are immiscible liquids,
but when we make emulsions, what we do is to mix these two
phases. The way it is done is by breaking up one of the phases
into small droplets, which are dispersed in the other phase. To the
naked eye, the emulsion looks like a homogeneous mixture of
the phases, but on a microscopic scale, the structure is heteroge-
neous. The mixing of the phases often results in a yellow-​whitish
product, such as mayonnaise, because all the small droplets in the
emulsion scatter the light.
There are two main groups of emulsions: those in which
oil droplets are suspended in water, which are called oil-​in-​
water emulsions (O/​W), and water-​in-​oil emulsions (W/​O), the
direct opposite; see Figure 73.1. Most food emulsions are O/​W
emulsions. Since oil or butter and water are immiscible liquids
due to their different densities, the force in the emulsion system
is towards separation into two distinct phases once more. This is
where we must start to apply our knowledge about emulsions in
order to encourage them to remain stable for a longer time, pref-
erably beyond the time of consumption.
Ratios
The first thing to control when making an emulsion is the ratio
between oil/​butter and water. In an O/​W emulsion, all oil droplets
need to be suspended in water, and the amount of water compared
with oil or melted butter is therefore crucial. Butter sauces and
mayonnaises are made with water-​containing ingredients such as
egg yolks, wine reductions, vinegar, lemon juice and mustard.
When a large amount of oil or butter is used, the water comprised
in these ingredients may not be enough to hold all the oil droplets.
If the emulsion separates as a result, many believe that they have
not used enough egg yolk, but the reality is that one can produce
over 20 liters of stable mayonnaise with only one egg yolk, as
long as supplementary water is added (McGee, 1990). More
information about the emulsifying properties of egg yolks will
follow later.
Guro Helgesdotter Rognså
It is therefore not wise to go far below 20 percent water by
weight compared with oil in an O/​W emulsion, although less
than 10 percent is possible. The limit depends on the droplet size
range. However, as water evaporates slowly from food, even at
room temperature, if it is too close to the water content limit, the
emulsion becomes more unstable. You should therefore have a
certain idea of how much water by weight compared with oil you
have in your emulsion. Unfortunately, it is not always easy to cal-
culate water content based on recipes, as measurement units are
often “spoonfuls” and “halfs”, which are not standardized units.
However, egg yolks contain approximately 50 percent water by
weight. Plant oils and butter fat have lower densities than water
and normally weigh from 0.8 to 0.9 grams per milliliter. Sum
up the amount of water in the ingredients, divide it by the total
weight of all ingredients and multiply by 100 to obtain the water
percentage in the emulsion.
Warm emulsions, such as butter sauces, require a higher water
percentage than cold emulsions, as the evaporation rate of water
is proportional to the temperature. If you are planning to pre-
pare a hollandaise sauce a while before serving, it may be wise
to whisk in some water during the holding time to keep the sauce
stable. If you have questions regarding water content in foods,
there are food databases available online where this kind of infor-
mation is readily available (USDA, nd). Just remember that there
may be variations from one product brand to another and between
products from different countries.
The Art of Mixing
An important type of emulsion separation happens when the
oil droplets in the O/​W emulsion merge and grow larger until
all droplets form one phase and the emulsion has completely
separated. This type of emulsion separation is called coales-
cence. If few and large droplets are present in the emulsion, it
will normally separate quite quickly. However, if each droplet
is smaller, the process will take longer, which is why mixing
an emulsion vigorously or for a longer time can drastically
increase the stability. Today we have different types of kitchen
machines at our disposal, like immersion blenders or stand
mixers, which can relieve our arms and make the droplets in
emulsions much smaller than by hand. Warm butter sauces are
still normally prepared by hand, but there is also equipment
today that allows whisking while heating, such as induction
stand mixers or induction blenders. Unfortunately, they are
often quite expensive. In Figure 73.1 you will find schematic
information about emulsion structure and the different types of
emulsion separation.
Stabilizers and Emulsifiers
Gravity is a major driving force in the separation of emulsions.
Because water is denser than oil, it slowly moves down-
wards, while the oil droplets travel to the top, like the cream
in unhomogenized milk. This type of emulsion separation is
called creaming (see Figure 73.1). Gravitational movement in
Sauces: Hollandaise Sauce 501

Oil

Water

Emulsion
formation by
mechanical treatment
Dispersed
phase

Continuous
phase

Interphase

Phase
inversion

Oil-in-water emulsion
(O/W)

Water-in-oil
emulsion (W/O)

Emulsion separation

Sedimentation Creaming Flocculation Coalescence

FIGURE 73.1 Emulsion structure and separation. Oil-​in-​water (O/​W) and water-​in-​oil (W/​O) emulsions are made by mechanical treatment. Most food
emulsions are O/​W emulsions. Three terms are used to characterize emulsion structure: the continuous phase, the dispersed phase and the interface (where the
two phases meet). An emulsion can separate in a number of ways: sedimentation, creaming, flocculation and coalescence.
(Figure adapted from McClements, 2005)

the emulsion should therefore be limited in order to increase
the stability and prolong an emulsion’s life. By limiting the
movement in emulsions, the number of oil droplet interactions
are also decreased. Increasing the viscosity or thickness of the
water phase provides this result. Products with such abilities are
called stabilizers in emulsion terminology. Everyday examples of
stabilizers are thickeners such as flour, potato flour, cornstarch or
gelatin, but the list of available stabilizers for the industry is long.
Although the term may not seem particularly appetizing, an
emulsifier is just a name given to a group of compounds able
502
to stabilize emulsions. Many ingredients naturally contain
emulsifiers, such as egg yolks and milk. Emulsifiers are able
to adsorb at the emulsion interface (see Figure 73.1), which is
where the oil and water phases meet. At the interface, the emul-
sifier covers the droplets like a protective film and ensures that
droplets coming into contact with each other do not merge and
create larger ones (by electrostatic repulsion). In this way, the
emulsifiers make emulsions more stable by preventing aggrega-
tion and coalescence.
Industrial emulsions normally contain both stabilizers and
emulsifiers for increased protection against emulsion separation.
Next time you have a jar of store-​bought mayonnaise in front
of you, take a glimpse at the declaration, and you will probably
find that it contains both emulsifiers and stabilizers. This is also
why numerous homemade emulsions contain egg yolks, in which
both stabilizers and emulsifiers are naturally present. Egg yolk
proteins act as emulsifiers, but they may also stabilize the water
phase, especially if they are heat treated and coagulated (a result
of denaturation, the scientific term for uncoiling of the egg yolk
proteins). The denaturation of egg proteins results in gelling of
the water phase, the same property we exploit when custards are
thickened with egg yolks by heating.
The Droplets’ Life in Solitude
Many of the measures taken to increase emulsion stability are
done to minimize movement and contact between oil droplets in
the system, because droplets that come into contact with each
other may merge as a result. This is also the reason why the
temperature as a general rule should be kept low in emulsions,
and temperature fluctuations should be avoided. However, there
are certain exceptions, such as warm butter sauces, where it is
important to keep the sauce temperature higher than the butter’s
upper melting temperature, normally 35–​40 °C. An explanation
will be given in the section about butter.
Hollandaise Ingredients
We have now gone through why it is essential to know about
emulsion properties when making hollandaise sauce. In the rest
of the chapter, we will have a closer look at hollandaise sauce
ingredients and preparation. Butter sauces may be prepared with
a wide range of flavours. The use of different types of flavours has
led to numerous different emulsion sauces. Hollandaise sauce is
normally made with a white wine and vinegar reduction, egg
yolks, butter, salt, pepper and lemon juice. Some purists have
strict ideas of which ingredients the sauce must contain in order
to be called a hollandaise sauce, but others just add what they
think will make the best sauce. For most chefs, the seasoning
with salt and lemon juice at the end is a necessity for a hol-
landaise sauce, but many also include shallots or bay leaf in the
reduction.
Chefs often disagree on which white wine to use in a wine
reduction. Wine type, acidity or sweetness level does not sig-
nificantly influence the sauce flavor, so just choose a wine you
Guro Helgesdotter Rognså
like. The fat in the sauce will normally mask any sweetness in
the wine, as long as you do not use a dessert wine. The degree
of reduction has much more influence on sauce flavor than the
wine choice (Rognså et al., 2017). The sauce flavor, and some-
times also the sauce texture, is influenced by the ingredients in
the aqueous phase.
Some chefs add salt from the beginning of the sauce-​making,
whereas others just season the sauce with salt before serving it.
The sauce texture is influenced by when you add the salt. Adding
it towards the end may produce a thicker sauce compared with
adding the salt at the start (Rognså, unpublished results). Butter
sauces can easily be made thinner by adding water or lemon
juice, and thicker by whisking in more butter.
The Problematic Butter
Butter is, by percentage, the most important ingredient in hol-
landaise sauce, and it is added to the sauce either in cold cubes,
warm or clarified. The different options do not necessarily result
in the same product, because if whole butter is used, water and
a small portion of milk proteins are also added to the sauce, as
whole butter contains up to 15–​20 percent water and about 1 per-
cent proteins. If you use cold butter, the sauce-​making process
will take longer, as you decrease the temperature of the sauce
each time you add butter to it. These temperature fluctuations
may also result in particle formation in the sauce (Rognså, 2014).
Most chefs today use warm and clarified butter when preparing
hollandaise sauce (Rognså et al., 2017).
Butter is, in fact, composed of many different types of fat, each
of them having a specific melting point. These melting points
vary from approximately −40 to 40 °C (Wright et al., 2001), and
they are responsible for butter being more or less spreadable at
different temperatures. Butter is said to be “semi-​crystalline” in
this temperature range, meaning that some of the fats are pre-
sent as solid fat crystals and some of the fats are liquids at the
same temperature. The composition of the fat is influenced by
numerous factors, including what the cow has eaten and the
cow’s breed (Kalač and Samková, 2010).
These properties of butter are, in fact, very important when
it comes to the stability of the hollandaise sauce. If a holland-
aise sauce is cooled to below approximately 40 °C, fat crystals
start to form in the fat droplets of the emulsion, and they may
even grow through the protective emulsifier film. When droplets
with fat crystals on the surface come into contact, the crystals
may entangle, which results in clotting of fat droplets, an effect
called partial coalescence (McClements, 2005). This is not good
from a stability point of view, because the aggregated droplets are
susceptible to merging, especially if the sauce is reheated above
40 °C. Merging of droplets creates larger droplets of different
sizes in the emulsion, which can quickly lead to separation of the
hollandaise sauce.
It is therefore very important to control the temperature of hol-
landaise sauce if it is supposed to stay stable for some time before
serving. A butter sauce is challenging to prepare long before
serving for several reasons, water evaporation and temperature
fluctuations among them. Butter sauces should, in any case, not
Sauces: Hollandaise Sauce
be made long before serving due to bacterial issues. Holding
temperatures for butter sauces promote bacterial growth, which
is why such sauces should, in any case, not be kept more than
approximately 2 hours before serving (and leftovers should be
discarded. Guidelines vary slightly between countries) (Gisslen
et al., 2006). It is also recommended to use pasteurized egg yolk
in countries where Salmonella contamination in eggs is a con-
tinuous challenge.
Hollandaise Preparation
It is wise to prepare hollandaise sauce in a steel bowl over a
pot containing hot water, as this facilitates temperature control
during the preparation. If you use a pan directly on the stove,
the temperature may become too high, which can result in curd-
ling/​coagulation of the egg yolk proteins, resulting in a grainy
and unappealing sauce. Although elevated temperatures are
necessary to make butter sauces, too high temperatures, in gen-
eral over 75 °C, may lead to large structural changes in the egg
yolk proteins. These temperature-​induced changes may result in
the yolk proteins lumping together in an uncontrolled manner
by coagulation, which is like the scrambling of your eggs for
breakfast. Coagulation of the egg yolk proteins first makes the
water phase thicker, which is a desired effect, but a high degree
of coagulation makes the sauce grainy and lumpy, normally
followed by sauce separation.
As mentioned before, there are several variations of how to
prepare hollandaise sauce, but today, chefs strive to make light-​
textured sauces as a contrast to the high butter content, which
may be achieved by building the hollandaise sauce on a sab-
ayon. A sabayon is a light and relatively stiff foam made from
a wine or vinegar reduction and egg yolks, and it is formed
when the egg yolk mixture is heated while whisking with a
balloon whisk. The temperature in the sabayon during prep-
aration should, as a general rule, not exceed 70 °C. When the
sabayon is made, the structure of some of the yolk proteins is
changed (denatured), which is what increases the thickness of
the foam. When the sabayon has stiffened, the heating is halted.
Then, the butter is incorporated slowly while whisking; warm
(60–​65 °C is recommended), clarified butter is normally used.
Seasoning with lemon juice and salt finishes the sauce. The
sabayon ensures the presence of air bubbles in the hollandaise
sauce (a foam). The aqueous phase of the sauce also contains
coagulated egg yolk proteins, which gels the phase (a suspen-
sion). This means that a hollandaise sauce in reality has the
characteristics of several types of systems: emulsion, foam and
suspension.
Summary of Success Points for Emulsions and
Warm Butter Sauces
Hollandaise sauce and its derivatives are loved because of their
subtle flavor and velvety texture. It is, however, important to
remember that many of the properties of hollandaise sauce are
linked to it being, among other things, an emulsion, and knowing
503
about emulsion characteristics may help you to know what
to do when preparing the sauce, or rather, what not to do. By
following these principles, you should be able to minimize the
times emulsions separate in your kitchen. Hopefully, knowledge
about these properties may encourage more people to make butter
sauces and emulsions at home.
To recapitulate the highlights for stable emulsions: When
making emulsions at home, you should first of all check and
be sure that you are using enough water. Do not just blindly
trust the recipe, as its writer may not be aware of the import-
ance of the water ratio in O/​W emulsions. If you do not want
to calculate, just add a bit more water, vinegar, lemon juice or
another water-​based ingredient to be on the safe side. Then mix
the emulsion –​thoroughly. Use an immersion blender or a stand
mixer if you want to keep your emulsion stable for longer. Hand-​
made emulsions may keep together sufficiently long if you also
use ingredients containing emulsifiers and/​or stabilizers, such
as egg yolks. Finally, maintain stable and low temperatures to
keep the emulsion stable, except for butter sauces, which require
higher holding temperatures. For butter sauces, such as holland-
aise sauce, keep the sauce above 40 °C, but make the sauce close
to serving time, as you do not want to serve your guests a food
poisoning together with the dinner.
Butter sauce-​making may not be an easy task, but practice
makes perfect. Making delicious hollandaise sauce is considered
an art by many chefs. By knowing these important points about
hollandaise sauce and emulsion properties, you may avoid the
most common pitfalls.
Hollandaise Sauce Recipe
Base:
4 egg yolks from large eggs (approximately 80 grams
in total)
40 grams liquid (water, white wine vinegar, dry white
wine, lemon juice or a reduction (see following recipe))
220 grams unsalted and clarified butter (from 300 grams
whole unsalted butter)
White wine and vinegar reduction:
60 grams dry white wine
60 grams white wine vinegar
15 peppercorns
1 shallot
Flavorings:
Salt and freshly ground pepper
Lemon juice
1. If you want to use a reduction for extra flavor in the
sauce, you start by making the wine reduction. If you
prefer only to use water, jump to step 2. For the reduc-
tion, start by finely chopping the shallot. Put all the
ingredients in a saucepan and reduce till you have
approximately 40 grams of liquid. Pour the reduction
504
through a sieve before use to eliminate the peppercorns
and chopped shallot.
2. Put about 300 grams whole butter in a pot and melt it
slowly over medium heat. When the butter has melted,
the water will be at the bottom of the pan, and you
can easily obtain 220 grams clarified butter. Keep the
butter warm, approximately 65 °C.
3. Isolate the yolks and place them in a heat-​resistant
bowl. Add 40 grams of liquid (wine reduction,
water and/​or lemon juice) and start whisking. It is
recommended to use a balloon whisk for this task.
4. Place the steel bowl over a pot containing hot water
(close to the boiling point), but make sure that the
bowl is not in contact with the water. Whip the sab-
ayon forcefully until it stiffens. You should not exceed
70 °C in the foam to avoid curdling of the yolks. An
average yolk contains about 50 percent water, and the
total water content in this recipe is therefore about
23 percent water (before evaporation), which will
ensure that the emulsion does not separate because of
too little water.
5. When the sabayon is stiff, remove the bowl from the
water bath, and start to incorporate warm (60–​65 °C)
clarified butter while whisking. Start adding the butter
carefully in the beginning.
6. When all the butter is incorporated, flavor the sauce
with lemon juice and salt according to your own taste,
and serve.
You have now made a hollandaise sauce. The sauce may be
developed further by adding ingredients such as tarragon and/
or chervil to obtain a béarnaise sauce, caviar or löjrom for an
exquisite sauce for baked fish, or beurre noisette for a complexly
flavored sauce to serve with vegetables, fish or meat. Bon appétit!
Guro Helgesdotter Rognså
REFERENCES
Battus C. 1593. Cock boeck. The Netherlands: Medecyn Boeck.
Carême MA. 1854. L’art de la cuisine française au dix-​neuvième
siècle, Tome III (Unabridged facsimile 2006 ed.). Paris:
Elibron Classics.
Escoffier A. 1907. A Guide To Modern Cookery. London: William
Heinemann.
Food Standards Australia and New Zealand. 2016. The use of time
control for potentially hazardous food, Appendix 2 3rd ed.
Gisslen W, Griffin ME, Le Cordon Bleu. 2006. Professional Cooking
for Canadian Chefs. Hoboken, NJ: John Wiley & Sons.
Kalač C, Samková E. 2010. The effects of feeding various forages
on fatty acid composition of bovine milk fat: A review. Czech
Journal of Animal Science, 12, 521–​537.
La Varenne FP. 1651. Le cuisiner françois. Lyon.
Larousse DP. 1993. The Sauce Bible –​Guide to the Saucier’s Craft.
John Wiley & Sons.
McClements DJ. 2005. Food Emulsions –​Principles, Practices, and
Techniques, 2nd ed. CRC Press.
McGee H. 1990. The Curious Cook –​More Kitchen Science and
Lore. North Point Press, 19.
Rognså GH, Rathe M, Paulsen MT, Petersen MA, Brüggemann DA,
Sivertsvik M, Risbo J. 2014. Preparation methods influence
gastronomical outcome of hollandaise sauce. International
Journal of Gastronomy and Food Science, 2, 32–​45.
Rognså GH, Rathe M, Petersen MA, Misje K-​E, Brüggemann DA,
Hersleth M, Risbo J. 2017. From wine to hollandaise sauce:
Does the nature of the wine or wine reduction influence sen-
sory attributes? International Journal of Gastronomy and Food
Science, 9, 75–​87.
This H. 2007. Kitchen Mysteries. Columbia University Press.
USDA. nd. Food Data Central by US Department of Agriculture, fdc.
nal.usda.gov (last access 18 June 2020).
Wright AJ, Scanlon MG, Hartel RW, Marangoni AG. 2001.
Rheological properties of milkfat and butter. Journal of Food
Science, 66, 1056–​1107.
Sauces and Purées: The Underside of Applesauce
Cassandre Leverrier
UMR SayFood, AgroParisTech, INRAE, Université Paris-​Saclay, Massy 91300, France
Apple puree is a common product that we consume from
childhood to old age. Everyone is able to cook applesauce, but the
best ones are certainly our grandmothers’! What do they intui-
tively know that we don’t?
Let’s start with a simple definition of applesauce. Applesauce
is defined as a fruit puree, which is made of apples (corresponding
to the standard of Malus domestica Borkhausen fruit). Processing
to obtain fruit purees is simple: fruits are peeled and seeded, and
their flesh is then cooked, with or without added sugar. Finally,
cooked fruits are sieved or ground through mechanical treatment
to obtain the fruit puree (de Reynal, 2008). This apparently simple
process is, however, able to give a wide range of textures of the
final product: from very smooth to granular textures, from very
liquid to highly viscous consistencies, with or without pieces. To
understand where the texture of applesauce comes from, and what
are the parameters that play a role in it, let’s go a step further.
Apple Fruit: an Organized Structure from
Macroscopic to Microscopic Scale
At the macroscopic scale, apple is composed of three main
tissues: the epidermis (most peripheral zone of the fruit, the
“skin”), the parenchyma (reserve area of the fruit, the “flesh”)
and the endocarp (the deepest tissue, which contains the seeds)
(Figure 74.1). Edible plant tissues are usually rich in parenchy-
matous cells, which are thin-​walled and nonlignified (Waldron
et al., 2003). Fruit tissues are composed of plant cells, the size
and shape of which depend on the fruit type, the variety and their
localization in the fruit. Indeed, plant cells can have various sizes
and shapes within one tissue. The distribution of the cells in apple
parenchyma is called anisotropic: parenchyma cells are smaller
and round at the periphery of the parenchyma, and bigger and
radially elongated in the centre of the parenchyma (Khan and
Vincent, 1990). Apple parenchyma cell size can be from several
dozens of micrometres to 300 micrometres.
Plant cells are delimited by a plant cell wall (Figure 73.2), which
has a fundamental role in the structure of the cells, since it gives
them their shape and rigidity. Plant cell walls are the major source
of dietary fibre. Cell walls are not uniform either: their thickness,
their composition and their rigidity depend on their role within the
plant tissue, and they have an impact on accessibility to digestion,
thick cell walls being less accessible than thin ones (McDougall
et al., 1996). The plant cell wall is structured in three parts: the
middle lamella, the primary cell wall and the secondary cell wall.
The middle lamella, mainly composed of pectins, ensures that the
cells cohere together (Jarvis, 1998; Jarvis et al., 2003).
Plant cell walls are composed of four main components
(McDougall et al., 1996):
• cellulose
• non-​cellulosic polysaccharides (mainly hemicellulose
and pectins)
• proteins
• phenolic compounds.
Polysaccharides (cellulose, hemicellulose and pectins) represent
up to 90–​95% of the components of the cell plant wall, the
remaining 5–​10% being proteins, glycoproteins and phenolic
compounds (Waldron et al., 2003; Kojima et al., 2004).
Apple fruit texture is known to depend on several parameters,
such as the mechanical properties of apple tissues, the organ-
ization of the cells within the tissues, the turgor pressure of the
cells and the thickness of the cell walls (Waldron et al., 2003;
Bourles, 2010). The rigidity of the plant wall is highly influenced
by the cellulo-​pectic complexes in the wall and therefore by the
structural characteristics of pectins. The mechanical properties of
the walls result from the viscoelastic properties of cellulose and
its interactions with other polymers in the wall, particularly the
structure of pectins (Agoda-​Tandjawa et al., 2012). The textural
characteristics of the raw apple fruit are directly linked to their
structure, the organization of their tissues and the composition of
their cell walls (Figure 74.3).
From Apple Fruit to Applesauce:
Thermo-​Mechanical Treatment
To obtain fruit purees, thermo-​mechanical treatments are com-
monly used. The fruits are therefore heat-​treated (cooked) before
being pureed by mechanical treatment (either grinding or passing
through a refining grid).
Heat treatment, in addition to extending the shelf life of the
product, allows the texture of the raw fruit to be modified. Two
505
506
FIGURE 74.1 Apple parenchyma: scheme and scanning electron microscopy.
(Khan and Vincent, 1990; 1993)
FIGURE 74.2 Parenchyma cell and cell wall.
(Leverrier, 2016)
main mechanisms are at the origin of this texture modification.
Firstly, the cells break down under the thermal treatment, which
will cause a decrease in the turgor pressure, allowing the water
contained in the cells to diffuse freely. The second one is the solu-
bilization of the pectins contained in the middle lamella and in
the cell walls through thermal treatment, leading to the separation
Cassandre Leverrier
of the cells and the weakening of the cell walls (Moelants et al.,
2014). These two consequences of heat treatment considerably
reduce the firmness of the tissues.
Alongside the pre-​treatment used, the intensity of the mech-
anical treatment will influence the particle properties (especially
their size) and will therefore influence the applesauce texture.
Sauces and Purées: The Underside of Applesauce
FIGURE 74.3 Structural organization of the fruit at microscopic and
macroscopic scales drives its texture.
(Waldron et al., 2003)
We will see how the intensity of the heat and mechanical
treatments can influence the texture of applesauce in the section
“What parameters influence the texture of the apple puree?”
Applesauce: A Soft Plant Particle Suspension
During processing, parenchyma cells are partly disrupted and
release their cell content. Plant-​based purees can be described
as “concentrated dispersions of soft, deformable insoluble
particles in an aqueous serum” (Rao et al., 1986). Purees are thus
composed of two distinct phases (Espinosa, 2012), schematically
represented in Figure 74.4:
• the dispersed phase, containing soft solid particles,
which is composed of cells, cell fragments and cell
aggregates from the cooked apple parenchyma;
• the serum phase, mainly composed of water, containing
cell contents, sugar and soluble pectins.
As a result of thermal and mechanical treatments, the plant cells
have lost their turgidity, and the walls have become porous. The
inside and outside of the cell are therefore in balance and com-
municating. This gives the plant particles the capability of being
compressible (i.e., their volume is able to change depending on
external stress), since they are only delimited by porous deform-
able plant walls, and the inside of the cell is filled by the con-
tinuous phase (Leverrier, 2016).
Which Parameters Influence the Texture of
Apple Puree?
Texture, as defined by Szczesniak (2002), is a sensory property
that is a multi-​parameter attribute, dependent on the structure of
the food and detected by different senses. It represents a major
507
FIGURE 74.4 (a) Schematic representation of a highly diluted carrot puree;
(b) confocal microscopy of a non-​diluted Royal Gala (apple) puree.
((a) Moelants et al., 2014; (b) Leverrier, 2016))
quality attribute of food products and is also the quality attribute to
which fruit and vegetable cell walls contribute the most (Waldron
et al., 2003). All parameters that influence the rigidity and cohe-
sion of cell walls, whether fruit-​specific or due to processing, will
therefore influence the texture of the purees. In this section, we
will discuss the structural parameters impacting the texture of
applesauce, focusing on instrumental texture. Perceived texture
and its relationship with measured texture will be discussed at the
end of the section.
Particle Concentration and Serum Viscosity
As described in the previous section, applesauce is a suspen-
sion of vegetable particles, consisting of two phases: a dispersed
phase, the cells, clusters of cells and fragments of cells, suspended
in the continuous phase, consisting mainly of water, sugar and
biopolymers. Each of these two phases will have an impact on
the texture of the puree.
As with all concentrated suspensions, whether food suspensions
or not, the first-​order parameter affecting the behaviour of the
product in flow is its volume fraction. The volume occupied by
suspended particles is the factor that has the greatest impact on
the properties of concentrated suspensions: the larger the volume
occupied by the particles, the higher the viscosity (consistency)
of the suspension. For suspensions of hard particles, a divergence
of viscosity with the volume fraction of particles is observed
(Figure 74.5).
For plant cell suspensions, this divergence of the viscosity with
the increasing volume fraction or concentration is not observed
(Figure 74.6). Just as it is difficult to evaluate the volume fraction
of soft particles (Adams et al., 2004), the volume fraction of plant
particles is not obvious to determine because of their deform-
able and compressible nature. As the literature does not yet agree
on a definition, we will talk here about particle concentration
(meaning the insoluble solids building the cell walls).
Whether on suspensions of apple (Leverrier et al., 2016),
carrot (Day et al., 2010a; Day et al., 2010b; Lopez-​Sanchez et al.,
2012), tomato (Lopez-​Sanchez et al., 2012) or broccoli (Day
et al., 2010a; 2010b; Lopez-​Sanchez et al., 2012) particles, the
literature describes three areas of concentration (Espinosa, 2012;
508 Cassandre Leverrier

FIGURE 74.5 Master curve of relative viscosity divergence with relative volume fraction for suspensions of monomodal and bimodal hard spheres.
(Chong et al., 1971)

FIGURE 74.6 Storage modulus G′ as a function of total solids for carrot cell wall particle dispersions containing “cluster-​cell”, “single-​cell” or “cell-​
fragment” obtained through different thermo-​mechanical treatments.
(Day et al., 2010)

Day et al., 2010b; Leverrier et al., 2017a) related to the organiza-
tion and the interactions of the particles within the suspension:
• the diluted domain (identified as I in Figure 74.6) cor-
responds to a domain where particles do not occupy
the entire volume but are free to move in the space and
are able to flow without interacting with each other.
Therefore, the flow behaviour of these suspensions is
generally Newtonian;
• the intermediate domain (identified as II) corres-
ponds to the building of a weak network between the
particles at rest and the appearance of non-​Newtonian
properties during flow (measurable elastic properties
at rest; yield stress and shear-​thinning behaviour at
flow). Particles are no longer free to flow and have
to adapt their behaviour to the neighbouring particles
during the flow;
• the concentrated domain (identified as III) is
characterized by an important congestion of the system.
Particles are filling the entire available space and have
to modify their shape and decrease their volume to
adapt to the available space and to allow the flow.
Sauces and Purées: The Underside of Applesauce
FIGURE 74.7 Concentration domains observed for plant particles.
(Leverrier et al., 2017b)
These three concentration domains and the organization of the
particles in the space are illustrated in Figure 74.7.
This concentrated domain is directly linked to the nature of
plant particle suspensions: plant particles have the peculiarity to
be composed of porous and soft walls; the inside and outside of
the particles therefore respond to each other, and the particles are
compressible, meaning that they are able to reduce their volume
under stress (stress such as pressure, for example, or during the
increase in concentration).
The viscosity of the continuous phase is a second-​ order
parameter impacting on applesauce. Indeed, suspensions of
vegetable particles (compotes, fruit and vegetable purees, and
soups) are generally very concentrated suspensions. Thus, the
contribution of the continuous phase is quite low. However,
in more diluted suspensions (fruit and vegetable juices),
the viscosity of the continuous phase can impact the prop-
erties of the product, as shown in Figure 74.8. In this study,
model suspensions of apple cells were reconstructed in six
different continuous phases of viscosities, varying from 1
to 108 mPa.s. In the diluted domain, the impact of the con-
tinuous phase is high, and the viscosity of the suspensions is
dominated by the viscosity of the continuous phase, which
is in good agreement with Einstein’s law (η = η0 (1 + kE φ )),
with η the viscosity of the suspension, η0 the viscosity of the
continuous phase, φ the volume fraction of suspended particles
and kE a variable parameter that depends on the particles’
morphology and rigidity). A remarkable point of this work is
that since all suspensions exhibit the same relative viscosity
in the diluted domain, the contribution of particles to the vis-
cosity of suspensions is the same whatever the composition of
the continuous phase. However, in the concentrated domain, the
rheological behaviour of the suspension is no longer dominated
by the continuous phase, and all the concentrated suspensions
have comparable viscosities despite two orders of magnitude of
difference in continuous phase viscosity (Table 74.1).
Particle Morphology
Under the term “particle morphology”, we will discuss the size
and shape of the particles as well as the rigidity of the cell walls
509
that compose them. The morphology of plant particles may be
affected not only by intrinsic parameters, such as the variety or
maturity of the fruit, but also by extrinsic parameters related to
the process applied.
Intrinsic Parameters
Literature has shown that the variety and maturity of the fruit
modify the chemical composition and the structural properties
of the cell walls (Kunzek et al., 1999). These modifications of
the raw fruit have a direct impact on the quality of the processed
product and therefore on the texture and consistency of the puree
(Voragen et al., 1995; Colin-​Henrion, 2008). Growing conditions
of the fruits, such as farming practices or environmental
conditions, may also influence the texture of the fruit product
(Waldron et al., 2003). The texture of fruits and vegetables during
growth, ripening and storage is strongly influenced by the quan-
tity and nature of the pectins present, and changes in the proper-
ties of fruits and vegetables are often linked to changes in pectic
components (Voragen et al., 1995).
Variety and Cultivars
The variety of the fruit or vegetable influences not only the
structure of the parenchyma, the morphology of the cells and
the rigidity of the walls, but also the chemical composition of
the fruit and of the puree made from it. Redgwell et al. (1997)
illustrated the variability in cell morphology among varieties
(Figure 74.11).
If we only focus on one species of fruit, the apple, the cultivar
will also play an important role in the structure and therefore,
the texture of raw and cooked fruits. Bourles (2010) reports that
the literature has highlighted significant differences in the texture
of raw apples from different cultivars. For example, it has been
shown that the Braeburn and Granny Smith varieties are firmer
than Golden Delicious. These textural differences may be related
to the structure of the tissues: the size of the cells differs from
one variety to another, and fruits made up of large cells with large
intercellular spaces are less firm than those made up of small cells
with small intercellular spaces.
510 Cassandre Leverrier

FIGURE 74.8 Influence of the continuous phase viscosity on the viscosity of single apple cell suspensions over a wide range of concentrations.
(Leverrier et al., 2017b)

TABLE 74.1
Viscosity of the Continuous Phases Described in Leverrier et al. (2017a)
Viscosity at 100 s−1 (m.Pa.s) Density at 25 °C Conductivity (μS/​cm)
Water 1.0 ± 0.2 1.000 0.80 ± 0.1
NaCl 0.1% 1.1 ± 0.2 1.000 1751 ± 15
Sucrose 48% 12.8 ± 0.3 1.219 18 ± 1
CMC 1% 12.6 ± 0.1 1.006 2000 ± 30
CMC 3.2% 108 ± 2 1.018 6000 ± 210
Apple serum 13 ± 2 1.051 1731 ± 22

Maturity/​Loss of Adhesion
The second intrinsic parameter that has a considerable influ-
ence on the texture of fruits is ripening. Ripening is a complex
natural process, which results in a modification of the structure
and composition of cell walls, and a decrease in cell adhesion,
which therefore leads to a softening of the fruit tissues. The
evolution of the state of the cell walls with maturity is illustrated
in Figure 74.11.
Under mechanical stress, fruit disruption will occur dif-
ferently depending on intercellular adhesion forces and wall
adhesion forces. If the intercellular adhesion forces are higher
than those of the cell walls, the rupture will occur at the walls.
Sauces and Purées: The Underside of Applesauce
FIGURE 74.9 Images of crunchy (a) and floury (b) apple cells by cryo-​
scanning electron microscopy.
(Segonne et al., 2014)
FIGURE 74.10 Pear cells undergoing separation.
(Waldron et al., 2003)
If the turgidity pressure is high at breaking, the fruit will be
considered as “juicy”. Moreover, in the opposite case, where
the intercellular adhesion forces are lower than those of the
cell walls, there is no rupture of the walls but a separation of
the cells (Bourles, 2010; Waldron et al., 1997). In general, for
mature fruits, intercellular adhesion is strong, and tissue rupture
occurs mainly by wall rupture, as illustrated in Figure 74.9a. In
overripe fruits, cell separation predominates, as illustrated in
Figure 74.9b, and these fruits are described as “floury” in tex-
ture (Waldron et al., 2003; Segonne et al., 2014). Softening
of fruits during ripening is associated with solubilization of
pectins (Billy et al., 2008).
The presence of active enzymes in the maturation process can
induce a loss of cell adhesion, up to and including separation. Cell
separation requires the dissolution of the middle lamella, but the
latter is not uniform, as the three-​cell junctions and corners of the
intercellular spaces are more resistant to degradation. Figure 74.10
illustrates the cellular separation of two pear parenchyma cells,
remaining attached only by a point corresponding to a punctuation
(an opening allowing communication between two cells).
Thermo-​Mechanical Treatment
In fruit and vegetable processing, the quality of finished products
is largely dependent on the structure and on the polysaccharide
511
FIGURE 74.11 Ripe (right) and unripe (left) raw fruits showing cell size
variability depending on the species of the fruits and wall swelling during
ripening. (A, B) avocado; (C, D) blackberry; (E, F) plum; (G, H) persimmon.
(Redgwell et al., 1997)
composition of the cell walls. Among polysaccharides that com-
pose the cell walls, pectins are the most hydrophilic and most
sensitive to degradation induced by heat treatment. Just as can
be observed during the natural ripening process of the fruit,
changes in the texture of fruits and vegetables are often linked
to changes in pectic components (Voragen et al., 1995). During
cooking, these pectic compounds are subject to depolymerization
reactions, which very often lead to softening and cell separation
in tissues. Figure 74.12 illustrates the effect of thermal treatment
on onion parenchyma cells with increasing cooking time. This
sensitivity to heat treatments leading to solubilization and cell
separation of pectic compounds is therefore influenced by the
varieties of fruit or vegetables, and the differences observed
from one variety to another can be attributed to the parenchyma’s
structure and the pectin composition of the fruits (Anantheswaran
et al., 1985).
Depending on the heat treatment, the separation of the cells
will not occur in the same way, as shown in Figure 74.13. If the
pre-​treatment has been intense enough, the pectins contained in
the average lamella will have been solubilized, causing the cells
to separate at the level of the average lamella during mechanical
treatment. If the heat treatment is low or non-​existent, then the
intercellular adhesion force is high, and the mechanical treatment
will cause tissue rupture through the cells. Figure 74.13 illustrates
these two types of separation. It is also observed that the shape
512 Cassandre Leverrier

FIGURE 74.12 Effect of thermal treatment on onion parenchyma cells. (A) raw, (B) cooked 20 min, (C) cooked 50 min.
(Sila et al., 2008)

FIGURE 74.13 Impact of process on the size and morphology of carrot (A) and broccoli (B) puree particles. (1) Raw parenchyma, (2) clusters of cells
obtained after bleaching (80 °C, 10 min), (3) individual cells obtained after cooking (100 °C, 30 min).
(Day et al., 2010b)

and size of the cells will be impacted by the intensity and the
duration of the process: less cooked tissues lead to the formation
of large angular particles, while more cooked tissues lead to the
formation of smaller and rounder particles (Day et al., 2010b;
Lopez-​Sanchez et al., 2011).
would be directly related and driven by the cell wall content, and
the granulometry directly related to the size of the particles in the
suspension. This work also highlights the second-​order role of the
continuous phase in concentrated suspensions, the increase in the
continuous phase by adding pectins to the purees (blue arrows)
having the effect of leading to more hearty purees.

Link between Perceived Texture (Sensory) and

Measured Texture (Instrumental)
There is great interest in predicting the perceived texture by
instrumental measurements, and many studies aim to link the
perceived texture to the measured texture. The following results
(Figure 74.14), conducted on apple purees, allow us to approach
this link. Indeed, in this work, the authors show that two major
dimensions are necessary to describe, in a sensory way, the proper-
ties of apple purees: consistency and granulometry. However, this
work highlights the possibility of controlling these two quantities
by the content of cell walls (concentration of insoluble elements)
and by the size of the particles, respectively. Thus, the consistency
Conclusions and Perspectives
The texture of fruit and vegetable purees is complex and depends
on many factors. The variety and maturity of the fruit used will
influence the size of the cells in suspension and their ability to sep-
arate during thermo-​mechanical treatments. The process and its
intensity will also influence the firmness of the particles, the ability
of the cells to disengage, and also the shape of the suspended
particles. All these parameters will therefore have a major influ-
ence on the final texture of the product. The research works cited
in this chapter allow us to understand better the three main factors
influencing the texture of products (in order of importance):
Sauces and Purées: The Underside of Applesauce 513

FIGURE 74.14 Sensory characteristics of apple purees can be described by two sensory axes (consistency and graininess), which are described by their
structural characteristics (pulp content and particle size).
(Espinosa-​Muñoz et al., 2012)

FIGURE 74.15 Using confocal microscopy, three-​dimensional mesh construction and physics modelling to approach the behaviour of plant particles
under shear.
(Plana-​Fattori et al., 2018)

• the particle concentration
• the particle morphology (which depends on the fruit
variety, growing conditions, post-harvest storage and
thermo-mechanical treatments undergone by the fruits)
• the continuous phase viscosity.
While it is now understood that particle size, shape and content
influence the texture, and therefore the viscosity, of fruit and
vegetable purees, the question of the behaviour of these particles
during flow is not yet resolved. Many questions remain open,
such as:
514
How do particles deform during shearing?
How does the wall stiffness influence the ability of the par-
ticle to deform?
How does the porosity of the wall influence the rigidity of
the particles?
Since the rheology of suspensions is strongly related to
the hardness of the particles, is it possible to measure
the mechanical properties of plant particles after the
thermo-​mechanical process?
Is it possible to measure the mechanical properties of
the walls?
To answer these questions, innovative work combining micro-
scopic observation by confocal microscopy, three-​dimensional
mesh construction and physics modelling is currently being
carried out in order to approach the behaviour of these particles
under shear, as illustrated in Figure 74.15.
REFERENCES
Adams S, Frith W J, Stokes JR. 2004. Influence of particle modulus
on the rheological properties of agar microgel suspensions,
Journal of Rheology, 48, 1195–​1213.
Agoda-​Tandjawa G, Durand S, Gaillard C, Garnier C, Doublier
JL. 2012. Properties of cellulose/​pectins composites: impli-
cation for structural and mechanical properties of cell wall,
Carbohydrate Polymers, 90, 1081–​1091.
Anantheswaran RC, McLellan MR, Bourne MC. 1985. Thermal
degradation of texture in apples, Journal of Food Science, 50,
1136–​1138.
Billy L, Mehinagic E, Royer G, Renard CMGC, Arvisenet G, Prost
C, Jourjon F. 2008. Relationship between texture and pectin
composition of two apple cultivars during storage, Postharvest
Biology and Technology, 47, 315–​324.
Bourles E. 2010. Aptitude variétale des pommes à la transformation.
Recherche d’incateurs biochimiques de l’évolution de la tex-
ture et de marqueurs d’intérêt nutritionnel. PhD Dissertation,
Université d’Angers. https://​tel.archives-​ouvertes.fr/​tel-​
00538900/​file/​BOURLES.pdf
Chong JS, Christiansen EB, Baer AD. 1971. Rheology of
concentrated suspensions, Journal of Applied Polymer Science,
15, 2007–​2021.
Colin-​Henrion M. 2008. De la pomme à la pomme transformée:
Impact du procédé sur deux composés d’intérêt nutritionnel.
ESA, UFR des Sciences Pharmaceutiques. https://​tel.archives-​
ouvertes.fr/​tel-​00351179/​document
Day L, Xu M, Øiseth SK, Hemar Y, Lundin L. 2010a. Control of
morphological and rheological properties of carrot cell wall
particle dispersions through processing, Food Bioprocess
Technology, 3, 928–​934.
Day L, Xu M, Øiseth SK, Lundin L, Hemar Y. 2010b. Dynamic
rheological properties of plant cell-​wall particle dispersions,
Colloids Surfaces B. Biointerfaces, 81, 461–​467.
de Reynal B. 2008. Livre Blanc –​Compotes, fruits au sirop et nutri-
tion. Vivactis.
Espinosa L. 2012. Texture de la purée de pomme: influence de la
structure sur les propriétés rhéologiques et la perception
sensorielle –​effet du traitement mécanique. PhD Dissertation,
AgroParisTech.
Espinosa-​Muñoz L, Symoneaux R, Renard CMGC, Biau N, Cuvelier
G. 2012. The significance of structural properties for the devel-
opment of innovative apple puree textures, LWT –​Food Science
and Technology, 49, 221–​228.
Cassandre Leverrier
Jarvis MC. 1998. Intercellular separation forces generated by intra-
cellular pressure, Plant, Cell & Environment, 21, 1307–​1310.
Jarvis MC, Briggs SPH, Knox JP. 2003. Intercellular adhesion
and cell separation in plants. Plant, Cell & Environment, 26,
977–​989.
Khan AA, Vincent JFV. 1990. Anisotropy of apple parenchyma,
Journal of the Science of Food and Agriculture, 52, 455–​466.
Khan AA, Vincent JFV. 1993. Anisotropy in the fracture properties
of apple flesh as investigated by crack-​opening tests, Journal
of Material Science, 8, 45–​51.
Kojima T, Fujita S, Tanaka M, Sirisomboon P, Mongkut K. 2004.
Plant compounds and fruit texture: The case of pear. In
Kilcast D (ed.) Texture in Food, Woodhead Publishing, Vol. II,
259–​294.
Kunzek H, Kabbert R, Gloyna D. 1999. Aspects of material science
in food processing: Changes in plant cell walls of fruits and
vegetables, Zeitschrift für Lebensmittel und -​forschung A, 208,
233–​250.
Leverrier C. 2016. Relations Structure /​Propriétés de suspensions de
particules végétales, PhD Dissertation, University Paris Saclay.
Leverrier C, Almeida G, Espinosa-​ Munoz L, Cuvelier G. 2016.
Influence of particle size and concentration on rheological
behaviour of reconstituted apple purees, Food Biophysics, 11,
235–​247.
Leverrier C, Almeida G, Menut P, Cuvelier G. 2017a. Design of
model apple cells suspensions: Rheological properties and
impact of the continuous phase, Food Biophysics, 12, 383–​396.
Leverrier C, Moulin G, Cuvelier G, Almeida G. 2017b. Assessment
of deformability of soft plant cells by 3D imaging, Food
Structure, 14, 95–​103.
Lopez-​Sanchez P, Chapara V, Schumm S, Farr R. 2012. Shear elastic
deformation and particle packing in plant cell dispersions,
Food Biophysics, 7, 1–​14.
Lopez-​Sanchez P, Nijsse J, Blonk H, Bialek L, Schumm S, Langton
M. 2011. Effect of mechanical and thermal treatments on the
microstructure and rheological properties of carrot, broccoli
and tomato dispersions, Journal of the Science of Food and
Agriculture, 91, 207–​217.
McDougall GJ, Morrison IM, Stewart D, Hillman JR. 1996. Plant
cell walls as dietary fibre: range, structure, processing and
function, Journal of the Science of Food and Agriculture, 70,
133–​150.
Moelants Katlijn RN, Cardinaels R, Van Buggenhout S, Van Loey
A, Moldenaers P, Hendrickx ME. 2014. A review on the
relationships between processing, food structure, and rheo-
logical properties of plant-​ tissue-​
based food suspensions,
Comprehensive Review in Food Science and Food Safety, 13,
241–​260.
Plana-​Fattori A, Doursat C, Almeida G, Moulin G, Leverrier C, Ou E,
Flick D. 2018. Estimation of intrinsic viscosity of apple cells
by numerical simulation. In 10th International Conference on
Simulation and Modelling in the Food and Bio-​Industry 2018,
FOODSIM 2018.
Rao MA, Cooley HJ, Nogueira JN, McLellan MR. 1986. Rheology
of apple sauce: Effect of apple cultivar, firmness, and pro-
cessing parameters, Journal of Food Science, 51, 176–​179.
Redgwell RJ, MacRae E, Hallett I, Fischer M, Perry J, Harker R.
1997. In vivo and in vitro swelling of cell walls during fruit
ripening, Planta, 203, 162–​173.
Segonne SM, Bruneau M, Celton JM, Le Gall S, Francin-​Allami M,
Juchaux M, Laurens F, Orsel M, Renou JP. 2014. Multiscale
investigation of mealiness in apple: An atypical role for a
pectin methylesterase during fruit maturation, BMC Plant
Biology, 14, 375.
Sauces and Purées: The Underside of Applesauce
Sila DN, Duvetter T, De Roeck A, Verlent I, Smout C, Moates GK,
Hills BP, Waldron KK, Hendrickx M, Van Loey A. 2008.
Texture changes of processed fruits and vegetables: Potential
use of high-​pressure processing, Trends in Food Science and
Technology, 19, 309–​319.
Szczesniak AS. 2002. Texture is a sensory property, Food Quality
and Preferences, 13, 215–​225.
Voragen AGJ, Pilnik W, Thibault JF, Axelos MAV, Renard CMGC.
1995. Pectins. In Stephen AM (ed.), Food Polysaccharides
and Their Applications, CRC Press –​Taylor & Francis Group,
287–​340.
515
Waldron KW, Parker ML, Smith AC. 2003. Plant cell walls and food
quality, Comprehensive Reviews in Food Science and Food
Safety, 2, 128–​146.
Waldron KW, Smith AC, Parr AJ, Ng A, Parker ML. 1997. New
approaches to understanding and controlling cell separation in
relation to fruit and vegetable texture, Trends in Food Science
and Technology, 8, 213–​221.
Seaweeds: Phycogastronomy –​the Culinary Science of Seaweeds
Ole G. Mouritsen
Department of Food Science, Taste for Life, Design and Consumer Behavior, University of Copenhagen, 26 Rolighedsvej,
DK-​1958 Frederiksberg C, Denmark
Whereas seaweed extracts like alginate, carrageenan, and agar
have long played an important role in molecular cooking as gel-
ation agents, the culinary science of whole seaweeds has only
recently gained momentum. The gastronomic value of seaweeds
has a focus on the texture and flavour of different species, in
particular the possibility of imparting umami to various dishes
involving vegetables.
Algae constitute a large heterogeneous kingdom ranging in size
from tiny unicellular species to large multicellular ones. Large
marine algae are known as seaweeds, of which there are about
12,000 different species, most of which are edible (Mouritsen,
2013a; 2013b; Pereira, 2016; Pérez-​ Lloréns et al., 2018).
A gallery of some seaweeds is shown in Figure 75.1. Together
with the microalgae, seaweeds generate more than two-​thirds
of the organic production on the planet, and since they perform
photosynthesis, they are also responsible for the major produc-
tion of molecular oxygen in the atmosphere (Pérez-​Lloréns et al.,
2018). Seaweeds are an important part of the world cuisine, in
particular in Southeast Asia and Polynesia, where they have been
part of the daily fare for millennia (O’Connor, 2017; Mouritsen
et al., 2018). There is still a living tradition of using seaweeds as
human foods in a number of coastal areas around the world, e.g.,
Scotland, Wales, Brittany, Iceland, Spain, and in particular Ireland
(Rhatigan, 2009; 2018). Moreover, the use of whole seaweeds as
food and for culinary advances appears to be on the rise around
the Western world (Chapman et al., 2015; Pérez-​Lloréns, 2019;
Lucas et al., 2019; Mouritsen et al., 2019a), not least fuelled
by an appreciation of the health benefits associated with eating
whole seaweeds (Arasaki and Arasaki, 1983; Brown et al., 2014;
Cornish et al., 2015; 2017; 2019; Pereira, 2018). Still, one should
not expect that consumers will increase their consumption of
seaweeds unless they become appreciated for their gastronomic
value, with a particular focus on flavour (Mouritsen, 2017).
The world cuisine of seaweeds was recently reviewed, and a
comprehensive list of various cookbooks dealing with seaweeds
was prepared (Mouritsen et al., 2019a). Only a few of these
cookbooks deal with aspects of the culinary science of seaweeds
(Mouritsen, 2013a; Kilinç et al., 2013; Druehl, 2016; Ibáñez
and Guerrero, 2017; O’Connor, 2017; Pérez-​ Lloréns et al.,
2018). The use of water-​soluble seaweed extracts in the form
of alginates from brown algae and carrageenans and agar from
red algae as thickening and gelation agents in local cuisines,
as well as in food and materials technology, has a long history.
Furthermore, seaweed hydrogels featured prominently in the rise
of molecular cooking (This, 2006; Barham et al., 2010; Youssef,
2013). Seaweeds used in this indirect way as extracts can add
texture to other foodstuffs and are often incorporated into, e.g.,
yogurt, ice cream, and desserts. In 2012, I suggested algal cuisine
as an interesting laboratory for exploring the emerging science of
gastrophysics (Mouritsen, 2012) by employing a wide range of
the most powerful theoretical, computational, and experimental
biophysical techniques to study the empirical world of cooking
and gastronomy of seaweeds. Since texture is so important for
the sensory perception and acceptance of seaweeds as food
(Mouritsen and Styrbæk, 2017), gastrophysics is particularly
essential to explore the science of algal cuisine (Mouritsen, 2012).
Sensory Perception of Seaweeds
The gastronomic value of seaweeds and seaweed dishes is judged
by five senses: vision, taste, olfactory sense, tactile sensation
(texture, mouthfeel), and hearing, all of which contribute. In
the following, we briefly go through the different contributions
to the overall ‘taste experience’ of seaweeds. Seaweeds’ ability
to appeal to all our five senses is the reason why they are used
widely in world cuisine as well as in high-​end gastronomy. They
can be used as a flavour-​giver and seasoning agent, as a condi-
ment, as a snack, and as part of a dish or as a whole dish. Due to
their varied shapes and colours, seaweeds are also used for decor-
ation and to impart a particularly appealing visual appearance to a
presentation, be it a salad, a soup, a dessert, or a drink (Mouritsen
et al., 2019a).
Visual Appearance and Aesthetics
Some seaweeds are valued for their visual appearance in a dish
in terms of both colours and shapes (Figure 75.1a–​e). The great
variety in morphology and colours of seaweeds imparts an aes-
thetic appeal to dishes such as salads or as toppings to seafood
servings. The actual colour of the seaweeds can vary substantially
depending on the species and the degree of bleaching. Branched
517
518
FIGURE 75.1 Gallery of edible seaweeds. (a) Dulse (Palmaria palmata); (b) carrageen, Irish moss (Chondrus crispus); (c) funori (Gloiopeltis spp.);
(d) ogonori (Gracilaria spp.); (e) tosaka-​nori (Meristotheca papulosa); (f) konbu (Saccharina japonica).
(Courtesy of Jonas Drotner Mouritsen)
and stringy morphologies are often considered attractive in
species where the fronds have some degree of stiffness and there-
fore are less likely to collapse flatly in the presentation. A variety
of seaweeds can be used to add a spectrum of colours to a simple
green salad. The easiest way to incorporate seaweeds into a green
salad is to crush toasted nori (Pyropia spp.), dulse (Palmaria
palmata), or wakame (Undaria pinnatifida) and sprinkle them
over the salad leaves or to mix in thin strips of cooked and
marinated algae. Green wakame is a welcome and colourful add-
ition to a standard miso soup.
Auditory Sensation
In several cases, eating seaweeds can also invoke the auditory
system; e.g., dried, fried, or toasted seaweed can appear crispy,
crackly, and crunchy. Crisp textures can pep up an otherwise
slightly dull salad that lacks crunch. When completely dry, nori
sheets wrapped around maki-​sushi give a pleasant crisp sensation
in the ears.
Aroma and Odour
The aroma and odour of seaweeds can sometimes be unpleasant
when common decomposition products like dimethyl sulfide
Ole G. Mouritsen
are present in larger amounts. However, in small amounts,
dimethyl sulfide is the odour we associate with the fresh
ocean. Fresh and well-​preserved seaweeds have the pleasant
mild odour that is associated with salty waters, often with
notes of halogenated compounds. Certain brown seaweeds,
like the Japanese konbu, are stored in controlled-​temperature
conditions in cellars and warehouses for at least two years, and
in some cases ten years, to mellow their odour (Mouritsen and
Styrbæk, 2014). Some seaweeds are often infused with special
flavours, e.g., by marinating, smoking, fermenting, or adding
various spices.
There are few published studies of volatile components in
brown seaweeds (Boonprab et al., 2006; Keng et al., 2013).
A recent study of 20 different species of brown seaweeds used
for human consumption from around the world, including the
12 genera Nereocystis, Macrocystis, Laminaria, Saccharina,
Undaria, Alaria, Postelsia, Himanthalia, Ecklonia, Sargassum,
Fucus, and Corda (Mouritsen et al., 2019b), identified a large
number of aroma compounds in aqueous extracts (dashi) and
found that 49 of them exhibited significant differences between
the samples. Among other compounds found in significant
amounts were typical degradation products of fatty acids, such
as propanal, nonanal, decanal, undecanal, acetone, ethyl acetate,
ethyl propanoate, methyl methacrylate, 3-​ heptanone, butanol,
butyrolactone, and acetophenone.
Seaweeds: Phycogastronomy
As described later, it has been found that certain volatile
compounds in brown seaweeds can enhance umami taste in
dashi with very low glutamate content prepared from, e.g., bull
kelp (Nereocystis leutkeana) and arame (Ecklonia bicyclis),
specifically (E)-​2-​nonenal, (E)-​2-​octenol, 2-​butenal, octanal,
2,3-​
butanedione, trimethylamine, and 3-​ methylnutanal (Frøst
et al., 2018).
Texture (Mouthfeel)
The appreciation of seaweed texture is essential for the use of
seaweeds in many Asian cuisines, not least the traditional Japanese
cuisine (Nisizawa, 2002; Japanese Culinary Academy, 2016).
Texture is that part of the foodstuff’s physical structure that our
sensory apparatus can detect (Mouritsen and Styrbæk, 2017). The
texture of seaweeds, either in their natural state or as prepared,
can be hard, soft, crunchy, crispy, slimy, chewy, tough, tender,
elastic, etc. The polysaccharides in some seaweeds, such as those
in sugar kelp, produce a slimy mouthfeel. It is interesting to note
that the Japanese language contains four times as many words
pertaining to mouthfeel as many Western languages (Mouritsen
and Styrbæk, 2017). It is likely that this particular attention to
texture is one of the reasons why the Japanese value seaweeds for
their mouthfeel and consequently use three different terms to this
effect: kuchi atari (palatability, mouthfeel), shitazawari (tongue
feel), and hagotae (crunchiness, tooth resistance).
The many different types of seaweeds used for human con-
sumption have a very wide range of textures, and the texture
depends on the age of the seaweed and the stage in its life cycle,
which part we are talking about, how it has been stored after
harvesting, and, last but not least, how it has been prepared in
the kitchen.
Except for very young sprouts and fronds, most seaweeds
are tough and chewy when freshly harvested, even the thin and
delicate fronds of purple laver (Porphyra/​Pyropia spp.) and
sea lettuce (Ulva spp.), despite the fact that the fronds of these
FIGURE 75.2 Fluorescence microscopy image of the cell structure of the red seaweed dulse (Palmaria palmata) before and after it has been cooked. The
cells are around 20 μm in diameter. Cooking loosens the cell structure so that the dulse becomes softer and easier to chew.
(Courtesy of Jinsoo Yi)
519
seaweeds are no thicker than two layers of cells. Drying, roasting,
and cooking will tenderize them. Some large brown seaweed
species have capillary systems that render them chewy, and some
species have a stem that is more tough than the rest of the frond.
Other species have stipes and floats that can be quite tough and,
in many cases, inedible. Sporophylls are often softer than the rest
of the frond, and due to their higher content of fatty acids, they
can also be tastier.
Stored seaweeds become tenderized by their own enzymes
even in their dry state. Speedier tenderization can be achieved
after light rehydration, which is, e.g., used for some products
of dulse (Palmaria palmata) (Erhart and Cerier, 2001)
(Figure 75.2). Some of the large brown seaweeds, e.g.,
Saccharina latissima (sugar kelp), exude significant amounts of
slimy polysaccharides, such as alginate, as well as mannitol.
Although these substances have useful applications in the kit-
chen as hydrogels, they are generally a nuisance when these
seaweeds are used in salads or in soups, which will become too
slimy or too viscous, respectively. Some Japanese products, like
konbu tsukedani, which is konbu simmered in soy sauce, com-
bine umami taste and a soft and slightly sticky texture with a
very pleasant tender mouthfeel.
Many commercial seaweed products focus on crispness, e.g.,
dried nori and toasted seaweed paper made of Porphyra/​Pyropia
spp. Nori is used for maki-​sushi or to sprinkle in small pieces in
the form of furikake over other dishes, e.g., cooked rice. Due to
the dry seaweed’s natural capacity for binding water, preparations
involving nori must be consumed right away before the seaweed
absorbs moisture and becomes soggy. Dried roasted and deep-​
fried seaweeds of green, red, and brown varieties are valued for
their crispness and can be used as snacks and candy. A special
preparation of dulse (Palmaria palmata) has been launched in
recent years as a novel product, although it actually follows an
old Scottish recipe: crisp seaweed ‘bacon’. It is based on dried
and smoked fronds of dulse that have been deep-​fried. The tex-
ture resembles fried pork bacon and is, with a proper smoked
flavour, reminiscent of bacon.
520 Ole G. Mouritsen

Taste range of brown seaweeds from around the world (Mouritsen

The most celebrated taste of seaweeds is umami, associated et al., 2019b). Table 75.1 shows a selection of results for free
with seaweed products based on Japanese konbu (Saccharina glutamate contents for some brown compared with some selected
japonica) (Figure 1f). We shall deal specifically with umami in red seaweed species; the contents vary dramatically for the
the following section. Apart from umami, seaweeds can have different species. In addition to this variation, there is also sub-
sweet notes due to sweet, free amino acids (alanine, glycine, pro- stantial variation for individual species depending on the place
line, and serine) and sugar alcohols, like mannitol in sugar kelp and time of harvest. Without comparison, the different kinds of
(Saccharina latissimi). Bitter notes are caused by polyphenols. Japanese konbu have by far the highest contents of free glutamate.
In some cases, harvesters (Druehl, 2016) let their seaweeds, e.g., Other local subspecies of Saccharina japonica not listed, such
Macrocystis pyrifera (macrokelp) and Nereocystis luetkeana as Rishiri-​konbu and Raushu-​konbu, are in between Ma-​konbu
(bullwhip kelp), dry in the open air in sunshine, whereupon ultra- and Hidaka-​konbu, but there are some differences for different
violet radiation leads to breakdown of some of the polyphenols. harvesting sites and vendors (Japanese Culinary Academy, 2016).
Most unwashed seaweeds are salty due to sea salts, and since It is noteworthy that the common sugar kelp, which is of the same
there is often a surplus of potassium salts, such as potassium genus Saccharina as konbu, and oarweed, which is in the same
chloride, in many seaweeds (Mouritsen, 2013a) compared with order Laminariales as konbu, do not contain much glutamate. The
sodium salts, this leads to a certain degree of bitterness. Some same is true of all the investigated species of Fucus.
brown seaweeds have a sour taste, and others roasted or astrin- Based on these findings, we can conclude that most of the
gent notes (Mouritsen et al., 2019b). investigated seaweeds, apart from konbu and dulse, do not
contain much free glutamate, and hence, they have a very low
umami potential. However, sensory studies using trained panels
of tasters show that there is no simple relationship between the
Umami and dashi actual content of free glutamate in a dashi and a taster’s percep-
Umami as a taste was first proposed as a true basic taste when it tion of umami. This is illustrated in Figure 75.3, which shows the
was discovered that Japanese konbu (Saccharina japonica), e.g., perceived umami intensity from a sensory analysis as a function
used in the Japanese soup stock dashi, contains large amount of of the amount of free glutamate in dashi prepared from different
free glutamate (2–​3% dry weight), which was pointed out to be brown seaweeds. Hence, even if a dashi is made of seaweeds that
the main component responsible for the delicious (umai) taste release very little free glutamate, it may be perceived by tasters to
of dashi (Ikeda, 2002). Dashi means ‘cooked extract’, and the have a substantial umami taste. It also appears that certain vola-
whole traditional Japanese cuisine revolves around dashi as a fla- tile compounds in the seaweeds enhance the perception of umami
vour provider to all kinds of dishes, not only soups (Tsuji, 1980; in tasters (Frøst et al., 2018).
Mouritsen and Styrbæk, 2014; Japanese Culinary Academy, To complete the description of dashi and umami from
2016). The ability of different varieties of konbu to elicit umami seaweeds, it should be mentioned that the umami sensation can
taste correlates with their content of free glutamate. Of the many be enhanced via the principle of synergy, as described in another
different variants of Japanese Saccharina japonica, ma-​konbu, chapter in the present volume (see chapter on Umami), according
rausu-​konbu, and rishiri-​konbu are the best for extraction to to which certain free nucleotides such as inosinate, guanylate,
dashi, and they lead to a very light dashi with a mild and some- and adenylate act in a simultaneous action on the umami receptor
what complex taste. Ma-​konbu has the largest amount of free (Mouritsen and Khandelia, 2012) to lead to a strong umami taste.
glutamate, 3,200 mg/​100 g, whereas rausu-​konbu has 2,200 mg/​ In a classical dashi, such free nucleotides would be delivered by
100 g and rishiri-​konbu has 2,000 mg/​100 g. The lower-​quality the fish product katsuobushi (inosinate) or shiitake mushrooms
hidaka-​konbu has 1,300 mg/​ 100 g. The red seaweed laver (guanylate). This pairing principle behind the umami synergy
(Porphyra/​Pyropia spp.) used to produce nori has comparable has been known phenomenologically for centuries by cooks and
amounts (1,378 mg/​100 g), whereas most other types of seaweed chefs in different food cultures (e.g., the well-​known pairings
are inferior in this respect (Ninomiya, 1998; Blumenthal et al., eggs–​bacon, vegetables–​meat, cheese–​ham, etc.).
2009, Mouritsen et al., 2012). The use of this scientifically based synergy principle and the
In the quest for umami in the New Nordic Cuisine, we magic behind dashi has now perfused the world cuisine, where
measured the glutamate content in dashi from seaweed species in it is being used in a large variety of dishes. Konbu has therefore
the Nordic terroir (Mouritsen et al., 2012; Mouritsen et al., 2013) also found a place as a common staple in the kitchen of many
and found that dulse (Palmaria palmata) has a significant poten- restaurants. A better-​informed use of the culinary sciences behind
tial to impart umami, whereas others, like sugar kelp (Saccharina seaweeds, umami, and dashi may hold promise for preparing
latissima), do not. Dashi from dulse has been used in a variety of more delicious dishes of greens and vegetables, which generally
preparations, such as ice cream, fresh cheese, and bread. contain very little free glutamate. The traditional Japanese pick-
It is often claimed, including by chefs, that other seaweeds, in ling techniques, tsukemono, are examples of such uses that impart
particular various brown seaweeds, also have umami taste. We umami to vegetables by using various pickling media and fer-
have recently measured the free amino acid contents of a wide mentation techniques (Mouritsen, 2018).
Seaweeds: Phycogastronomy 521

TABLE 75.1
Content of Free Glutamate (mg/​100 g) in Some Selected Aqueous Seaweed Extracts (Dashi). All Data are Based on Extraction at 60 °C of
5 g Dry Seaweed in 250 g Water
Seaweed extract (dashi) Glutamate content Origin
Ma-​konbu (Saccharina japonica) 37a Hokkaido, Japan
Hidaka-​konbu (Saccharina japonica) 18a Hokkaido, Japan
Dulse (Palmaria palmata) 10b Stykkisholmur, Iceland
Graceful red weed (Gracilaria verrucosa) 6b Grenå, Denmark
Bull kelp (Nereocystis leutkeana) 3a Vancouver Island, Canada
Oarweed (Laminaria digitata) 1a Grindavik, Iceland
Winged kelp (Alaria esculenta) 1a Grindavik, Iceland
Sugar kelp (Saccharina latissima) 1a Lillebælt, Denmark
Bladderwrack (Fucus vesiculosus) 1a Lillebælt, Denmark
Wakame (Undaria pinnatifida) 0a Shimane, Japan

a Mouritsen et al. (2019b); b Mouritsen et al. (2012).

FIGURE 75.3 Perceived umami intensity (scores on a scale 1–​10) from a sensory analysis as a function of the amount of free glutamate in dashi prepared
from different seaweeds.

Acknowledgements
This work was supported in part by a grant to Smag for Livet
(Taste for Life) from Nordea-​fonden.
REFERENCES
Arasaki S, Arasaki T. 1983. Low Calorie High Nutrition Vegetables
from the Sea, Japan Publications Inc., Tokyo.
Barham P, Skibsted LH, Bredie WLP, Frøst MB, Møller P, Risbo
J, Snitkjær P, Mortensen LM. 2010, Molecular gastronomy:
a new emerging scientific discipline, Chemical Review, 110,
2313–​2365.
Blumenthal H, Barbot P, Matsushisa N, Mikuni K. 2009. Dashi and
Umami: The Heart of Japanese Cuisine, Cross Media Ltd,
London.
Boonprab K, Matsui K, Akakabe Y, Yoshida M, Yotsukura N,
Chirapart A, Kajiwara T. 2006. Formation of aldehyde flavor
(n-​hexanal 3Z-​nonenal and 2E-​nonenal) in the brown alga
Laminaria angustata, Journal of Applied Phycology, 18,
409–​412.
Brown EM, Allsopp PJ, Magee PJ, Gill CIR, Nitecki S, Strain CR,
McSorley EM. 2014. Seaweed and human health, Nutrition
Review, 72, 205–​216.
Chapman A, Stevant P, Larssen W. 2015. Food or fad? Challenges
and opportunities for including seaweeds in a Nordic Diet,
Botanica Marina, 58, 423–​433.
Cornish ML, Critchley AT, Mouritsen OG. 2015. A role for dietary
macroalgae in the amelioration of certain risk factors associated
with cardiovascular disease, Phycologia, 54, 649–​666.
Cornish ML, Critchley AT, Mouritsen OG. 2017. Consumption of
seaweeds and the human brain, Journal of Applied Phycology,
29, 2377–​2398.
Cornish ML, Mouritsen OG, Critchley AT. 2019. A mini-​review on
the microbial continuum: consideration of a link between judi-
cious consumption of a varied diet of macroalgae and human
health and nutrition, Journal of Oceanological Limnology, 37,
790–​805.
Druehl LD. 2016. Pacific Seaweeds, Harbour Publication, Madeira
Park, British Columbia, Canada.
Erhart S, Cerier L. 2001. Sea Vegetable Celebration, Book Publishing
Company, Summertown, Tennessee.
522
Frøst MB, Mouritsen OG, Hartmann ALMA, Petersen MA,
Duelund L. 2018. Odour-​induced umami –​olfactory contri-
bution to umami taste in seaweed extracts (dashi) by sensory
interactions. Abstract presented at Eurosense 2018: Eighth
European Conference on Sensory and Consumer Research,
Sept 2–​5, 2018.
Ibáñez E, Guerrero M. 2017. Las algas que comemos, Consejo
Superior de Investigaciones Científicas, Madrid.
Ikeda I. 2002. New seasonings, Chemical Senses, 27, 847–​ 849
[Translation from the original article in Journal of the Chemical
Society of Japan, 30, 820–​836 (1909)].
Japanese Culinary Academy. 2016. Flavour and Seasonings: Dashi,
Umami, and Fermented Foods, Shuhari Initiative Ltd, Tokyo.
Keng FSL, Phang SM, Rahman NA, Leedham EC, Hughes C,
Robinson AD, Harris NRP, Pyle JA, Sturges WT. 2013. Volatile
halocarbon emissions by three tropical brown seaweeds under
different irradiances, Journal of Applied Phycology, 25,
1377–​1386.
Kilinç B, Cirik S, Turan G, Tekogul H, Koru E. 2013. Seaweeds for
food and Industrial applications food industry, In: Muzzalupo I
(ed.) Food Industry, InTech, Rijeka, 735–​748.
Lucas S, Gouin S, Lesueur M. 2019. Seaweed consumption and
label preferences in France, Marine Resource Economics, 34,
143–​162.
Mouritsen OG. 2012. The emerging science of gastrophysics and its
application to the algal cuisine, Flavour, 1, 6.
Mouritsen OG. 2013a. Seaweeds Edible, Available & Sustainable,
Chicago University Press, Chicago.
Mouritsen OG. 2013b. The science of seaweeds. Marine macroalgae
benefit people culturally industrially nutritionally and eco-
logically, American Scientist, 101, 458–​465.
Mouritsen OG. 2017. Those tasty weeds, Journal of Applied
Phycology, 29, 2159–​2164.
Mouritsen OG. 2018. Tsukemono –​crunchy pickled foods from
Japan: a case study of food design by gastrophysics and nature,
International Journal of Food Design, 3, 103–​124.
Mouritsen OG, Khandelia H. 2012. Molecular mechanism of the
allosteric enhancement of the umami taste sensation, FEBS
Journal, 279, 3, 3112–​3120.
Mouritsen OG, Styrbæk K. 2014. Umami: Unlocking the Secrets of
the Fifth Taste, Columbia University Press, New York.
Mouritsen OG, Styrbæk K. 2017. Mouthfeel: How Texture Makes
Taste, Columbia University Press, New York.
Ole G. Mouritsen
Mouritsen OG, Williams L, Bjerregaard R, Duelund L. 2012.
Seaweeds for umami flavour in the New Nordic Cuisine,
Flavour, 1, 4.
Mouritsen OG, Dawczynski C, Duelund L, Jahreis G, Vetter W,
Schroder M. 2013. On the human consumption of the red sea-
weed dulse (Palmaria palmata (L) Weber & Mohr), Journal of
Applied Phycology, 25, 1777–​1791.
Mouritsen OG, Pérez Lloréns JL, Rhatigan P. 2018. World cuisine of
seaweeds: science meets gastronomy, International Journal of
Gastronomy and Food Science, 14, 66–​76.
Mouritsen OG, Pérez Lloréns JL, Rhatigan P. 2019a. The rise of sea-
weed gastronomy: phycogastronomy, Botanica Marina, 62,
195–​209.
Mouritsen OG, Duelund L, Petersen MA, Hartmann AL, Frøst MB.
2019b. Umami taste free amino acid composition and volatile
compounds of brown seaweeds, Journal of Applied Phycology,
31, 1213–​1238.
Ninomiya K. 1998. Natural occurrence, Food Review International,
14, 177–​211.
Nisizawa K. 2002. Seaweeds, Kaiso Bountiful Harvest from the Seas,
Japan Seaweed Association, Tosa Kochi, Japan.
O’Connor K. 2017. Seaweed: A Global History, Reaction Books,
London.
Pereira L. 2016. Edible Seaweeds of the World, CRC Press, Boca
Raton, Florida.
Pereira L. 2018. Therapeutic and Nutritional Uses of Algae, CRC
Press, Boca Raton, Florida.
Pérez-​Lloréns JL. 2019. Seaweed consumption in the Americas,
Gastronomica, 19, 49–​59.
Pérez-​Lloréns JL, Hernández I, Vergara JJ, Brun FG, León Á. 2018.
Those Curious and Delicious Seaweeds, A Fascinating Voyage
from Biology to Gastronomy, Servicio de Publicaciones de la
Universidad de Cádiz, Cádiz.
Rhatigan P. 2009. Irish Seaweed Kitchen, Booklink, Co. Down,
Ireland.
Rhatigan P. 2018. Prannie Rhatigan’s Irish Seaweed Christmas
Kitchen, Booklink, Co. Down, Ireland.
This H. 2006. Molecular Gastronomy: Exploring the Science of
Flavor, Columbia University Press, New York.
Tsuji A. 1980. Japanese Cooking, A Simple Art, Kodansha Intl, Tokyo.
Youssef J. 2013. Molecular Gastronomy at Home, Firefly Books,
Richmond Hill, Ontario.
Size Reduction

José M. Aguilera
Department of Chemical Engineering and Bioprocesses, Pontificia Universidad Católica de Chile,
Avenida Libertador Bernardo O’Higgins 340, Santiago, Chile

Introduction methods to effect drastic physical changes in the structure of

foods, some of which are shown in Figure 76.1.
How often do you need to reduce the size of a food in the kitchen?
Size reduction (grinding, milling, crushing, etc.) is a frequent
Almost every time you prepare a dish. In the kitchen, manipula-
intermediate stage to prepare materials for further uses. Grinding
tion of size is mostly needed to accelerate heat and mass transfer,
reduces ingredients to a finer size, resulting in a powder or a
to separate different parts of a material, to mix ingredients, and to
collection of particles that can be as small as a few micrometres.
obtain uniform cooking (as in the wok). Size reduction continues
Reduction in size facilitates an intimate mixing, improves the
on your plate as you portion your bite and then in the oral cavity
appearance of the product, increases the surface/​volume ratio
during mastication (Chen, 2009).
(e.g., for more even cooking), and accelerates the transfer of heat
In engineering, size reduction is one of many “unit operations”
and mass (as in frying, extraction, impregnation or moisture con-
or actions that must be taken on a material as part of a process.
ditioning). For example, roasted coffee grains are ground in a mill
Size reduction is divided into two major categories depending on
to speed up and improve extraction in hot water.
whether the material is a solid (grinding and cutting) or a liquid
Food engineers know that not all grinding machines produce
(emulsification or atomization) (Earle, 1983). Technically, forces
the same type of particles. Wheat is ground first between serrated
acting on a body produce the fracture or shearing of the material.
cylindrical rolls separated by a narrow gap to break open the grain.
While engineers are mostly concerned with the energy involved
Successive passes through smooth cylindrical rolls “scrape”
in grinding, cooks appreciate the effects of size reduction in
off particles of a frail white endosperm, leaving the plastic and
cooking (i.e., colour, flavour and texture).
deformable bran (fibrous outer layers) intact. This process can
go on until approximately 78% of the weight of the whole grain
becomes white flour.
Cutting Roasted coffee grains are pulverized at home with high-​speed
Cutting is the controlled separation of a material into two or more rotating blades, but commercially, they are ground between two
parts using a blade or a sharp tool (Atkins, 2009). Cutting takes rotating hard plates that rub together. The particle size of powders
many names in the kitchen: slicing, flaking, dicing, shredding, is often controlled by periodic sieving and removal of “fines”
mincing, etc., leading to several types of cuts. For example,
julienne are basically thin strips; batonnet refers to rectangular
sticks; brunoise are small dices; and chiffonade is a cut for vege-
table leaves and fresh herbs whereby they are stacked, rolled and
then sliced as thin strips. Cutting is quite relevant in cooking,
since the time for doneness is proportional to the square of the
thickness of the piece (e.g., in a steak or roast).

Grinding
Some structures produced by nature must be destroyed, or at least
significantly modified in their size and shape, in preparation for
processing, cooking or consumption. If food structures were not
destroyed, there would be no fruit juices or wine to drink and
no flour, sugar or oil to make baked products. There are several FIGURE 76.1 Mechanisms of size reduction for solids and liquids.

523
524 José M. Aguilera

from the already disintegrated fraction, while the coarse particles of its Spanish designer, Pequeñas Industrias Mecanico-​Electricas
continue the grinding process (see the chapter on coffee in this Reunidas (PIMER).
book). Parmesan cheese should be shredded and garlic cloves
crushed with a press. Size reduction of cellular tissue causes the
fracture of cells, exposing the internal contents. Breakage of the
Measuring Particle Size
cells of plants releases enzymes that can participate in biochem-
ical reactions altering the colour and flavour (e.g., browning of Often, materials containing pieces, particles or droplets need
potatoes or the pungent flavour of garlic). Therefore, it is often to be characterized for the size (and sometimes the shape) of
recommended to apply a mild heat treatment beforehand to the dispersed phase. There are several means of measuring
inactivate enzymes (blanching) and later protect the tissues from particle size:
air once they are cut (e.g., by putting them under water).
• Sieve analysis is the oldest, cheapest and easiest way,
and utilizes standard sieves with calibrated grids to
classify the dry or wet material into size fractions as
Emulsions and Suspensions they pass or are retained in progressively smaller mesh
sizes. The weight of material retained over a sieve is
A food emulsion is a dispersion of a liquid into another one,
expressed as a fraction of the total mass.
produced by the application of energy in the form of shearing
• Image analysis utilizes a digital image acquired with a
forces (Figure 76.1, far right). As a result, one of the liquids
camera, either directly or using a light microscope if
becomes the dispersed phase (droplets), while the other is the particles are relatively small (e.g., less than 0.1 mm).
surrounding medium, or continuous phase. The droplets in a Particles should be dispersed so that they do not touch
liquid emulsion generally vary in size from 1 to 100 micrometres or overlap each other. Then, software identifies, counts
and have some sort of emulsifier on their interfaces. Creams and and determines different sizing parameters (Saragoni
many sauces are oil-​in-​water emulsions with soft, creamy, pal- et al., 2007). This method is simple and can be used
atable consistencies. A food suspension is a two-​phase system for small dry or wet particles, as well as for droplets
in which rigid or soft particles are dispersed within a liquid, as in emulsions. In the latter case, phase contrast may be
in fruit juices and some salad dressings. The sedimentation rate enhanced using chemical stains. It also provides infor-
of particles depends on their density and is proportional to the mation on the shape of particles (e.g., roundness, aspect
square of their size. ratio, etc.).
• Laboratory methods. There are several laboratory
instruments that measure and count particles down to
the nanometre size (Robins, 2006).
Processes and Equipment
Equipment used in food engineering for size reduction of solids
and emulsion formation is thoroughly described in Brennan Particle Size Distributions
(2006). Moisture content, food composition and tempera-
ture are key factors in the fracture properties of solid foods. One must be aware that any size reduction method will yield a
Usually, dry material is easier to grind than wet material, and distribution of particle sizes, which can span several decades in
lower temperatures increase brittleness and favour the breakage size. A typical size distribution graph obtained by sieve analysis
of otherwise sticky, oily or fatty foodstuffs. Since, during size (histogram) is a plot of the relative percentage fraction versus the
reduction, energy is introduced to the system, a large proportion respective size interval (Figure 76.2).
of it is converted into heat, and chefs have to be aware that this
influences odour retention and colour.
The Pacojet™ is a kitchen appliance that was patented in 1992
and that allows grinding of pre-​frozen materials as well as the
preparation of frozen desserts and ice cream. Cryogenic grinding
of frozen brittle material produces finer particles and lower vola-
tilization of aroma compounds than conventional milling. It has
been used advantageously to grind spices (Balasubraman et al.,
2012). Although, in industrial practice, many different types of
mills are used, the kind of appliance used in the kitchen depends
on the type of material (wet tissue, dry solids, liquids, frozen
foods, etc.) and the final expected result (coarse or fine paste,
powder, emulsion, etc.). Specific apparatus and applications are
dealt with in extenso in Myhrvold et al. (2011). The mini-​pimer
or hand-​held blender with a high-​speed set of rotating blades
is one of the most used appliances in home kitchens to disinte-
grate all kind of foodstuffs; its name comes from the acronym FIGURE 76.2 A particle size distribution curve (histogram and curve).
Size Reduction
A curve may be adjusted using curve-​ fitting techniques
(Barbosa-​Cánovas et al., 2005).
Most particulate systems are characterized by a single param-
eter: the average size (or the mode). Cooks should be aware that
finer and coarser material than the mean size are also present in
the product. Stored food powders may change their size distribu-
tion curves, particularly those containing amorphous components
such as sugars and protein hydrolysates. Unfavourable storage
conditions of high temperature and high moisture content favour
the formation of lumps and agglomerates, a phenomenon called
caking (Aguilera et al., 1995).
REFERENCES
Aguilera JM, del Valle JM, Karel M. 1995. Caking phenomena
in amorphous food powders. Trends in Food Science &
Technology, 6, 149–​154.
Atkins T. 2009. The Science and Engineering of Cutting: The
Mechanics and Processes of Separating, Scratching and
Metals. Oxford,
Puncturing Biomaterials, Metals and Non-​
UK, Butterworth-​Heinemann.
525
Balasubraman S, Gupta MK, Singh KK. 2012. Cryogenics and its
application with reference to spice grinding: a review. Critical
Reviews in Food Science and Nutrition, 52, 781–​794.
Barbosa-​Canóvas GV, Yan H, Harte F. 2005. Particle size distri-
bution in food powders. In GV Barbosa-​Canóvas, ed., Food
Engineering (EOLSS), Paris, UNESCO Publications, 119–​130.
Brennan JG. 2006. Food Processing Handbook. Weinheim, Wiley-​
VCH, 537–​558.
Chen J. 2009. Food oral processing –​A review. Food Hydrocolloids,
23, 1, 1–​25.
Earle RL. 1983. Size reduction. In Unit Operations in Food
Processing, Oxford, Pergamon Press, ­chapter 11.
Myhrvold N, Young C, Bilet M. 2011. Modernist Cuisine: The
art and science of cooking (Volume 2). Bellevue, WA, The
Cooking Lab.
Robins MM. 2006. Particle size analysis in food. In P Worsfold, C
Poole, A Townshend, M Miró, eds., Encyclopedia of Analytical
Chemistry. Hoboken, NJ, John Wiley & Sons, Ltd.
Saragoni P, Aguilera JM, Bouchon P. 2007. Changes in particles of
coffee powder and extensions to caking. Food Chemistry, 104,
122–​126.
Smoked Foods

Jane K. Parker and Alice Pontin

Department of Food and Nutritional Sciences, University of Reading, United Kingdom

Introduction 2007). Salmon is the most popular example of a cold-​smoked

product and retains its raw translucent deep pink colour, whereas
Smoking to impart flavour is an ancient gustatory practice –​from
hot-smoked salmon takes on a pale pink opaque appearance.
Roman townspeople’s smoking of sausages over hay or wood-
During hot-​smoking, the food is both cooked and smoked at
chip to medieval Europe’s use of smoking (of fish, after salting,
the same time, and the food acquires flavour from the smoking
in particular) to impart much-​needed flavour in the absence of
process, from the glycation reaction, from lipid degradation
condiments as well as effecting preservation. Smoking required
and from the interactions between these processes. As in cold-​
a flexible and functional architecture –​from the open central
smoking, the food is usually cured and seasoned prior to smoking,
hearth of the northern European Bronze Age roundhouse to the
and the process is typically long and slow, being carried out
use of funneled brick chimneys in the 17th and 18th centuries
between 50 and 80 °C for up to 24 hours, often in a drum fuelled
to smoke hams and other meats. Outdoor smoking –​often com-
by a firebox containing woodchips/​sawdust or a propane gas
munity smoking –​of foods depended on the vagaries of wea-
burner. Meats (beef or pork ribs, or larger cuts) are very popular
ther but remained popular in northern Europe into the late 19th
hot-smoked foods.
century.
Liquid smoke (Figure 77.1) is the leading technology used by
Whereas the tradition of smoking was born out of necessity,
the food industry, particularly for the large-​scale production of
these days, its role in preservation is secondary, and it is sought
bacon and ham. It is obtained by condensation of smoke from
after for its distinctive and multidimensional flavour. Smoke fla-
fast pyrolysis reactors. Once mixed with water, the product can
vour is a complex cocktail of smoky and burnt notes, sweet notes
be further fractionated and refined to obtain the desired flavour.
such as vanilla and caramelized sugar, more subtle fragrant notes
such as clove, woody and spicy, as well as phenolic and medicinal
notes. Smoke is used by chefs and culinary enthusiasts all over
the world in an ever-​expanding array of both sweet and savoury Odorants in Smoke
dishes, in snacks, soups, sauces, rubs and marinades in the food Given the range of methods and smoking parameters (typic-
industry, and not forgetting its role in whisky. The American chef ally 400–​500 °C for 12–​24 hours) and the different sources of
Victor Albisu describes smoke as being “as important as salt or smoke (hardwoods, softwoods, peat or corncobs), quite a range
garlic or onion … Without it, the food always feels a little empty of flavours can be achieved. The odorant compounds respon-
to me” (Shahin, 2014). sible for the flavour (Figure 77.2) are formed from the burning
of the cellulose, hemicellulose and lignin in the wood or peat.
One group of compounds that is very important for the odour
Smoking Methods of smoke is the guaiacols. They are formed from lignin, which
is a polymer of phenylpropanoid alcohols such as coniferyl and
Traditionally, food can either be cold-​smoked or hot-​smoked.
synapyl alcohols. When these break down, they form guaiacols
In the case of cold-​smoking, the raw food is usually cured to
and syringol, respectively, of which the former tend to have low
reduce the moisture content before it comes into contact with
thresholds and impart a variety of odours (Figure 77.2).
a stream of cold smoke (20–​30 °C). This usually lasts for sev-
eral hours, or even days, during which time further moisture is
• Guaiacol is one of the most important odorant
lost from the surface. No direct heat is applied to the food; the
compounds generated in smoke. Its odour is described
flavour develops from the smoke particles deposited on the sur-
as phenolic, smoky, spicy, woody and medicinal, and it
face, which reduces surface bacteria, and the odorant and taste has a very low odour threshold of 3 ppb in water.
compounds penetrate into the food. Wood chips or sawdust can • Methylguaiacol (threshold 21 ppb) is structurally
be pyrolysed on electrically thermostated plates, by dropping related, with a similar odorant character and threshold,
onto an electrically heated ring, or by friction (Varlet et al.,

527
528
FIGURE 77.1 Liquid smoke and wood used for smoking.
FIGURE 77.2 Important compounds produced by smoking.
but it is present in much lower amounts than guaiacol
and therefore has less of an impact.
• 4-​vinylguaiacol (threshold 3 ppb), again structurally
related, is far more mellow, providing warm spicy
notes, and can be reminiscent of smoky bacon.
• Figure 77.2 shows the structural resemblance of these
guaiacols to vanillin (20 ppb). Vanillin is present at
much lower levels but gives the sweet creamy vanilla
notes present in some smokes.
• Similarly, eugenol (6 ppb) and isoeugenols impart a
characteristic woody clove note.
• Syringol makes a major contribution to smoke odour,
but its odour threshold is much higher than that of
the guaiacols (1850 ppb). It gives smoke desirable
characteristics such as vanilla, powdery, balsamic and
spicy notes.
Jane K. Parker, Alice Pontin
• 4-​methylphenol (or p-​cresol) (50 ppb) is derived from
the coumaryl alcohol component of the hemicelluloses
and has a more phenolic and medicinal note than the
guaiacols.
A second group of compounds, generally referred to as furans,
are formed primarily from hemicellulose, which is a branched
polymer of pentose and hexose units. At high temperatures, these
dehydrate to give a variety of furans and pyranones similar to
those formed during caramelization.
• 2-​
furfural is quantitatively the major volatile com-
ponent in smoke. It is described as sweet, woody and
bread-​like, but its threshold (3000 ppb) is relatively
high and its contribution tends to be less than that of the
Smoked Foods
FIGURE 77.3 Changes in smoked salmon during curing and smoking.
guaiacols. This five-​carbon molecule is derived from the
pentose units in the hemicellulose, as well as from the
hexoses. It is one of the more reactive components of
smoke and is likely to undergo condensation reactions
with other aldehydes, many of which are generated in
the glycation reactions.
• Maltol has a sweet caramel candy-​floss aroma, but with
an odour threshold of 35,000 ppb, it only contributes a
sweet nuance to smoke odour.
Different Smoking Materials
Hardwoods are produced by deciduous trees such as oak, beech,
hickory, mesquite, pecan or applewood and are generally pre-
ferred to softwoods produced from evergreen trees such as
pine and fir. Softwoods have a higher lignin content (24–​33%)
compared with hardwoods (19–​28%), but also contain a resin,
which burns to give a harsh note to the smoke. Oak (62% cellu-
lose/​hemicellulose, 24% lignin) (Toledo, 2008) gives a strong but
sweet smoke due to the high hemicellulose content. On the other
hand, applewood gives a mild smoke, as the cellulose/​hemicellu-
lose fraction is only 28% and the lignin 38%.
Grasses, which are the basis for peat, have 10–​30% lignin,
and this varies with the geographical origin and the depth of the
peat in the ground (Harrison and Priest, 2009). The less stable
hemicelluloses degrade over time, so the deeper the peat is in
the ground, the longer the degradation period, and the lower the
hemicellulose content becomes. This reduces the formation of
the sweeter caramellic furans and pyranones in the smoke but
maintains the characteristic smoky guaiacols, which are formed
from the more stable lignins. Those that have been shown to be
particularly important for the peaty note are 5-​methylguaiacol
and 3-​ethylphenol (Mall and Schieberle, 2019).
Taste
As well as providing a very desirable odour, smoked ingredients
can also enhance the umami character of foods. Low
concentrations of smoke itself have little taste, but it has been
shown that low levels of smoke odour can enhance the perception
of umami in low-​salt soups.
529
Texture
Smoking also produces changes in texture. During the curing
step of smoked fish production, dry salt is layered over the sur-
face of the fillets and dissolves into the flesh, drawing mois-
ture out. Throughout the drying step, moisture evaporates
(Figure 77.3), and the surface proteins denature and coagulate,
producing a tough and sticky layer, i.e., the “pellicle”, to which
smoke compounds can readily adhere (Flick and Kuhn, 2012).
This layer contains a higher concentration of phenolics in fattier
smoked fish because they are deposited onto the surface, but in
leaner fish, 60% of phenolics can penetrate deeper than the sur-
face layer (Stolyhwo and Sikorski, 2005). Moreover, the salmon
tail contains a lower lipid level and higher moisture level than the
neck (Robb et al., 2002) resulting in a hydrophilic environment,
so the salt can diffuse readily through the flesh. The higher lipid
level in the neck and belly produces a hydrophobic environment,
slowing the rate of salt diffusion through the fillet and resulting in
a lower salt concentration below the surface layer. In the neck, the
surface evaporation rate is faster than the rate of water diffusion
to the surface due to the hydrophobic conditions, causing the
surface to dry out rapidly, trapping the moisture in the flesh and
producing a softer pellicle. These textural changes also influence
the rate of absorption and penetration of the volatiles. During the
hot-​smoking of meat, a similar “bark” is formed, which is also
full of glycation-​derived flavours.
Health Aspects
As we know from the tobacco industry, smoke also contains
known carcinogens. A range of polyaromatic hydrocarbons (PAH)
are formed during the smoking process, of which benzo[a]‌pyrene
is classified as a Group 1 carcinogen by the International Agency
for Research on Cancer (IARC). In Iceland, where traditionally
meat is smoked by the indigenous population for months at a
time, high levels of PAHs and a high incidence of stomach cancer
were observed (Bailey and Dungal, 1958). Similar associations
were observed in smoke-​dried meat in Nigeria (Alonge, 1988).
However, such extreme smoking conditions are exceptions, and
almost all smoked foods conform to regulatory requirements. The
EU regulatory limit for benzo[a]pyrene in smoked fish products
is 2 ppb. Most smoked salmon samples come 100 times lower
530
than that, but hot-​smoked salmon can occasionally reach levels
up to 200 ppb. However, concerns about the occurrence of PAHs
in smoke flavourings (liquid condensates) led the EU to initiate
an evaluation of smoke flavourings, and as a result, several were
removed from the market. In 2015, Parker et al. (2018) developed
a zeolite-​based filter that could remove up to 95% of PAHs
from a smoke stream, which allowed the preparation of smoked
ingredients with greatly reduced levels of PAHs. Aiming to show
no adverse effect on the aroma, they discovered that the flavour
was more rounded and balanced when the new filter was applied
(Chua et al., 2019).
Applications of Smoke
From its origin as an ancient method of “shelf-​life” extension,
smoke has featured for many centuries in traditional artisanal
products, whether they be traditionally smoked meats, fish or
cheese. Many are specific to the country or region, with smoked
sausages being specialties of Germany, Poland and Lithuania,
smoked paprika from Hungary, the chipotle (a traditional smoked
chili pepper) from Mexico, and kippers (herring) and Arbroath
smokies (haddock) being famous in Scotland. Extending into
beverages, whisky and lapsang souchong tea are classic smoky
products. Over the last few decades, there has been a marked
growth in the consumption of smoked foods. We have experienced
an increase in the use and sophistication of the enthusiast’s home-​
smoking apparatus, but more notably, the food industry is using
smoked ingredients to enhance flavour and fulfil the increased
customer demand for smoky BBQ flavour snacks, marinades
and condiments. Common smoked ingredients are smoked salt,
smoked pepper, smoked water for brining and smoked oils. What
is exciting today is the innovation from chefs and culinary experts
worldwide as they seek to expand the use of smoke into novel
ingredients, recipes and restaurant theatre, where a dish can be
presented under a cloche, which provides a waft of the heady
and fragrant smoky aroma when it is lifted. Smoke is making its
way into the sweet categories, with chefs exploring its use in ice
cream, syrups, chocolate, ales and ginger beer, and its creative
use in cocktails has even brought us the smoked Martini, with its
own trail of smoke. This versatile ingredient presents a world of
opportunity.
Jane K. Parker, Alice Pontin
REFERENCES
Alonge DO. 1988. Carcinogenic polycyclic aromatic hydrocarbons
(PAH) determined in Nigerian kundi (smoke-​ dried meat),
Journal of the Science of Food and Agriculture, 43, 167–​172.
Bailey EJ, Dungal N. 1958. Polycyclic hydrocarbons in Icelandic
smoked food, British Journal of Cancer, 12(3), 348–​350.
Chua XL, Uwiduhaye E, Tsitlakidou P, Lignou S, Griffiths HD,
Baines DA, Parker JK. 2019. Changes in aroma and sensory
profile of food ingredients smoked in the presence of a zeolite
filter. In: Guthrie B, Beauchamp JD, Buettner A, Toth S, Qian
MC (Eds). Sex, Smoke and Spirits: The Role of Chemistry,
Washington, DC: ACS Publications, 6, 67–​89.
Flick GJ, Kuhn DD. 2012. Smoked, cured and dried fish. In: Granata
LA, Flick GJ Jr, Martin RE (Eds). Seafood Industry: Species,
Products, Processing and Safety, Chichester: Blackwell
Publishing Ltd.
Harrison BM, Priest FG. 2009. Composition of peats used in the
preparation of malt for scotch whisky production –​Influence
of geographical source and extraction depth, Journal of
Agricultural and Food Chemistry, 57(6), 2385–​2391.
Mall V, Schieberle P. 2019. On the importance of phenol derivatives
for the peaty aroma attribute of scotch whiskies from Islay. In:
Guthrie B, Beauchamp JD, Buettner A, Toth S, Qian MC (Eds).
Sex, Smoke and Spirits: The Role of Chemistry, Washington,
DC: ACS Publications, ­chapter 9, 107–​116.
Parker JK, Lignou S, Shankland K, Kurwie P, Griffiths HD, Baines
DA. 2018. Development of a zeolite filter for removing poly-
cyclic aromatic hydrocarbons (PAHs) from smoke and smoked
ingredients while retaining the smoky flavor, Journal of
Agricultural and Food Chemistry, 66(10), 2449–​2458.
Robb DHF, Kestin SC, Warriss PD, Nute GR. 2002. Muscle lipid
content determines the eating quality of smoked and cooked
Atlantic salmon (Salmo salar), Aquaculture, 205, 345–​358.
Shahin J. 2014. Smoke: why we love it, for cooking and eating,
Washington Post, 20 May.
Stolyhwo A, Sikorski ZE. 2005. Polycyclic aromatic hydrocarbons
in smoked fish –​a critical review. Food Chemistry, 91(2),
303–​311.
Toledo RT. 2008. Wood smoke components and functional prop-
erties, International Seafood Conference Alaska Sea Grand
College Program, Fairbanks.
Varlet V, Serot T, Cardinal M, Knockaert C, Prost C. 2007.
Olfactometric determination of the most potent odor-​active
compounds in salmon muscle (Salmo salar) smoked by using
four smoke generation techniques. Journal of Agricultural and
Food Chemistry, 55 (11), 4518–​4525.
Sous Vide Cooking
Douglas Baldwin
Breville Pty Ltd, 19400 South Western Ave, Torrance, California 90501, United States
Sous vide (French for “under vacuum”) cooking has a few
working definitions. The FDA Food Code (2013) defines it as
“sous vide, in which raw or partially cooked food is placed in a
hermetically sealed, impermeable bag, cooked in the bag, rap-
idly chilled, and refrigerated at temperatures that inhibit the
growth of psychrotrophic pathogens.” This reduced-​oxygen-​
packaging sous vide processing can extend the shelf-​lives of
minimally processed foods. Since the 2000s, sous vide also
describes any precise temperature-​controlled cooking in a water
bath, especially for low-​temperature and relatively long-​time
cooks for immediate service (Baldwin, 2012). Both for imme-
diate service and for extended shelf-​life, sous vide seeks to
maximize consistency and taste.
Restaurant and home cooks who cook sous vide often prefer it
over other methods for meat, fish, and poultry. Suppose you want
to cook a 30 mm-​thick pork chop to 65 °C for a medium level of
doneness. Using traditional methods, you might do the following
to cook the pork chop: heat some butter and oil to 140 to 160 °C
in a skillet; add the pork chop and flip it every few minutes while
spooning hot butter and oil over it; when the core reaches 62 °C,
remove the chop and let it rest while the residual heat brings the
core to around 65 °C.
With sous vide cooking, you might do the following: place the
chop in an impermeable, food-​safe bag with some oil; put the
bagged chop in a 65 °C water bath for about an hour (you can
hold it for many additional hours without affecting safety, tex-
ture, or appearance); remove it from the bag and pat it dry with
paper towels; quickly sear it in a 200 to 300 °C cast-​iron skillet
for 20 seconds on each side. While faster, the traditional method
requires attention and careful timing to keep the pork chop from
being under-​or overcooked and skill to keep the surface from
burning before it’s cooked through. Sous vide cooking, by separ-
ating the cooking and the searing, gives much greater consistency
and control over texture, safety, and appearance with no worries
about overcooking.
First, select foods that benefit from precise temperature con-
trol at temperatures below the local boiling point. Chefs often
cook fish at 35 to 55 °C, tender meat and poultry at 50 to 65 °C,
tough meat at 55 to 95 °C, starchy vegetables at 75 to 90 °C, other
vegetables at 80 to 95 °C, and beans and legumes at 90 to 95 °C.
Use bath temperatures over 75 to 80 °C to extend shelf-​life or
create a braised texture. While industrial processors cook beans
and grains sous vide, restaurant and home kitchens mainly use
sous vide cooking for meat, fish, and poultry. Tender cuts, such
as tenderloin, are usually just heated through and then browned
with high heat for flavour and appearance. Other tender cuts, like
chicken breasts, are cooked and pasteurized. For tough cuts, such
as beef short ribs, a long time at a low temperature, say two to
three days at 58 °C, gives a tender, steak-​like texture; a shorter
time at a higher temperature, say 12 to 24 hours at 85 °C, gives
a succulent, braised texture – see Figure 78.1 for a map of times
and temperates used in fine-dining restaurants.
Whether to pasteurize or not depends on the hazard posed by
the raw food and the risk to the individuals being served. For
example, consider the hazards associated with a seared boneless
steak with intact muscle: with proper handling, it has no physical,
chemical, or microbiological hazards for a healthy adult, because
intact muscle is effectively sterile inside, and searing destroys
surface microorganisms. Compare this with ground poultry:
grinding spreads surface microorganisms throughout, posing
a microbiological hazard, and bone fragments pose a physical
hazard. But both would need pasteurizing if serving a high-​risk,
immunocompromised individual because of the high risk posed
by even a few active food pathogens. Cooking for extended shelf-​
life also requires pasteurization: pasteurized food in a hermetic-
ally sealed bag can be rapidly cooled and refrigerated for 30 days
below 3 °C or frozen indefinitely.
After cooking sous vide, meats are often finished with high
heat to produce the Maillard reaction for flavour and appearance.
This is often done with a hot pan or grill or under a broiler for a
short time to minimize overcooking.
Packaging
Most sous vide cooking is done in food-​ safe plastic bags.
Separating the food from the water has several benefits:
1. Cooking in a bag reduces nutrients and flavour leaching
from the food into the water.
2. Removing most of the air improves heat transfer from
the water to the food, because air is a good insulator,
and too much air causes bags to float and the food to
cook unevenly.
531
532
3. Cooking in a hermetically sealed bag prevents
recontamination after pasteurizing.
4. Pulling a vacuum before sealing can reduce fat oxida-
tion, warmed-​over flavours in pre-​cooked meat, and
microbial growth.
In larger kitchens, the food is sealed in multi-​layer hermetic-
ally sealed bags using a chamber vacuum sealer. For a chamber
vacuum sealer, the food is put in a bag, the bag is placed in
a chamber with one edge over a seal-​bar, and then most of
the air is removed from the chamber; the seal-​bar uses heat
to hermetically seal the bag, and air is allowed to reenter the
chamber. In smaller kitchens without a chamber vacuum sealer,
there are a variety of options for cooks wanting to cook by sous
vide, especially if extended shelf-​life is not important. Zipper-​
topped food-​storage bags designed for reheating food, such
as those which are microwavable, are also suitable for sous
vide cooking for immediate service: put the food in the bag,
then quickly dunk the bag in the water with the top open and
above the water to displace most of the air, then seal the bag
and submerge completely under the water. For short cooking
processes that don’t require pasteurization, it can be convenient
to clip the bag on the side of the container, but the whole bag
should be submerged when pasteurizing, since food pathogens
might have been smeared on the bag’s inside surface above the
water level.
Since these zipper-​topped bags sometimes leak or have seams
that fail, a multi-​layer textured bag sealed using a clamp-​style
vacuum sealer is more robust; when sealing food with liquid in
the bag, it is important to stop the vacuum and seal the bag before
the liquid reaches the seal-​bar and the vacuum pump, and so it
helps to have the vacuum sealer above the food. When cooking
for immediate service, reusable silicone bags designed for food
storage and cooking can be used; when using reusable bags, it’s
important to thoroughly clean them and to separate those used for
people with food allergies. Jars are also used for foods like eggs,
custards, and cheesecakes.
What is in the bag with the raw meat, fish, or poultry can affect
how it cooks. For example, adding dry salt or a concentrated salt
brine will draw water out of the flesh, while a less concentrated
brine can draw water into the flesh, and all these mass transfers
occur more rapidly as the temperature increases. An acidic mar-
inade will result in less water loss compared with cooking in a
bag without additional ingredients, with oil, or with a brine at a
similar salt concentration as the food, which all give similar water
losses. Cooking in a sauce or with strongly flavoured ingredients
can enhance the meat. Often, cooking in sauce precludes post-​
cook searing, and the sauce should be thickened so that lost juices
don’t make it too thin. For herbs and spices, vacuum sealing them
without a liquid in the bag will press them unattractively against
the food. Adding sufficient oil, brine, or stock to allow movement
also usually improves the appearance of meat, fish, poultry, and
vegetables; if the food can’t move within the bag, the edges often
look crimped from the bag. Oils and sauces also often prevent
albumins, water-​soluble proteins that denature and coagulate,
from coagulating on the surface. In general, small molecules, like
Douglas Baldwin
some volatile flavour compounds, salts, and sugars, can diffuse
into muscle, while larger molecules, like fats, cannot.
Cooking Sous Vide
In restaurant and home kitchens, most often, an immersion cir-
culator precisely controls the temperature of the water that the
packaged food cooks in. Unlike most heat sources in the kitchen,
some immersion circulators can keep the water’s temperature to
within ±0.1 °C of a set point. For meat, fish, and poultry, tempera-
ture control within ±0.5 °C is sufficient between about 35 and
70 °C and ±1.0 °C above about 75 °C. While less precise cooking
can produce noticeable differences, animal-​to-​animal variation
masks differences from more precise cooking. Sous vide cooking
is also done in convection steam ovens, in water ovens without
forced convection, and on stovetops. For steam ovens, the air
needs to be saturated with water to cook efficiently; this is diffi-
cult to achieve outside 65 to 90 °C. Water ovens and uncirculated
water baths have a lower heat-transfer coefficient than forced
convection, but it is high enough that recipes usually do not need
adjustment; the popularity of immersion circulators over water
ovens mainly comes from the circulator’s compact size and its
flexibility to use multiple bath sizes. Stove-​top sous vide is best
done with a large pot, a relatively short cooking time, a good
thermocouple or thermistor thermometer, and a lot of patience
from whoever is adjusting the burner’s dial. For cooking times
longer than a few hours or water temperatures over about 70 °C,
it is best to consider covering the bath to reduce evaporation.
Meat, Poultry, and Fish
It is convenient to divide meat, fish, and poultry into tender and
tough cuts: tender cuts just need heating through for acceptable
appearance and texture, and are sometimes held longer to pas-
teurize them for safety. Tough cuts, usually from large, powerful
muscles, benefit from longer cooking times or higher temperatures
to transform them so that they are acceptably tender.
Doneness, Appearance, and Texture
Meat is about 75% water, 20% protein, and 5% fat and other
substances. During cooking, some of these proteins denature;
which proteins have denatured determines the doneness. It’s
helpful to divide the proteins into muscle fibres, soluble proteins,
and connective tissue (see chapter on meat cooking).
The soluble proteins mostly determine the appearance, espe-
cially for red meat. When they start to denature and scatter light,
the muscle changes from translucent to opaque, and red hues
become pinker and white hues whiter. At higher temperatures, the
soluble red or purple myoglobin denatures to brown, unless cured
with nitrate or nitrite so that the myoglobin is a stable pink colour.
The texture mainly depends on the muscle and what tempera-
ture it’s heated to and for how long. Large, powerful muscles have
more fat and connective tissue than smaller, weaker muscles.
Sous Vide Cooking
With higher cooking temperatures and longer cooking times, this
fat melts and the connective-​tissue collagen dissolves into gel-
atine; this gelatine and fat let the muscle fibres easily slide against
each other when chewing, along with saliva, and make tough
cuts succulent. Tender cuts, without this abundant fat and con-
nective tissue, do not become succulent at higher temperatures;
they become dry and stringy because muscle fibres and collagen
shrink and squeeze out water, and denatured soluble proteins gel
and trap water. But at lower temperatures, tender cuts are juicy,
because biting into them gives a burst of moisture.
Tender Cuts
Since tender cuts just need heating through, it’s helpful to under-
stand how they heat. In general, the rate of heat transfer is pro-
portional to the difference in temperatures. Thus, this gives two
heat-transfer rates to estimate: how heat gets from the water to
the food’s surface and how it gets from the food’s surface to its
core. In many foods, it can be assumed that either the food is a
uniform temperature or the surface temperature equals the fluid
temperature. However, in sous vide cooking, both these effects
are important; so it takes sophisticated mathematical modelling
to determine when the food will be perfectly cooked.
Most sous vide cooking is so-​ called equilibrium cooking,
where the water bath is set to just above the meat’s desired core
temperature. The core cannot exceed the water bath’s tempera-
ture, so in that sense, the food can never overcook. Thus, while a
25 mm steak cooked in water at 55 °C is ready in about an hour, it
can be held for an additional two to three hours without a notice-
able difference in colour or texture. Since the heat-​transfer rate is
proportional to the difference between the water’s and the food’s
temperature, heating the core from 50 to 54 °C takes longer than
it took to heat from 5 to 50 °C. So, restaurants often use so-​called
high-​ΔT sous vide cooking to reduce the heating time; rather than
cook a 25 mm steak to 54 °C in a 55 °C water bath, they might
use a 65 °C water bath and use a thermometer to remove the steak
when the core reaches 53 °C after 15–​20 minutes. This high-​ΔT
sous vide cooking cuts heating time but removes the ability to
hold the food in the water bath without overcooking it.
Tough Cuts
The changes in water-​ soluble proteins and connective tissue
depend on both time and temperature. While the changes happen
quickly at a higher temperature, they occur more slowly at lower
temperatures. Above 53 to 55 °C, where food pathogens stop
growing, the holding time needed for such changes halves every 5
to 10 °C increase in temperature. However, they don’t all change
at the same rate as the temperature changes, so you might cook
beef shank at 80 °C for 16 hours for a braised-​like texture or at
60 °C for 48 hours for a steak-​like texture.
Traditionally, tough cuts like pork shoulder and beef brisket are
smoked for long periods at relatively low temperatures to trans-
form their connective tissue and fat to make them fork-​tender;
533
similarly, braising transforms tough cuts in pot roast, goulash,
and so on into tender and succulent feasts. Compared with rib
and back muscles, these leg, shoulder, butt, and neck muscles are
strong and powerful in four-​legged animals. The same muscles
in different animals require similar time-​and-​temperature com-
binations; beef, sheep, pig, deer, and buffalo shoulders have a
steak-​like texture after 24 hours at 55 to 60 °C and show a braised
texture after 12 hours at 80 to 85 °C.
Storing and Reheating
Sous vide cooking for extended shelf-​life involves rapidly cooling
the cooked food and then either refrigerating it at low enough
temperatures to inhibit pathogen growth or freezing it. The tem-
perature required to inhibit food pathogen growth depends on
the pathogen, the food’s composition, and how the food was
cooked. Let’s consider spoilage microorganisms, pathogenic
microorganisms, and spore-​ forming bacteria. Some spoilage
microorganisms are true psychrophiles and can grow in most
freezers, but cooking usually destroys them. Other microorganisms,
like Listeria monocytogenes and Yersinia enterocolitica, can grow
slowly at refrigerator temperatures below 5 °C if not destroyed
during cooking. Other bacteria, like Clostridium botulinum,
Clostridium perfringens, and Bacillus cereus, can form spores
that aren’t destroyed during pasteurization for active (vegetative)
pathogens; the spores can germinate and grow if cooled too slowly
or stored for too long (see Jay’s Modern Food Microbiology, 7th
ed. (2005) for more on food microbiology).
Let’s consider three example recipes to illustrate these cases.
Salmon fillet heated to 40 °C for 60 minutes: While 40 °C
is too low to pasteurize the salmon, a salt and sugar cure can
enhance taste, texture, and safety: the cure makes it harder
for cold-​growing food pathogens to grow. Soak a skinned and
trimmed salmon loin in a cold 10% salt, 5% sugar brine for 45
minutes. Then remove from the brine and pat dry with a paper
towel before vacuum sealing and cooking at 40 °C for 60 minutes.
Then, rapidly chill the salmon, still in the bag, in ice water before
refrigerating for up to a week and serving cold. Without the salt-​
and-​sugar cure, limiting Yersinia enterocolitica growth would
give a maximum storage time of just one day. For recipes that
combine several hurdles, or partial but not complete controls for
biological hazards, challenge tests are often necessary to demon-
strate acceptable risks.
Chicken breast cooked at 65 °C for 1 hour: This will pas-
teurize a typical boneless, skinless chicken breast for Listeria
monocytogenes and will also destroy other vegetative food
pathogens. But it will not destroy the spore forms of some
foodborne bacteria. Thus, storing below 3 °C, the minimum
temperature for C. botulinum growth, gives a 30-​day shelf-​life,
though regulations vary.
Beef short ribs cooked at 70 °C for 24 hours: The time required
to get the desired texture at 70 °C exceeds the time required to
destroy non-​proteolytic C. botulinum spores, the types of C. botu-
linum that can germinate and grow below 10 °C. However, this
gives little practical increase in shelf-​life, because the minimum
534 Douglas Baldwin

FIGURE 78.1 An illustrative map with sous vide cooking times and temperatures used in fine-​dining restaurants. Since many chemical reaction rates approxi-
mately double with a 5–10 °C increase, the time axis has a log-​scale from less than 10 minutes for kale to more than 3 days for pork cheeks.
(Copyright Breville from Baldwin et al., 2014, used with permission)

growth temperature for Bacillus cereus is 4 °C. To have signifi-
cantly longer shelf-​life requires more than just sous vide cooking;
you would need hurdles similar to those used in canning (such
as a pH below 4.6, enough salt or sugar to reduce water activity
below 0.91, or pressure cooking to destroy all spores), fermenta-
tion, or curing.
Finishing for Service
While some food is ready to serve immediately after cooking sous
vide, other foods benefit from searing, because the browning or
Maillard reaction adds savoury and roast flavours. Meat, poultry,
and fish that has been seared or grill-​marked, especially when
cooked with a sauce in the bag, does not need searing again after
cooking for flavour or appearance. Food with delicate flavours
that searing would overwhelm, such as skinless finfish, skinless
poultry, shellfish, vegetables, and fruit, is also ready for service
right from the bag.
Searing in the shortest practical time has several benefits: it
limits overcooking already cooked food and may reduce mutagen
formation. While mutagens formed in the Maillard reaction may
not be carcinogenic in humans, reducing their formation reduces
risk. Over about 150 °C, the reaction rate for forming mutagens
doubles about every 25 °C; the amount increases almost linearly
in time before levelling off after 5–​10 minutes; so using high
temperatures for just 5 to 30 seconds is unlikely to cause signifi-
cant mutagen formation. We can estimate the depth of overcooking
Sous Vide Cooking
using meat’s thermal diffusivity of between 0.1 to 0.2 mm²/​s: in
5
to 30 seconds, the heat diffuses (5 s) (0.1 mm 2
)
/s ≈ 0.7 mm to
(30 s) (0.2 mm2 /s) ≈ 2.4 mm. Thus, searing for a short time at REFERENCES
a high temperature maintains sous vide cooking’s characteristic
edge-​to-​edge doneness. For example, many cooks will dry a steak
with paper towels, lightly coat the steak with oil and season with
salt and pepper, then heat a heavy cast-​iron pan to 200 to 250 °C
and sear for 10 to 20 seconds on each side until the beef is a rich
mahogany brown. Other cooks use a blow torch, a gas or charcoal
grill or even a grate on a chimney charcoal starter, a salamander
535
or broiler, or a deep-​fat fryer; the best method depends mostly on
the recipe’s goal.
Baldwin D. 2012. Sous vide cooking: A review, International Journal
of Gastronomy and Food Science, vol. 1, pp. 15–​30.
Baldwin D. et al. 2014. Sous Vide: Beyond the Basics, ChefSteps,
Seattle, WA.
FDA. 2013. Food Code. Technical report, U.S. Department of Health
and Human Services.
Jay J, Loessner M, Golden D. 2005. Modern Food Microbiology. 7th
ed. Springer, New York.
Myhrvold N, Young C, Bilet M. 2011. Modernist Cuisine: The Art
and Science of Cooking. The Cooking Lab, Bellevue, WA.
Spherification
Linda A. Luck
Professor of Chemistry-​Emeritus, State University of New York at Plattsburgh, Plattsburgh, 12901 New York, United States
Introduction
The spherification technique is a process that was introduced
in kitchens by Hervé This in a molecular gastronomy research
program called Inicon and has been further developed and used
by the chef Ferran Adria at El Bulli since 2003. This technique
consists of creating spheres of liquid consistency encapsulated
by a thin, hydrocolloid gel-​like membrane. These spheres can be
constructed in many sizes and from many different foods. The
spheres formed are known among chefs as caviar, eggs, gnocchi
and ravioli. When they are eaten, the flavorful liquid inside the
gel membrane explodes, giving the diner an amazing culinary
experience. In addition, it is possible to encapsulate solid moi-
eties within the liquid center, expanding the culinary horizons.
Shown are caviars where maple syrup is encapsulated in the
interior of the sphere (Figure 79.1).
The spherification method exploits a chemical reaction,
namely, gelling with a hydrocolloid such as sodium alginate and
a divalent metal ion such as calcium. There are three different
strategies used in the spherification process to produce spheres
that encapsulate a variety of liquids (see Part III, which includes
a chapter with recipes using all of the three different strategies).
Basic spherification is the procedure whereby a hydrocolloid
such as sodium alginate is mixed with the food, which is liquid
in consistency. The liquid is then dropped into a bath that has
FIGURE 79.1 Maple syrup caviars on Abel Skivers. An example of reverse
spherification.
(Moltzen et al., 2017)
available calcium ions. The exterior of the liquid drop gels as
soon as it hits the calcium bath. Since all the liquid inside the gel
membrane contains alginate, the gellification process will con-
tinue as the calcium ions from the bath migrate into the liquid.
The time that the liquid spheres sit in the calcium bath is critical;
if it is too long, the entire drop solidifies. The spheres are then
removed from the calcium bath and washed in water to remove
any residual calcium on the outside of the spheres. The timing
and washing of the spheres are important in order to prevent the
solidification of the entire liquid within the membrane of the
sphere.
One drawback of this technique is that, even with the water
rinse, the spheres tend to become solidified and thus need to be
served immediately. A second drawback to this technique is the
need for the liquid center material to be devoid of calcium, which
limits the scope of foods that can used with this methodology.
Milk is an example of a product that contains calcium and cannot
be used in this particular scheme for spherification.
The second method is reverse spherification, which is much
more versatile for chefs, as many foods have inherent calcium
content. Furthermore, this method is also compatible with
alcohol. The spheres produced by this process are longer lasting
and can be stored for later consumption. For this approach, the
calcium is added to the liquid that is to be encapsulated by the
hydrocolloid membrane. The liquid is then immersed in a hydro-
colloid bath such as sodium alginate. This is the reverse of the
basic spherification technique, and is superior to that method. The
outside thickness of the encapsulating membrane of the liquid
spheres can be better controlled, because the gellification is
halted when the spheres are removed from the hydrocolloid bath.
A slight downside of this reverse method is that there is a trans-
lucent layer on the surface of the sphere. Attention must also be
paid to the thickness of the outer membrane in sphere formation,
so that it will not be evident to the palate of the diner.
Other hydrocolloids have been exploited for the bath, including
gellan, low methoxy pectin and carrageenan, of which the latter
two can be used for more acidic foods. (Sanchez, 2016). There
are two different calcium molecules that can be added to the
liquid in the interior of the sphere, calcium lactate or calcium
lactate gluconate, both of which have a neutral flavour. Calcium
chloride is not recommended, as it has a very salty taste and can
over-​power the food product.
537
538 Linda A. Luck

Cryospherification or molded spherification is an approach acid group is removed, leaving the carboxyl group with a nega-
that is useful in both basic and reverse techniques, although use tive charge. The pKa of the carboxyl group is important, as it
with the reverse method is more prevalent. In this system, the determines what charge is on that group, i.e., whether it is neutral
material that will be spherified is frozen in a silicon mold. In or negatively charged at a specific pH. In the case of gelation, it is
the basic technique, this is the liquid containing alginate, and in important to know that the gels will precipitate at a pH below the
the reverse technique, the liquid with the calcium added. The pKa of the carbohydrates, which is why the spherification of very
frozen molds are added to the subsequent baths, and as the frozen acidic liquids is difficult in this process.
molds thaw, the outer layers of the molds slowly gellify. This Full hydration of the alginate is important for functionality
practice allows tight control over the shape of the end product and as a thermally stable cold setting gel agent. For the purposes
can be used to make larger spheres. of spherification of a food product, the alginate is packaged in
powder form as sodium alginate, and thus, the monovalent ion
Na+ is complexed with the acid. This formulation will allow the
alginate to dissolve in water at lower temperatures with no need
The Science behind the Technique
to heat the gel material. Dry alginates, however, can be obtained
All the methods described so far are based on the premise that in complexed form with other cations such as K+, NH + and pro-
4
gelling occurs when a hydrocolloid such as sodium alginate, pylene glycol for purposes other than culinary (FMC, 2003).
a complex carbohydrate, sequesters a divalent ion such as cal- Gel formation of the alginate takes place when a divalent ion
cium, inducing crosslinking of the long, fibrous sugar strands. such as calcium replaces the sodium ion complexed to the carbo-
The crosslinking induces a phase transition in the alginate from hydrate moiety on the alginate molecule. Since the calcium ion
a viscous liquid to a semi-​solid to a solid form. Alginates, which is divalent, with two positive charges, it has the ability to bind to
were discovered by the British chemist E. C. Stanford, are water-​ two negatively charged units from neighboring alginate strands.
soluble biopolymers extracted from brown seaweed (Draget, With a multitude of alginate molecular strands and calcium ions
2009). These seaweed extracts have been purified and used in crosslinking, a gel network is formed. The calcium ion trapped in
many processes, such as gelling, film-​forming, or as thickening the mesh of alginate has been described as an “egg-​box” model
and stabilizing agents for cooking, and have a number of (Figure 79.3) due to the ionic interactions between the anions and
applications in the health industry (Parreidt, 2018). cations of the system, mainly on the G units (Grant et al., 1973).
Alginates are composed of the sugar components, b-​ D-​ The gel strength is contingent on the composition of the G, M and
mannuronic (M) and a-​L-​guluronic (G) acids, linked by glyco- MG blocks in the alginate. The higher the number of G blocks,
sidic bonds in linear, not branched, configurations. The number the more rigid the gel (Morris et al., 1973). Other divalent ions
and sequence of these sugars can vary depending on the alginate can be utilized, and alginate has an affinity for the alkaline earth
(Figure 79.2). There are G blocks within the alginate, consisting metals in the following order: Ca++ < Sr ++ < Ba++ (Sutherland,
of regions of solely G units, M blocks consisting of only M units, 1991). Other divalent, trivalent and tetravalent metals induce
and regions that have a combination of M and G (Skurtys et al., gelation (and spherification) but are not suitable for use in food
2010). The proportions and length of these blocks determine the items due to their toxicity (Konieczny et al., 2017). Monovalent
chemical and physical properties of the alginates, including the ions such as sodium do not cause gelation (Kohn, 1975).
charge, viscosity and solubility. The most important property of Even though the gel appears rigid, there is significant
the alginates in the process of spherification is the charge on the movement of water and other chemical molecules in and out of
carbohydrate units. The M and G carbohydrate units have carb- the gel. Molecules may be trapped within the matrix by chemical
oxyl groups, which can be neutral or charged depending on the capillary forces, while others may diffuse freely, depending on
pH of the solution/​food. In neutral water at pH 7, the carboxyl their size. This consideration must be addressed when forming
groups on M and G units will be negatively charged, as the pKa the encapsulating barrier around the sphere and storage of the
of the carboxyl groups is 3.38 for mannuronic acid and 3.65 spheres before consumption. Molecules of color and flavor
for guluronic acid. At a pH above the pKa, the hydrogen on the may diffuse out of the interior of the caviars. For example,

–OOC

O OH
–OOC
OH
O
O
O
O
HO O

OH

Guluronic acid (G) Mannuronic acid (M)

FIGURE 79.2 Chemical structures of guluronic acid (G) and mannuronic acid, the two carbohydrate structures that make up the alginate molecule.
Spherification
FIGURE 79.3 The “egg-box” model for the binding of the calcium ions
with the G units on the alginate strands.
encapsulation of green cream de menthe in a caviar loses its color
after time, as the green molecule escapes through the gel matrix
when it is stored in water.
In addition to the properties described, the gel formation is
also dependent on the source of calcium ions. It has been shown
that calcium chloride produces a much faster-​forming gel than
calcium lactate or calcium gluconate (Soazo, 2015). This may be
in part due to the solubility of the salt, as calcium chloride is the
most soluble.
In the case of spherification, we are specifically interested in
forming a very thin layer of gel on the outside of the sphere and
keeping the interior of the sphere liquid. Thus, we also need to
carefully optimize the timing of the sphere in the gelation bath.
To this end, a very important chemistry principle underpinning
this spherification process is the movement of the calcium ions
at the molecular level. This movement can be described in terms
of a “random walk” of the ions in solution. The alginate strands
in solution are associated with the monovalent sodium ions when
divalent calcium ions are introduced. The calcium ions move in
the water or other liquid, interact with the alginate, and replace the
sodium ions associated with the G groups. They do this in a random
way, similarly to a gas smell permeating a room or thermal energy
passing through meat on a grill. This diffusion or movement can
be calculated by an equation discovered by Bachelier (1901) and
later developed by Einstein (1905). Using this equation, we are
able to determine how long it will take for the calcium ions to
diffuse specific distances into the alginate network, i.e., how long
to leave the liquid in the setting bath to give a specific thickness
of the membrane surrounding the liquid for the spherification pro-
cess. The following equation is used for this calculation:
L= (4 Dt )
539
where L is the distance that the calcium ions move. For example,
in reverse spherification, the calcium must move from the liquid
into the alginate layer on the surface of the forming sphere. L is
the measure of the distance over which it is desirable for the cal-
cium ion to move and hence is proportional to the thickness of the
membrane to be formed. D is the diffusion constant for the ion
in the liquid or material, and lastly, t is the time for the process.
The units for D are meters2/​seconds, L is in meters and t is in
seconds. The D value for a calcium ion is 8 × 10−10 m2/​s, and this
range is similar for most divalent ions. If a 0.5 mm or 0.0005 m
thick gel is desired for the spheres, which is L, then 78 seconds
is calculated as the time required for the process (Brenner et al.,
2015). Using this equation, one can calculate how long (t) the
sphere must remain in the bath in order to produce the desired
thickness of the membrane.
(See the chapter on ionic diffusion in spherified calcium alginate
gels by Berardi et al. in the education section of this book.)
The Future
The future is bright with new ideas for using this methodology
to help humankind. One of the most interesting applications
of spherification is the edible water bottle (Bealsscience, n.d.;
iflscience, n.d.; CNET, 2015). Water and other fruits and liquid
can be encapsulated exploiting this spherification methodology.
As we produce an enormous amount of plastic water bottles a
year, this would surely decrease the waste and help clean up the
environment. The specific application of these eco-​friendly edible
containers for juices and other liquids in schools and nursing
homes could revolutionize our way of delivering liquids to large
groups of clients without waste, including paper cups, straws and
plastic bottles, and under more sanitary conditions.
REFERENCES
Bachelier L. 1901. Theorie Mathématique du Jeux, Ann. Sci. Ec.
Normale Super, 143–​210.
Bealsscience. n.d. Available at www.bealsscience.com/​single-​post/​
2018/​02/​23/​How-​to-​make-​Edible-​Water-​Bottles
Brenner MP, Sorensen P, Weitz DA. 2015. Science and Cooking.
A companion to the Harvard Course. Harvard University,
Boston, MA.
CNET. 2015. Available at www.cnet.com/​news/​appliance-​science-
edible-​water-​bottles-​and-​the-​strange-​chemistry-​of-​spherifi
cation/​
Draget KI. 2009. Alginates. In: GO Phillips and PA Williams, eds.
Handbook of Hydrocolloids, 2nd ed. Woodhead Publishing,
Sarston, UK, 807–​828.
Einstein A. 1905 On the motion of small particles suspended in
liquids at rest required by the molecular kinetic theory of heat.
Annalen der Physik, 17, 549–​560.
FMC. 2003. Online pamphlet. Available at https://​mafiadoc.com/​
alginates-​fmc-​biopolymer_​59bde7011723dd0c40f86d1e.html
Grant GT, Morris ER, Rees DA, Smith PJC, Thom D. 1973.
Biological interactions between polysaccharides and divalent
cations: The egg-​box model. FEBS Letters, 32, 195–​198.
Iflscience. n.d. Available at www.iflscience.com/​chemistry/​how-​
make-edible-​water-​bottles/​
Kohn R. 1975. Ion binding on polyuronates –​alginate and pectin.
Pure and Applied Chemistry, 42, 371–​397.
540
Konieczny A, Moltzen C, Luck LA. 2017. The role of metals in
reverse spherification: A molecular gastronomy project. Sigma
Xi Meeting. SUNY Plattsburgh, NY.
Moltzen C, Konieczny A, Luck LA. 2017. Hydrocolloid
Spherification Using Pure Maple Syrup. NYS Academy of
Nutrition and Dietetics Annual Meeting and Exposition. Lake
Placid, NY.
Morris ER, Rees DA, Thom D. 1973. Characterization of polysac-
charide structure and interactions by circular dichroism: Order-​
disorder transition in the calcium alginate system. Journal of
the Chemical Society, Chemical Communications, 7, 245–​246.
Parreidt TS, Müller K, Schmid M. 2018. Alginate-​based edible films
and coating for food. Food, 7, 170–​208.
Sanchez J. 2016. Molecular Gastronomy: Scientific Cuisine
Demystified. J. Wiley and Sons, Inc., Hoboken, NJ.
Linda A. Luck
Skurtys O, Acevedo C, Pedreschi F, Enrione J, Osorio F, Aguilera
JM. 2010. Food hydrocolloid edible films and coatings. In:
CS Hollingworth, Ed. Food Hydrocolloids Characteristics,
Properties and Structures. Nova Science Publishers, Inc.,
New York, NY, 41–​80.
Soazo M, Báez G, Barboza A, Busti PA, Rubiolo A, Verdini R,
Delorenzi NJ. 2015. Heat treatment of calcium alginate films
obtained by ultrasonic atomizing: Physicochemical character-
ization. Food Hydrocolloids, 51, 193–​199.
Sutherland IW. 1991. Alginates. In: D Byrom (ed.). Biomaterials:
Novel Materials from Biological Sources, 1st ed. Palgrave
Macmillan, Basingstoke, UK, 307–​331.
Squid: Gastrophysics of Squid –​from Gastronomy to Science and
Back Again
Ole G. Mouritsen1, Charlotte Vinther Schmidt1, Peter Lionet Faxholm1 and Mathias Porsmose Clausen2
1 Department of Food Science, Taste for Life, Design and Consumer Behavior, University of Copenhagen,
26 Rolighedsvej, DK-​1958 Frederiksberg C, Denmark
2 Department of Green Technology, SDU Biotechnology, University of Southern Denmark, Campusvej 55, DK-​5230
Odense M, Denmark
The history of physics has demonstrated repeated ventures into
other disciplines turning them into physics, e.g., astronomy,
geology, chemistry, and biology, and this is possibly about
to happen with gastronomy, turning some gastronomy into
gastrophysics. We will demonstrate how this works by using
squid as an example.
Introduction
Gastrophysics is an interdisciplinary field of science that can
be defined as qualitative reflections and quantitative studies of
all gastronomic aspects pertaining to food, including culinary
precisions and transformations, preparation techniques, texture,
and organoleptic properties, with a focus on physical effects
and physico-​chemical characterization (Mouritsen 2012; 2016;
Mouritsen and Risbo, 2015). The empirical basis of gastrophysics
is gastronomy itself, as well as food and food preparations of
specific gastronomical value and potential. It is possible that
gastrophysics in combination with neurogastronomy (Shepherd,
2012) can furnish the scientific underpinnings of gastronomy at
large. This definition of gastrophysics is different from the one
recently adopted by Spence (2017), who put less emphasis on the
physical sciences and more on psychology.
The starting point for a gastrophysical study is a gastronomic-
ally inspired question. In the context of cephalopod gastronomy
(Mouritsen and Styrbæk, 2018; 2020), which deals with the
culinary aspects of squid, cuttlefish, and octopus, key gastronomical
questions pertain to texture and flavour. A relevant question could
be: how does one treat the muscular tissues of cephalopods, spe-
cifically the mantle and arms, and for squid also the tentacles and
retractor muscles, to obtain a particular structure that leads to a
desired texture and mouthfeel? Although many cookbooks contain
recipes for dishes with cephalopods, and there are a couple of older
cookbooks dedicated to octopus and squid (Cronin, 1981; Schultz
and Regardz, 1987), there is very little published about the culinary
sciences of cephalopods. We shall here demonstrate how we have
carried out a gastrophysical investigation of the texture and flavour
of Danish squid, leading to new scientific insights, new culinary
precisions, and new products of gastronomical value (Faxholm
et al., 2018; Schmidt et al. 2020, 2021).
The Challenge of Tough Muscular Fibres
Cephalopods such as squid contain about four times as much
collagen as finfish, and the collagen is much more cross-​bound
and hence much stronger in cephalopods. Cross-​binding is a
determining factor for the toughness of the tissues. In addition,
the muscle fibres in cephalopods are longer and typically ten
times thinner than in finfish, rendering the muscular structure of
cephalopods smoother than in finfish. These facts, together with
the three-​dimensional organization of the cephalopod muscles,
are the main reasons why these organisms can use the principle
of a muscular hydrostat to exhibit an artistic degree of mobility
in all directions. The structure of these fibres controls the texture
and mouthfeel of cephalopods and is hence of utmost importance
for how texture makes taste (Mouritsen and Styrbæk, 2017). The
challenge in the culinary techniques applied to squid is how to
avoid toughness and provide a desirable mouthfeel.
The organization of the muscular fibres in different cephalopods
and in different parts of the cephalopods, i.e., arms, tentacles, and
mantle, are different (Mizuta et al., 2003; Kier and Stella, 2007;
Kier, 2016). These differences are crucial for the use of ceph-
alopod meat as food and how tender it will be. For example, squid,
because of their special ability to perform jet-​repulsion swimming,
have particularly strong muscles circularly organized around the
mantle. This implies for fried squid dishes that cutting the conven-
tional rings across the mantle is, in fact, the worst possible way of
achieving mechanical tenderization. Cutting along the long direc-
tion of the mantle severs more muscle fibres and leads to a more
tender product with a more appealing mouthfeel (Melendo et al.,
1997). Some insight into the physical structure of the muscles in
cephalopods is therefore a useful guide for the field of gastronomy.
541
542 Ole G. Mouritsen et al.

Texture of Squid laser light, such a material can ‘combine’ two incoming photons
There are many different ways of preparing squid, from raw, into one and generate light with exactly half the wavelength of
raw-​marinated (ceviche), cooked, steamed, grilled, smoked, and the incoming light. This can be used to image the structure of
fried to fermented in its own intestinal enzymes (Mouritsen and squid muscles and determine how the microscopic organization
Styrbæk, 2018). We have investigated the application of sous of collagen changes during preparation. The technique allows
vide techniques for tenderizing squid mantle (Faxholm et al., non-​invasive imaging of the structure of squid muscles, thereby
2018; Schmidt et al., 2021) and studied the resulting texture by revealing how the microscopic organization of collagen changes
texture analysis and the collagen structure by second-​harmonic-​ during preparation. Furthermore, since it uses far-​red laser light,
generation microscopy. it can penetrate deep inside tissue (Brüggemann et al., 2010).
Texture analysis is a mechanical technique that involves Examples of micrographs of squid mantle before and after
applying force, e.g., by compression or shearing, to the material sous vide treatment are shown in Figure 80.1. In the raw tissue
using different probes (cylinder, knife-​blade, etc.). The result is a (Figure 80.1a), the collagen organizes in a dense network of long
graph showing the relationship between the applied force of the linear bundles that entangle in the lateral direction in a roughly
probe and the deformation of the tested material over time for perpendicular manner. After sous vide treatment (Figure 80.1b),
a given loading rate. Proper interpretation of the data can yield the signal intensity decreases, indicating that the heat treatment
information about textural parameters like hardness, springi- has changed the molecular structure of the collagen as the squid
ness, toughness, stickiness, and crispiness. Figure 80.1 shows an becomes more tender. A few intact collagen fibres are seen,
example of such a measurement of mantle from squid (Loligo surrounded by what appears to be partly or fully denatured collagen.
forbesii) before and after sous vide treatment. From the figure, Texture analysis and microscopy do not reveal any information
we can determine the maximum force needed to shear the squid about what textural profile and mouthfeel may be perceived by
meat (maximum shear force) as well as calculate the area below humans. Hence, we performed a sensory evaluation using a so-​
the force versus time curve (toughness) in a defined interval. The called quick method developed and modified for the present pur-
maximum shear force and toughness reveal information about the pose with a focus on attributes of taste, texture, and mouthfeel.
muscle tenderness, since they are inversely correlated. The results of this test revealed that, even if the longest cooking
Texture analysis profiles may be compared with results from time resulted in a more tender texture profile, this preparation
microscopy, as the applied force of the probe depends on the was not the most liked. Rather, our evaluation demonstrated that
structure and state of the collagen network inside the meat. The a relatively short cooking time (but not raw) was preferred, as
highly ordered molecular structure of collagen fibrils and the it yielded a better taste and a more stimulating texture during
lack of optical symmetry give collagen the ability to interact with mastication, whereas longer cooking times yielded an over-​
light in a non-​linear fashion and behave as a so-​called second-​ cooked surface, a bitter taste, and a dull mouthfeel throughout
harmonic-​ generation material. When illuminated with strong mastication.

FIGURE 80.1 Force versus time at loading rate 2 mm/​s for a squid mantle sample before and after sous vide treatment (55 °C for 0.5, 1.0, and 1.5 h). Insets
show second-​harmonic microscopy images of the collagen structure when raw (a) and after sous vide treatment (b). The scale bars correspond to 50 μm.
Squid: Gastrophysics of Squid
Umami Taste of Squid
One of the most characteristic features of the flavour profile of
seafood is umami, the fifth basic taste (Mouritsen and Styrbæk,
2014). This is true of fish, shellfish, and cephalopods, as well as
seaweeds. Umami is elicited by free glutamate interacting with
the glutamate receptor, often in a synergistic manner with free
nucleotides such as inosinate, adenylate, and guanylate (Zhang
et al., 2008; Mouritsen and Khandelia, 2012). There is, how-
ever, very little information available about the contents of these
compounds in squid. Japanese squid has been found to contain
high levels of adenylate, up to 184 mg/​100 g (Yamaguchi and
Ninomiya, 2000).
We have measured the levels of free glutamate in Danish
squid by chemical analysis (Faxholm et al., 2018; Schmidt et al.,
2021) and found that squid mantle contains 63 mg/​100 g (wet
weight) (Figure 80.2). This value is somewhat higher than the
content reported from Japanese squid (UIC, 2018) of 20–​30 mg/​
100 g, a value that presumably is based on whole Japanese flying
squid (Todarodes pacificus) and not only mantle. We have also
measured the total amount of glutamic acid in Danish squid and
found that the relative potential for forming free glutamate is
much larger than indicated by the content of free glutamate. This
potential may be released by culinary processes, in particular
hydrolysis and fermentation. Further investigations are needed to
indicate which preparation techniques will yield the most desir-
able amino acid profile outcome possible.
The data in Figure 80.2 for raw squid mantle shows that it
contains sweet-​tasting free amino acids (Ala, Gly, Thr, and Ser)
and a relatively high content of umami-​potent free amino acids
(Glu and Asp). This suggests that the sweet taste known from
fresh raw squid is caused to some extent by the presence of sweet
free amino acids. Furthermore, squid may have a great potential
for umami taste when prepared, through hydrolysis and liberating
the relatively high amount of Glu and Asp.
FIGURE 80.2 Free and total amino acid composition (mg/​100 g) of raw squid mantle. Ala, Gly, Thr, and Ser taste sweet, and Glu and Asp taste umami.
543
Some New Squid Products of Gastronomical
Value
We have developed a number of prototype products from Danish
squid using mantle, arms, and tentacles (Faxholm et al., 2018).
As an example, where the focus has been on developing a tex-
tural profile that is assumed to be acceptable to Western palates,
we have cooked squid mantle both to tenderize the mantle and to
preserve it as a confit. By cutting the mantle into thin strips along
the long direction of the mantle, which is across the strongest
(circular) muscle fibres, it is possible to obtain a product with
an appealing curly appearance when the squid strips have been
cooked in sunflower seed oil at 85 °C for 35–​40 min, depending
on the thickness of the mantle. The curling is caused by the diffe-
rence in the collagen structures at the surface compared with the
inner part of the mantle. By adding fat to the very lean squid
meat in this preparation, a much smoother texture is obtained
compared with equivalent low-​fat cooking methods. The curly
mantle confit is silky soft and can be served as is with lemon
juice, chopped parsley, salt, and pepper (Figure 80.3). This curly
squid can also be served cold, like pasta, in a cold salad, or in
most ways in which pasta is used.
We believe that, as demonstrated in this chapter, by applying
scientific reasoning and methodologies (gastrophysics) along
with gastronomic exploration of Nordic squid, it is possible to
promote a more varied use of an underappreciated and underused
food resource from Nordic waters.
Acknowledgements
This work was supported in part by a grant to Smag for Livet
(Taste for Life) from Nordea-​fonden and by a grant from
VILLUM Foundation (Grant number 00025414).
544
FIGURE 80.3 ‘Silky squid’ (curly mantle confit) of squid mantle strips.
(Courtesy of Jonas Drotner Mouritsen)
REFERENCES
Brüggemann DA, Brewer J, Risbo J, Bagatolli L. 2010. Second har-
monic generation microscopy: a tool for spatially and tem-
porally resolved studies of heat induced structural changes in
meat. Food Biophysics, 5, 1–​8.
Cronin I. 1981. The International Squid Cookbook. Aris Books,
New York.
Faxholm PL, Schmidt CV, Brønnum LB, Sun YT, Clausen MP, Flore
R, Olsen K, Mouritsen OG. 2018. Squids of the North: gas-
tronomy and gastrophysics of Danish squid. International
Journal of Gastronomy and Food Science, 14, 66–​76.
Kier WM. 2016. The musculature of coleoid cephalopod arms and
tentacles. Frontiers in Cell Development Biology, 4, 10.
Kier WM, Stella MP. 2007. The arrangement and function of
octopus arm musculature and connective tissue. Journal of
Morphology, 268, 831–​843.
Melendo JA, Beltán JA, Roncalés P. 1997. Tenderization of squid
(Loligo vulgaris and Illex coindetii) with bromelain and a
bovine spleen lysosomal-enriched extract. Food Research
International, 30, 335–341.
Mizuta S, Tanaka T, Yoshinaka R. 2003. Comparison of collagen
types of arm and mantle muscles of the common octopus
(Octopus vulgaris). Food Chemistry, 81, 527–​532.
Ole G. Mouritsen et al.
Mouritsen OG. 2012. The emerging science of gastrophysics and its
application to the algal cuisine. Flavour, 1, 6.
Mouritsen OG. 2016. Gastrophysics of the oral cavity. Current
Pharmaceutical Design, 22, 2195–​2203.
Mouritsen OG, Khandelia H. 2012. Molecular mechanism of the
allosteric enhancement of the umami taste sensation. FEBS
Journal, 279, 3112–​3120.
Mouritsen OG, Styrbæk K. 2014. Umami: Unlocking the Secrets of
the Fifth Taste. Columbia University Press, New York.
Mouritsen OG, Risbo J (eds.). 2015. Special issue on the emerging
science of gastrophysics. Flavour, 4.
Mouritsen OG, Styrbæk K. 2017. Mouthfeel: How Texture Makes
Taste. Columbia University Press, New York.
Mouritsen OG, Styrbæk K. 2018. Cephalopod gastronomy –​a
promise for the future. Frontiers in Communication, 3, 38.
Mouritsen OG, Styrbæk K. 2021. Octopuses, Squid & Cuttlefish:
Seafood for Today and for the Future. Springer, New York.
Schmidt CV, Poojary MM, Mouritsen OG, Olsen K. 2020. Umami
potential of Nordic squid (Loligo forbesii). International
Journal of Gastronomy and Food Science, 22, 100275.
Schmidt CV, Plankensteiner L, Faxholm PL, Olsen K, Mouritsen
OG, Frøst MB. 2021. Physicochemical characterisation of
sous vide cooked squid (Loligo forbesii and Loligo vulgaris)
and the relationship to selected sensory properties and hedonic
response. International Journal of Gastronomy and Food
Science, 23, 100298.
Schultz J, Regardz B. 1987. Calamari Cookbook: Exploring the
World’s Cuisines Through Squid. Celestialarts, Berkeley, CA.
Shepherd G. 2012. Neurogastronomy. Columbia University Press,
New York.
Spence C. 2017. Gastrophysics: The New Science of Eating. Penguin,
New York.
UIC. 2018. Umami Information Center, Tokyo. www.umamiinfo.
com/​richfood/​foodstuff/​seafood.php#ANCHOR08. Accessed
on 21 November 2018.
Yamaguchi S, Ninomiya K. 2000. Umami and food palatability.
American Society for Nutrition Science, 130, 921S–​926S.
Zhang FB, Klebansky B, Fine RM, Xu H, Pronin A, Liu H, Tachdjian
C, Li X. 2008. Molecular mechanism for the umami taste syn-
ergism. Proceedings of the National Academy of Science of the
United States of America, 105, 20930–​20934.
Sugars: Soft Caramel and Sucre à la Crème –​an Undergraduate
Experiment about Sugar Crystallization

Irem Altan1, Patrick Charbonneau1,2 and Justine de Valicourt3,4

1 Department of Chemistry, Duke University, Durham, North Carolina, United States
2 Department of Physics, Duke University, Durham, North Carolina, United States
3 Vestjyllands højskole, Ringkøbing, Denmark
4 Les Colibris –​Entreprise permacole, St-​Jean-​de-​Matha, Québec, Canada

Introduction presence in parts of the United States, finding room, for instance,
in Frye’s Practical Candy Maker (Frye, 1885), and in the revised
Candy-​ making entails finely controlling the phase transitions
edition of The Practical Cook (Sears, 1892). The candy even
of aqueous solutions of sugar. Both kinetic and thermodynamic
captured the attention of an anonymous Englishman visiting
aspects of sugar crystallization are indeed at play in the art of
Vermont in the 1870s (Anonymous, 1873). Whatever the origin of
confectionery (Hartel, 2013). As part of a course on the chem-
sucre à la crème may be, it long predates that of fudge (Benning,
istry and physics of cooking taught at Duke University, we have
1990; Larousse 2009), a closely related North American delicacy
illustrated these undergraduate-​level concepts by preparing and
that only appeared late in the 19th century.
contrasting soft caramel and sucre à la crème recipes that, remark-
ably, differ only in their tempering. This laboratory experience
allows students to explore the fine dependence of look and mouth
feel of candies on the microstructure of sugar. In this chapter, we Physical Chemistry of the Confections
provide some background on these two candies and detail the
Recipes for soft caramel and sucre à la crème can vary, but, from
physical chemistry involved in controlling their sucrose compos-
a materials standpoint, both candies are straightforward mixtures
ition and microstructure.
of (maple) sugar and dairy fat. The exact source of sugar and fat
may change; condensed milk, cream and butter all do the trick,
and maple syrup (or cheaper substitutes) nowadays can replace
Historical and Cultural Background maple sugar (Table 81.1). The deliciousness of the candy relies
While soft caramel is a fairly generic candy (Larousse, 2009), more on the precise control of material properties than on the
sucre à la crème is associated with a geographically spe- refinement of its constituting ingredients.
cific region. Its precise origin remains uncertain, but its deep Specifically, to obtain soft caramel or sucre à la crème, one
connection with the French Canadian experience is undebatable: must properly tune the concentration and microstructure of sugar.
one of its key ingredients, maple sugar, is indeed indigenous to The former is set by the terminal boiling point temperature, the
the north east of the American continent. Although recipes for second by shearing (or not) the mixture as it is tempered. In order
sucre à la crème did not appear in the earliest French-​language to better understand both processes, we refer to the temperature–​
cookbook written in Montreal (Perrault, 1840; Driver, 2008) –​ concentration sucrose–​water binary phase diagram at standard
possibly because of a certain contemporary elitism towards the pressure (Figure 81.1). Because maple syrup is a simple aqueous
maple sugar used in rural households (Parker, 2006) –​by the turn solution of sucrose, 68% by weight, i.e., near the eutectic point,
of the 20th century it had become a staple of the genre (Caron, with <1% other sugars and compounds (Ball, 2007), this refer-
1883; Ogilvie, 1905; Perrault, 1984; Driver, 2008). The candy ence offers a useful framework for interpreting its phase behavior.
even captures the essence of the French Canadian experience A caveat about that phase diagram should, however, be
for Patriote François-​Xavier Prieur, who fondly reminisced of highlighted. The version presented here comes from a critical
early 19th-​century settlers’ huts with “some twists of tobacco, compilation of various experimental observations (Starzak and
pipes, small bottles containing pepper, cinnamon, nutmeg, sticks Mathlouthi, 2006), but estimates of the boiling point elevation
of sucre à la crème for the children etc. etc.” (Prieur, 1864; line above around 85% sucrose vary by as much as 5 °C (Hartel,
Prieur, 1949). 2011). This regime is subject to a reasonably large uncertainty
However, sucre à la crème is not exclusively found in French because it falls in a metastable, out-​of-​equilibrium range (Brady,
Canada. Maple cream, as it is also known, has a marked cultural 2009), and hence the measurements are preparation-​dependent.

545
546 Irem Altan

TABLE 81.1
Typical Recipes for Soft Caramel and Sucre à la Crème
Ingredients 540 g maple syrup
240 g heavy cream (35% milk fats)
45 g butter
Preparation (i) Mix all the ingredients and bring to a boil until reaching 118 °C. To avoid the formation of sucrose crystals from the splatters, avoid
stirring the mixture once boiling is attained and keep the sides of the pot clean.
(iia) Soft caramel: pour delicately into a featureless container and cool to room temperature (or below).
(iib)  Sucre à la crème: place the pot into a cold water bath to quench the mixture to 55–​75 °C, then whisk vigorously. Transfer the mixture
into a mold, trace the squares for individual servings and cool down further.
(iii) When it is around room temperature, slice into individual servings. Dipping the knife in warm water can help.

Maple syrup being 68% sucrose, it can be replaced with 370 g of maple sugar, which is nearly pure sucrose (and the ingredient historically used for this recipe),
or with 385 g of brown sugar, which is 95% pure sucrose (and is much more affordable).

fully controlled by the sucrose concentration. This suggests that

FIGURE 81.1 Binary phase diagram for sucrose–​water under standard
pressure conditions. Vaporization and crystallization lines are drawn from
compiled data (Starzak and Mathlouthi, 2006). The broadening of the
boiling point elevation line at high sucrose concentration reflects the experi-
mental data scatter in this regime. Note that the glass transition line (Hartel,
1993) should not be viewed as a thermodynamic phase transition but rather,
as an out-​of-​equilibrium marker of dynamical arrest for the (metastable)
supercooled solution. The line reported here is for pure sucrose; milk fats
shift it to higher temperatures. The red path denotes a typical protocol for
preparing soft caramel and sucre à la crème.
In this regime, a careful determination of sucrose concentra-
tion from the boiling point alone is thus inherently somewhat
imprecise. Despite this ambiguity, for the sake of this chapter,
Figure 81.1 provides a sufficient experimental guide.
In both soft caramel and sucre à la crème recipes, ingredients
are heated up to 18 °C above the boiling point of pure water,
i.e., 118 °C under standard pressure conditions. Because the
molecular weight of milk triglycerides (Breckenridge, 1967) is
about three times that of sucrose (~800 g/​mol or more vs. 342 g/​
mol) and only about a third by weight is used in the recipe (130 g
vs. 370 g), the molarity of sucrose is about an order of magnitude
larger than that of triglycerides. The contribution of the latter to Conclusion
the boiling point elevation is thus likely to be minimal. For sim- In this chapter, we have briefly reviewed how basic notions of
plicity, we here approximate this colligative property as being physical chemistry relate to making soft caramel and sucre à la
a sugar concentration of about 85–​90% is reached at the target
boiling temperature of both preparations.
Once the target sucrose concentration is reached, the mixture is
tempered to room temperature or lower. Cooling the mixture suf-
ficiently quickly, within a smooth container and without external
shocks, inhibits heterogeneous crystal nucleation. This protocol
permits the viscous supercooled fluid to persist without the for-
mation of macroscopic crystals (Figure 81.2). The presence of
milk fats and other solutes helps suppress sucrose nucleation.
If this soft caramel is then kept at room temperature for a few
days, the metastable fluid does eventually crystallize, hence
reaching thermodynamic equilibrium. Although further lowering
temperature, e.g., below 0 °C, increases the drive for sucrose
to crystallize (by growing the difference in chemical potential
between the supercooled liquid and the crystal), the caramel is
then nonetheless better preserved. The reduction in mass trans-
port at this temperature more than compensates for the reduced
barrier to nucleation.
If, in contrast, the mixture is sheared upon cooling, then
crystallites quickly form. Because most of these small crystals
grow as the mixture is cooled, the degree of shearing and the
temperature at which it takes place determine the degree of
polycrystallinity, and thus the mouth feel, of the final sucre à la
crème (Figure 81.2). Interestingly, it is also possible to obtain
a product intermediate between a soft caramel and sucre à la
crème by seeding the nucleation process with sucrose crystals
(Jeffery, 2001).
The final confections –​grained and ungrained, in confectionary
speak –​differ markedly from one another. The shine of soft
caramel reflects its molecularly homogeneous nature, while the
matte finish of sucre à la crème captures the intense scattering off
its constituting micron-​scale crystals (Figure 81.2). The former is
chewy, while the latter typically crumbles in the mouth. Despite
their obvious similarities, the two products please differently; the
authors are split on which one is better.
Sugars: Soft Caramel and Sucre à la Crème
FIGURE 81.2 (a) Soft caramel has a smooth and shiny appearance; (b) sucre à la crème contains microcrystals that heavily scatter light, giving the confection
a matte appearance; (c) a lightly sheared mixture seeds few sucrose nuclei, which grow to be macroscopically coarse.
crème. Coupling the colligative properties of aqueous sucrose
solutions with shear-​induced crystallization results in a char-
acteristic microcrystalline candy with broad appeal. These
concepts, being fairly elementary, are well within reach of first-​
year college students with a high-​school science background.
Interestingly, this simplicity is also reflected in other culinary
cultures transforming similar ingredients by similar approaches.
Because no obvious historical link relates sucre à la crème with
the Scots tablet (McNeill, 2010; Pagrach-​ Chandra, 2013) or
the Dutch borstplaat (Berkum, 1900; Pagrach-​Chandra, 2013),
among others, their parallel emergence is likely. The appreciation
for sugary confections is almost as universal as the underlying
physical chemistry.
Acknowledgements
The authors acknowledge the help of Steven Berdnarski, Benoit
Charbonneau, Robert Charbonneau, Wouter Ellenbroek, Kathryn
Harvey, Julie Parker, Nicoline van der Sijs, and Haley Scholz
as well as the staff of the Duke University Library and the
Bibliothèque et Archives nationales du Québec for getting hold
of some of the historical references consulted for this chapter.
We also thank Michel Lambert and Jean-​Marie Francoeur for
stimulating exchanges about the origin of sucre à la crème, and
Brigitte Lavoie for providing the material and supplies for the
cooking experiments. PC and IA acknowledge support from the
National Science Foundation Grant no. NSF DMR-​1749374.
REFERENCES
Anonymous. 1873. An Englishman in Vermont, Macmillan’s
Magazine, 28, 171–​180.
Ball DW. 2007. The chemical composition of maple syrup, Journal
of Chemical Education, 84, 1647–​1650.
Benning LE. 1990. Oh, Fudge!: A Celebration of America’s Favorite
Candy, New York, Henry Holt and Company.
Berkum Nv. 1900. De Hollandsche tafel in Indië, Gorinchem,
Noorduyn & Zoon.
Brady JB. 2009. Magma in a beaker: analog experiments with water
and various salts or sugar for teaching igneous petrology,
Canadian Mineralogy, 47(2), 457–​471.
Breckenridge WC, Kuksis A. 1967. Molecular weight distributions
of milk fat triglycerides from seven species, Journal of Lipid
Research, 8(5), 473–​478.
547
Caron M. 1883. Directions diverses données en 1878 par la
Révérende mère Caron alors supérieure générale des soeurs
de charité de la Providence pour aider ses soeurs à former de
bonnes cuisinières, 2nd Edition, Montréal, Sœurs de charité de
la Providence.
Driver E. 2008. Culinary Landmarks: A Bibliography of Canadian
Cookbooks, Toronto, University of Toronto Press, 1825–​1949.
Frye GV. 1885. Frye’s Practical Candy Maker, Chicago, Press of
E. J. Decker.
Hartel RW. 1993. Controlling sugar crystallization in food products,
Food Technology, 47(11), 99–​107.
Hartel RW, Ergun R, Vogel S. 2011. Phase/​state transitions of con-
fectionery sweeteners: Thermodynamic and kinetic aspects,
Comprehensive Review in Food Science and Food Safety,
10, 17–​32.
Hartel RW. 2013. Advances in food crystallization, Annual Reviews
in Food Science and Technology, 4, 277–​292.
Jeffery MS. 2001. Grained and ungrained confections, Manufacturing
and Confectionery, 73(11), 97–​110.
Larousse. 2009. Larousse Gastronomique: The World’s Greatest
Culinary Encyclopedia, New York, Clarkson Potters Publishers.
McNeill FM. 2010. The Scots Kitchen: Its Traditions and Lore,
Edinburgh, Birlinn Limited.
Ogilvie. 1905. Ogilvie’s Book for a Cook: A Selection of Recipes and
Other Things Adapted to the Needs of the Average Housekeeper,
Some Entirely New, and All Have Been Thoroughly Tested,
Montreal, Ogilvie.
Pagrach-​Chandra G. 2013. Sugar & Spice: Sweets and Treats from
around the World, Northampton, Interlink Books.
Parker J. 2006. «La Cuisinière Canadienne. Contenant tout ce
qu’il est nécessaire de savoir dans un ménage…» : Témoin
de l’émergence d’une cuisine nationale au carrefour des
cultures au XIXe siècle. MA Thesis in History, UQAM. https://​
histoire.uqam.ca/​wp-​content/​uploads/​sites/​21/​2017/​03/​Julie-
Parker.pdf.
Perrault L. 1840. La Cuisinière canadienne, contenant tout ce qu’il
est nécessaire de savoir dans un ménage […], Montréal, Louis
Perrault.
Perrault L. 1984. La Cuisinière Canadienne, Montréal, Edi-​Courtage.
Prieur FX. 1864. Notes d’un condamné politique de 1838, Soirées
Canadiennes, 4, 167–​407.
Prieur FX. 1949. Notes of a convict of 1838, Translated by George
Mackaness, Australian historical monographs, 18.
Sears DS. 1892. The Practical Cook : A Collection of Tested Recipes,
Revised Edition, Omaha, Young Men’s Journal Co.
Starzak M, Mathlouthi M. 2006. Temperature dependence of water
activity in aqueous solutions of sucrose, Food Chemistry, 96,
346–​370.
Sugars: Sugar (and Its Substitutes) in Pastries
Anne Cazor1 and Ramon Morató2
1 Scinnov, France
2 Creative Director for Cacao Barry brand
Sucrose is the most commonly used type of sugar in sweetened
baked products due to its availability, cost and suitability in baking
(Nip, 2007). However, governments and health professionals are
continuously putting pressure on the food industry to reduce
the amount of added sugar and the mean caloric value of their
products (Navarro et al., 2012).Two different strategies are used:
(1) reducing the sucrose content and/​or (2) using ingredients that
mimic the functional properties of sucrose (Alija et al., 2012).
In order to formulate pastries with a good amount of sucrose,
their different roles in pastries must be understood beforehand.
Sugar, mostly sucrose, plays multiple roles in pastries, such as
sweetness, colour, texture and preservation.
Sweet Taste
The most notable function of sucrose in food is its sweet taste.
A 25% sugar reduction is perceived as significantly less sweet
than a standard biscuit (Drewnowski et al., 1998; Biguzzi et al.,
2014). Polyols have been promoted as sucrose substitutes
because of their low glycaemic index. The sweetness of chiffon
cakes prepared with xylitol, a polyol, was similar to that of rice
chiffon cake with sucrose (Kim et al., 2014). In a study by Gao
et al. (2017), 50% sucrose was replaced by rebaudioside A and
erythritol in cocoa and vanilla-​flavoured muffins, and there was
no significant effect on the sensory analysis results.
FIGURE 82.1 Cakes with stevioside and liquid sorbitol (SO).
(Manisha et al., 2012)
Colour and Flavour Profile
Maillard and caramelization reactions produce pleasant odours
and colours using sugars.
Texture Properties
Sucrose plays an important role in tenderizing pastries. For
exemple, Licciardell et al. (2012) have studied the influence of
sugar/​egg white ratio on meringues. A low sugar/​egg white ratio
determines an increase in the average pore size and hence, higher
porosity. Sucrose acts as a stabilizer in egg white foam. Manisha
et al. (2012) studied interactions between stevioside, liquid sorb-
itol, hydrocolloids and emulsifiers for the replacement of sugar in
cakes (Figure 82.1).
Preservation
The hygroscopic nature of sugars plays a crucial role in reducing
the water activity (aw) in foods. In simple terms, aw is a measure
of the amount of free water available in a foodstuff. The average
aw of a cake is between 0.8 and 0.9 (Figure 82.2).
As sucrose plays different roles according to the pastry, a
sucrose alternative may require a combination of reducing the
sucrose quantity to the minimum to deliver essential properties
549
550
FIGURE 82.2 Water activity (aw) ranges for various types of food, and the
aw ranges at which microorganisms can grow.
(Voysey et al., 2014)
and selecting a sweetener and a bulking agent with synergistic
effects (Luo et al., 2019). Table 82.1 gives some examples in spe-
cific pastries.
Experimentation
A two-​day workshop was organized with the Cacao Barry com-
pany and its Cacao Barry Lab, Ramon Morato, creative director,
and some pastry chefs from the Relais Dessert Association. The
theme was exploring new ways for a healthier patisserie. Tests
were conducted and led to the organoleptic validation of recipes
for lemon tart, Paris-​Brest and a chocolate tart. The reduction of
sugar could reach up to 30% in certain sub-​elements of each of
the three recipes.
If we focus on lemon tart, three recipes and their nutritional
values are shown in Figures 82.3–​82.9.
Anne Cazor, Ramon Morató
TABLE 82.1
Sugar Reduction Approaches in Specific Food Categories
Pastry category Alternative sweeteners Bulking agents
cookies sucralose, maltitol, erythritol, inulin, maltodextrin
stevioside
muffins and cakes rebaudioside A, sorbitol, inulin, polydextrose,
maltitol, isomalt, erythritol dietary cocoa fibre
Source: Di Monaco et al. 2018.
REFERENCES
Alija J, Talens C. 2012. New concept of desserts with non added
sugar. International Journal of Gastronomy and Food Science,
1, 116–​122.
Biguzzi C, Schlich P, Lange C. 2014. The impact of sugar and fat
reduction on perception and liking biscuits. Food Quality and
Preference, 35, 41–​47.
Di Monaco R, Miele NA, Cabisidan EK, Cavella S. 2018. Strategies
to reduce sugars in food. Current Opinion in Food Science,
19, 92–​97.
Drewnowski A, Nordensten K, Dwyer J. 1998. Replacing sugar and
fat in cookies: Impact on product quality and preference. Food
Quality and Preference, 9(1), 13–​20.
Gao, J, Brennan MA, Mason SL, Brennan CS. 2017. Effects of sugar
substitution with “Stevianna” on the sensory characteristics of
muffins. Journal of Food Quality, 2017.
Kim JN, Park S, Shin WS. 2014. Textural and sensory characteristics
of rice chiffon cake formulated with sugar alcohols instead of
sucrose. Journal of Food Quality, 37, 281–​290.
Licciardell F, Frisullo P, Laverse J, Muratore G, Del Nobile MA.
2012. Effect of sugar, citric acid and egg white type on the
microstructure and mechanical properties of meringues.
Journal of Food Engineering, 108, 453–​462.
Luo X, Arcot J, Gill T, Louie JCY, Rangan A. 2019. A review of food
reformulation of baked products to reduce added sugar intake.
Trends in Food Science & Technology, 86, 412–​425.
Manisha G, Soumya C, Indrani D. 2012. Studies on inter-
action between stevioside, liquid sorbitol, hydrocolloids
and emulsifiers for replacement of sugar in cakes. Food
Hydrocolloids, 29, 363–​373.
Navarro V, Serrano G, Lasa D, Luis Aduriz A, Ayo J. 2012. Cooking
and nutritional science: Gastronomy goes further. International
Journal of Gastronomy and Food Science, 1, 37–​45.
Nip WK. 2007. Sweeteners. In Hui YH (Ed.) Bakery Products,
Blackwell Publishing, 137–​160.
Voysey PA, Legan JD. 2014. Confectionery products, cakes and pas-
tries. In Batt CA and Tortorello ML (Eds.) Encyclopedia of
Microbiology, Elsevier, 497–​503.
Sugars: Sugar in Pastries 551

FIGURE 82.3 Regular lemon tart.

552 Anne Cazor, Ramon Morató

FIGURE 82.4 Slim lemon tart.

Sugars: Sugar in Pastries 553

FIGURE 82.5 Skinny lemon tart.

554 Anne Cazor, Ramon Morató

FIGURE 82.6 Comparison of the three tarts.

FIGURE 82.7 Almond shortcrust pastry.

Sugars: Sugar in Pastries 555

FIGURE 82.8 Lemon cream.

FIGURE 82.9 Meringue.

Sugars: Erythritol–​Sucrose Mixtures out of Equilibrium –​Exciting
Thermodynamics in the Mouth
Hannah M. Hartge, Birgitta I. Zielbauer and Thomas A. Vilgis
Max Planck Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany
Erythritol (see Figure 83.1) is a sweetener that is used as a
sugar substitute because its properties are quite similar to those
of sucrose. Nevertheless, its strong tendency to crystallize and
inability to form a stable glass at room temperature restrict its
possible application. In this chapter, the influences of different
sucrose contents on the crystallization behavior of undercooled
erythritol melts are presented. We will also suggest two different
ways of using this knowledge for new culinary purposes.
Currently, for most sweet creations, sucrose is used as a main
ingredient because of its taste, relative cheapness and broad
range of possible textures, which are mainly controlled through
heat treatment and the concentration ratio between sucrose and
water (Hartel et al., 2011). Polyols like isomalt ((2R,3R,4R,5R)-​
6- ​ [ [(2S,3R,4S,5S,6R)- ​ 3 ,4,5- ​ t rihydroxy- ​ 6 - ​ ( hydroxymethyl)-​
2-​t etrahydropyranyl]oxy]hexane-​1 ,2,3,4,5-​p entol),     xylitol
((2R,3R,4S)-​pentane-​1,2,3,4,5-​pentol) and erythritol ((2R,3S)-​
butane-​1,2,3,4-​tetrol) are gaining in popularity as low-​caloric
sweeteners, though (Rice et al., 2019; Regnat et al., 2018), being
used for treats where less sweetness, crunchy crystallinity, a
hard glassy texture with less hygroscopy or simply better health
properties are desired (Fontana and González-​Cabezas, 2012;
Grembecka, 2015; Rice et al., 2019).
One of these sweeteners is erythritol. This is a polyol with
interesting properties: it offers a range of different advantages
over table sugar and other common sweeteners. Erythritol is crys-
talline and about 60–​70% as sweet as sucrose, it is not cario-
genic, it has almost no physiological calorific value or effects on
insulin levels, and unlike other sugar alcohols like sorbitol, it has
high digestive tolerance (Grembecka, 2015). Figure 83.1 shows
FIGURE 83.1 The chemical structures of sucrose (a) and erythritol (b). The
hydroxyl groups allow strong hydrogen bonding.
the molecular structure of erythritol in comparison with that of
sucrose, which can explain differences in physical properties like
viscosity, crystallization speed or dissolution properties.
Its strong tendency to crystallize rapidly and the cooling
effect upon dissolution of erythritol crystals in the mouth can be
interesting for culinary uses, but displeasing if these effects are
not specifically wanted: erythritol often causes a sandy mouthfeel
when consumed in creams or baked goods, for example, which
is undesirable in most processed foods and prevents sucrose-​
analogous use in some. In particular, no edible “sugar glass” can
be produced with pure erythritol, as will be shown later (Jesus
et al., 2010).
On the contrary, when sucrose is completely molten and
quickly cooled down again without agitation, the highly viscous
mass hardens at around 60 °C without crystallization, because the
molecules are dynamically too slow to find perfect positions in
a crystal order and eventually are immobilized in their arbitrary
positions. This is called a “glass” in a physical sense: it refers to a
solid, “frozen-​in” state of an undercooled liquid (Scholze, 1991),
which therefore shows no periodical order of the molecules; the
glassy sugar remains transparent. This property of table sugar
is what allows us to produce and store lollipops or pulled sugar
sculptures. Thermodynamically, glasses are semi-​stable, because
the molecules are not in their lowest-​energy state –​which for
temperatures below the melting temperature is reached when
they are ordered in crystals. This is also the reason why glassy
hard candy made of pure sucrose will tend to crystallize during
storage. Crystallization out of the glass can theoretically be
inhibited if the material is cooled down sufficiently, reducing the
mobility of the molecules, or by incorporating a different type of
molecule that hinders the movement and formation of sucrose
crystals, as glucose syrup does in the case of candy manufacture
(Hartel et al., 2011). The addition of different molecules will also
change the final glass transition temperature of the mixture, so
water (as present in glucose syrup) will lower the point at which
the mass hardens.
With only 4 compared with the 12 carbon atoms in sucrose
molecules and a molecular mass of 122.12 g/​mol (de Cock,
2012), erythritol molecules are smaller and possess only half as
many OH groups. Due to the higher mobility of the molecules in
557
558 Hannah M. Hartge et al.

FIGURE 83.2 Molecular model of an erythritol melt (blue molecules) with 30% sucrose (red molecules) in solution.

the erythritol melt, an erythritol glass can only be produced if it The melting point of such a solution is only about 6 °C lower
is quenched down to temperatures around −50 °C by −20 °C per than that of pure erythritol (see later measurements). What does
minute or more (Jesus et al., 2010). And still, this glass would change significantly, however, is the time necessary for complete
start to crystallize long before it reaches room temperature –​ crystallization. Because this is in part due to the reduced mobility
which makes it inedible as a glass. However, in combination of the molecules, it is now possible to keep the amorphous state
with sucrose, the properties can be adapted for some specific for several hours after melting, at least when cooled.
purposes.
A first interesting observation is that sucrose dissolves in
molten erythritol (Hartge, 2017). While sucrose is commonly Experimental Observation of Glass Transitions and
said to melt at temperatures around 160 °C to 180 °C (Roos, Crystallization
2013) –​although thermodynamically, it is argued that sucrose
This effect can systematically be shown by calorimetric
does not actually melt, but loses crystalline structure due to
measurements of such mixtures. In the following example (Hartge,
thermal degradation of the molecules (Lee et al., 2011) –​eryth-
2017), different erythritol–​sucrose mixtures are measured by dif-
ritol starts to melt at nearly 120 °C and degrades at slightly
ferential scanning calorimetry. This means that the heat flow of
higher temperatures. If this is not wanted, erythritol should not
a sample, which changes when energy is set free during crystal-
be brought to temperatures of more than 125 °C for a long time.
lization or used up for melting, is measured with respect to time
In the hot liquid erythritol melt, about 40 g of sucrose can be
and temperature. Specifically, the previously mixed samples were
dissolved in 100 g of erythritol. This requires some time for
first heated at 5 °C per minute from 20 °C to 125 °C, then kept at
heating and stirring, since the mobility of molecules in molten
this temperature for 15 minutes to ensure complete melting and
erythritol is quite low (compared with water, for example).
dissolution, and then cooled down by −20 °C per minute from
Figure 83.2 shows a visualization of such a melt containing 30%
125 °C down to −60 °C. Finally, the samples were heated up
of sucrose, where the spheres shown surrounding the molecules
again at 5 °C per minute to 125 °C. This last heating segment is
correspond to their molecular mass.
Sugars: Mixtures out of Equilibrium 559

FIGURE 83.4 Glass transition temperatures of the mixtures with respect to

FIGURE 83.3 Measurement of the heat flow of different erythritol–​sucrose
their sucrose content.
mixtures by differential scanning calorimetry.

of the greatest interest within these considerations, since it allows Glass Transitions
one to take a closer look at how glass transition and crystalliza-
If we focus on the glass transition temperatures and plot them
tion temperatures are affected in our examples.
with respect to the sucrose concentration, we receive the result
The graph in Figure 83.3 shows the results of the rapid cooling
shown in Figure 83.4.
and final heating. The first two segments are not shown here for
First, we can see that the glass transition is found at about
the sake of better visibility and because they do not show any
−42 °C for pure erythritol. The graph clearly shows that the glass
interesting changes except for the melting. The upper curves,
transition temperature rises with increasing sucrose content.
which result from cooling the sample by −20 °C per minute, just
This matches empirical quantitative models for glass transitions
confirm that our cooling rate is high enough to produce a glass:
of mixed systems, like the Gordon Taylor model (Hartel et al.,
no crystallization peaks or other significant changes in the heat
2011). It also shows that the rise in glass transition temperature
flow are visible, except for a slight curve that is caused by the
here is not high enough to make it suitable for the kitchen: even
measuring device and a step in heat flow that shows we reached
with a concentration of 30% sucrose, it still is below −30 °C.
the glass transition between −50 °C and −25 °C. Because of
According to the Gordon Taylor model (Seo et al., 2006), the
the rapid cooling, the erythritol molecules are not fast enough
glass transition temperature will continuously increase up to the
to organize into an equilibrium crystal structure before they are
transition temperature of pure sucrose with rising concentration.
kinetically “frozen” into an amorphous glass, even if no sucrose
This was also empirically shown by Hadjikinova and Marudova
is added.
(2016), but it is not suitable for the simple preparation method of
The last segment represented in the lower curves is showing
dissolving sucrose in molten erythritol discussed here.
the slow heating of our now solid erythritol–​sucrose glasses and
Because the glass transition temperature of erythritol is so
deserves a closer look. Again, we can find glass transition steps
low, and because an erythritol glass will crystallize as soon as
between −45 °C and −30 °C, indicating that here the mixture
it gets warmer, pure erythritol glass is of no great interest for
passes from the solid to the liquid state, while still remaining
culinary purposes. A mixture of erythritol with sugar, though,
viscous and amorphous. At higher temperatures, at least one
will not only slightly increase the glass transition tempera-
distinct exothermic peak is visible in all samples; its position
ture but more importantly, inhibit crystallization at warmer
clearly shifts with the sucrose concentration towards higher
temperatures.
temperatures, and it becomes lower and broader. These peaks can
be associated with a crystallization process, since here energy
is set free: the binding of the periodically ordered molecules
lowers their energy state, and this energy difference is released Crystallization
in the form of heat. Endothermic peaks can also be found in this The concentration of sucrose has a significant effect on crystal-
segment, each of which lies between 90 °C and 130 °C and cor- lization out of the amorphous phase, as shown by the graph in
responds to a melting process, where energy is needed to break Figure 83.5.
up the molecular bindings in the crystals. As the sucrose con- While in our measurement, crystallization occurs at 0 °C
centration increases, they shift towards lower temperatures and for pure erythritol, it is shifted up to about 80 °C for a mixture
become flatter, although the effect is less pronounced than for containing 30% of sucrose. Note that this does not imply that
crystallization. crystallization will only take place at these high temperatures for
560
FIGURE 83.5 Crystallization temperatures of the mixtures with respect to
their sucrose content.
such mixtures, although crystallization temperatures were very
consistent among repetitions. Instead, all the mixtures observed
here will also crystallize at room temperature, but nucleation,
as well as crystal growth, will take longer for higher sucrose
concentrations. Since these are primarily statistical processes,
they are dependent on energy (i.e., temperature) as well as on
time, with higher temperatures speeding up the arrangement of
the molecules due to their higher mobility.
Melting
Looking at the minimums of the downward peaks, plotted in
Figure 83.6, it can clearly be seen that the melting temperature
decreases almost linearly with increasing sucrose concentration.
This is even more true for the onset temperature of the melting
process (which would be more reliable for conclusions on the
necessary temperature), but this was not investigated here,
because it is harder to compare in the given measurements. For a
system of molecules that are incompatible in their crystal struc-
ture, as is to be expected for erythritol and sucrose, a decrease
of the melting temperature with increasing “contamination” of
sucrose in the erythritol crystals is to be expected up to some min-
imum point, because here the binding energy of the molecules is
reduced due to the disturbance by the sucrose molecules. In fact,
the reduction of the melting point below the melting temperatures
of both pure substances is an interesting observation, since
it suggests that erythritol and sucrose might form a so-​called
eutectic mixture for a specific ratio.
Inspiration for the Kitchen
Can these thermodynamic considerations be useful for confec-
tionery and molecular cooking? Apart from getting clues for fur-
ther investigations on how to control crystallization in products
containing erythritol, we can already get some practical inspir-
ation from these findings.
Hannah M. Hartge et al.
FIGURE 83.6 Melting temperatures of the mixtures with respect to their
sucrose content.
Ice Blossoms
Using a mixture of erythritol and 10% to 20% sucrose, it is
possible to grow blossom-​or star-​shaped crystals with the typ-
ical cooling effect of erythritol in the mouth. Carefully heat the
erythritol and add the sucrose until all sugar is dissolved. The
liquid should be at least 1 cm high in the pot to prevent the grown
crystals from sticking to the bottom. Turn off the heat and wait
until crystals start to form at the surface; this should take about
10 to 15 minutes. To speed up the process, you can sprinkle a few
single grains of erythritol as seeding crystals on the melt. As soon
as the crystal “blossoms” are large enough, take them out with
tweezers and carefully rinse with hot water. Let them dry and
store in an airtight container. The pictures in Figure 83.7 show
crystals resulting out of this process.
Here, sucrose is necessary to slow down the crystallization
growth to such an extent that easy handling is possible, but
crystals are still able to form in regular shapes within a short
period of time –​about 10 to 15 minutes until crystallization starts,
and about as much time again until several crystals of about 1 cm
in diameter are obtained. If 10% sucrose is used, the “petals”
of the crystal blossoms will be more pronounced with sharper
edges, but will also result in flatter and more fragile shapes. For
20% sucrose, mostly dense, half-​spherical blossoms will grow,
and part-​caramelization is more likely to occur. Especially if
seed crystals are used, the crystals tend to build more fine radial
spikes. A yellow hue will appear if the sugar is heated. A slight
coloration of the melt is no problem, but if the liquid turns darker
yellow, the sugar might be caramelized to an extent that inhibits
proper crystallization.
Hot Syrup
With the right erythritol/​sugar ratio, it is possible to shift crystal-
lization and dissolution of the mixture into a small temperature
range. This finding leads to a simple idea for a nice heat-​and-​
cooling sensation by purely physical processes suitable for the
tongue. Carefully heat a mixture of 70% erythritol and 30%
Sugars: Mixtures out of Equilibrium 561

FIGURE 83.7 Crystals grown out of an erythritol melt with 10% (a) and 20% (b) sucrose.

FIGURE 83.8 Erythritol–​sucrose-​melt close to room temperature when freshly prepared (a) and after stirring with a spoon (b). The agitation in the middle
of the plate causes the undercooled melt to crystallize where it was moved, while the rest stays viscous but liquid.

sucrose until all crystals have disappeared. A heating temperature
of 140 °C will be fine but requires caution if no significant cara-
melization is wanted. Pour the hot liquid onto a cold plate and let
it cool down. If it is not meant to be served immediately, store it
in the freezer to prevent crystallization for some hours, and let
the “syrup” warm up to room temperature immediately before
serving. Figure 83.8 shows the freshly produced mixture right
after pouring onto a cooled plate and after inducing crystalliza-
tion by stirring with a spoon.
Although the liquid looks much like a very thick syrup, it is
actually an undercooled melt, since it does not contain any water.
If taken into the mouth with a small spoon, the sugar mixture
will heat up to body temperature and experience agitation by
the tongue. Both lead to instant crystallization, which can be felt
in the mouth, not only because the smooth, viscous liquid sud-
denly becomes hard and grainy but also because a remarkable
amount of heat is released. Thus, the sugar should be tried care-
fully with a small spoon and moved inside the mouth to avoid
burning the tongue, especially for people who are heat sensitive.
As soon as the erythritol crystals just formed start to dissolve in
the saliva, its typical cooling effect can be felt, in contrast to the
preceding heat.
For this interesting effect of first heating up inside the mouth
and then cooling down again on the tongue, erythritol is respon-
sible for the big changes in heat flow. The sucrose is needed in
order to inhibit crystallization until the liquid reaches the mouth.
Higher concentrations of erythritol would crystallize sooner,
while sugar glass dissolves in the mouth without crystallizing.
Conclusion
Calorimetric measurements show that crystallization in
supercooled erythritol–​sucrose melts is inhibited with increasing
sugar content. With a higher sucrose concentration up to 30%,
the melting point decreases and the glass transition temperature
rises. Both effects are clearly measurable but far from the usual
room temperatures. As the sugar content increases, the time it
takes for crystallization to occur in the undercooled melt is also
greatly increased. Thus, it is possible to keep a mixture with 30%
562
sucrose in an amorphous state for some time without crystalliza-
tion occurring, and after delayed nucleation, crystal growth is
also significantly slowed down.
These observations allow better control of crystal growth
in erythritol mixtures. For example, by choosing a suitable
erythritol–​
sucrose mixture, decorative edible crystals can be
grown selectively within one hour of preparation. It is also pos-
sible to produce a syrup-​like melt that allows the thermodynamic
effects of the heat flow during crystallization and dissolution of
the erythritol to be directly experienced in the mouth.
REFERENCES
De Cock P. 2012. Erythritol. In O’Donnell K and Kearsley M (eds.)
Sweeteners and sugar alternatives in food technology, John
Wiley & Sons, Chichester, West Sussex, UK, 215–​241.
Fontana M and González-​Cabezas C. 2012. Are we ready for defini-
tive clinical guidelines on xylitol/​
polyol use? Advances in
Dental Research, 24(2), 123–​128.
Grembecka M. 2015. Sugar alcohols –​their role in the modern
world of sweeteners: a review, European Food Research and
Technology, 241(1), 1–​14.
Hadjikinova R and Marudova M. 2016. Thermal behaviour of
confectionery sweeteners’ blends, Bulgarian Chemical
Communications, 48, 446–​450.
Hartel RW, Ergun R, Vogel S. 2011. Phase/​state transitions of con-
fectionery sweeteners: thermodynamic and kinetic aspects,
Comprehensive Reviews in Food Science and Food Safety,
10(1), 17–​32.
Hannah M. Hartge et al.
Hartge HM. 2017. Glasübergänge und Kristallisation von
Erythrit-​Saccharose-​Mischungen, Bachelor thesis, Johannes
Gutenberg-​Universität Mainz, Germany.
Lee JW, Thomas LC, Jerrell J, Feng H, Cadwallader KR, Schmidt
SJ. 2011. Investigation of thermal decomposition as the kin-
etic process that causes the loss of crystalline structure in
sucrose using a chemical analysis approach (part II), Journal
of Agricultural and Food Chemistry, 59(2), 702–​712.
Jesus AL, Nunes SC, Silva MR, Beja AM, Redinha JS. 2010.
Erythritol: Crystal growth from the melt, International Journal
of Pharmaceutics, 388(1–​2), 129–​135.
Regnat K, Mach RL, Mach-​Aigner AR. 2018. Erythritol as sweet-
ener –​wherefrom and whereto? Applied Microbiology and
Biotechnology, 102(2), 587–​595.
Rice T, Zannini EK, Arendt E, Coffey A. 2019. A review of polyols
–​biotechnological production, food applications, regulation,
labeling and health effects, Critical Reviews in Food Science
and Nutrition, 60(12), 2034–​2051.
Roos YH, Karel M, Labuza TP, Levine H, Mathlouthi M, Reid D,
Shalaev E, Slade L. 2013. Melting and crystallization of sugars
in high-​solids systems, Journal of Agricultural and Food
Chemistry, 61(13), 3167–​3178.
Scholze H. 1991. Glass: nature, structure, and properties, Springer,
New York.
Seo JA, Kim SJ, Kwon HJ, Yang YS, Kim HK, Hwang YH. 2006. The
glass transition temperatures of sugar mixtures, Carbohydrate
Research, 341(15), 2516–​2520.
Sugars: Intramolecular Dehydration of Hexoses
Marie-​Charlotte Belhomme, Stéphanie Castex and Arnaud Haudrechy
Institut de Chimie Moléculaire de Reims, UMR CNRS 7312, SFR Condorcet FR CNRS 3417, Université de Reims,
BP 1039, F-​51687 REIMS Cedex, France
If the definition of saccharides (or sugars) is rather unclear,
specialists usually agree on saying that chemists can describe
saccharides as structures derived from chiral entities bearing
hydroxyl or diversely substituted hydroxyl, incorporating in this
family molecules with nitrogen, sulphur or even alkyl, inspired
by a Fischer-​ type skeleton. Because this is only limited by
our imagination, chemists may generally say that saccharide
derivatives should look like saccharides (joke). However, we
agree on an unanimous definition for “hexoses”: simple sugars
containing six carbon atoms and the equivalent of one molecule
of water per carbon atom (C,H2O)6 or C6H12O6 in their general
formula. All D-​hexoses are naturally occurring, with the excep-
tion of D-​altrose, the major ones being D-​glucose, D-​galactose,
D-​mannose and D-​fructose (Figure 84.1).
Hexoses are widely present in food and undergo partial trans-
formations during the process of cooking, one of the common
transformations being dehydration. Thus, the purpose of this
FIGURE 84.1 D-​altrose, D-​glucose, D-​galactose, D-​mannose and D-​fructose.
C6H12O6 C6H10O5
H2 O H2O
6C
H2 O
FIGURE 84.2 Consecutive dehydrations of a hexose.
chapter is to enumerate the reported and potential chemical trans-
formations occurring through dehydration of hexoses. The prop-
erties of the most common observed hexose dehydration products
will also be described.
Arithmetic of Hexose Dehydration
A general consideration of the definition of dehydration (water
release) gives a guiding principle for discussion. Figure 84.2
shows that it is possible to remove six molecules of water via
sequential transformations:
Even if the digestive properties of vegetable coal (mostly
composed of pure carbon) are commonly admitted to be benefi-
cial when used for short periods of time (Huizen, 2018), we will
restrict our discussion to shorter unpolymerized organic species,
which could eventually be formed during heating of food.
C6H8O4 C6H6O3
H2O
H2O
C6H2O C6H4O2
H2O
563
564 Marie-Charlotte Belhomme et al.

Inspection of Systematic Possibilities of Hexose
Dehydration
The consideration of an aldohexose (with no precision on
configurations) as a preliminary model for possible dehydrations
will show the amazing complexity of this apparently trivial trans-
formation (Feather and Harris, 1973).
D-​hexoses have common features, as they possess a carbonyl
function (located on position 1 for D-​glucose, D-​galactose and
D-​mannose, on position 2 in the case of D-​fructose). Figure 84.3
shows the different possible structures of aldohexoses (opened
and cyclized forms) that exist in different proportions depending
on the saccharide and various parameters. Please note that
hexoses can also rarely adopt a four-​membered ring or a seven-​ Figure 84.5 shows examples of hypothetical dehydration
membered ring, named oxetoses and septanoses, respectively.
All these forms need to be considered for the subsequent
dehydrations. Dehydration can occur according to a classical
elimination (enhanced by heating or catalysed by an acid or a
base) or a cyclization, as exemplified in Figure 84.4.
Consequently, even a simple aldohexose (with no consider-
ation of R/​S configurations) offers several possibilities in the con-
text of only one dehydration:
• The furanose form A can eliminate water coming from
OH-​1 and H-​2, OH-​2 and H-​1 (or H-​3), OH-​3 and H-​
2 (or H-​4), OH-​5 and H-​4 (or H-​6), or OH-​6 and H-​
5. Considering the cyclization mode for loss of water,
each OH, namely, OH-​2, OH-​3, OH-​5 and OH-​6, can
react on carbon 1, and this is true for all positions
(except carbon 4), giving 20 possible derivatives for a
monocyclization!
products.
• The opened form B can eliminate water coming from
OH-​2 and H-​1 of the aldehyde (or H-​3), OH-​3 and H-​2
(or H-​4), OH-​4 and H-​3 (or H-​5), OH-​5 and H-​4 (or

HO
OH OH 6
O 5 O 1 OH
5 4 1 OH HO
6
HO 5 3
HO 4 2 CHO 2
6 3
2 1 4
3 HO OH
OH OH
HO OH
OH
Furanose form A Opened form B Pyranose form C

FIGURE 84.3 Equilibrium between structures of aldohexoses (with carbon-​numberings).

OH OH
O
HO
n n
H Classical elimination Cyclization

FIGURE 84.4 Two modes of dehydration (classical elimination and cyclization).

HO HO HO
5 4 O Elimination of O O
1 OH
HO OH-1 and H-2 HO HO
6 H
3 2
HO OH O HO OH HO O
H H
Furanose form A

HO HO
5 4 O Cyclization of O
1 OH
HO OH-2 on C-1 HO
6 O
3 2
HO OH O HO
H H
Furanose form A Brigl-type anhydride

FIGURE 84.5 Examples of products resulting from loss of water with a possible keto-​enol equilibrium (from furanose form A).
Sugars: Intramolecular Dehydration
O O
HO
O O
O
CHO
HO
HO OH
OH
1 2
FIGURE 84.6 Cyclic structures resulting from dehydration of aldohexoses.
H-​6), or OH-​6 and H-​5. As discussed before in the case
of furanose form A, 20 other possibilities of dehydration
products could result from these cyclization processes.
• The pyranose form C can eliminate water coming from
OH-​1 and H-​2, OH-​2 and H-​1 (or H-​3), OH-​3 and H-​2
(or H-​4), OH-​4 and H-​3 (or H-​5), or OH-​6 and H-​5.
In addition, 20 possible cyclization products have to be
considered.
Taking into account elimination and cyclization processes for the
loss of only one water molecule, the total number of possible dehy-
dration compounds resulting from a single aldohexose is conse-
quently 85. Following an analogous hypothetical approach, 927
structures should exist for only two water molecule eliminations!
Most Commonly Observed Hexose Dehydration
Products and Properties
Even though hexose dehydration products are detected in small
amounts in food, they prove to be particularly important, notice-
ably affecting the flavour, colour and nutritional value (Olsson,
1977). Quite surprisingly, isolated and characterized dehydration
derivatives coming from hexoses are limited to:
-​Cyclic structures like 2,5-​ anhydro-​ D-​
mannose (1) (also
named chitose) (Feather, 1973; Goekjian, 2017), levoglucosan
(2) (Luque, 2014; Le Brech, 2015) and levogalactosan (3) (Le
Brech, 2015), all of them resulting from one dehydration, and
levoglucosenone (4), resulting from three dehydrations (Luque,
2014). Cyclic structures of these dehydration compounds appear
in Figure 84.6.
2,5-​anhydro-​D-​mannose (1): This hygroscopic compound
appears as a usual ending moiety of many active ingredients,
without any report of harmful effects (Zamani et al., 2008;
Nowicki et al., 2008).
Levoglucosan (2) and galactosan (3): These bicyclic derivatives
belong to a category of tracers for biomass burning (Aiken, 2010)
and are produced during the caramelization of sugar. No harmful
activity has been reported. Interestingly, the presence of (2) has
been correlated with several diseases (https://​pubchem.ncbi.nlm.
nih.gov/​compound/​Leucoglucosan#section=Cellular-​Locations).
Levoglucosenone (4): Classically used as a platform chem-
ical (Comba et al., 2018), this highly functionalized, constrained
structure apparently does not reveal any harmful effects
(https:// ​ p ubchem.ncbi.nlm.nih.gov/ ​ c ompound/ ​ L evoglucose
none#section=2D-​Structure).
565
O
O
OH HO OH
O
OH
3 4
O O O
HO OH OH HO OH
O O O
5 6 7
FIGURE 84.7 Pyran-​4-​one structures from dehydration of aldohexoses.
Pyranone structures like 2,3-​dihydro-​3,5-​dihydroxy-​6-​methyl-​
4H-​ pyran-​4-​
one (5) (resulting from two dehydrations), 3-​
hydroxy-​2-​methyl-​pyran-​4-​one (6) (Feather, 1973) and 3,5-​
dihydroxy-​2-​methyl-​pyran-​4-​one (7) (Olsson, 1977) are very
similar (Figure 84.7).
2,3-​dihydro-​3,5-​dihydroxy-​6-​methyl-​4H-​pyran-​4-​one (5):
This derivative (Kim, 1996), found in several cooked and stored
foods (vegetables, bread and dehydrated juices) has been iden-
tified as a strong antioxidant (Cechovska, 2011; Yu, 2013) and
has shown antifungal activity (Teoh, 2014) and antiproliferative
effects (Ban, 2007).
3-​hydroxy-​2-​methyl-​pyran-​4-​one (6): Also called maltol,
(6) is a flavouring agent improving mouthfeel, with the odour of
caramel and cotton candy, which can be found in many foods,
such as roasted malt, barley, peanuts, breads, milk and heated
butter, cocoa and coffee (Lee, 2000; De Rovira, 2008). Also
used as an intermediate in pharmaceuticals, (6) helps to flavour
cosmetics and has been recently found to attenuate diabetic per-
ipheral neuropathy (Guo, 2018).
3,5-​dihydroxy-​2-​methyl-​pyran-​4-​one (7): This compound
has been found in honeys (D’Arcy et al., 1997) and heated oak
(Cutzach et al., 1997) and was also identified as a decomposition
derivative in stored orange juices (Shinoda, 2004).
• Opened structures like 4,5,6-​trihydroxy-​2-​oxo-​hexanal
(8) (resulting from one dehydration), 3,4-​dihydroxy-​
hex-​3-​en-​2,5-​dione (9) (also named acetylformoin,
resulting from two dehydrations), and (Z)-​and (E)-​5,6-​
dihydroxy-​2-​oxo-​hex-​3-​enals (10 and 11) (resulting
from two dehydrations) (Feather, 1973) are very prob-
ably intermediates in the course of transformation into
cyclic structures (Jadhav, 2011) (Figure 84.8).
4,5,6-​trihydroxy-​2-​oxo-​hexanal (8): The compound (8) (one
of its diastereoisomers being named 3-​deoxyglucosone) reacts
easily with amino groups, modifying long-​lived proteins (like
566
OH O OH
HO CHO
OH O OH
8 9
FIGURE 84.8 Opened structures resulting from dehydration of aldohexoses.
HO
O O
CHO
12
FIGURE 84.9 Furan derivatives resulting from dehydration of aldohexoses.
collagen), and is consequently involved in diseases such as ath-
erosclerosis, diabetes and Alzheimer’s disease (Monnier, 1984).
3,4-​dihydroxy-​hex-​3-​en-​2,5-​dione (9) and (Z)-​and (E)-​5,6-​
dihydroxy-​ 2-​oxo-​hex-​3-​
enals (10) and (11): No specific data
exist in the literature concerning the biological properties of these
compounds.
-​Furan derivatives (resulting from three dehydrations), with
the emblematic example of 5-​ hydroxymethylfurfural, 5-​HMF
(12) (from D-​fructose (Hansen, 2008; Queneau et al., 2008;
Hansen et al., 2011; Lopes de Souza, 2012; Van Putten, 2013;
Despax et al., 2014; Luque and Balu, 2014; Jia et al., 2018; Ma
et al., 2018; Makhubela and Darkwa, 2018; Shinde and Rode,
2018; Sudarsanam et al., 2018; Suleiman, 2018), D-​mannose
(Van Putten, 2013; Jia et al., 2018; Zhong, 2018), D-​galactose
(Van Putten, 2013; Zhong, 2018), D-​ glucose (Olsson, 1977;
Hansen, 2008; Lopes de Souza et al., 2012; Van Putten, 2013;
Luque and Balu, 2014; Jia et al., 2018; Ma et al., 2018; Suleiman,
2018; Zhong, 2018), and from unusual carbohydrates L-​sorbose
(Van Putten et al., 2013; Despax et al., 2014) and D-​tagatose (Van
Putten et al., 2013), 2-​hydroxy-​acetyl-​furan (13) (Olsson et al.,
1977) and 2-​acetyl-​3-​hydroxy-​furan (14), also named isomaltol
(obtained in low yield from D-​glucose and D-​fructose for (13) and
from D-​fructose for (14), respectively, under acidic conditions)
(Feather and Harris, 1973) are described in Figure 84.9.
5-​hydroxymethylfurfural, 5-​ HMF (12), considered to be
harmful, is present, for example, in old honeys (Martinez et al.,
2017), in fruit juices (Thakur, 2018) and in carob powder, a substi-
tute for cocoa (Gunel et al., 2018). (12) is considered an indicator
of heating and aging of food (jams, alcoholic beverages). (12) is
normally not identified in fresh food, but because its formation
occurs after microwave treatment or under a hot air atmosphere,
drying and cooking can produce this compound in processes such
as caramelization. 5-​HMF (12) is naturally observed in coffee
(Murkovic and Pichler, 2006), beers (Husoy et al., 2008) or dried
fruits.
2-​hydroxy-​ acetyl-​
furan (13): This compound has been
observed during acid hydrolysis of Jerusalem artichoke when
trying to perform ethanol fermentation (Kim and Hamdy, 1986),
Marie-Charlotte Belhomme et al.
O O
HO CHO
OH
10 and 11
O O
O
OH
OH
13 14
in roasted chicory root (Sannai et al., 1982), and in concentrated
aroma of jam (Sugawara et al., 1982).
2-​acetyl-​3-​hydroxy-​furan (14) (also called isomaltol): This fla-
vour component, smelling like caramel and found in bread crust,
is produced by thermal degradation of sugars (DeMan, 1999) and
is a classical food additive.
Conclusion
The apparent paradox in this discussion is that only a few
dehydrated structures have been isolated, compared with the
great number of possible organic species (85 for one dehydra-
tion transformation and 927 for only two dehydration transform-
ations, per aldohexose …). In fact, it is difficult to find something
if you are not looking for it, and if you have not developed spe-
cific targeted methods for detection (especially if structures are
present in very minor quantities in your sample)! This does
mean that dehydrated products probably contaminate food in
very small amounts whenever carbohydrates are present before
cooking, due to various physico-​chemical treatments including
heating, exposure to slightly acidic conditions, etc.
REFERENCES
Aiken AC, De Foy B, Wiedinmyer C, Decarlo PF, Ulbrich IM, Wehrli
MN, Szidat S, Prevot ASH, Noda J, Wacker L, Volkamer R,
Fortner E, Wang J, Laskin A, Shutthanandan V, Zheng J, Zhang
R, Paredes-​Miranda G, Arnott WP, Molina LT, Sosa G, Querol
X, Jimenez JL. 2010. Mexico city aerosol analysis during
MILAGRO using high resolution aerosol mass spectrom-
etry at the urban supersite (T0) –​Part 2: Analysis of the bio-
mass burning contribution and the non-​fossil carbon fraction,
Atmospheric Chemical Physics, 10, 5315–​5341.
Ban JO, Hwang IG, Kim TM, Hwang BY, Lee US, Jeong HS, Yoon
YW, Kim DJ, Hong JT. 2007. Anti-​ proliferative and pro-​
apoptotic effects of 2,3-​dihydro-​3,5-​dihydroxy-​6-​methyl-​4H-​
pyranone through inactivation of NF-​kappaB in human colon
cancer cells, Archives in Pharmaceutical Research, 30, 11,
1455–​1463.
Sugars: Intramolecular Dehydration
Cechovska L, Cejpek K, Konecny M, Velisek J. 2011. On the role
of 2,3-​dihydro-​3,5-​dihydroxy-​6-​methyl-​(4H)-​pyran-​4-​one,
European Food Research and Technology, 233, 367–​376.
Comba MB, Tsai Y, Sarotti AM, Mangione MI, Suarez AG, Spanevello
RA. 2018. Levoglucosenone and its new applications: valor-
ization of cellulose residues, European Journal in Organic
Chemistry, 5, 590–​604.
Cutzach I, Chatonnet P, Henry R, Dubourdieu D. 1997. Identification
of volatile compounds with a “toasty” aroma in heated oak used
in barrelmaking, Journal of Agricultural and Food Chemistry,
45, 2217–​2224.
D’Arcy BR, Rintoul GB, Rowland CY, Blackman AJ.
1997. Composition of australian honey extractives. 1.
Norisoprenoids, monoterpenes, and other natural volatiles
from blue gum (Eucalyptus leucoxylon) and yellow box
(Eucalyptus melliodora) honeys, Journal of Agricultural and
Food Chemistry, 45, 1834–​1843.
DeMan JM. 1999. Principles of Food Chemistry, Springer, 286.
De Rovira Sr D. 2008. Dictionary of Flavors, 2nd edition,
Wiley-​Blackwell, 403.
Despax S, Maurer C, Estrine B, Le Bras J, Hoffmann N, Marinkovic
S, Muzart J. 2014. Fast and efficient DMSO-​mediated dehydra-
tion of carbohydrates into 5-​hydroxymethylfurfural, Catalysis
Communications, 51, 5–​9.
Feather MS, Harris JF. 1973. Dehydration reactions of carbohydrates,
Advances in Carbohydrate Chemistry and Biochemistry, 28,
161–​224.
Goekjian P, Haudrechy A, Menhour B, Coiffier C. 2017. C-​
Furanosides: Synthesis and stereochemistry, 1st edition,
Academic Press, 771.
Gunel Z, Tontul I, Dincer C, Topuz A, Sahin-​Nadeem H. 2018.
Influence of microwave, the combined microwave/​ hot air
and only hot air roasting on the formation of heat-​induced
contaminants of carob powders, Food Additives and
Contaminants: Part A. doi: 10.1080/​19440049.2018.1544720.
Guo N, Li C, Liu Q, Liu S, Huan Y, Wang X, Bai G, Yang M, Sun S,
Xu C, Shen Z. 2018. Maltol, a food flavor enhancer, attenuates
diabetic peripheral neuropathy in streptozotocin-​induced dia-
betic rats, Food Functions Food Additives and Contaminants:
Part A. doi: 10.1039/​c8fo01964a.
Hansen JR. 2008. Fuel and chemicals from renewable alcohols:
Part 1+2. PhD Thesis.
Hansen TS, Mielby JJ, Riisager A. 2011. Synergy of boric acid
and added salts in the catalytic dehydration of hexoses to
5-​hydroxymethylfurfural in water, Green Chemistry, 13,
109–​114.
Huizen J. 2018. What are the benefits of activated charcoal? www.
medicalnewstoday.com/​articles/​322609.php
Husoy T, Haugen M, Murkovic M, Jöbstl D, Stolen LH, Bjellaas T,
Ronningborg C, Glatt H, Alexander J. 2008. Dietary exposure
to 5-​ hydroxymethylfurfural from Norwegian food and
correlations with urine metabolites of short-​term exposure,
Food Chemistry and Toxicology, 46, 3697–​3702.
Jadhav H, Pedersen CM, Solling T, Bols M. 2011. 3-​ Deoxy-​
glucosone is an intermediate in the formation of furfurals from
D-​glucose, ChemSusChem, 4, 1049–​1051.
Jia S, He X, Ma J, Wang K, Xu Z, Zhang ZC. 2018. Efficient syn-
thesis of 5-​ hydroxymethylfurfural from mannose with a
reusable MCM-​41-​supported tin catalyst, Catalysis Science
and Technology, 8, 5526–​5534.
Kim K, Hamdy MK. 1986. Acid hydrolysis of Jerusalem artichoke
for ethanol fermentation, Biotechnology and Bioengineering,
28, 138–​141.
567
Kim MO, Baltes W. 1996. On the role of 2,3-​dihydro-​3,5-​dihydroxy-​
6-​methyl-​4(H)-​pyran-​4-​one in the Maillard reaction, Journal
of Agricultural and Food Chemistry, 44, 282–​289.
Le Brech Y. 2015. Analyse des mécanismes primaires de pyrolyse de
la biomasse. Université de Lorraine. https://​hal.univ-​lorraine.
fr/​tel-​01751801/​document.
Lee KG, Shibamoto T. 2000. Antioxidant properties of aroma
compounds isolated from soybeans and mung beans, Journal
of Agricultural and Food Chemistry, 48, 4290–​4293.
Lopes de Souza R, Yu H, Rataboul F, Essayem N. 2012. 5-​
Hydroxymethylfurfural (5-​ HMF) production from hexoses:
limits of heterogeneous catalysis in hydrothermal conditions
and potential of concentrated aqueous organic acids as reactive
solvent system, Challenges, 3, 212–​232.
Luque R, Balu AM. 2014. Producing fuels and fine chemicals
from biomass using nanomaterials, CRC Press, Taylor and
Francis Group.
Ma Y, Wang L, Li H, Wang T, Zhang R. 2018. Selective dehydra-
tion of glucose into 5-​hydroxymethylfurfural by ionic liquid-​
ZrOCl2 in isopropanol, Catalysts, 8, 467–​478.
Makhubela BCE, Darkwa J. 2018. The role of noble metal catalysts
in conversion of biomass and bio-​derived intermediates to
fuels and chemicals, Johnson Matthey Technology Reviews,
62, 4–​31.
Martinez RA, Schvezov N, Brumovsky LA, Pucciarelli Roman
AB. 2017. Influence of temperature and packaging type on
quality parameters and antimicrobial properties during Yatei
honey storage, Food Science and Technology, doi: 10.1590/​
1678-​457x.17717.
Monnier VM, Kohn RR, Cerami A. 1984. Accelerated age-​related
browning of human collagen in diabetes mellitus, Proceedings
of the National Academy of Science of the United States of
America, 81, 583–​587.
Murkovic M, Pichler N. 2006. Analysis of 5-​hydroxymethylfurfural
in coffee, dried fruits and urine, Molecular Nutrition Food
Research, 50, 842–​846.
Nowicki MW, Tulloch LB, Worrall L, McNae IW, Hannaert V,
Michels PAM, Fothergill-​ Gilmore LA, Walkinshaw MD,
Turner NJ. 2008. Design, synthesis and trypanocidal activity
of lead compounds based on inhibitors of parasite glycolysis,
Bioorganic Medicinal Chemistry, 16, 5050–​5061.
Olsson K, Pernemalm PA, Popoff T, Theander O. 1977. Formation
of aromatic compounds from carbohydrates. V. Reaction of D-​
glucose and methylamine in slightly acidic, aqueous solution,
Acta Chemica Scandinavica, 31, 469–​474.
Queneau Y, Jarosz S, Lewandowski B, Fitremann J. 2008. Sucrose
chemistry and applications of sucrochemicals, Advances in
Carbohydrate Chemistry and Biochemistry, 61, 217–​292.
Sannai A, Fujimori T, Kato K. 1982. Studies on flavor components
of roasted chicory root, Agricultural Biology and Chemistry,
46, 429–​433.
Shinde SH, Rode CV. 2018. An integrated production of diesel fuel
precursors from carbohydrates and 2-​methylfuran over Sn-​
Mont catalyst, ChemistrySelect, 3, 4039–​4046.
Shinoda Y, Murata M, Homma S, Komura H. 2004. Browning and
decomposed products of model orange juice, Bioscience
Biotechnology and Biochemistry, 68, 529–​536.
Sudarsanam P, Zhong R, van den Bosch S, Coman SM, Parvulescu
VI, Sels BF. 2018. Functionalised heterogeneous catalysts for
sustainable biomass valorisation, Chemical Society Reviews,
47, 8349–​8402.
Sugawara E, Ito T, Odagiri S. 1982. Sweet aroma components in
three kinds of jam, Nippon Nogeikagaku Kaishi, 56, 101–​108.
568 Marie-Charlotte Belhomme et al.

Suleiman A. 2018. Synthesis, characterisation and catalytic Yu X, Zhao M, Liu F, Zeng S, Hu J. 2013. Identification of 2,3-​
applications of novel iron N-​heterocyclic carbenes immobilised dihydro-​3,5-​dihydroxy-​6-​methyl-​4H-​pyran-​4-​one as a strong
on renewable resources. PhD thesis. York University. http://​ antioxidant in glucose-​histidine Maillard reaction products,
etheses.whiterose.ac.uk/​20467/​ Food Research International, 51, 397–​403.
Teoh YP, Don MM. 2014. Mycelia growth and production of total Zamani A, Jeihanipour A, Edebo L, Niklasson C, Taherzadeh MJ.
flavonoids and 4H-​pyran-​4-​one, 2,3-​dihydro-​3,5-​dihydroxy-​ 2008. Determination of glucosamine and N-​acetyl glucosa-
6-​methyl-​ by Schizophyllum commune using a bubble column mine in fungal cell walls, Journal of Agricultural and Food
bioreactor considering aeration effect and mass transfer Chemistry, 56, 8314–​8318.
study, Chemical and Biochemical Engineering Quarterly, 28, Zhong N. 2018. Sulphite treatment to enhance the ethanol production
553–​559. of softwood whole slurries. University of British Columbia,
Thakur JS. 2018. HMF as a quality indicator in Garcinia indica fruit Vancouver. https://​open.library.ubc.ca/​cIRcle/​collections/​
juice concentrate, Current Research in Nutrition and Food ubctheses/​24/​items/​1.0367000
Science Journal, 6, 227–​233.
van Putten RJ, Soetedjo JNM, Pidko EA, van der Waal JC,
Hensen EJM, de Jong E, Heeres HJ. 2013. Dehydration of
different ketoses and aldoses to 5-​ hydroxymethylfurfural,
ChemSusChem, 6, 1681–​1687.
Taste and Sound
Bruno A. Mesz
Universidad Nacional de Tres de Febrero (UNTREF)
Instituto de Investigación en Arte y Cultura (IIAC). Sáenz Peña, Argentina
Throughout history, food has been associated with music.
However, only recently have researchers started to consider the
impact of what we hear on the experience of eating and drinking,
be it consumption sounds, soundscapes or music. Crossmodal
correspondences between taste and sound have been used to
modify the experience of many different food and drink products
by changing the music or soundscape that people listen to. Chefs
and musicians are also becoming increasingly interested in
exploiting the expressive potential of combining music and food.
This so-​called “sonic seasoning” can be boosted by the use of
specially designed sensory apps.
Food has been associated with music since Antiquity: think of
Trimalchio’s dinner in the Satyricon by Petronius, where meat is
carved to the rhythm of music, or of the references to music in The
Physiology of Taste by Brillat-​Savarin (2009), where it is stated
that “music and song were made to increase the pleasures of the
table”; in Thomas More’s Utopia (1965), the ideal city should have
music played during evening suppers to aid digestion. Beyond these
and many other literary references, let us also mention the long
and important tradition of Tafelmusik (literally, “table-​music”) that
developed from the late Medieval period through the Enlightenment,
in which music accompanied meals of all types (Zohn, 2016).
Food-​Intrinsic Sounds
However, only recently have researchers, chefs and musicians
started to consider in a systematic way the impact of what we
hear on the experience of eating and drinking, be it consumption
sounds, soundscapes or music. In one of the most-​cited studies
in this area of research, Zampini and Spence (2004) showed
that mastication sounds affect the perception of crispness and
freshness of potato chips. They put earphones on the participants
of their study and manipulated their biting noises in real time
while they ate Pringles’ potato chips, so that for some chips, the
participants heard the actual noises, while for others, the sound
volume and the high-​frequency components were boosted or
attenuated. Their key finding was that boosting volume and/​or
high frequencies enhanced participants’ ratings of crispness and
freshness. Another study (Demattè et al., 2014) showed similar
results for the crispness and hardness of apples. Textural qualities
such as creaminess seem to depend in part on subtle perceptions
of sounds coming from the interior of the mouth (Van Aken,
2013): if you rub the tongue against the palate after drinking
coffee with cream, it will sound different than if you do the same
after drinking just black coffee. Perhaps our brains use such audi-
tory cues in order to ascertain the texture of that which we have
put into our mouths.
Of course, the experimental method of Zampini and Spence
can be used creatively to deliver any kind of pre-​recorded sounds
in synchrony with mastication, as has been done by Japanese
researchers (Masuda and Okajima, 2011). They played back
over headphones the sounds of mastication of crunchy and sticky
textures to people eating a variety of foods, resulting in altered
texture perception in all cases.
Food-​Extrinsic Sounds
Besides the sound of food in the mouth, loud noise has also been
shown to affect taste perception. In a study in which participants
sampled solutions of five prototypic tastants, and where
conditions of airplane cabin noise were simulated with broad
spectrum auditory stimulation, perception of sweetness was
suppressed when the participants were exposed to 85 dB of noise
intensity, while the taste of umami was rated as more intense
instead (Yan and Dando, 2015). Note, however, that there is some
conflicting evidence on this matter (Spence, 2014). Moreover,
the pleasantness of food-​related odours can be diminished by
the presence of noise (Velasco et al., 2014). Also, loud back-
ground music or noise in a bar can stimulate drinking (but
impair the perception of the alcohol content of drinks) (Spence
et al., 2019). This observation has been explained by means of
an “arousal hypothesis”: louder music tends to produce greater
levels of emotional arousal, which in turn increases consump-
tion (Guéguen et al., 2008). The arousal hypothesis may also
help to explain the effect of high musical tempo on consumer
behaviour. Researchers have observed that high-tempo music,
which tends to increase arousal (Husain et al., 2002), may make
people drink more rapidly (Spence et al., 2019). On the other
hand, slow-tempo music may induce customers to linger for
569
570
longer in a bar (Spence et al., 2019). Music style and genre can
also exert a marked effect on food evaluation by priming certain
aspects of the taste experience. Classical music has been shown
to increase the purchased amount of food and beverages in sev-
eral real-​world cases (see, for instance, North and Hargreaves,
1998; North et al., 2003), an effect that has been explained by
the semantic connection that customers may make between clas-
sical music and notions of high quality and class. In other cases,
music is able to bias judgements on specific aspects of taste,
as in a study by North (2012) showing that playing “Carmina
Burana” by Karl Orff to a group of university students while
they tasted a red or a white wine made both wines taste more
“powerful and heavy”, while playing “Just Can’t Get Enough”
by Depeche Mode made the wines more “zingy and refreshing”.
Sonic Ingredients
These findings have stimulated culinary creativity, and some
chefs, such as Heston Blumenthal in his dish named “Sound of
the Sea”, have used soundscapes as an essential ingredient: in this
dish, “seafood, seaweed and edible ‘sand’ are used to create what
looks like the edge of the seashore, all of which is accompanied
by an iPod and earphones so that the diner can hear the sounds of
the waves lapping up against the shore while eating” (Spence and
Piqueras-​Fiszman, 2014). In an experimental tasting conducted
by Blumenthal and Spence, they had two groups of people eating
a bacon and egg ice-​cream; one group simultaneously listened to
the sounds of bacon sizzling in the pan, the other to the sounds of
chickens clucking; in each case, sound intensified the congruent
flavour (Spence and Piqueras-​Fiszman, 2014).
Other chefs, like Grant Achatz of Chicago’s Alinea, are now
starting to consider having a musician come into the restaurant
to play something to accompany one of the dishes on the menu
(Ulla, 2011). In his collaborative gastronomic project “Kitchen
Theory”, Jozef Youssef organizes multisensory dinners where
he uses soundscapes to enhance particular dishes, for example
liquid nitrogen-​poached grapefruit and vodka meringues, made at
the table and served to the sound of rustling wind. Interestingly,
he has also experimented with changing the music while diners
consumed a dish entitled “Believe Nothing of What you Hear”,
playing first a glitchy and distorted song and then a more con-
sonant and melodic piece. He reports that “when consulted, the
guests consistently reported that the glitchy music brought out the
more bitter (cocoa), sour (passion fruit) and crunchy (crumble)
elements of the dish, whereas the more melodic music brought
out the sweeter and creamier (mango chocolate ganache) elem-
ents” (Youssef, 2015).
Technology for “Sonic Seasoning”
The expression “sonic seasoning” refers to altering people’s taste
experiences by the use of sound. This can be boosted by the use
of specially designed sensory apps. For instance, a device built by
Japanese designers called the “Chewing Jockey” can detect when
a user’s jaws move while eating and play back a pre-​recorded
Bruno A. Mesz
FIGURE 85.1 Interactive wine glasses.
(Mesz et al., 2017)
sound in synchrony (Spence and Piqueras-​Fiszman, 2014). As
said before, this can have an effect on texture perception and also
on the pleasure or surprise while eating –​think of matching mas-
tication to the sounds of thunder or a breaking glass. At our lab,
we made an analogue device for drinks: a wine glass equipped
with sensors that are able to detect three main gestures of the user:
when the glass is taken by the hand, when the wine is aired, and
when the liquid contacts the mouth while drinking (Figure 85.1).
At a recent presentation, we used these glasses with a young and
an aged wine to synchronize sounds designed to match the age of
the wine (Mesz et al., 2017a; Mesz, 2019).
The musical spoons by food designers Sam Bompas and Harry
Parr (Anonymous, 2013), with tiny mp3 players embedded in
them, are specifically for Heinz baked beans. Each baked bean
flavour gets a different type of music, which is heard via bone
conduction to the ear when users bite the spoons –​for instance,
cheddar cheese gets an Elgar-​inspired sound, while garlic and
herb has a track made by rustling garlic skins together. The
SmartPlate by designer Julian Caraulani (Seth, 2012) is a con-
ceptual project of a plate that detects the ingredients of the food
placed over it and generates a musical composition based on each
of them.
For reviews on sound technology for “sonic seasoning”, see
Velasco et al. (2016; 2018); see also Spence (2019) on sensory
apps for musical augmentation of the wine drinking experience.
Crossmodal Correspondences between Taste
and Sound
Crossmodal correspondences refer to the associations that many
of us appear to make between seemingly unrelated attributes,
features or dimensions in two or more different senses, as, in our
case, taste and hearing (Spence, 2011). Empirical research shows
that sweet taste tends to be matched with sounds that are high in
pitch, with slow-​tempo music that is “legato” in articulation (i.e.,
continuous and without separation between successive sounds)
and soft in dynamics, and with consonant harmonies (Mesz et al.,
Taste and Sound
TABLE 85.1
Summary of Crossmodal Correspondences between Basic Tastes and Sound Dimensions in the Literature to Date
Sweet Acid
Pitch Medium Very high
High-​high
Articulation Legato Mean
Consonance High Low (dissonant)
Pitch range (ambitus) Small Large
Rhythm Regular
Tempo Slow Fast
Sharpness (high-​frequency content) Low High
2011; Bronner et al., 2012). By contrast, sour taste tends to be
matched with extremely high-​pitched sounds, fast tempo and dis-
sonant music. Bitter taste is matched with sounds that are low in
pitch and more likely to be brassy (Crisinel and Spence, 2012;
Wang et al., 2015). Salty taste is mostly associated with “stac-
cato” music (i.e., music with clearly detached successive notes)
(Mesz et al., 2011; Knöferle and Spence, 2012) (Table 85.1).
Interestingly, the same correspondences have been documented
in non-​Western cultures (Knoeferle et al., 2015).
Importantly, these preferred pairings of taste and sound,
when presented together, tend to enhance the pleasure of what
one is tasting (Spence et al., 2014b; Wang and Spence, 2015).
A growing body of empirical research now shows that our experi-
ence of many different food and drink products can be modi-
fied by changing the music or soundscape that people listen to
(Crisinel et al., 2012; Spence and Deroy, 2013; Velasco et al.,
2013; Spence et al., 2014a; Reinoso-​Carvalho et al., 2015a,b;
Wang and Spence, 2015a,b; Wang and Spence, 2016; Hauck and
Hecht, 2019); see a summary of recent studies on this subject in
Spence et al. (2019).
These effects of “sonic seasoning” tend to be more pronounced
for foods with complex flavours, which could be explained by an
attentional account. According to this approach, taste-​congruent
soundtracks work by drawing the listener’s attention towards the
taste/​flavour that happens to correspond to the soundtrack rather
than to another taste/​flavour. Attention can enhance the sali-
ence of the attended stimulus/​feature relative to when the same
stimulus/​feature is unattended, and the effects of attention on
awareness tend to become more pronounced as the perceptual
input becomes more challenging or complex (Wang, 2017). Note
that “sonic seasoning” may also work comparatively better with
unfamiliar food products, in which case the role of memories of
previous experiences with their taste/​flavour will presumably not
dominate over the actual tasting situation.
Besides attentional biases, there are several other plausible
mechanisms accounting for the effect of sound in taste percep-
tion/​evaluation (Wang, 2017), such as emotion and sensation
transference. The idea of emotion transference is that “if you like
the music more, you like the food more”: people will prefer a food
or drink consumed while listening to music they enjoy in com-
parison to eating/​drinking it with music they don’t; that is, pref-
erence for the music is transferred to taste preferences. In fact,
571
Salty Bitter
Medium Low
Staccato Legato
Medium Low (dissonant)
Medium Medium
Medium Slow
Low Low
several studies have shown precisely this for a variety of products
such as fruit juice, chocolate and beer (Reinoso-​Carvalho et al.,
2015a,b; Wang, 2017; Reinoso-​Carvalho et al., 2019).
More generally, sensation transference happens when what
we think about what we hear (and the ideas/​concepts primed
by such music or soundscapes) is transferred to whatever it is
we happen to be tasting. For instance, in the already-​mentioned
study by North on music and wine (North, 2012), a music that is
judged as heavy and powerful can make a wine taste more heavy
and powerful. It is worth bearing in mind that such effects may
depend on the whole service environment; a music that is judged
congruent with the environment can improve customers’ evalu-
ation of its quality (Demoulin, 2011), and in turn, this sensation
can be transferred to food evaluations (Spence et al., 2019). Of
course, these influences of auditory stimuli on food perception
also operate in the kitchen and may affect how food and drinks
are seasoned and prepared (Kontukoski et al., 2015).
A favourite pairing in the literature on crossmodal matching
and multisensory perception of sound and taste is combining
music with wine (Spence and Wang, 2015a,b,c). The existing
research shows that sweetness, acidity, fruitiness, astringency
and length of the flavour sensation can all be modified by playing
appropriate musical selections.
Since music and the consumer experience of food/​drink are
both time-​varying in nature, it would seem appropriate to take
temporality into account when studying the impact of music on
the eating/​drinking experience. A recent study used a time-​based
method of sensory analysis called time-​intensity to measure tem-
poral changes in sweetness and sourness evaluations of an off-​dry
white wine when the music stimulus changed from a soundtrack
commonly associated with sweetness to one associated with
sourness instead, and vice versa (Wang et al., 2017a). The results
revealed that a change of soundtrack results in a change in taste
intensity (for both sweetness and sourness) in the same direc-
tion as the change in the soundtrack. More specifically, a switch
from the sweet to the sour soundtrack enhanced the intensity of
sourness, whereas a switch from the sour to the sweet soundtrack
enhanced the perceived intensity of sweetness (this experiment
is similar to Youssef’s idea of changing music while savouring
a dish). More complex shifts in the taste of red wine presented
together with classical and pop music were measured with the
method of temporal dominance of sensations (Wang et al., 2019).
572
For considerations on the aesthetics of temporal sequences in
cuisine and music, see Rozin and Rozin (2018).
Multisensory Gastromusical Art
Inspired in part by the growing research on the effects of sound
on taste perception, musicians and chefs are increasingly
interested in exploiting the expressive potential of combining
music and food.
Irene and Per Moneeo are Swedish composers of electronic
art music that create “Taste of Sound Dinner Experiences”; these
spectacles, which they have been presenting since 2012, are large-​
scale dinners for all the senses with sound, music, food, drinks,
fragrances and other sensory stimuli (Monneo and Monneo,
2020). The music works like a soundtrack to a movie, enhan-
cing the taste of the dishes and creating specific atmospheres
(Figure 85.2).
Ben Houge, professor at Berklee College of Music in Boston,
together with designer Jutta Friedrichs, has been developing a
“food opera” project (Houge and Friedrichs, 2018a). They dis-
tribute sound among diners using individual speakers at each
seat in the dining area, or by networked mobile devices, which
allows them to use real-​time software to render a customized
soundtrack based on each diner’s behaviour. They have also been
working with sensors for systems of networked plates and glasses
to increase the extent to which diners’ activities and gestures
can influence the music that accompanies a meal. Some recent
performances of the project were a dinner at Symphony Hall in
Boston (Arnett, 2018) and the “XX aniversario” dish at Mugaritz
restaurant in Spain (Houge and Friedrichs, 2018b).
Jo Burzynska is an Australasia-​based sound artist and wine
writer whose work in these areas has increasingly converged
in the production of multisensory art. Her work “Carbonic
Oscillation” in the Mishearings exhibition at the Auricle Sonic
Arts Gallery in Christchurch, New Zealand was an immersive
FIGURE 85.2 Taste of sound dinner experience.
Bruno A. Mesz
chamber in which sparkling wine formed the focus of a fusion of
all the senses. An effervescing environment was created through
“a sound work made from recordings of fizzing wine reinforced
by a bubbling light projection and the taste and tactile elements
of the sparkling wine consumed within it” (Burzynska, 2015).
“Amazuppai” is a work that sets out to sonically modulate the
sensory and conceptual perception of sweet and sour in wine,
which was first exhibited at the Auricle in Christchurch in 2017
(Burzynska, 2017). Her “Risonanze di Vino” project, conducted
in Campania in 2018 during the vintage, was a cultural study that
used winegrowers’ sensory responses to guide the six wine and
sound compositions that resulted (Burzynska, 2018).
Composer and chef Ysanne Spevack creates sensorial events
called “Yntegriti”. In one of them, she presented “Cacao in E
Major,” “an immersive five-​course symphony of chocolate syn-
aesthesia designed to inspire deep inner happiness” (Halter,
2016). Each of the courses included single-​estate raw cacao from
Ecuador, and Spevack served the dinner to the sounds of a very
primal place, the rainforest, together with her own compositions
and those of Billy Corgan (Smashing Pumpkins) and Shepard
Fairey (Nøise) (Spevack, 2017).
In “Cena Emocional” (Emotional Dinner), María Zegna, Janice
Wang and Bruno Mesz built four scenes where food, smells and
music responded to different emotional states: “Narcotic” (music:
“Clair de lune” by Gabriel Fauré, food: “arroz con leche”, smell:
roses), “Irritant” (music: “Child of Tree” by John Cage, food:
spicy bread served on the spines of a cactus, smell: an irritant
monomolecular odour), “Depressive” (music: sound texture
by Mesz, food: blueberries floating in fish tanks containing icy
water that people had to catch with their hands, dance by Zegna,
smell: humidity and earth) and “Stimulant” (music: “Rebonds B”
by Iannis Xenakis, drink: champagne mixed with Fanta Orange
Soda) (Zegna et al., 2015). Other performances by Zegna and
Mesz (MoT and ToM) were organized in four parts corresponding
to the basic tastes bitter, sweet, sour and salty. In each part, a
piece of contemporary art music was presented together with
Taste and Sound
visual live performance and dance, while at a given moment
(indicated on a screen) the spectators had to eat a small canape
of the corresponding taste, made from ingredients used by the
originary people of Latin America (MoT) or typically Finnish
(ToM) (Mesz, 2012a,b). The music for each taste was selected on
the basis of crossmodal taste–​music correspondences.
Artist and researcher Janice Wang has directed many
performances and multisensory dinners combining music and
gastronomy (Wang, 2019a). In “Aros far from home multisensory
dinner”, five courses were served in the Aros museum restaurant
in Aarhus, Denmark. Each course emphasized the role of one
particular sense in flavour perception, while the overall progres-
sion of the evening took the diners further from home. Each dish
was served with an assortment of accompaniments including
beverages, lighting, sound and aroma (Wang, 2019b).
Conclusion
Sound is often described as the forgotten flavour sense (Spence
et al., 2011); however, there is now a large volume of research on
auditory contributions to the experience of eating and drinking.
These investigations show that the effects of sound and music
on taste perception and consumer behaviour related to food are
surprising and profound; this fact, documented both in the labora-
tory and in real-​world contexts, offers exciting opportunities to
conceive new kinds of multisensory gastronomic events. It can
be expected that multisensory technologies and apps will have a
relevant role to play in enriching dining experiences, in aspects
such as the selection and control of sound events for “sonic
seasoning” (Spence, 2019) or the synchronization of the music
with the act of eating/​drinking.
REFERENCES
Brillat-​Savarin JA. 2009. The physiology of taste: or meditations on
transcendental gastronomy. Vintage.
Bronner K, Frieler K, Bruhn H, Hirt R, Piper D. 2012. What is the
sound of citrus? Research on the correspondences between
the perception of sound and flavour. In Camboroupoulos E,
Tsougras C, Mavromatis P, Pastiadis K (eds.), Proceedings
of the 12th International Conference of Music Perception
and Cognition (ICMPC) and the 8th Triennial Conference
of the European Society for the Cognitive Sciences of Music
(ESCOM), Thessaloniki, Greece, 142–​148.
Crisinel AS, Cosser S, King S, Jones R, Petrie J, Spence C. 2012.
A bittersweet symphony: Systematically modulating the taste
of food by changing the sonic properties of the soundtrack
playing in the background. Food Quality and Preference, 24,
201–​204.
Crisinel AS, Spence C. 2012. The impact of pleasantness ratings on
crossmodal associations between food samples and musical
notes. Food Quality and Preference, 24, 136–​140.
Demattè ML, Pojer N, Endrizzi I, Corollaro ML, Betta E, Aprea E,
Charles M, Biasioli F, Zampini M, Gasperi F. 2014. Effects
of the sound of the bite on apple perceived crispness and
hardness. Food Quality and Preference, 38, 58–​64.
Demoulin NT. 2011. Music congruency in a service setting: The
mediating role of emotional and cognitive responses. Journal
of Retailing and Consumer Services, 18(1), 10–​18.
573
Guéguen N, Jacob C, Le Guellec H, Morineau T, Lourel M. 2008.
Sound level of environmental music and drinking behavior: A
field experiment with beer drinkers. Alcoholism: Clinical and
Experimental Research, 32(10), 1795–​1798.
Hauck P, Hecht H. 2019. Having a drink with Tchaikovsky: The
crossmodal influence of background music on the taste of
beverages. Multisensory Research, 32(1), 1–​24.
Husain G, Thompson WF, Schellenberg EG. 2002. Effects of musical
tempo and mode on arousal, mood, and spatial abilities. Music
Perception: An Interdisciplinary Journal, 20(2), 151–​171.
Knöferle KM, Spence C. 2012. Crossmodal correspondences
between sounds and tastes. Psychonomic Bulletin & Review,
19, 992–​1006.
Knoeferle KM, Woods A, Käppler F, Spence C. 2015. That sounds
sweet: Using crossmodal correspondences to communicate
gustatory attributes. Psychology & Marketing, 32(1), 107–​120.
Kontukoski M, Luomala H, Mesz B, Sigman M, Trevisan M, Rotola-​
Pukkila M, Hopia AI. 2015. Sweet and sour: Music and taste
associations. Nutrition & Food Science, 45, 357–​376.
Masuda M, Okajima K. 2011. Added Mastication Sound Affects
Food Texture and Pleasantness. Poster presented at the 12th
International Multisensory Research Forum meeting in
Fukuoka, Japan.
Mesz B, Herzog K, Amusátegui JC, Samaruga L, Tedesco S. 2017.
Let’s drink this song together: Interactive taste-​sound systems.
In Proceedings of the 2nd ACM SIGCHI International
Workshop on Multisensory Approaches to Human-​ Food
Interaction, 13–​17. ACM.
Mesz B, Trevisan MA, Sigman M. 2011. The taste of music.
Perception, 40(2), 209–​219.
More T. 1965. Utopia, translated from the Latin by Paul Turner.
Penguin, 83.
North AC. 2012. The effect of background music on the taste of
wine. British Journal of Psychology, 103, 293–​301.
North AC, Hargreaves DJ. 1998. The effect of music on atmosphere
and purchase intentions in a cafeteria. Journal of Applied
Social Psychology, 28(24), 2254–​2273.
North AC, Shilcock A, Hargreaves DJ. 2003. The effect of musical
style on restaurant customers’ spending. Environment and
Behavior, 35(5), 712–​718.
Reinoso-​Carvalho F, Dakduk S, Wagemans J, Spence C. 2019. Not
just another pint! The role of emotion induced by music on the
consumer’s tasting experience. Multisensory Research, 32(4–​
5), 367–​400.
Reinoso-​ Carvalho F, Van Ee R, Rychtarikova M, Touhafi A,
Steenhaut K, Persoone D, Spence C. 2015a. Using sound-​taste
correspondences to enhance the subjective value of tasting
experiences. Frontiers in Psychology, 6, 1309.
Reinoso-​ Carvalho F, Van Ee R, Rychtarikova M, Touhafi A,
Steenhaut K, Persoone D, Leman M. 2015b. Does music influ-
ence the multisensory tasting experience? Journal of Sensory
Studies, 30(5), 404–​412.
Rozin P, Rozin A. 2018. Advancing understanding of the aesthetics
of temporal sequences by combining some principles and
practices in music and cuisine with psychology. Perspectives
on Psychological Science, 13(5), 598–​617.
Spence C. 2011. Crossmodal correspondences: A tutorial review.
Attention, Perception, & Psychophysics, 73(4), 971–​995.
Spence C. 2014. Noise and its impact on the perception of food and
drink. Flavour, 3(1), 9.
Spence C. 2015. Music from the kitchen. Flavour, 4(1), 25.
Spence C. 2019. Multisensory experiential wine marketing. Food
Quality and Preference, 71, 106–​116.
574
Spence C, Deroy O. 2013. On why music changes what (we think)
we taste. i-​Perception, 4, 137–​140.
Spence C, Piqueras-​ Fiszman B. 2014. The perfect meal: the
multisensory science of food and dining. John Wiley & Sons.
Spence C, Reinoso-​Carvalho F, Velasco C, Wang QJ. 2019. Extrinsic
auditory contributions to food perception & consumer behav-
iour: An interdisciplinary review. Multisensory Research, 1(4–​
5), 275–​318.
Spence C, Shankar MU, Blumenthal H. 2011. “Sound bites”:
Auditory contributions to the perception and consumption
of food and drink. In Bacci F, Melcher D (eds.), Art and the
senses. Oxford University Press, 207–​238.
Spence C, Velasco C, Knoeferle K. 2014a. A large sample study on
the influence of the multi-​sensory environment on the wine
drinking experience, Flavour, 3, 8.
Spence C, Velasco C, Vanne M, Hopia A. 2014b. Can you taste
the music? In Hopia A, Ihanus S (Eds.), 5D cookbook, 73.
Seinäjoki, Finland, KUMURU-​project. www.5Dcookbook.fi.
Spence C, Wang QJ. 2015a. Wine and music (I): On the crossmodal
matching of wine and music. Flavour, 4(1), 34.
Spence C, Wang QJ. 2015b. Wine and music (II): Can you taste the
music? Modulating the experience of wine through music and
sound. Flavour, 4(1), 33.
Spence C, Wang QJ. 2015c. Wine and music (III): So what if music
influences the taste of the wine? Flavour, 4(1), 36.
Van Aken GA. 2013. Acoustic emission measurement of rubbing and
tapping contacts of skin and tongue surface in relation to tactile
perception. Food Hydrocolloids, 31, 325–​331.
Velasco C, Balboa D, Marmolejo-​ Ramos F, Spence C. 2014.
Crossmodal effect of music and odor pleasantness on olfactory
quality perception. Frontiers in Psychology, 5, 1352.
Velasco C, Jones R, King S, Spence C. 2013. Assessing the influ-
ence of the multisensory environment on the whisky drinking
experience. Flavour, 2, 23.
Velasco C, Obrist M, Petit O, Spence C. 2018. Multisensory tech-
nology for flavor augmentation: A mini review. Frontiers in
psychology, 9, 26.
Velasco C, Reinoso-​ Carvalho F, Petit O, Nijholt A. 2016. A
multisensory approach for the design of food and drink enhan-
cing sonic systems. In Proceedings of the 1st Workshop on
Multi-​sensorial Approaches to Human-​ Food Interaction, 7,
ACM.
Wang QJ. 2017. Assessing the mechanisms behind sound-​ taste
correspondences and their impact on multisensory flavour
perception and evaluation (Doctoral dissertation, University
of Oxford).
Wang QJ, Mesz B, Spence C. 2017. Assessing the impact of music
on basic taste perception using time intensity analysis. In
Proceedings of the 2nd ACM SIGCHI International Workshop
on Multisensory Approaches to Human-​ Food Interaction,
18–​22, ACM.
Wang QJ, Mesz B, Riera P, Trevisan M, Sigman M, Guha A, Spence
C. 2019. Analysing the impact of music on the perception of
red wine via Temporal Dominance of Sensations. Multisensory
Research, 32(4–​5), 455–​472.
Wang QJ, Spence C. 2015a. Assessing the effect of musical con-
gruency on wine tasting in a live performance setting. I-​
Perception, 6(3), 1–​13.
Wang QJ, Spence C. 2015b. Assessing the influence of the
multisensory atmosphere on the taste of vodka. Beverages,
1(3), 204–​217.
Bruno A. Mesz
Wang QJ, Spence C. 2016. “Striking a sour note”: Assessing the
influence of consonant and dissonant music on taste percep-
tion. Multisensory Research, 29(1–​3), 195–​208.
Wang QJ, Woods AT, Spence C. 2015. What’s your taste in music?
A comparison of the effectiveness of various soundscapes in
evoking specific tastes. i-​Perception, 6(6), 2041669515622001.
Yan KS, Dando R. 2015. A crossmodal role for audition in taste
perception. Journal of Experimental Psychology: Human
Perception and Performance, 41(3), 590.
Zampini M, Spence C. 2004. The role of auditory cues in modulating
the perceived crispness and staleness of potato chips. Journal
of Sensory Science, 19, 347–​363.
WEB REFERENCES
Anonymous. 2013. “Musical spoons to go with your Heinz beans”.
Advertising Age, 28 March, http://​adage.com/​article/​creativity-​
pick-​of-​the-​day/​bompas-​parr-​design-​musical-​spoons-​heinz-​
beans/​240605/​
Arnett A. 2018. “Food operas elevate the art of eating”. Boston
Globe, 22 January 2018, www.bostonglobe.com/​ lifestyle/​
food-​ d ining/​ 2 018/​ 0 1/​ 2 2/​ f ood-​ o peras-​ e levate-​ a rt- ​ e ating/​
wlBXNrU8603ukdJtwqLdRI/​story.html
Burzynska J. 2015. “Carbonic Oscillation”. http://​joburzynska.com/​
portfolio/​
Burzynska J. 2017. “Amazuppai”. http://​joburzynska.com/​amazuppai/​
Burzynska J. 2018. “Risonanze di Vino”. http://​joburzynska.com/​
risonanze-​di-​vino/​
Halter, R. 2016. “Happiness: a chocolate symphony”. www.huffpost.
com/entry/happiness-a-chocolate-symphony_b_57ea9daae4b
07f20daa0fbf9
Houge B, Friedrichs J., 2018a. “Food Opera”. www.audiogustatory.
com/​
Houge B, Friedrichs J., 2018b. “Mugaritz XX Aniversario”. www.
audiogustatory.com/​mugaritz-​xx-​aniversario
Mesz B. 2012a. https://​apps.utu.fi/​media/​nakoislehdet/​5d-​cookbook/​
en/​desktop/​index.html?article=20&page=1
Mesz B. 2012b. “ToM recipes”. www.academia.edu/​7053828/​To_​
M_​recipes
Mesz B, Herzog K, Amusátegui JC, Samaruga L, Tedesco S. 2017b
“Let’s drink this song together”. https://​drive.google.com/​open
?id=1NACoGZNT2ivZg1KhhjEvUDT4o6tcipz7
Mesz B. 2019. “Copas”. https://​vimeo.com/​338451637
Monneo I, Monneo P. 2020. “Taste of Sound”. www.tasteofsound.se
Seth R. 2012. “Musical dinner”. www.yankodesign.com/​2012/​09/​
27/​musical-​dinner
Spevack Y. 2017. “Series”. https://​yntegrity.com/​series/​
Ulla G. 2011. “Grant Achatz plans to ‘overhaul the experience’ at
Alinea”. http://​eater.com/​archives/​2011/​11/​23/​grant-​achatz-​
planning-​major-​changes-​at-​alinea.php#more
Wang QJ. 2019a. “Projects”. www.qianjanicewang.com/​projects
Wang QJ. 2019b. “Aros multisensory dinner”. www.qianjanicewang.
com/​aros-​multisensory-​dinner-​2019
Youssef J. 2015. “Sonic seasoning: how does sound affect what we
eat?” www.greatbritishchefs.com/​features/​sonic-​seasoning-
jozef-​youssef
Zegna MC, Wang J, Mesz B. 2015. “Emotional Dinner”. www.
youtube.com/​playlist?list=PLXX0DvM2mUJ4S7iAlq8mR9N
ZpBw_​NcZ35
Zohn SD. 2016. “Telemann’s Musique de table and the Tafelmusik
Tradition”. www.oxfordhandbooks.com/​view/​10.1093/​oxford
hb/ ​ 9 780199935321.001.0001/ ​ o xfordhb- ​ 9 780199935321-
e-​120
Temporal Domination of Sensation: When Building Dishes,
Let’s Take Temporality into Account

Pascal Schlich
Centre des Sciences du Goût et de l’Alimentation, AgroSup Dijon, CNRS, INRAE, Université Bourgogne
Franche-​Comté, F-​21000 Dijon, France
CNRS, INRAE, ChemoSens Facility, PROBE Infrastructure, F-21000 Dijon, France

The time course perception of dishes is important for their them require the panellists to be able to score intensities and thus
appreciation, and the culinary design of flavour by chefs could require training.
be improved if based on careful analysis of systems. Temporal When the author and collaborators designed Temporal
Dominance of Sensations (TDS) is an analytical technique that Dominance of Sensations (TDS) (Pineau et al., 2003), they also
fills the gap between static multidimensional sensory profiling and aimed to record intensities over time, but chose to do this only
dynamic unidimensional Time–​Intensity (TI) by offering a way when an attribute attracts the attention of the taster. That is the
to assess simultaneously several attributes dynamically over time. concept of “dominance”. In its present form, TDS cannot realis-
The core idea of TDS is to no longer score intensities but to elicit tically deal with more than approximately 12 attributes (Pineau
dominances, which is simpler for untrained individuals. Recently, et al., 2012). Profiling products with so few attributes sometimes
TDS was paired with liking, wanting and/​or satiation evaluated requires gathering several sensory attributes into a single attribute
quantitatively and dynamically during the intake of a full portion (for instance, “fruity” to gather the several fruits one can per-
of a food, beverage or combination of the two. Further, Temporal ceive in that type of product). It is thus possible to define briefly
Dominance of Emotions (TDE) has been recently proposed as a (using words and/​or references) each attribute to untrained con-
TDS based on emotional rather than sensory attributes. sumers. Then came the idea of asking panellists only for these
successive dominances without associating each of them with an
intensity score, and TDS thus became feasible for a consumer
panel (Albert et al., 2012; Brachet et al., 2014, Visalli et al.,
Origin of TDS: From Intensity to Dominance
2016). Indeed, it only requires a quick familiarization with the
Sensory perception while eating a food product is a temporal data acquisition device, thanks to the evaluation of one warm-​up
process. This was acknowledged by sensory evaluation approxi- product.
mately 60 years ago (Neilson, 1957) with the so-​called Time–​ In practice, the computerized TDS system shows the panellist
Intensity (TI) technique (Lee, 1986). It is usually done by moving the entire list of attributes on a computer screen (Figure 86.1).
a computer mouse up or down according to the intensity perceived. The panellist is then asked to click on the start button as soon
TI focuses on a single attribute, such as sweetness. If several as the product enters his/​her mouth and then to consider which
attributes are of interest, then several TI-​tastings are necessary, of the attributes is perceived as dominant. Thereafter, each time
making the experiment very difficult and time-​consuming. the panellist feels the perception has changed, he/​she selects the
Dual Attribute Time Intensity (DATI) was proposed 20 years new dominant attribute, until the perception ends. During the
ago to extend TI to two attributes simultaneously (Duizer et al., testing of one product, the panellist is free to select an attribute
1997a; Duizer et al., 1997b), but again, it was too complicated several times. Conversely, another attribute may not be selected
for panellists to handle. More recently, a so-​ called Multiple at all. During the tasting, the computer records a series of events
Attribute Time–​Intensity technique was proposed to extend TI (clicks), each composed of two items: time of the click (0 being
to a larger number of attributes (Kuesten et al., 2013), but it is the time at which the panellist starts to taste) and name of the
the opinion of the author that tracking the intensity of more than attribute clicked. From these discrete data, a sequence of con-
one attribute continuously over time is almost impossible. This tinuous dominances is defined by considering that an attribute
is certainly the reason why researchers developed discontinuous remains dominant until another one is clicked.
solutions to more or less repeat a sensory profiling method (Clark The procedure plots each attribute as a curve (Figure 86.2). For
and Lawless, 1994; Jack et al., 1994; Methven et al., 2010) sev- each point of time, the proportion of runs (subject × replication) for
eral times during the consumption of a mouthful or portion of a which the given attribute was assessed as dominant is calculated.
food product. Depending on the products and list of attributes, These proportions are smoothed over time (e.g., using moving
each of these various techniques has its own value, but all of averages or splines) and displayed as curves of the evolution

575
576 Pascal Schlich

of the dominance rate for each attribute. The product depicted
in Figure 86.2b can be described as dominated by crunchiness
perception at the beginning of the sequence, followed by brittle-
ness and finally stickiness. The time-​standardized representation
(Figure 86.2b) takes individual differences in onset and duration
of evaluation into account. To illustrate this, one can think about
a beer in which the first sensation quoted might be bitterness, but
after several seconds, a rising perception of an aroma (say, malty
aroma) occurs, whereas the bitterness remains the most intense.
In that case, the panellist perceiving the malty aroma must click
on the corresponding attribute in the list, because this aroma did
trigger his attention.
During panel training, this concept can also be illustrated
through sounds, the attributes being the different instruments of a
band. For the panellists, it is generally easy to understand that the
most triggering/​dominant perception is not necessarily the loudest
one but the one bringing the biggest change in the melody. TDS
with intensity being also a tool for the trained panel, the question
of monitoring panellist performances has been raised by a few
authors (Meyners, 2011; Hutchings et al., 2014; Lepage et al.,
2014), but none of these approaches has been universally accepted.
However, TDS with no intensity as a technique for consumers is
TDS evaluation of product n° 236
START Crunchy
STOP
Dry
Sticky
not concerned with this request for performance investigation.
Furthermore, we have unpublished data suggesting that trained
panellists may use fewer attributes than untrained consumers and
may take more time to pick the first one, as if they were thinking
too much during the task and/​or possibly looking for some learned
sequences in which the attributes are expected to be perceived.
Similar results were recently suggested in a study (Rodrigues
et al., 2016) in which the same panellists did TDS first without
and then after some training. Such data, if confirmed, would even
suggest that TDS is not well adapted to trained panels.
Today, TDS has been included in most software for sensory
analysis. Readers interested in a review describing some of this
literature can refer to Di Monaco et al. (2014). Although a large
majority of these papers report studies based on trained panels,
we speculate that TDS will be used more and more for consumer
studies with no intensities. To some extent, TDS could be seen
as a Check-​All-​That-​Apply (CATA) method, making sense of
the sequence and times at which people check the attributes. The
popularity of CATA in consumer studies may soon propagate to
TDS with no intensity. In fact, Temporal Check-​All-​That-​Apply
(TCATA) has been proposed recently as an alternative to TDS
(Castura et al., 2016). TCATA only differs from TDS in that an
attribute selected remains applicable until the panellist unclicks
it. Therefore, several attributes could be applicable at a given
time, which is perfectly acceptable from a sensory point of view.
However, we personally do not believe that a consumer could
look for new applicable attributes and, at the same time, wonder
Hard whether the attributes currently selected are still applicable.
Indeed, in most cases, the perception of one intake of a product
lasts for 30–​60 s, which is too short for a complicated task of
Gritty
selecting and deselecting attributes. We found that the TDS task
was at approximately the right level of difficulty for an untrained
Crispy consumer, while the TCATA task is obviously more complicated.

Brittle Light
Pairing TDS with Liking, Wanting and Satiation
The justification for using TDS with consumers is, of course,
FIGURE 86.1 TDS evaluation screen of the texture of a cereal snack to relate temporal descriptive data to liking evaluation of the
product. products given by the same consumers. Several studies (Bouteille

Non-standardized TDS curves Standardized TDS curves

0.8 0.8
0.7 0.7
Brittleness
0.6 0.6
Dominance rate [%]

Dominance rate [%]

Crispiness
0.5 0.5 Crunchiness
0.4 0.4 Dryness
Grittiness
0.3 0.3
Hardness
0.2 0.2 Lightness
0.1 0.1 Stickiness

0 0
0 10 20 30 40 50 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Time [s] Time [0,1]
(a) (b)

FIGURE 86.2 TDS curves of a cereal snack product (a). TDS curves of the same product after left–​right time standardization (b).
Temporal Domination of Sensation
et al., 2013; Paulsen et al., 2013; Bemfeito et al., 2016) used a
trained panel for TDS and a consumer panel for liking. Other
studies (Oliveira et al., 2015; Ares et al., 2017) used panels of
consumers who did TDS or TCATA and provided a static liking
score afterward. However, the challenge was to use a dynamic
liking evaluation in order to correlate sensory and hedonic kin-
etics. Indeed, there are sufficient data suggesting that liking is
temporal (Lee, 1986; Taylor and Pangborn, 1990; Veldhuizen
et al., 2006) and further, that its temporality allows product dis-
crimination (Sudre et al., 2012; Delarue and Loescher, 2004).
An extension of TDS was proposed under the name Temporal
Drivers of Liking (TDL) (Thomas et al., 2015) in two different
variations. In the alternated-​ TDL (Thomas et al., 2016), the
liking score is given between two successive intakes (e.g., sips
or mouthfuls), whereas in simultaneous-​TDL (S-​TDL) (Thomas
et al., 2017), the liking scale is available on the same computer
screen as the TDS panel of attributes. The alternated-​TDL is
easier than the S-​TDL, during which the panellist has to perform
TDS and temporal liking concurrently during the perception
of a mouthful of food or sip of a beverage. Additionally, more
research and data are necessary for validating the ability of con-
sumers to use TDL protocols.
The alternated-​TDL of a full portion of the product involves
a large number of intakes and thus a large number of liking
scores in order to show a potential between-​intake dynamic of
liking. Evaluating the full portion of a product opens the door
to monitoring the evolution of wanting and/​or satiation. These
measures can be as interesting as, if not more interesting than,
monitoring the liking. Examples of such studies can be found
in Thomas et al. (2016); Thomas et al. (2017); Thomas et al.
(2018). These studies emphasize a double level of temporality:
within versus between intakes. We found that the between-​intake
scale often induced more temporal variation in both perception
and liking. However, more research is necessary to investigate
whether asking for several variables between numerous intakes
is not too difficult for consumers and does not result in cognitive
correlations between these variables. The need for developments
of multiple-​intake protocols was clearly stated by Delarue and
Blumenthal in their thorough review of temporal aspects of
consumer preferences (Delarue and Blumenthal, 2015). In line
with this recommendation, a recent study (Rocha-​Parra et al.,
2016), based on three consecutive sips of a new beverage,
combined a dynamic TI evaluation of liking with facial expres-
sion and concluded that liking increased over sips as a result of a
decreasing intensity of negative emotions.
Temporal Dominance of Emotions (TDE)
TDS was used recently to evaluate modulations of the temporality
of chocolate perception induced by emotional states mediated by
different types of music listened to by consumers during choc-
olate consumption (Kantono et al., 2016). In this study, emotions
were assessed only at the end of the tasting. However, it was
suggested that emotions elicited when tasting a food product
may also be temporal (Jager et al., 2014). These authors conse-
quently proposed the Temporal Dominance of Emotions (TDE)
577
as a straightforward extension of TDS by simply replacing sen-
sory with emotional attributes. In their example, it was clear
that flavoured chocolates elicited more arousal and stimulating
attributes than plain chocolates, which elicited more ‘boring’ and
calm attributes. However, is this difference truly due to emotion
or just an expression of differences among chocolates in their
level of novelty? Whether emotions can be measured by a ques-
tionnaire is debatable, but those who adopt this idea should also
consider TDE.
No Dominance and Dual-​TDS
Assuming that a single attribute is dominant at any time and
remains dominant until a new one is selected, TDS also assumes
that it is not possible to have a period of no dominance or to
have two attributes dominant at the same time. However, a pan-
ellist might no longer perceive the attribute as dominant without
having a new dominant attribute in mind or might perceive two
dominant attributes simultaneously. The latter often occurs when
attributes belong to two different sensory modalities. In that case,
a panellist is likely to choose the dominant attribute from the
sensory modality that is easier to describe. For instance, it was
observed for many products that texture attributes are more likely
to be selected as dominant than flavour attributes, resulting in
poor temporal flavour profiles. Hold-​Down and Dual-​TDS were
proposed recently to overcome these problems.
In Hold-​Down TDS, panellists are instructed to hold down the
button of the mouse or, in the case of a tactile screen, to keep
their finger on the button until dominance ends. When the pan-
ellist releases the mouse button, or removes their finger from
the screen, a period of no-​dominance starts automatically until
the panellist presses another button or eventually, the original
one. In classical TDS, another simple way of allowing for a no-​
dominance period is simply to make it possible to re-​click on the
highlighted button to switch it off, thus starting a no-​dominance
period. In both solutions, a period of no-​dominance is initiated,
and it ends when the panellist selects a new dominant attribute.
In Dual-​TDS, attributes belonging to two different modalities are
arranged on the computer screen in two columns, and panellists
are instructed that they can have one attribute dominant in each
column at the same time. In practice, the selection of an attribute
switches off only the dominant attribute from the same column
and not the other one. Dual-​TDS can also be combined with the
two ways of accepting no-​dominance periods. If Hold-​Down
TDS is used in Dual-​TDS, then panellists have to use one finger
from each of their two hands to click and re-​click on attributes of
both columns. Software for sensory data acquisition is expected
to offer these new protocols in the near future, and at least one
already does (TimeSens, 2019).
Using TDS-​Liking for Descriptive and Hedonic
Evaluation of Food Pairing
In an experiment where consumers sampled a piece of cheese
between sips of wine, which was TDS-​profiled and liking-​scored,
578
Galmarini and co-​workers provided strong data suggesting that
cheese makes the wine taste different and better (Galmarini et al.,
2016). In another experiment, in which intakes of cheese were
evaluated and sips of wine were consumed between the intakes
of cheese, wines moderately changed the taste of cheeses but
definitely not their liking (Galmarini et al., 2017a). This raised
interesting questions about the certainty of liking and further-
more, offered a new and useful technique for studying any kind of
food combination. Further, the same authors also proposed a way
to freely assess a pair of products in an S-​TDL way (Galmarini
et al., 2018). In this protocol, each panellist is free to take a sip
of wine and a mouthful of cheese at any time and in any order, as
he/​she will do in natural conditions of wine and cheese tasting. Of
course, TDS curves are no longer meaningful on this data; how-
ever, analysis of variance (ANOVA) and multivariate ANOVA
(MANOVA) of durations of dominance by periods allowed ana-
lysis of this data and knowledge to be gained on perception of
wine-​cheese combinations.
A Need for TDS Data Analysis Refinements
Typical analysis of TDS data relies mostly on the visual inspec-
tion of the TDS curves that represent, for each attribute, the evo-
lution across time of the proportion of panellists selecting that
attribute as dominant (Pineau et al., 2009). Randomization tests
were proposed for making a statistical inference based on TDS
data (Meyners and Pineau, 2010), but these tests require very
long computations that have to be adapted to each situation,
making them almost impossible to use routinely by a sensory sci-
entist. We also do not support the idea of extracting TDS curve
parameters for analyzing them by principal component analysis
(PCA) or analyzing the curves by a multivariate analysis method
such a PARAFAC (Rodrigues et al., 2015) because, in contrast
to TI, those curves are not produced by individuals but are an
approximate visual summary of the panel.
It is the opinion of the author that a more appropriate way of
testing product differences from TDS data is to use ANOVA and
MANOVA of dominance durations (time between a citation and
the next one). A nice refinement consists of incorporating into
those linear models a ‘period’ effect splitting the total time into a
few (usually three: beginning, middle and end of perception) con-
secutive periods (Galmarini et al., 2017b). It is thus possible to
test product differences by period and to assess whether product
perception evolves differently among samples. Finally, sensory
trajectories can be visualized from a PCA biplot based on a data
table containing dominance durations of products by periods as
observations and attributes as variables. If product trajectories are
not parallel on that biplot, then differences in temporal perception
of the products are evident (Pineau and Schlich, 2015).
It should be noted that all these methods implicitly assume
panellist homogeneity in temporality perception. In cases of poor
homogeneity (which is likely to occur with a consumer panel),
curves should not reach significance very often, and ANOVA
and MANOVA should not exhibit period and product by period
interaction effects. To circumvent this limitation, Lecuelle et al.
Pascal Schlich
(2018) recently proposed analysing TDS data with an adaptation
of semi-​Markov chains. This technique estimates the probability
of transition from one dominant attribute to another and allows
those probabilities to be different among a number of periods
automatically determined to maximize a likelihood function.
This stochastic framework allows researchers to further segment
the consumer panels into clusters of homogeneous consumers
in terms of temporal perception (Cardot et al., 2019). We hope
that this new framework will be a new deal for the analysis of
TDS data.
Last But Not Least: What Is Dominance?
The extensive literature on TDS validates its usefulness in many
areas of applications. Panellists who have done TDS usually like
it because they find it very easy to do. Overall, we can say that
TDS is easily measuring something useful. However, do we really
know what is measured? We defined dominance as what attracts
your attention at a given time, but the reasons for raising your
attention may be diverse among individuals and also across types
of product. It can be due to the perception of a new attribute, the
extinction of the perception of a former attribute, a sudden vari-
ation in the perceived intensities of one or several attributes, a
cognitive search for novelty, or a willingness to sample as many
attributes as possible. We have to admit that we lack basic know-
ledge on what dominance is, and we hope that psychologists and
physiologists will address this topic sooner or later.
REFERENCES
Albert A, Salvador A, Schlich P, Fiszman S. 2012. Comparison
between temporal dominance of sensations (TDS) and key-​
attribute sensory profiling for evaluating solid food with
contrasting textural layers: Fish sticks. Food Quality and
Preference, 24, 111–​118.
Ares G, Alcaire F, Antunez L, Vidal L, Gimenez A, Castura JC.
2017. Identification of drivers of (dis)liking based on dynamic
sensory profiles: Comparison of Temporal Dominance of
Sensations and Temporal Check-​all-​that-​apply. Food Research
International, 92, 79–​87.
Bemfeito RM, Rodrigues JF, Silva JG, Abreu LR. 2016. Temporal
dominance of sensations sensory profile and drivers of liking
of artisanal Minas cheese produced in the region of Serra da
Canastra, Brazil. Journal of Dairy Science, 99, 7886–​7897.
Bouteille R, Cordelle S, Laval C, Tournier C, Lecanu B, This H,
Schlich P. 2013. Sensory exploration of the freshness sensa-
tion in plain yoghurts and yoghurt-​like products. Food Quality
and Preference, 30, 282–​292.
Brachet S, Galmarini M, Chatelet B, Loiseau AL, Cordelle S,
Lempereur V, Schlich P. 2014. Evaluation de la Dominance
Temporelle des Sensations de vins rouges de Bourgogne et
du Beaujolais par un panel d’experts et de consommateurs.
In 37th World Congress of Vine and Wine and 12th General
Assembly of the OIV; Mendoza, Argentina, 9-​14th November,
2014:06006.
Cardot H, Lecuelle G, Schlich P, Visalli M. 2019. Estimating finite
mixtures of semi-​Markov chains: An application to the segmen-
tation of temporal sensory data. Journal of the Royal Statistical
Society, Applied Statistics, Series C, 68(5), 1281–​1303.
Temporal Domination of Sensation
Castura JC, Antúnez L, Giménez A, Ares G. 2016. Temporal Check-​
All-​That-​Apply (TCATA): A novel dynamic method for char-
acterizing products. Food Quality and Preference, 47, 79–​90.
Clark CC, Lawless HT. 1994. Limiting response alternatives in time–​
intensity scaling: An examination of the halo-​dumping effect.
Chemical Senses, 19, 583–​594.
Delarue J, Blumenthal D. 2015. Temporal aspects of consumer
preferences. Current Opinion in Food Science, 3, 41–​46.
Delarue J, Loescher E 2004. Dynamics of food preferences: A case
study with chewing gums. Food Quality and Preference, 15,
771–​779.
Di Monaco R, Su C, Masi P, Cavella S. 2014. Temporal Dominance
of Sensations: A review. Trends in Food Science & Technology,
38, 104–​112.
Duizer LM, Bloom K, Findlay CJ. 1997a. Dual-​ attribute time-​
intensity sensory evaluation: A new method for temporal meas-
urement of sensory perceptions. Food Quality and Preference,
8, 261–​269.
Duizer LM, Bloom K, Findlay CJ. 1997b. Dual-​ attribute time–​
intensity measurement of sweetness and peppermint percep-
tion of chewing-​gum. Journal of Food Science, 61, 636–​638.
Galmarini M, Dufau L, Loiseau AL, Visalli M, Schlich P. 2018. Wine
and cheese: Two products or one association? A new method
for assessing wine-​cheese pairing. Beverages, 4(1), 13.
Galmarini MV, Loiseau AL, Visalli M, Schlich P. 2016. Use of multi-​
intake Temporal Dominance of Sensations (TDS) to evaluate
the influence of cheese on wine perception. Journal of Food
Science, 81, S2566–​S2577.
Galmarini MV, Loiseau AL, Visalli M, Schlich P. 2017a. Use of
multi-​intake Temporal Dominance of Sensations (TDS) to
evaluate the influence of wine on cheese perception. Journal
of Food Science, 82, 2669–​2678.
Galmarini MV, Visalli M, Schlich P. 2017b.Advances in represen-
tation and analysis of mono and multi-​ intake Temporal
Dominance of Sensations data. Food Quality and Preference,
56, 247–​255.
Hutchings SC, Foster KD, Grigor JMV, Bronlund JE, Morgenstern
MP. 2004. Temporal dominance of sensations: A comparison
between younger and older subjects for the perception of food
texture. Food Quality and Preference, 31, 106–​115.
Jack FR, Piggott JR, Paterson A. 1994. Analysis of textural changes
in hard cheese during mastication by progressive profiling.
Journal of Food Science, 59, 539–​543.
Jager G, Schlich P, Tijssen I, Yao JL, Visalli M, de Graaf C, Stieger
M. 2014. Temporal dominance of emotions: Measuring
dynamics of food-​related emotions during consumption. Food
Quality and Preference, 37, 87–​99.
Kantono K, Hamid N, Shepherd D, Yoo MJ, Grazioli G, Carr BT.
2016. Listening to music can influence hedonic and sensory
perceptions of gelati. Appetite, 100, 244–​255.
Kuesten C, Bi J, Feng Y. 2013. Exploring taffy product consump-
tion experiences using a multi-​attribute time–​intensity (MATI)
method. Food Quality and Preference, 30, 260–​273.
Lecuelle G, Visalli M, Cardot H, Schlich P. 2018. Modeling Temporal
Dominance of Sensations with semi-​ Markov chains. Food
Quality and Preference, 67, 59–​66.
Lee WE, III, Pangborn RM. 1986. Time–​intensity: The temporal
aspects of sensory perception. Food Technology, 40, 71–​82.
Lepage M, Neville T, Rytz A, Schlich P, Martin N, Pineau N. 2014.
Panel performance for Temporal Dominance of Sensations.
Food Quality and Preference, 38, 24–​29.
Methven L, Rahelu K, Economou N, Kinneavy L, Ladbrooke-​Davis
L, Kennedy OB, Mottram DS, Gosney MA. 2010. The effect
of consumption volume on profile and liking of oral nutritional
579
supplements of varied sweetness: Sequential profiling and
boredom tests. Food Quality and Preference, 21, 948–​955.
Meyners M. 2011. Panel and panelist agreement for product
comparisons in studies of Temporal Dominance of Sensations.
Food Quality and Preference, 22, 365–​370.
Meyners M, Pineau N. 2010. Statistical inference for temporal dom-
inance of sensations data using randomization tests. Food
Quality and Preference, 21, 805–​814.
Neilson AJ. 1957. Time-​Intensity studies. Drug & Cosmetic Industry,
80, 452.
Oliveira D, Antúnez L, Giménez A, Castura JC, Deliza R, Ares G.
2015. Sugar reduction in probiotic chocolate-​flavored milk:
Impact on dynamic sensory profile and liking. Food Research
International, 75, 148–​156.
Paulsen MT, Næs T, Ueland Ø, Rukke EO, Hersleth M. 2013.
Preference mapping of salmon–​sauce combinations: The influ-
ence of temporal properties. Food Quality and Preference, 27,
120–​127.
Pineau N, Cordelle S, Imbert A, Rogeaux M, Schlich P. 2003.
Dominance temporelle des sensations –​Codage et analyse d’un
nouveau type de données sensorielles. In 35èmes Journées de
statistique; Lyon, France, 2-​6th June.
Pineau N, Goupil de Bouillé A, Lepage M, Lenfant F, Schlich P,
Martin N, Rytz A. 2012. Temporal Dominance of Sensations:
What is a good attribute list? Food Quality and Preference, 26,
159–​165.
Pineau N, Schlich P. 2015. Temporal dominance of sensations
(TDS) as a sensory profiling technique. In Delarue J, Lawlor
B, Rogeaux M (eds.), Rapid Sensory Profiling Techniques
(Woodhead Publishing), 269–​306.
Pineau N, Schlich P, Cordelle S, Mathonnière C, Issanchou S,
Imbert A, Rogeaux M, Etiévant P, Köster EP. 2009. Temporal
Dominance of Sensations: Construction of the TDS curves and
comparison with time–​intensity. Food Quality and Preference,
20, 450–​455.
Rocha-​Parra D, García-​Burgos D, Munsch S, Chirife J, Zamora MC.
2016. Application of hedonic dynamics using multiple-​ sip
temporal-​liking and facial expression for evaluation of a new
beverage. Food Quality and Preference, 52, 153–​159.
Rodrigues JF, Condino JPF, Pinheiro ACM, Nunes CA. 2015.
Temporal Dominance of Sensations of chocolate bars with
different cocoa contents: A multivariate approach to assess
TDS profiles. Food Quality and Preference, 54, 51–​57.
Rodrigues JF, de Souza VR, Lima RR, Carneiro JDS, Nunes CA,
Pinheiro ACM. 2016. Temporal dominance of sensations
(TDS) panel behavior: A preliminary study with chocolate.
Food Quality and Preference, 54, 51–​57.
Sudre J, Pineau N, Loret C, Martin N. 2012. Comparison of methods
to monitor liking of food during consumption. Food Quality
and Preference, 24, 179–​189.
Taylor DE, Pangborn RM. 1990. Temporal aspects of hedonic
responses. Journal of Sensory Studies, 4, 241–​247.
Thomas A, Chambault M, Dreyfuss L, Gilbert CC, Hegyi A,
Henneberg S, Knippertz A, Kostyra E, Kremer S, Silva AP,
Schlich P. 2017. Measuring temporal liking simultaneously
to Temporal Dominance of Sensations in several intakes. An
application to Gouda cheeses in 6 European countries. Food
Research International, 99, 426–​434.
Thomas A, van der Stelt AJ, Prokop J, Lawlor JB, Schlich P. 2016.
Alternating temporal dominance of sensations and liking
scales during the intake of a full portion of an oral nutritional
supplement. Food Quality and Preference, 53, 159–​167.
Thomas A, van der Stelt AJ, Schlich P, Ben Lawlor J. 2018. Temporal
drivers of liking for oral nutritional supplements for older
580
adults throughout the day with monitoring of hunger and thirst
status. Food Quality and Preference, 70, 40–​48.
Thomas A, Visalli M, Cordelle S, Schlich P. 2015. Temporal drivers
of liking. Food Quality and Preference, 40, 365–​375.
TimeSens. 2019. www.timesens.com
Veldhuizen MG, Wuister MJP, Kroeze JHA. 2006. Temporal
aspects of hedonic and intensity responses. Food Quality and
Preference, 17, 489–​496.
Pascal Schlich
Visalli M, Lange C, Mallet L, Cordelle S, Schlich P. 2016. Should
I use touchscreen tablets rather than computers and mice in
TDS trials? Food Quality and Preference, 52, 11–​16.
Wang HY, Qian H, Yao WR. 2011. Melanoidins produced by the
Maillard reaction: Structure and biological activity. Food
Chemistry, 128, 573–​584.
Texture: The Physics of Mouthfeel –​Spreadable Food and
Inulin Particle Gels
Thomas A. Vilgis
Max-​Planck-​Institut für Polymerforschung, Ackermannweg 10, 55128 Mainz, Germany
Introduction
Most food can be classified as “soft matter”, which has to be
processed in the mouth by chewing, mastication and bolus forma-
tion. Even “hard matter”, such as crystalline chocolates or glasses,
and amorphous matter, like hard caramels or potato chips, show
“soft properties”: either they melt at slightly lower temperatures
present in the mouth, they dissolve relatively quickly in the
saliva, or their ultimate fracture properties are weak enough for
them to be broken between tongue and palate. Consequently, pure
physical processes induce a dominant part of the consistency
and “mouthfeeling”, particularly mechanical processes; how do
solids break?; how do the particles, through mastication, become
smaller and smaller?; how are they wetted by the saliva?; how do
the plastic deformation properties of the forming bolus change
with time?; and what about the friction in the mouth? Apart from
the composition of the foods, these mechanical properties also
determine the release of water-​soluble taste and volatile odorant
components during oral processing, and finally lead to the desired
sensations associated with indulgence (Chen and Engelen, 2012).
However, to understand these processes better, it must be noted
that most of the sensory aspects have to be correlated with the
hierarchical and multiscale (molecular) structure of the foods.
When, for example, food materials break, the crack propaga-
tion depends on a number of molecular properties, the water
activity, the resulting surface structure, and the molecular elas-
ticity of proteins and carbohydrates, to name only a few. Even
small differences in the physical properties yield distinguishable
reactions. The most well-​known examples are fresh and slightly
stale potato crisps, where only a minimal uptake of humidity
completely changes the fracture behaviour as well as taste and
odour release. In addition, most processed foods are emulsions,
which are sufficiently stable on long time scales. Well-​known
examples are liquid (e.g., mayonnaise) or solid (e.g., sausages,
purees, and semi-​solid foods) emulsions. In such emulsions, taste
and odorant compounds can be exactly partitioned in their appro-
priate solvents to provide their controlled release during oral pro-
cessing (Vilgis, 2014; Vilgis, 2015).
In this chapter, we concentrate mainly on the textural prop-
erties, which have a physical origin. One interesting and typical
example is liver sausage, which is based on a mixture of pork fat,
meat and liver, which forms a semi-​solid emulsion. Its mouthfeel
is dominated by a combination of plastic deformation and, when
its temperature rises in the mouth, melting of the partially crys-
talline pork fat. Since such sensations are very much liked, it is
also tempting to create fat-​free, vegetarian or vegan alternatives,
which might provide a similar sensation in the mouth. This is,
however, a non-​trivial challenge for soft matter physics.
Liver Sausages: The Role of Proteins and Fats
The mechanical deformation of liver sausage is dominated by its
plasticity. Figure 87.1 shows a typical force–​deformation curve
of liver sausages when measured using a texture profile ana-
lyser. The increase at small deformation corresponds to elastic
responses, mainly due to the protein matrix, whereas the con-
stant force on increasing deformation corresponds to the plastic
behaviour, when fat particles in the emulsion are pushed away
from each other without any force required, which is easy to
understand.
At room temperature, most of the fat is crystalline and becomes
emulsified during processing inside a meat and liver matrix. In
traditional liver sausages, the meat and fat are heated and cooled
down, whereas the liver is not heated when processed in the
cutter. In this case, the (native) liver proteins act as emulsifiers
and enclose the still molten fat, which crystallises in the
droplets during cooling. The water-​(and salt-​)soluble proteins
of the liver (as well as mono-​and diglycerides, free fatty acids
and phospholipids present in the organ) act as highly effective
surfactants and induce selective interactions between the fat
particles and the emulsifiers in the sausage.
The surface activity (Lauridsen, 1976) of the proteins has its
origin in the distribution of the hydrophilic and hydrophobic
amino acid residues in the primary structure. When the sequence
is sufficiently “blocky”, the proteins are highly surface-​active,
and stable emulsions can be formed easily during processing.
Some proteins are denatured mechanically by the energy impact
and distributed accordingly between fat and water phases. Such
proteins emulsify oil droplets (Figure 87.2). Consequently, the
liver sausage with its naturally high fat content forms an emulsion
(Figure 87.3) in which the fat droplets form an irregular network.
581
582
FIGURE 87.1 Typical force–​deformation curve for liver sausages.
FIGURE 87.2 Emulsified fat droplets. Hydrophilic amino acids are drawn
blue, hydrophobic red.
The droplets interact with specific interactions between the emul-
sifying proteins.
In many practical cases, such networks may obey random
fractal structures; i.e., the size of the cluster R (part of a larger
particle network) scales characteristically with the number
of droplets, N; for example, R d f ∝ N , where d f is the “fractal
dimension” (Joshi et al., 2018). When clusters (Figure 87.3) are
mechanically stressed, the force is transmitted only through the
“simple connecting path” from the top droplet to the last droplet;
the dangling parts of the clusters do not transmit forces. As long
as this path resists the increasing force, the response is “elastic”.
Once the connections between the droplets are broken, only
local rearrangements of the droplets with no further increase of
the force are necessary to change the form. The deformation is
purely plastic, as can be seen in Figure 87.1, and makes a signifi-
cant contribution to the texture and mouthfeel of liver sausages
and patés.
It is tempting to reproduce this behaviour of particle gels in
vegetarian and vegan products, and some consumers also demand
strongly fat-​reduced or even fat-​free versions. Often, the fibre
inulin, an extract from plant roots, is used as a “fat replacer”
in such semi-​solid foods, and indeed provides a very fat-​like
mouthfeel. Given the fact that inulin consists of fructose-​based
Thomas A. Vilgis
FIGURE 87.3 Simple model of a cluster of emulsified fat droplets. The
background is hydrophilic (matrix).
oligomers, between 2 and 120 fructose units, the question is why
and how does it work?
Inulin: Selectively Crystallisable Molecules
Nutritionally, inulin is a dietary fibre, exhibiting low energetic
content and prebiotic properties, and is suitable for diabetic
nutrition. Inulin belongs to the family of the fructans, built up
of β-(2,1) linked D-​fructose units and, most often, a terminal
D-​glucose residue (Joshi et al., 2018; Beccard et al., 2019). Its
physical strength is linked to its broad molecular weight dis-
tribution. Inulin consists of small units (of slightly sweet taste)
and of long units (tasteless). When inulin is immersed in water,
the chains dissolve at a given temperature that depends only on
the chain length. Short units and chains dissolve first, while long
chains do not dissolve at all. Only at temperatures close to 100 °C
do all molecules dissolve in water. Consequently, when inulin
becomes dissolved at some higher temperature, chains dissolve
and crystallise again during cooling according to the chain length;
the undissolved material acts as crystallisation nuclei. The result
is again an irregular particle network (Bot et al., 2004), of very
similar structure to that depicted in Figure 87.3, and, not surpris-
ingly, the texture profile measured in a texture profile analyser
appears very similar. The network of crystalline inulin particles
is fractal at certain length scales, as shown by the red cluster at
the left of Figure 87.4.
At the yield point of the force, the fractal network is destroyed,
and the plastic regime takes over. Almost no force is needed
apart from constant friction between the particles on increasing
deformation. Apart from the details, inulin particle gels resemble
very much the behaviour of pasty food like liver sausages.
Texture: The Physics of Mouthfeel
FIGURE 87.4 Structural changes of inulin particle gels under compression.
Thus, inulin is a perfect candidate for building controlled
textures that mimic plastic deformations in the mouth to give a
“fatty” mouthfeel. However, the mouthfeel in foods containing
fats such as pork, goose or duck fat, as well as coconut oil or
cocoa butter, also contributes to the melting of the partially
crystalline triacylglycerols (“fat molecules”), which cannot be
provided by the inulin. The crystals need to be dissolved by the
saliva, which has usually different time scales. Nevertheless, the
fractal nature of inulin helps here as well.
Fractal Particle Gels and “Simulated Melting”
In the mouth, both mechanically induced sensations from com-
pression and shear forces add up to the resulting physical tex-
ture. The resulting shear deformations between tongue and palate
during oral processing consist of small deformations as well as
large amplitudes. It is therefore necessary to study such particle
gels under shear, especially in amplitude sweeps, which cover the
linear response as well as the non-​linear deformation regime of
the gels. Whereas the gel supports linear elastic modes quite well
(see Figure 87.4 in the low deformation regime), the gel becomes
destroyed at larger amplitudes. Fractal particle gels under large
deformation amplitudes show a very distinctive behaviour, which
has features of melting under mechanical forces, as shown in
Figures 87.5 and 87.6.
A typical experimental result is shown in Figure 87.5. The
oscillation strain was applied in a rheometer, and the storage
modulus was measured (at a frequency of 1 Hz). Small
deformations show the linear elastic regime, where the modulus
is practically constant, showing only a small decrease. At larger
strains, above 20%, the modulus decreases, and the particle gel
becomes weaker and weaker. Obviously, the gel structure breaks
down. Interestingly, the modulus follows a scaling law, which has
been developed in a completely different context, the physics of
reinforced polymer materials such as rubbers (Vilgis and Huber,
1999). These experimental results allow for a more quantitative
583
FIGURE 87.5 Amplitude sweeps of inulin particle gels underline the fractal
nature of the structures. The steep decrease of the modulus corresponds to the
“mechanical melting”.
analysis and the development of a physical model for the break-
down and mouthfeel of particle gels in general.
The breakdown of the gel particles with increasing amplitude
supports the fractal nature of the structure. As in liver sausages,
the gel breaks successively at the weakest points. Since, on
average, the broken pieces statistically resemble a similar struc-
ture to the large original network, a “scaling law” can be expected.
The physical concept of this is shown in Figure 87.6, where
the breakdown of the network is schematically illustrated. The
large network breaks with increasing deformation into smaller
and smaller self-​similar pieces. This behaviour can be described
mathematically, and the modulus can be predicted (Huber and
Vilgis, 1999).
To do this, it is necessary to introduce several “fractal
dimensions”. The first is, of course, the fractal dimension df itself
(see earlier), and the second is the “connectivity dimension” C,
which describes the connectivity of the network and contains, for
example, information on the “weakest path”, where the particle
gel is likely to break. The connectivity dimension has two clear
limits; the lower limit is 1, corresponding to a linear polymer,
for example, while the upper limit is about 1.33 for randomly
branched clusters, as in the present case.
After carrying out some mathematical “fractal calculus”, the
shear modulus G′ for the experiment shown in Figure 87.5 can
be described as
G' − G0 1 1
= , m=
G0' − G∞' 1 + K 2γ 2m C − df + 2
This model for the storage modulus thus provides a simple
scaling relation between the shear amplitude γ and the structural
parameters of the inulin particle gel. K is a numerical constant,
which reflects the local interactions between the particles forming
the gel. When it is fitted to the experimental results, it is possible
to extract the fractal dimension df with a value of 2.3. The clusters
are not space-​filling in three dimensions.
584
FIGURE 87.6 Model of the systematic breakdown of the network of a gel.
Conclusions
The breakdown of the particle gels under oral processing by
compression and shear determines to a great extent the physical
contribution to the mouthfeel. In addition, the gradual lowering
of the friction between tongue, the forming bolus and the palate
enhances the pleasure simultaneously with the release of the
flavour. Underlying the chemical senses triggered by the simul-
taneous release of taste and odorant compounds, a large part of
the mouthfeel and the texture is driven by physical processes,
which are ruled by quite fundamental physical properties. Simple
model systems can help to indicate some underlying universal
physical concepts. Clearly, the preparation of pure inulin gels
and their investigations under ideal laboratory conditions are,
indeed, a long way from packing inulin into vegetarian or vegan
replacements for pasty sausages for spreading. However, even
in complex matrices, the selective crystallisation of hydrophilic
oligofructose molecules needs to take place in the water-​rich
domains, which leads to cluster-​rich regions, which are respon-
sible for the “fatty sensation”. Thus, this investigation leads to
two basic conclusions. In many pasty and semi-​solid foods, the
interplay between the destruction of the structure seems to be one
key issue for the mouthfeel. It is indeed coupled to many other
processes, such as the increase of temperature of the food in the
mouth associated with the dissolution of inulin droplets and the
coupled accelerated release of flavour compounds. The structure
Thomas A. Vilgis
of the particle gel, the size of the secondary particles and the
properties of the food matrix define in the end the strength of the
oral coating, too, and thus, the time span of the flavour sensation
in the oral cavity.
REFERENCES
Beccard S, Bernard J, Wouters R, Gehrich K, Zielbauer B, Mezger
M, Vilgis TA. 2019. Alteration of the structural properties of
inulin gels. Food Hydrocolloids, 89, 302–​310.
Bot A, Erle U, Vreeker R, Agterof WG. 2004. Influence of crystal-
lization conditions on the large deformation rheology of inulin
gels. Food Hydrocolloids, 18(4), 547–​556.
Chen J, Engelen L. 2012. Food oral processing: fundamentals of
eating and sensory perception. John Wiley & Sons.
Joshi B, Beccard S, Vilgis TA. 2018. Fractals in crystallizing food
systems. Current Opinion in Food Science, 21, 39–​45.
Lauridsen JB. 1976. Food emulsifiers: Surface activity, edibility,
manufacture, composition, and application. Journal of the
American Oil Chemists’ Society, 53(6–​2), 400–​407.
Vilgis TA, Huber G. 1999. Universal properties of filled rubbers:
Mechanisms for reinforcement on different length scales.
Kautschuk Gummi Kunststoffe, 52(2), 102–​107.
Vilgis TA, Lendner I, Caviezel R. 2014. Ernährung bei
Pflegebedürftigkeit und Demenz –​Lebensfreude durch Genuss.
Springer.
Vilgis TA. 2015. Soft matter food physics –​the physics of food and
cooking. Reports on Progress in Physics, 78(12), 124602.
Texture: How Texture Makes Flavour
Ole G. Mouritsen
Department of Food Science, Taste for Life, Design and Consumer Behavior, University of Copenhagen,
26 Rolighedsvej, DK-​1958 Frederiksberg C, Denmark
The tactile sensation of food, mouthfeel, is an important but
often overlooked part of the total flavour experience of food.
Mouthfeel is determined by the texture of the food, the texture
being a multifaceted quality defined as that part of the food struc-
ture our sensory apparatus can detect. Texture often determines
our food preferences, and modifying texture is therefore a key
culinary exercise. An insight into the mechanics and molecular
underpinnings of food structure and how it can be modified by
various culinary transformations can be used to enhance the
cooking experience and lead to new creations.
Preparing food and the culinary arts involve working with
raw materials to produce food that is tasty, interesting, and nutri-
tious. All of this necessarily involves finding ways to work with
mouthfeel in the course of an often long series of transformations,
first in the kitchen and then in the mouth. Many of the processes
carried out in the kitchen are irreversible. Once a potato has been
boiled, it is not possible to cool it and return it to its raw state.
Similarly, a cooked egg that has set cannot easily become liquid
again (This, 1996). In addition, subjecting a raw ingredient to a
FIGURE 88.1 The sensory system in the skin, which registers heat, cold, pressure, and pain: (1) pain, (2) temperature, (3) touch or light pressure, and
(4) pressure.
combination of processes will not necessarily lead to the same
result if the processes are carried out in a different sequence.
Mouthfeel refers to sensory inputs form the so-​called somato-
sensory system (Shepherd, 2011; Mouritsen and Styrbæk, 2017).
This system is found not only in the mouth, but everywhere in the
body. Physical stimuli, including pain, temperature, and tactile
sensations, such as pressure, touch, stretching, and vibrations, are
detected by the somatosensory system. Technically, mouthfeel is
mediated by certain receptors in the epithelial cells of the oral
cavity that are associated with four different types of somatosen-
sory nerve endings that are sensitive to temperature, pain, touch,
and pressure, respectively (Figure 88.1). The somatosensory
system also detects the position and movements of the body and
parts of the body (kinaesthesia), like the tongue. The motions of
the tongue help us to explore and assess the size, shape, and tex-
ture of a piece of food while we are chewing. The nerve endings
in the teeth also provide information about the structure of the
food –​its hardness, whether it is crunchy or elastic, and the size
of the food particles.
585
586
Transforming the properties of raw ingredients to achieve a
desired texture and preserving it until the food is about to be
eaten can be challenging. If prepared commercially, the food
might need to be transported and may be kept on supermarket
shelves for a long time. It is also necessary to take into account
the changes that occur in the final stages of preparation. In all
cases, it is crucial to make allowances for what happens when the
food goes into the mouth and mouthfeel comes into play. There,
the texture is affected by temperature, saliva, enzymes, and the
way in which we chew.
The flow properties of liquids and soft solids, as can be
measured by rheology experiments, are also important for
mouthfeel. We can feel how the food flows in our oral cavity on
its own or after we have worked on it with our tongue, palate,
and teeth. We notice whether it is sticky (like syrup), slimy (like
a thin jelly), or fatty (like oil). Liquid foods such as drinks flow
into the mouth more or less quickly on the time scale of a second
and fill it up. Semisolids, such as emulsions, on the other hand,
flow more slowly and can behave in a viscoelastic manner. Solids
do not flow at all, but sometimes they can be dissolved in saliva or
melt, or can be reduced to very small bits by the action of tongue
and teeth.
Interplay between Mouthfeel and Other Sensory
Perceptions
A ‘taste sensation’ is the brain’s interpretation of a complex
multimodal integration of all sensory inputs, not only taste
proper: flavour is in the brain (Small, 2012). It is useful to know
how mouthfeel interacts with the other sensory inputs. Whereas
much is known about how sight, hearing, and somatosensory
impressions are integrated, the understanding of the interplay
between chemical sensory impressions (taste and smell) and
mouthfeel is more limited (Saint-​ Eve et al., 2011). Most of
the existing knowledge is rather phenomenological, with little
understanding of its neurological basis. However, some of the
knowledge can come in very handy in the kitchen. The following
presentation is based on a series of research results compiled by
Verhagen and Engelen (2006) and is adapted from a review by
Mouritsen (2016).
To stimulate the five basic tastes experimentally, certain clas-
sical taste stimulators are used: sweetness is derived from ordinary
table sugar (sucrose), saltiness from sodium chloride, bitterness
from quinine or caffeine, sourness from citric acid, and umami
from monosodium glutamate (MSG). Irritation and pain are typ-
ically induced by capsaicin (from chili) or piperin (from black
pepper). Mouthfeel can be evoked by changing the viscosity of
the food, by direct contact with the food, or by the mechanical
movements of the tongue in the mouth. In the following, ‘taste’
refers to chemical taste, i.e., proper taste perceived by the taste
buds. ‘Touch’ refers to a mechanical stimulus, ‘impression’ is a
somatosensory perception, and ‘irritation’ is the consequence of
chemesthesis. For references to the original literature regarding
the observations reviewed in the following, see Verhagen and
Engelen (2006).
Ole G. Mouritsen
Taste → Touch
Taste can affect the way in which viscosity of a food is perceived.
Sweetness enhances it, sourness lessens it, bitterness has no
effect, and it is not clear whether salt has any effect.
Touch → Taste
The viscosity of a food raises the taste threshold for, and decreases
the intensity of, sour, sweet, salty, and bitter tastes. The effect
is also dependent on the medium in which the taste substances
are embedded. For example, oil limits the access of the taste
compounds to the taste buds. In contrast to other tastes, umami is
intensified by the movement of the tongue.
Temperature → Taste
The taste thresholds for sour, sweet, salty, and bitter are lowest in
the temperature range of 22–​37 °C, which is, therefore, the range
within we are most sensitive to the taste of food. It would appear
that this is due to the temperature of the tongue rather than that
of the food. Moreover, some experiments have shown that chan-
ging the temperature over a small area of the tongue can evoke
different taste sensations. For example, warming a spot on a
tongue that has been cooled brings out a sweet taste, and cooling
the tongue to 10–​15 °C can result in sour and salty tastes, but the
effect varies from one person to another. Warming the tip of the
tongue results in enhancement of sensitivity to sweetness from
sucrose, but not to other tastes. This indicates that the perception
of sweetness due to warming is less susceptible to adaptation.
Taste → Irritation
Sweetness (of sucrose) can reduce the burning sensation produced
by piperin and capsaicin, whereas sourness (citric acid) and water
have only a minor effect, and saltiness (NaCl) and bitterness
(quinine) have no effect. Here, it should be noted that substances
such as salt and quinine in large concentrations could in them-
selves lead to irritation.
Irritation → Taste
Irritation and pain caused by capsaicin and piperin reduce the
taste intensity of sweet, bitter, and umami tastes, whereas sour
and salty tastes have little or no effect. In the case of irrita-
tion caused by piperin, sour, sweet, salty, and bitter tastes are
decreased; it would appear that its effect on umami has not yet
been investigated. The special irritation and prickly sensation
produced by carbonated drinks can reduce bitterness and enhance
sourness. These results are complicated, given that some of the
test subjects experienced capsaicin as a moderately intense bitter
taste and that CO2 forms carbonic acid, which is sour, in water.
Smell → Touch
The few results that are available indicate that smell can influ-
ence the experience of textural impressions such as creaminess,
consistency, and melting qualities of a food. For example, a sub-
stance like vanilla, which releases intense odour substances and
Texture: How Texture Makes Flavour
is associated with creaminess, enhances the perception of creami-
ness of a vanilla pudding.
Touch → Smell
Increasing the viscosity of a food attenuates the intensity of the
smell perception. In this experiment, the actual release of odour
substances from the food is controlled so that it is not dependent
on viscosity.
Smell → Temperature
Some odours are perceptually associated with temperature. This
interaction is due to neural binding. For example, the retronasal
aroma given off by many spices is associated with warmth.
Temperature → Smell
In general, the intensity of odours increases with temperature.
This is most probably a purely physical-​chemical effect, as the
higher temperature leads to the vaporization of a greater quantity
of volatile odour substances and makes the food less viscous, also
facilitating the release of these substances. This sensory enhance-
ment is independent of the neurological system.
Smell → Irritation
Normally, smells suppress the feeling of irritation. On the other
hand, many smells can, in themselves, trigger irritation.
Irritation → Smell
Irritation, for example, as induced by capsaicin, depresses the
intensity of smells such as those of oranges and vanilla. There
are also indications that astringency in the nose contributes to the
sense of smell.
Touch → Temperature
Experiments have shown that applying vibration to the lips raises
the threshold for experiencing warmth or depresses sensitivity to
temperature. Increasing the viscosity, for example, by increasing
the fat content, increases the perceived temperature of a cold
substance, even though the temperature remains constant. An
explanation for this is that fat has an insulating effect. A further
indication of this effect is that a food with a high fat content at
a low temperature is sensed as being less cold than its low-​fat
counterpart. Conversely, a food with a high fat content at a high
temperature is experienced as being less warm than one with a
low fat content.
Temperature → Touch
Temperature has a marked influence on the mucous membranes
and the tongue and on the physical properties of the food,
especially its coarseness and texture. Hence, it is likely that
the influence of temperature on mouthfeel is predominantly a
physical-​chemical effect rather than a neurological one.
587
Touch → Astringency
Astringency due to the tannins from grape seeds is decreased
when the viscosity of the media in which they are dissolved is
increased. This is in agreement with other studies showing that
lubricants, for example, cooking oil, reduce the feeling of astrin-
gency. The reason for this effect is probably that the friction that
accompanies astringency is reduced by the lubricant.
Temperature → Irritation
Heat increases irritation and pain, for example, from capsaicin,
piperin, and alcohol, while cold does the opposite.
Irritation → Temperature
Irritation due to capsaicin is experienced as heat and subsequently
suppresses the feeling of cold.
Texture
Alina Surmacka Szczesniak has studied texture and how it affects
our choice of what to eat. She defines texture, and concomitantly,
mouthfeel as follows (Szczesniak, 2002):
1. A sensory quality of food, and consequently, a quality
that only a human (or another living being) can recog-
nize and describe. Only certain properties of texture can
be measured by physical means and the results of these
measurements require a sensory interpretation;
2. A multifaceted quality that cannot be described by a
single parameter, such as hard or creamy;
3. A quality that depends on the structure of the food at all
levels, from the molecular to the microscopic;
4. A quality that is recognized by several senses, of which
touch and pressure are the most important.
Bourne (2002) has condensed the definition of texture to:
‘The textural properties of a food are that group of physical
characteristics that arise from the structural elements of the food,
are sensed primarily by the feeling of touch, are related to the
deformation, disintegration, and flow of the food under a force,
and are measured objectively by functions of mass, time, and dis-
tance.’ In accordance with this definition, one ought to refer to
‘textural elements’ to acknowledge that texture is a set of sen-
sory properties. Many different parameters may therefore be
involved in the description of texture, and we use a large number
of popular expressions to describe them. What has made it so dif-
ficult to define texture and, along with it, mouthfeel is that some
textural elements have measurable physical dimensions, whereas
others take on meaning only in connection with human sensory
impressions and perceptions of the structure of the food. Hence, it
is difficult to correlate the texture as it is experienced by the sen-
sory apparatus of a human being with physical properties that can
be defined accurately and measured quantitatively. Tables 88.1
and 88.2 provide lists of parameters and popular expressions for
588
TABLE 88.1
Classification of the Textural Properties of Solid Food
Parameter Popular expressions
Mechanical
Hardness Soft, firm, hard
Cohesiveness Crunchy, brittle, tender, tough, mealy,
pasty, gummy
Viscosity Thin, thick
Springiness Plastic, elastic
Stickiness Sticky, gooey
Geometrical
Particle size Grainy, coarse, fine, grating, sandy
Particle shape Fibrous, stringy
Spatial orientation of particles Crystalline
Other
Moisture content Dry, wet, watery, moist
Fat content Oily, fatty, greasy
Source: Szczesniak, 2002; Mouritsen and Styrbæk, 2017.
the textural properties of solid and liquid food, respectively. Note
that some expressions indirectly reflect the multimodal character
of the sensations; e.g., crunchy is a tactile sensation interacting
with an auditory signal.
Texture and Food Culture
It is interesting to note that different countries and different food
cultures have significantly different numbers of terms to describe
texture (Mouritsen and Styrbæk, 2017). These differences most
likely reflect different food cultures’ appreciation of mouthfeel
as an important characteristic of food. The language with one
of the richest vocabularies for texture appears to be Japanese,
with more than 400 discrete terms, whereas most European and
North American countries only have 100 or fewer. Drake (1989)
has pointed out, from studying 22 languages, that 54 words for
textural elements have an identical meaning. These words fall
into six broad categories: viscous, plastic, elastic, compressible,
cohesive, and adhesive. The most frequently used word regarding
texture in the United States and in many European countries
is ‘crisp’. In Japan, the word that occurs most often is ‘hard’.
‘Crisp’, ‘juicy’, ‘soft’, ‘creamy’, ‘crunchy’, and ‘hard’ are among
the terms for texture most commonly used in the United States,
Austria, and Japan.
Jurafsky (2014), using big-​data techniques and data mining
of information on the internet regarding food, recipes, menus,
restaurant reviews, and people’s eating habits, has found, e.g.,
for desserts, that the reviewers’ descriptions of the desserts are
full of expressions that can be interpreted as having thinly veiled
sexual connotations. This connection manifests itself most prom-
inently in the focus on mouthfeel, rather than those for aroma,
taste, sound, or appearance. Some typical sensuously laden words
that appear frequently include ‘silky’, ‘satiny’, ‘juicy’, ‘wet’,
‘creamy’, ‘sticky’, ‘smooth’, ‘oozing’, ‘spongy’, ‘melting’,
and ‘hot’.
Ole G. Mouritsen
TABLE 88.2
Categories of Expressions of Texture for Liquid Foods
Category Typical expression
Viscosity-​related terms Thick, thin, viscous, consistency
Feel on soft tissue surfaces Smooth, pulpy, creamy
Carbonation-​related Bubbly, tingly, foamy
Body-​related terms Heavy, watery, light
Chemical effect Astringent, burning, sharp
Coating of oral cavity Mouthcoating, clinging, fatty, oily
Resistance to tongue movement Slimy, syrupy, pasty, sticky
After-​feel –​ mouth Clean, drying, lingering, cleansing
After-​effect –​ physiological Refreshing, warming, thirst-​quenching,
filling
Temperature-​related Cold, hot
Wetness-​related Wet, dry
Source: Szczesniak, 2002; Mouritsen and Styrbæk, 2017.
Texture and the Recognition of Food
Experiments have surprisingly demonstrated (Stuckey, 2012) that
for a large range of raw ingredients that have been pureed, only
about 40% of a group of younger, blindfolded test subjects were
able to identify correctly the raw ingredient. In the case of older
test subjects, the success rate decreased to about 30%. There is
a vast difference in the extent to which some of the ingredients
were identified. For example, more than 80% of the younger test
subjects were able to pick out pureed apples, strawberries, and
fish, while the comparable numbers were 40% for beef, 8% for
cucumber, and a meagre 4% for cabbage. These experiments
highlight that taste proper is not a strong sensation and, even in
cases combined with aroma, cannot on its own be a sure guide
to telling one vegetable from another. There are a number of
socially driven and gender-​related differences with respect to
the weight placed on mouthfeel. It appears that well-​educated,
wealthy people, especially the women in that category, place the
most emphasis on mouthfeel.
Food Complaints Often Relate to Mouthfeel
When customers complain about products available in food stores
or at the market, or when diners in a restaurant express dissatis-
faction with a dish, it can very often be traced back to mouthfeel.
Anecdotally, customers and restaurant guests seldom complain
that something tastes ‘bad’. Instead, it is legitimate to say that the
soufflé has collapsed, the meat is not tender, the French fries have
gone soggy, the bread is dry, the coffee is tepid, and so on. It is
much easier convincingly to describe an unfulfilled expectation
related to texture than one that pertains to chemical sensations,
such as taste and aroma. In addition, a certain texture is often
associated with the freshness of the raw ingredients and their
proper preparation.
Texture: How Texture Makes Flavour 589

TABLE 88.3
Gelling Agents and Gums
Gelling agent or gum Properties
Alginate (sodium Sets in the presence of calcium ions. Soluble
alginate) in cold water. Thermoreversible gel
formation.
Agar Sets on cooling. Thermoreversible gel
formation. Cloudy and very fragile gels.
Carrageenan Forms somewhat clear and fragile gels
containing proteins. ι-​carrageenan forms
elastic gels in the presence of calcium ions.
Locust bean gum Forms cloudy, elastic gels in the presence
of ions, especially when combined with
xanthan gum.
Guar gum Easily soluble in cold water. Forms opaque
liquids that flow slowly.
Gum arabic Easily soluble in water. At high concentrations Thickener (wine gums, soft candies, syrups),
in an acidic environment forms opaque,
viscous liquids. Inhibits sugar from
crystallizing in candies.
Xanthan gum Soluble in both cold and warm water. Forms
clear gels (together with locust bean gum)
and slowly flowing, complex liquids that
exhibit shear thinning. Thermoreversible.
Gellan gum Gelation properties resemble those of agar,
carrageenan, and alginate. Stable at
temperatures up to 120 °C.
Pectin Forms clear gels in sour foods with a high
sugar content. Some types of pectin form
strong gels in the presence of calcium ions.
Methyl cellulose Swells and thickens when heated. Clear.
Gelatine When cooled forms very clear, pliable, and
elastic gels. Thermoreversible.
Starch Swells up and dissolves in warm water.
Opaque.
Uses Mouthfeel
Thickener, stabilizer (for example, in Clean, lingering.
ice cream and frozen desserts), and
gelling agent (marmalade). Used for
spherification.
Thickener, stabilizer, and gelling agent. Clean.
Thickener, stabilizer, and gelling agent. Used, Creamy, clean, lingering.
for example, in dairy products such as
yogurt and chocolate milk.
Thickener and stabilizer. Improves freezing Lingering, sticky.
and thawing properties, for example, of
ice cream. Adds softness and elasticity to
bread dough.
Thickener (ketchup, dressings) and stabilizer Lingering, smooth.
(ice cream, dough).
Lingering, sticky.
emulsifier, and stabilizer. Binding agent in
glazes. Medium for aroma additives and
food dyes.
Versatile thickener (sauces, salad dressings) Lingering, smooth to sticky.
and stabilizer.
Thickener and stabilizer. Clean, creamy.
Thickener (ice cream, desserts, ketchup) and Clean, lingering.
gelling agent (marmalades, jams, candies).
Stabilizer in emulsions and certain drinks.
Stabilizer, emulsifier, and thickener (ice Clean to lingering and sticky.
cream).
Gelling agent used in a wide range of food Clean to lingering and sticky.
products.
Thickener used in a wide range of food Sticky and lingering.
products.

Source: Barham et al., 2010; Myhrvold et al., 2010; Mouritsen and Styrbæk, 2017.

Modifying Texture
A common goal of the culinary arts is to alter the textural proper-
ties of the food, e.g., so that a dish fractures when chewed to the
extent that it breaks up into appropriate pieces, thereby enhan-
cing its taste and maximizing its nutritional value; or the food
deforms, flows, and coats the oral cavity in a pleasant way and
releases the taste and odorant compounds in a desired fashion.
A prominent example is cooking (heat-​ treating) raw
ingredients so that they are easier to chew. In the case of both
vegetables and meat, the cellular structure in the connective
tissue is ruptured, but paradoxically, this is not why they are
easier to chew. Cooked vegetables deform more readily when
chewed because the cooking process has softened the cellulose in
the plant fibres, making them less stiff. In contrast, meat that has
been heated becomes stiffer due to actin and myosin coagulation,
and at the same time, the connective tissue loosens when collagen
denatures. Enzymes and various methods of fermentation are
other ways to modify the texture of food. Moreover, temperature,
acidity, and the presence of specific minerals (salts) can be abso-
lutely crucial for the texture.
Common ways of modifying the mouthfeel of food involve
various natural or artificial agents that change the texture of
liquid foods, e.g., egg, bread, dairy products, emulsifiers, gel-
ation agents, thickeners, and stabilizers. A list of some of these is
provided in Table 88.3.
Once the desired texture of a foodstuff has been achieved,
it may change over the course of time by a variety of natural
physico-​ chemical, chemical, and enzymatic processes, as
illustrated in Table 88.4 (Bourne, 2002). Some of these changes
can be remedied; others cannot.
The Challenge in Designing Texture
In principle, one can classify food into three levels. The top level
contains the biological organisms we eat (e.g., animals, plants,
insects, algae, fungi, bacteria, etc.). Below that is the level of
590 Ole G. Mouritsen

TABLE 88.4
Texture Changes in Food
Food Texture change Cause
Bread crumbs Firmness increases, springiness decreases Starch retrogradation, moisture transfer from starch to gluten
Bread crust Crispness decreases, toughness increases Moisture migrates from crumb to crust
Butter and margarine Firmness and graininess increase, spreadability Growth of fat crystals, change in crystal form, strengthening of
decreases network bonds
Cheese, ripe Firmness and fracturability increase, elasticity Enzymatic changes
decreases
Chocolate Graininess develops Change of the crystal structure of the cocoa butter
Surface ‘bloom’ (white spots) Sugars and fats crystallize on surface
Crackers Loss of crispness Moisture absorption from air
Fruit, fresh Softening, wilting, loss of crispness, loss of juiciness Pectin degradation, respiration, bruising, loss of moisture and turgor,
weakening of middle lamella
Ice cream Coarseness increases Ice crystals enlarge
Butteriness Clumping of fat globules
Sandiness Crystallization of lactose
Crumbliness Poor protein hydration
Mayonnaise Emulsion breaks Fat crystallization
Meat, fresh Toughness increases at first Rigor mortis
Toughness decreases later Autolysis
Meat, frozen Freezer burn, drip Surface desiccation, reduced water-​holding capacity
Mustard, prepared Leakage of water (syneresis) Aggregation of particles
Pickles Softening Breakdown caused by enzymes and microorganisms
Pies Crust loses crispness, filling becomes dry Moisture migrates from filling to crust
Filling seeps out Leakage of water from the gelling agent (syneresis)
Shellfish Softening and mushiness Enzymatic breakdown
Sugar confections Crystallinity, stickiness Sugars change from amorphous to crystalline state
Vegetables, fresh Toughening Deposition of lignin in the cell walls, e.g., asparagus, green beans
Softening Conversion of sugar to starch, e.g. green beans, sweet corn
Pitting Pectin degradation, loss of water, e.g., tomatoes
Chilling injury, e.g., bell peppers, green beans
Loss of crispness Moisture loss and turgor loss, e.g., lettuce, celery

Source: Adapted from Bourne (2002).

so-​called formulated food, that is, highly processed food (e.g.,
bread, dairy products, sauces, fermented food and beverages,
etc.), whose biological origin is difficult to recognize without
some preconceived knowledge. Below that level is still another
more basic level, which could be called the ‘elementary particles’
of food, that is, pure proteins, carbohydrates, nucleic acids, fats
and oils, minerals, water, vitamins, etc. Since all kinds of food
are built of these elementary particles, many of which are macro-
molecular entities or assemblies, it is in principle possible from
these particles to build and construct foodstuffs on the two higher
levels. Whereas this is a feasible and in some cases manageable
task for, e.g., bread, some dairy products, and sauces, it would be
very difficult, if not impossible, to construct a chicken or a mush-
room. One of the major problems is to build the foodstuff in the
hierarchical way living organisms are built by nature (Mouritsen
and Styrbæk, 2017). A main challenge here is to get the texture
right. This is exactly the challenge many companies are now
facing when attempting to grow artificial meat. Whereas it may
be possible to get the flavour right (as well as the nutritional con-
tent), it is much more difficult to provide a proper mouthfeel.
A mode of cooking has been coined note-​by-​note cuisine by
This (2014). This is an ultimate strategy to build dishes and a
meal from the elementary particles of food, i.e., pure components
of carbohydrates, proteins, and fats together with water, minerals,
etc. (see examples in the application/​ cooking section of this
book). The challenge here is less one of imparting a desired fla-
vour to the food and more of how to create a particular mouthfeel.
Whereas it is reasonably straightforward to construct a jelly, a
cheese-​like structure, and a bread from the elementary particles
of food, it is much more complicated to build more intricate
structures, e.g., those of meat and vegetables, whose texture is
multifaceted and often determined by a hierarchical architecture.
Acknowledgements
This work was supported in part by a grant to Smag for Livet
(Taste for Life) from Nordea-​fonden.
REFERENCES
Barham P, Skibsted LH, Bredie WL, Frøst MB, Møller, P, Risbo,
J, Snitkjaer, P, Mortensen, LM. 2010. Molecular Gastronomy:
a new emerging scientific discipline, Chemical Review, 110,
2313–​2365.
Texture: How Texture Makes Flavour
Bourne M. 2002. Food Texture and Viscosity, Academic Press,
London.
Drake B. 1989. Sensory textural/​rheological properties: A polyglot
list. Journal of Texture Studies, 20, 1–​27.
Jurafsky D. 2014. The Language of Food: A Linguist Reads the
Menu, Norton, New York.
Mouritsen OG. 2016. Gastrophysics of the oral cavity, Current
Pharmaceutical Design, 22, 2195–​2203.
Mouritsen OG, Styrbæk K. 2017. Mouthfeel: How Texture Makes
Taste, Columbia University Press, New York.
Myhrvold N, Young C, Bilet M. 2010. Modernist Cuisine The Art and
Science of Cooking, The Cooking Lab Publications, Bellevue,
WA, USA, 4.
Saint-​Eve A, Déléris I, Panouillé M, Dakowski F, Cordelle S, Schlich
P, Souchon I. 2011. How texture influences aroma and taste
perception over time in candies, Chemosens Perception, 4, 32.
591
Shepherd G. 2011. Neurogastronomy, Columbia University Press,
New York.
Small D. 2012. Flavour is in the brain, Physiology and Behavior,
107, 540–​552.
Stuckey B. 2012. Taste What You’re Missing: The Passionate
Eater’s Guide to Why Good Food Tastes Good, Atria Books,
New York.
Szczesniak AS. 2002. Texture is a sensory property, Food Quality
and Preference, 13, 215–​225.
This H. 1996. Can a cooked egg white be uncooked? The Chemical
Intelligencer, 10, 51.
This H. 2014. Note-​by-​Note Cooking: The Future of Food, Columbia
University Press, New York.
Verhagen JV, Engelen L. 2006. The neurocognitive bases of human
multimodal food perception: Sensory integration, Neuroscience
and Biobehavioral Reviews, 30, 613–​650.
Texture: Tsukemono –​the Art and Science of Preparing Crunchy
Vegetables
Ole G. Mouritsen
Department of Food Science, Taste for Life, Design and Consumer Behavior, University of Copenhagen,
26 Rolighedsvej, DK-​1958 Frederiksberg C, Denmark
Tsukemono is a Japanese term for ‘pickled or steeped things’ and
mostly applies to vegetables. It covers a wide range of one or
more conservation techniques, involving ingredients such as salt,
sugar, vinegar, alcohol, and herbs, in combination with methods
including dehydration, marinating in salt and acidic liquids, fer-
mentation, and curing. Apart from conserving the vegetables,
these techniques lead to very crunchy and flavourful products.
One of the best-​kept secrets of Japanese cuisine, which the
wider world has yet to discover in depth, is a range of side dishes
known as tsukemono (つ つけもの, 漬物, pronounced like ‘tskay-​
moh-​noh’) and literally meaning ‘something that has been steeped
or marinated’ (tsuke –​ steeped; mono –​things). While they may
not yet have appeared over the horizon in Western cuisines, these
pickles are just as common a part of every traditional Japanese
meal –​breakfast, lunch, and dinner –​as cooked rice and miso
soup. Tsukemono are usually made from vegetables, some
fruits, and flowers, and a few rhizomes are also preserved in
this way; it is, therefore, more accurate to characterize them as
‘pickled foods’. However, the process of making tsukemono is
more than just a simple way of preserving otherwise perishable
fresh produce (Richie, 1985; Yamaguchi, 1988; Shimizu, 1993;
Andoh, 2010; Hachisu, 2015; Mouritsen and Styrbæk, 2021;
Mouritsen, 2018).
Preparing tsukemono involves different salts, sugar, vinegar,
alcohol, enzymatic and bacterial fermentation, as well as various
pickling beds of miso, soy sauce, and sake lees. The pickling not
only aids in conservation and enhancement of nutritional value;
it also has significant effects on taste, aroma, and particularly
mouthfeel. In many cases, the umami taste is increased, but the
most striking feature of most tsukemono is its unique mouthfeel
of crunchiness (Mouritsen, 2018). An understanding of the scien-
tific mechanisms behind this texture involves the effect of ions,
in particular divalent ions such as calcium and magnesium ions
(from sea salt), pH, and enzymes. In many cases, the crunchy
mouthfeel is enforced by first drying vegetables before marin-
ating. It is generally a matter of changing the water activity in
the vegetables.
Tsukemono are normally prepared without any cooking and
are eaten cold, and they are generally easy to make at home.
Apart from their nutritional value, their contribution is to stimu-
late the appetite, add delicious flavour sensations, and improve
digestion, all while remaining an exceptionally elegant study in
simplicity and aesthetic presentation.
It is worth noting that, in the Japanese kitchen, tsukemono are
not just pickled foods that are served arbitrarily as condiments.
They are an essential element of Japanese cuisine. On the surface,
they have a Zen-​like simplicity, but both their tastes and textures
are very sophisticated. They are as much part of an ordinary meal
as of a formal, classical kaiseki dinner. Commercially prepared
tsukemono are sold in inexpensive plastic bags in supermarkets,
but they can also be bought as much sought-​after regional and
seasonal specialities, elaborately packaged for presentation as
highly prized gifts.
Tsukemono play a special role in relationship to cooked rice.
In Japan, rice is usually prepared without salt and has a rather
bland taste. Traditionally, tsukemono were served with a cup of
green tea and a bowl of cooked rice at the conclusion of a meal or,
possibly, as a snack at afternoon tea time. Things have changed,
and a small serving of tsukemono may make an appearance
before, during, or after a Japanese meal. In addition, they are
often combined with other types of food.
It is said that there are about 4000 different kinds of tsukemono
in Japan and more than 100 different ways of preparing them.
Some examples are shown in Figure 89.1. This overabundance of
choice can be overwhelming, especially when compared with the
number of pickled products that are common in Western cuisines.
Preparation of Tsukemono
Tsukemono can be made from virtually any kind of vegetable,
as well as fruit, seaweeds, fish, shellfish, and even flowers. The
different varieties and the techniques related to preparing them
are both denoted by the same word, which always ends with
–​zuke. For example, miso-​zuke refers to both a method and its
end product. The many different ways of preparing tsukemono
are a reflection of the need to conserve these fresh ingredients
for extended periods of time, up to several years. The advent of
593
594 Ole G. Mouritsen

Shiba-zuke Kyuri-zuke Daikon nara-zuke Shiba-zuke

Takuan-zuke Fukujin-zuke Matcha takuan-zuke Senmai-zuke

Kabocha-zuke Iburi-gakko Gari Nasu nara-zuke

Nasu shio-zuke Shiba-zuke Takana-zuke Uri nara-zuke

Kyuri-zuke Kabu shio-zuke Shiba-zuke Shiba-zuke

Takama-zuke Shiba-zuke tsukudani Shiba-zuke Takama-zuke

FIGURE 89.1 An assortment of tsukemono.

(Courtesy of Jonas Drotner Mouritsen)
Texture: Tsukemono
freezers and refrigerators has completely altered the landscape,
and there are now many types of tsukemono that have a much
shorter shelf life than earlier versions (Mouritsen, 2018).
Tsukemono are often classified into two different categories
that are distinguished by the duration of their pickling time. Some
of these require little effort and result in simple foods that will
keep for only a few days. Those that can be prepared quickly, for
example, asa-​zuke, or ‘quick pickles’, are marinated in salt or in
weak brine for a few hours or days. They usually need to be kept
refrigerated and are eaten within a few days. The others, called
furu-​zuke, require long, elaborate preparation, conservation, and
ageing for periods that range from a few weeks to a year or more,
and will keep for months and up to several years. As long as they
are stored in a dark place and in the original pickling crocks to
keep them from drying out, they do not need to be cooled. The
differences in preparation and ageing times are reflected in dis-
tinct variations in flavour and smell. A common trait, however,
is that most tsukemono made from vegetables are crisp and so
crunchy that every bite releases an explosion of taste and aroma.
Some Japanese chefs describe this experience as a taste with long
duration, much as the French use longueur en bouche to describe
the finish and aftertaste of wine.
The various techniques for preparing tsukemono can be divided
into ten basic types that are sometimes used in combination
(Table 89.1). These types correspond to three different procedures:
marinating in liquid, preserving in a paste, where there may also
be a simultaneous fermentation process, and a genuine fermen-
tation process mediated by enzymes and microorganisms such
as fungi, yeasts, and bacteria. With all three methods, the add-
ition of salt, either directly or indirectly from other ingredients,
is the most essential element. If a paste is involved, it is normally
wiped off, either completely or partially, before the preserve is
eaten. Both the liquid marinade and the paste can be flavoured
with spices and additives, such as Japanese chili (togorashi),
Japanese pepper (sansho), mustard (karashi), ginger, yuzu, shiso,
and seaweeds (konbu, wakame, nori, ao-​nori). The preparation
technique for a particular vegetable or fruit is chosen in order to
enhance its distinctive character, texture, and flavour.
TABLE 89.1
Ten Generic Ways of Preparing Tsukemono
Type of tsukemono Preparation technique
Shio-​zuke curing in salt
Su-​zuke curing in vinegar
Sato-​zuke curing in sugar
Amazu-​zuke curing in vinegar and sugar
Shoyu-​zuke curing in soy sauce
Karashi-​zuke curing in mustard
Nuka-​zuke fermenting in rice bran
Kasu-​zuke curing in sake lees
Miso-​zuke curing in miso
Koji-​zuke fermenting in koji
Source: Mouritsen and Styrbæk, 2021.
595
It is, by far, easiest to work with liquid marinades consisting, for
example, of brine, soy sauce, sake, or vinegar. Using fermentation
media containing pastes and fungus cultures is more complicated,
but the reward is a considerably more nuanced flavour. In some
instances, several preservation media, with different effects, are
used in combination. For example, a marinade may consist of
soy sauce, sake, and sugar or may be a mixture of miso and sake
lees. In the case of some vegetables, dehydrating the vegetables
before they are preserved results in a much crisper and crunchier
mouthfeel. Those that are to be immersed in a thick paste, gener-
ally for several weeks or months, must be desiccated first, other-
wise the preservation medium becomes too moist. Dehydration
can be carried out using salt, according to old-​fashioned methods
out of doors, or with the help of a dehydrator.
Salt, Salts, and Texture
With the exception of sato-​zuke, which are sweet, every type of
tsukemono is made with salt that is added either directly or indir-
ectly from other sources such as soy sauce and miso. Originally, a
high salt concentration, often as great as 8−10%, was an important
factor in ensuring that the pickles would keep for a long time.
And as a further precaution, the tsukemono were traditionally
stored in the coldest, darkest part of the house. Refrigeration and
freezing have radically altered the picture. With the addition of
preservatives and the use of effective cooling techniques, the salt
content has now been reduced to 3−4%, even for the tsukemono
that are intended to have a long shelf life. Some modern varieties
are also pasteurized, vacuum packed in plastic bags, and kept
refrigerated. Here, the primary role of salt is its effect on taste,
mouthfeel, and digestibility. In all circumstances, the salt content
of a given product is a question of striking a balance between fla-
vour and conservation.
The simplest way to make tsukemono is basically just to
marinate the ingredients in brine containing 5–​25% salt. Ordinary
table salt (impure NaCl), which has a dependably uniform com-
position, is not used exclusively. In Japan, it is traditional to
choose instead from a variety of sea salts. These may contain
additional salt compounds, e.g., potassium, magnesium, and
calcium salts, as well as substances, such as seaweed ash, that are
actually desirable impurities. It is, therefore, crucial to select
the right type of salt, as these differences completely determine
the outcome of the pickling process with respect to both the taste
and texture of the resulting tsukemono.
The salt used to prepare tsukemono draws liquid out of the
cells and destroys them, rendering the vegetables rather soggy, as
shown in Figure 89.2. This releases their enzymes, which assist
in the process of degrading many of the bitter substances present
in raw vegetables, leaving them with a milder, sweeter taste. At
the same time, microorganisms, such as yeast and bacteria, may
begin their work of breaking down the plant matter. This brings
about the formation of a whole new series of taste substances,
counteracts the impression of saltiness, and leads to rounder,
softer taste nuances. In addition, the salt prevents the growth of
unwanted bacteria.
596
FIGURE 89.2 Daikon before and after they have been brined, a process that
reduces their volume by 40% and makes them very crunchy.
FIGURE 89.3 Schematic illustration of the strengthening of the bonds
between pectin molecules in a vegetable in the presence of divalent ions, such
as calcium or magnesium ions. The bonds between the individual polysac-
charide molecules are represented by the jagged ‘egg box’ structure. The red
circles represent the divalent ions that can bind different strands of the pectin
molecules to stabilize the structure.
Tsukemono can become extra crisp if they are preserved
using sea salt that contain calcium and magnesium salts. Ions
from these two salts cause the carbohydrates in the vegetables,
for example, pectin, to form strong bonds, a process known as
cross-​binding (Figure 89.3), resulting in a very firm and crisp
mouthfeel. This is exactly the same mechanism used to make
jelly and other hydrogels from alginate, a gelation agent extracted
from seaweeds.
Many of the soluble dietary fibres, for example, pectins, are
found in the cell walls of plants. Pectin molecules in a aqueous
environment are, depending on pH, negatively charged and repel
each other. In order to bind the pectin molecules tightly together
Ole G. Mouritsen
FIGURE 89.4 Drying vegetables (cucumber, daikon, turnip, kohlrabi)
before pickling them for tsukemono.
(Courtesy of Jonas Drotner Mouritsen).
and form a firmer structure together with the water, it is neces-
sary to counteract this repulsion. One way to do so is to add sugar
(sucrose), which binds water and, as a consequence, forces the
pectin molecules to link together a little more strongly. Another
possibility is to add acid, which reduces the extent of the elec-
tric repulsion. Still another option is to introduce calcium ions
or magnesium ions, which have a positive charge and in this
way bind the negatively charged pectin molecules more strongly
together. The ability of calcium ions to stiffen foodstuffs that
contain pectin can be exploited to ensure that cooked or pickled
vegetables remain firm. It is simply a matter of adding sea salt or
a pure calcium salt such as calcium citrate.
Drying and Texture
Preserving vegetables is partly a matter of reducing their water
content, but it is not necessary to remove it entirely. Surprisingly,
one of the most effective ways of imparting crispness to tsuke-
mono involves pre-​drying of the vegetables so that they lose 40–​
80% of their fresh weight (Figure 89.4). As described in another
chapter in the present volume (on Imaging by Clausen et al.),
dehydration is hard on the cells of the vegetables; it changes their
shape, and the cell walls can be broken, altering the texture of the
vegetable. The sugar content of the dried vegetable also affects its
texture. One with more sugar will be chewier and more flexible.
Sugar binds the water and will also cause water to be absorbed
from the surroundings. Vegetables with a great deal of soluble
dietary fibre, such as pectin, that binds water less strongly than
sugar have a tendency to turn out more crunchy than those with
a high sugar content.
Typically, the temperature in a vegetable dehydrator is kept in
the range of 40−60 °C in order to preserve colour as well as taste
and aroma. The partial dehydration of the vegetables diminishes
the growth of bacteria and enzymatic activity on their surfaces,
thus lessening the risk of deterioration. Nevertheless, drying in
itself is not sufficient to preserve the vegetables; this will require
the addition of salt.
Texture: Tsukemono
FIGURE 89.5 Microscopy images of the cell structure in a radish: fresh (a),
dried (b), and rehydrated (c). Each image corresponds to 0.6 mm × 1.6 mm.
(Courtesy of Mathias Porsmose Clausen)
Figure 89.5 shows microscopy images of the cellular structure
of a radish in the fresh state, after drying (dehydration), and after
rehydration (in a marinade). During the dehydration, the cells
shrink, and the stiff and regular network of cell walls becomes
crumpled and more flexible. During subsequent rehydration (in a
marinade), the cells take up water again, but the network remains
irregular and flexible. This implies that the perceived texture in
the mouth is one where the vegetable is pliable and deformable,
but when the teeth cut through the vegetable, it fractures with a
crunchy mouthfeel and a noise, an auditory sensation: the unmis-
takable sound of tsukemono.
Flavour
It is the very special crunchy texture and mouthfeel that furnishes
the unique flavour experience of tsukemono (Mouritsen and
Styrbæk, 2017). Apart from this special mouthfeel of tsuke-
mono, the main common flavour characteristic is umami. Most
vegetables have little umami in their fresh state (Mouritsen and
Styrbæk, 2014, 2020). However, via the various ways of pre-
serving them, in particular via pickling beds involving fermented
products like soy sauce, miso, and sake lees, or by active lactic
bacteria fermentation in, e.g., a nuka-​doko fermentation bed,
the vegetables are imparted or develop a host of odour and taste
compounds, including glutamate, which leads to the umami taste.
Aesthetics and Health
Tsukemono evolved in Japan as tasty, nutritious condiments for a
simple bowl of cooked rice. Although tsukemono could be placed
on top of the rice, they are now presented separately on a small
dish or in a small bowl. The presentation is simple and consists
of only very small amounts; it is just as much a matter of what
597
is not there as of what is there –​it is a matter of wabi. They
are usually offered as a small selection of pieces, arranged separ-
ately on a dish or in a bowl. The empty spaces between them are
just as important as the actual tsukemono. The choice of colours,
sizes, and shapes is of key importance, in the same way as it is
in the traditional Japanese meals washoku, kansha, and kaiseki.
Harmony in all sensory impressions is centre stage.
Tsukemono are meant to be eaten in small quantities
selected from a variety of different types. These tasty, nutritious
condiments were, in the first instance, a simple staple that was
used to round out an ordinary meal of plain boiled white rice
and miso soup. But despite their humble origins, they were also
considered sufficiently sophisticated to be integrated into temple
foods and very elaborate dinners. Virtually no meal in Japan is
now served without at least a little tsukemono, and even in the
most modern households, a few pieces usually find their way onto
the plate. It is the contention of the present author that tsukemono
has a lot to offer to Western gastronomy.
In Asia, and particularly in Japan, tsukemono are considered
healthy foods (Murooka et al., 2008), because, apart from their
content of macronutrients like proteins and carbohydrates, they are
a source of important minerals, vitamins, and antioxidants, as well
as soluble and insoluble dietary fibres. The fact that these pickles,
whether or not they are fermented, are produced without heating
implies that many nutrients, vitamins, and active enzymes are better
preserved than those in vegetables that have been cooked. A pos-
sible drawback may be that some of these desirable components,
such as proteins, are less bioavailable during digestion than those
in cooked vegetables. Finally, it should be pointed out that the salt
content in tsukemono has been linked to health issues regarding
hypertension and certain cancers (Ren et al., 2012).
Acknowledgements
This work was supported in part by a grant to Smag for Livet
(Taste for Life) from Nordea-​ fonden. Mariela Johansen is
thanked for her diligent work on translating some of the authors’
texts from Danish into English.
REFERENCES
Andoh E. 2010. Kansha: Celebrating Japan’s Vegan and Vegetarian
Traditions. Ten Speed Press, Berkeley.
Hachisu NS. 2015. Preserving the Japanese Way: Traditions of
Salting, Fermenting, and Pickling for the Modern Kitchen.
Andrews McMeel Publishing, Kansas City.
Mouritsen OG. 2018. Tsukemono –​crunchy pickled foods from
Japan: a case study of food design by gastrophysics and nature.
International Journal of Food Design, 3, 103–​124.
Mouritsen OG, Styrbæk K. 2014. Umami: Unlocking the Secrets of
the Fifth Taste. Columbia University Press, New York.
Mouritsen OG, Styrbæk K. 2017. Mouthfeel: How Texture Makes
Taste. Columbia University Press, New York.
Mouritsen OG, Styrbæk K. 2020. Design and ‘umamification’ of
vegetables for sustainable eating. International Journal of
Food Design, 5, 9–42.
Mouritsen OG, Styrbæk K. 2021. Tsukemono: Decoding the Art
and Science of Japanese Pickling. Springer Nature, Cham,
Switzerland.
598
Murooka Y, Yamshita M. 2008. Traditional healthful fermented
products of Japan. Journal of Industrial Microbiology and
Biotechnology, 35, 791–​798.
Ren JS, Kamangar F, Forman D, Islami F. 2012. Pickled food and
risk of gastric cancer − a systematic review and metaanalysis
of English and Chinese literature. Cancer Epidemiology
Biomarkers & Prevention, 21, 905–​915.
Ole G. Mouritsen
Richie D. 1985. A Taste of Japan. Kodansha, Tokyo.
Shimizu K. 1993. Tsukemono. Japanese Pickled Vegetables.
Shufunotomo Co., Ltd, Tokyo.
Yamaguchi E. 1988. The Well-​Flavored Vegetable. Kodansha
International, Tokyo.
Thickeners: Cellulose and Its Derivatives
Rachel Edwards-​Stuart
Culinary Science Department, Westminster Kingsway College, 76 Vincent Square, London SW1P 2PD, United Kingdom
Cellulose is said to be the most abundant organic compound on earth:
it is a naturally occurring material that makes up about one-​third
of all plants (Granstrom, 2009). It is the main constituent of plant
cell walls, along with hemicellulose, pectin and lignin (Belitz et al.,
2004). Cellulose molecules are linear polymers consisting of D-​
anhydroglucopyranose units (AGU), linked by β-​(1→4) glycosidic
bonds formed between C1 and C4 atoms of adjacent monomers.
Even though cellulose shares similarities with amylose, in
that it is a linear plant polysaccharide made up solely of glucose
monomers, there is a difference in properties because of the way
the monomers are linked together. As a result, thermal treatments
dissolve starch granules but leave cellulose fibres intact; and
humans can digest starch but not cellulose (McGee, 2004).
Within plant tissues, the cellulose molecules pack together
in a highly ordered manner, kept in place by hydrogen bonding
between the chains to form microfibrils several micrometres in
length and about 10–​20 nanometres in diameter. It is the stability
of this ordered structure that gives cellulose its strength as well as
its insolubility in practically all solvents (Medronho et al., 2012;
Coultate, 2009).
Cellulase enzymes, which break down cellulose, are absent
in the human digestive tract as well as that of most animals,
so cellulose contributes to the indigestible component of plant
food, which is referred to as dietary fibre (Belitz et al., 2004).
Herbivorous animals, however, particularly ruminants, can digest
cellulose because of microflora in their rumen that can hydrolyse
cellulose (Belitz et al., 2004).
Cellulose material also cannot be softened by normal kit-
chen techniques; when cooked in boiling water, the cellulose
fibres remain intact, but the amorphous materials in which they
are embedded, which include the pectins, are partly extracted
into fluids from within the cells, thus weakening the walls and
tenderising the vegetables or fruits (McGee, 2004).
Cellulose derivatives are produced from cellulose, in par-
ticular the cellulose in wood and cotton –​wood contains about
40–​50% cellulose, while cotton fibres contain about 90% cellu-
lose (Hamad et al., 2016). Cellulose derivatives are commonly
grouped into four classes: sodium carboxymethylcellulose, also
known as cellulose gum, or CMC; methylcellulose (MC) and
hydroxypropylmethylcellulose (HPMC), collectively known as
modified cellulose gums; hydroxypropylcellulose (HPC); and
ethylcellulose (EC).
The number of AGUs linked together by β-​(1→4) glyco-
sidic bonds is known as the degree of polymerisation (or DP)
of the cellulose, or chain length, and this varies depending on
the source (Belitz et al., 2004; Granstrom, 2009). Each AGU
contains three hydroxyl groups at C2, C3 and C6 positions. These
are the sites where substitution takes place to form the cellulose
derivatives. The number of hydroxyl groups that are substituted
when forming the modified celluloses is known as the degree of
substitution (DS).
Cellulose, in its unmodified form, is insoluble in water
(Medronho et al., 2012). Each of the hydroxyl groups mentioned
is capable of hydrogen bonding to an adjacent molecule, and
because of the abundance of hydroxyl groups, the chains are bound
tightly together. Water molecules, whether hot or cold, cannot
force their way in between the chains to hydrate them, making
cellulose insoluble in water (Hoefler, 2002). When some of these
hydroxyl groups are substituted by hydrophobic groups, such as
methyl or hydroxypropyl groups, in making modified celluloses,
the number of intermolecular hydrogen bonds between units is
reduced, and the cellulose becomes water-​soluble (Cummings,
1984; Xu et al., 2004).
Cellulose derivatives are commonly used in food applications,
where they are effective as thickening agents, stabilisers and rhe-
ology modifiers (Imeson, 2010).
Carboxymethylcellulose
Structure
CMC is an anionic water-​ soluble cellulose derivative that is
hygroscopic. To make CMC, the cellulose is first suspended in
an alkali to open up the tightly bound cellulose chains, allowing
water to enter. Once the water has entered, the cellulose is then
reacted with sodium monochloroacetate to produce sodium
carboxymethylcellulose (Hoefler, 2002).
Commercially available grades of CMC have a DS of up to
1.5. CMC can theoretically have a maximum DS of 3; how-
ever, such a high DS value is neither practical nor useful. The
substituted carboxymethyl groups in CMC protrude from the
cellulose backbone, which means that the remaining OH groups
on the sugar backbone cannot get close enough together to form
intermolecular hydrogen bonds. This means that water can fit in
599
600 Rachel Edwards-Stuart

between the CMC molecules and hydrate them, making CMC Rheological behaviour is also affected by DS; unsubstituted
water-​soluble. regions of low-​DS cellulose or non-​uniformly substituted cellu-
lose gum grades tend to associate and cause thixotropic behaviour.
Solubility Thixotropy is defined as the progressive decrease in viscosity with
time for a constant applied shear stress, followed by a gradual
Reducing DS results in a decrease in water solubility. When recovery when the stress is removed (Cullen et al., 2012). The
the DS is below 0.4, the CMC is insoluble due to the associa- non-​substituted regions of a non-​uniformly substituted molecule
tive behaviour between cellulose units along the unsubstituted will behave just like cellulose and will hydrogen bond to a similar
regions of the cellulose. When substitution occurs uniformly region on an adjacent molecule, allowing a loose gel network to
along the backbone, associative behaviour is prevented, and build up. This loose gel network can be disrupted with shear, but
solubility increases. Solubility is also affected by chain length; when left alone without shear, the network will reform. High-​
decreasing the DP will increase water solubility (Imeson, 2010). DS grades of CMC and those with uniform substitution, where
there is little associative behaviour between polymer chains, tend
Viscosity to exhibit pseudoplastic flow with no thixotropy (Hoefler, 2002;
CMC is a common thickener used in the food industry. The vis- Imeson, 2010).
cosity of solutions of CMC is affected by molecular weight and
DS: longer chain lengths result in higher viscosity, and reduced Compatibility with Other Ingredients
DS or less uniform substitution leads to associative behaviour,
Solutions of cellulose gum maintain viscosity over a wide pH
resulting in higher viscosity. The viscosity of aqueous solutions
range, although viscosity will be at its maximum at near-​neutral pH
of CMC, like that of most other water-​ soluble polymers,
values; at pH levels below 3, the acid form of CMC predominates
decreases with increasing temperature. Under normal conditions,
and will be less soluble. If CMC is to be used in a low-​pH solution,
this effect is reversible; however, prolonged heating at very high
the cellulose gum should be allowed to fully hydrate before adding
temperatures will permanently degrade the cellulose gum due
the acidic ingredients. The situation is similar with salt, where
to depolymerisation, which will result in a permanent decrease
divalent salts, such as those containing calcium ions, will inhibit
in viscosity. This makes it unsuitable for use as an additive in
cellulose hydration, affecting viscosity and producing a hazy solu-
products destined to be sterilised in a retort (Hoefler, 2002).
tion, although adding the salt after hydration reduces the impact,
and monovalent salts added like this will have almost no effect.
Hydration Trivalent ions such as aluminium will actually interact with CMC
CMC is soluble in either hot or cold water. However, because it and can be used to form gels (Imeson, 2010), which produces an
hydrates so rapidly, particles tend to clump together when the apparent increase in viscosity; however, for taste reasons, this has
powder is introduced to the water. Using a high-​speed mixer to little application in the food industry (Hoefler, 2002).
create a vortex will help keep the particles separate when intro- CMC is more tolerant to ethanol than most food gums,
ducing them to the water, allowing them to hydrate separately making CMC useful for applications containing alcohol, where
and reducing lumping. When preparing a CMC solution like the required transparency of the liquid can be maintained in the
this, the powder should be added quickly, because it is extremely presence of CMC.
hard to thicken an already viscous solution by adding more dry Cellulose gum displays synergistic viscosity with other
powder; however, the rate of addition should be slow enough to hydrocolloids such as locust bean gum and guar gum, because
keep the particles separate. Alternatively, the powder can be pre-​ there are more average “collisions per second” between unlike
blended with dry ingredients such as sucrose, where the added molecules, and various authors have reported starch–​cellulose
ingredients function to keep the CMC particles away from each gum synergies too (Hoelfer, 2002; Imeson, 2010).
other. Alternatively, CMC may be first dispersed in glycerol,
ethanol or propylene glycol and this mixture then added to water Applications
(Hoefler, 2002).
The main functions of CMC in the food industry are to provide
viscosity, “organize” the water, form flexible films, and control
Rheological Behaviour the rate and size of crystal growth (Hoefler, 2002). The anionic
Low concentrations and low-​molecular-​weight CMC solutions nature of cellulose gum can be beneficial in stabilising proteins
display Newtonian-​ like flow behaviour. Solutions of higher in low-​pH protein beverages like acidified yoghurt or soy drinks,
concentration and higher-​ weight gums display where it will form complexes with the positive charges on the
molecular-​
pseudoplastic behaviour, where the apparent viscosity decreases proteins, preventing acid-​induced flocculation of caseins and sub-
in response to increasing shear rate. This is because the long sequent whey separation (Du et al., 2007; Imeson, 2010).
molecules will line up in the direction of the shear, giving less CMC is also used to maintain the quality of high-​ sugar
resistance to flow. This effect is completely reversible, and once applications like syrups and frostings, where the control of water
the shear force is removed, the CMC molecules will assume their in the continuous phase by the CMC molecules helps to keep
random positions relative to each other, and the viscosity will the crystals small and prevents larger crystals from forming,
return to its original value (Hoefler, 2002; Imeson, 2010). which results in improved texture by reducing graininess. It is
Thickeners: Cellulose and Its Derivatives
also effective at controlling ice crystal growth in ice cream and
frozen desserts (Imeson, 2010), and this today remains one of
the largest single uses for the gum; indeed, CMC use was origin-
ally pioneered in ice cream (Hoefler, 2002). Other applications
include cake mixes, pie fillings, dry mix beverages, pancake
syrup, pet foods and animal feed.
Methylcellulose and Hydoxypropylmethylcellulose
Structure
With MC (which has been assigned the European code for
additives E461), some of the hydroxyl groups on the glucose
ring have been converted to methyl ether groups, and with
HPMC (which has been assigned the E-​ number E464), the
hydroxyl groups are substituted with both methyl ether groups
and hydroxypropyl groups. Hydroxypropyl groups can also be
attached to other hydroxypropyl groups, in addition to the glu-
cose ring hydroxyl sites, resulting in stacking (Hoefler, 2002).
This means that both are non-​ ionic water-​soluble cellulose
derivatives, which are less hygroscopic than anionic CMC,
FIGURE 90.1 The structure of methylcellulose (a) and hydroxypropylmethylcellulose (b).
601
since the methyl ether groups are far more hydrophobic than the
carboxymethyl groups of CMC.
Cellulose is derivatised with methyl chloride to create MC,
and with both methyl chloride and propylene oxide to create
HPMC. A large number of different types of modified celluloses
exist, which vary in their chain length (which affects viscosity)
and their level of substitution (which affects their gelling and
melting temperatures). The relative number of substituted sites
on the AGU –​or the DS –​is normally 1.4–​2.2 for MC and 1.0–​
2.3 in HPMC (Imeson, 2010). The addition of these groups gives
MC and HPMC a slightly hydrophobic character, which means
they are surface active, and allows them to act to stabilise both
emulsions and foams. The degree of surface activity displayed
depends on the level and type of substitution.
Structures of methylcellulose and hydroxypropylmethyl­
cellulose are shown in Figure 90.1.
Dispersion/​Solubility
MC and HPMC are soluble in cold water; however, as with CMC,
particles tend to clump together when the powder is introduced
to the water. Using a high-​speed mixer to create a vortex, or
602
pre-​blending the modified cellulose with dry ingredients such
as sucrose, can help keep particles separate, reducing lumping.
Unlike CMC, MC and HPMC are insoluble in hot water, and this
can make it a useful method for dispersing the powders –​some of
the water required in the finished product is heated to above the
gelling temperature of the particular MC or HPMC, and then the
polymer is added and mixed to disperse it. A portion of cold water
is then added and the mixture cooled under agitation to create the
solution. An alternative method is to disperse the powder into the
total amount of hot water and mix constantly while cooling in a
refrigerated room (Edwards-​Stuart, 2009). Solubility is affected
by DP; higher-​DP grades will be less soluble.
Rheological Behaviour
Aqueous solutions of MC and HPMC are pseudoplastic, with
pseudoplasticity increasing with concentration and molecular
weight, and show little thixotropy. While low concentrations of
MC and HPMC in water, like CMC, have been reported to exhibit
almost Newtonian behaviour (Imeson, 2010), viscosity and flow
behaviour were studied extensively by Edwards-​Stuart (2009),
who found that flow behaviour was more dependent on molecular
weight than concentration; the low-​molecular-​weight grades of
HPMC gave solutions that were more Newtonian in their behav-
iour even at high concentrations, whereas high-​molecular-​weight
grades showed pronounced shear thinning behaviour at 37 °C
even at low concentrations.
Gelation Characteristics
As solutions of MC and HPMC are heated, viscosity slightly
decreases, as is typical for most hydrocolloid solutions. When the
gelling temperature is reached, viscosity increases sharply as the
ingredients gel and become a solid mass. Thermal gelation of the
modified celluloses is completely reversible, and when removed
from the heat source, the now cloudy gel will start to melt as it
cools and eventually become a clear liquid again, retaining its ori-
ginal viscosity. The temperature at which this occurs is known as
the “melt-​back” temperature. If heated too far above its gelation
point, the gel will flocculate (precipitate) and settle out, although
upon cooling, the flocculent material will redissolve in the water
(Hoefler, 2002).
Gelling temperature and gel texture depend on the substitu-
tion level and type; the MC family will gel at lower temperatures
and form firmer gels, whereas the HPMC family will gel at
higher temperatures and produce softer gels. According to Dow
Chemical Company, a major producer of modified celluloses,
MC solutions will gel between 38 and 55 °C, whereas HPMC
solutions will gel between 58 and 90 °C.
A number of researchers have studied the thermogelation
characteristics of MC and HPMC solutions, and various different
methods have been used to measure gelling temperatures; when
using rheological methods, the gelling temperature is normally
characterised as the point at which the elastic and viscous
module cross on heating, and when using differential scanning
calorimetry (DSC), gelling temperature is characterised by the
temperature of the endotherm peak on heating. When using
Rachel Edwards-Stuart
calorimetric methods, Edwards-​ Stuart (2009) found that the
gelling temperatures for the HPMC solutions studied ranged
from 60 to 70 °C, which is in agreement with the Dow Chemical
Company guidelines, and melt-​back temperatures ranged from 45
and 55 °C. However, values obtained from calorimetric methods
did not agree with those obtained from rheological methods, and
therefore, determining the exact gelling and melting temperatures
of different modified cellulose solutions using these methods
proved challenging (Edwards-​Stuart, 2009).
Effect of Other Ingredients on Gelling Temperatures
The thermal gelling temperature of MC and HPMC solutions can
be affected by the presence of salt and sucrose; Imeson (2010)
reports that the addition of 2% sodium chloride lowers the gel
temperature of a 2% modified cellulose solution by 10–​15 °C,
and a “salting out” effect can be observed with higher levels
of salt. Medium-​molecular-​weight HPMC is more tolerant to
salt than MC and high-​molecular-​weight HPMC. The addition
of 10% sucrose will also depress gelling temperatures by up to
10 °C, while higher amounts of sugar have an even more severe
impact, and solutions containing 40% sucrose will have gelling
temperatures as much as 30 °C lower than that obtained when no
sugar is added (Imeson, 2010).
Mechanism of Gelation
Thermal gelation is caused by the weakening of water–​polymer
interactions and the strengthening of polymer–​ polymer
interactions upon heating (Imeson, 2010). In their solution
state, there is very little interaction between the modified cel-
lulose polymer chains, and they exist as aggregated “bundles”,
held together by packing of unsubstituted/​sparingly substituted
regions and by hydrophobic clustering of methyl groups in more
densely substituted regions. In the early stages of heating, there
is a partial dissociation, along with swelling, of the clusters,
followed by the separation of strands at the ends of the bundles,
exposing methyl groups to the aqueous environment, where-
upon they form structured water cages. At higher temperatures,
these cages are disrupted, and the polymer chains gradually lose
water, which is followed by the formation of the final gel net-
work by hydrophobic association of strands radiating from the
different bundles (Haque and Morris, 1993). The temperature at
which this transformation occurs is known as the gelling tem-
perature. A schematic representation of this process is shown in
Figure 90.2.
Hoefler (2002) describes the initial stages slightly differ-
ently, stating that there is a layer of “organised” water molecules
surrounding the MC molecule even at room temperature. This
“water of hydration” is associated with the hydroxyl groups
on the backbone of the cellulose and extends outwards by the
thickness of several water molecules. This organised water
layer is large enough to prevent the hydrophobic groups from
getting close enough to each other to start associating together.
It is the thinning of this “organised” water layer around the MC
molecules as the temperature is raised that causes the solution
to initially lose viscosity, and then the water layer reaches a
Thickeners: Cellulose and Its Derivatives
FIGURE 90.2 The postulated structures and processes involved in the
thermogelation of methylcellulose.
(Reproduced with kind permission from Haque and Morris, 1993)
critical “thinness”, at which point the methyl ether groups can
get close enough to each other to try and cling together to avoid
the remaining water and reduce entropy. These methyl group
“clumps” form junction zones, which results in the build-​up of a
gel structure, in an analogous manner to the hydrogen bonding
of unsubstituted regions of CMC molecules to form a thixotropic
gel structure.
Applications
One of the key functions of MC and HPMC in the food industry
is to provide thickening when cold (i.e., below body tempera-
ture), and these ingredients can also be useful in creating diet
products, where they can significantly diminish the calories of
food by replacing polysaccharides or fat (Hamad et al., 2016).
The ability of the modified celluloses to gel on heating makes
them useful in forming bake-​stable sauces, fillings, and formed
and extruded foods. Their best-​known application in the food
industry involves deep fat frying, as their thermal gelation prop-
erties can lead to greater mechanical strength when hot. They are
603
therefore often used in battered and breaded fried products such
as chicken nuggets and reformed onion rings (Hoefler, 2002).
MC and HPMC constitute a water-​holding gel that inhibits oil
access during frying, with a reduction in up to 50% oil absorption
according to Hamad et al. (2016).
The modified celluloses are also commonly used in gluten-​
free products, where they can provide highly acceptable bread
products when combined with rice flour and potato starch
(Imeson, 2010).
The presence of both hydrophobic and hydrophilic groups
along the chain of modified cellulose molecules results in surface
activity, which makes them excellent whipping agents, a prop-
erty common to protein-​based ingredients such as egg whites
and gelatine but fairly uncommon amongst the hydrocolloids.
They are also used with other stabilisers and emulsifiers in
liquids consisting of fat systems, where they gather at oil
droplet interfaces and hinder oil droplet association (Hamad
et al., 2016).
Gelation on heating rather than cooling is fairly uncommon
amongst gelling agents. While some protein-contain­ing ingre­dients
such as eggs will “gel” on heating as the egg proteins denature and
coagulate, trapping the water within, these will not melt again on
cooling, so the reversible nature of heat-​induced gelation appears
to be a property exclusive to the modified celluloses.
The ability of the modified celluloses to gel on heating has
meant that these ingredients have previously been used by chefs
for more unusual applications; for example, Wylie Dufresne,
in his restaurant WD50 in New York, served his customers a
flavoured modified cellulose solution, which they were instructed
to pipette into a hot aromatic liquid; the instant gelation allowed
customers to make their own noodles at the table.
The unique phenomenon of melting on cooling was studied
extensively as part of Edwards-​Stuart’s PhD, which was sponsored
by the Fat Duck Restaurant in Bray, UK, in order to understand
whether it might be possible to use the modified celluloses to
create hot gels, which could form the basis of a dish such as hot
ravioli, which could melt in the mouth as the mouth cooled it.
The ability of gels to melt in the mouth is a property currently
only displayed by gelatine; gels made from this ingredient melt as
the mouth acts to warm it, and this melt-​in-​the-​mouth sensation
is thought to be a major factor in determining its popularity as a
gelling agent. Mimicking this sensation by cooling a hot gel in
the mouth was therefore thought to be of interest by the Fat Duck
restaurant’s proprietor, Heston Blumenthal. Results showed that
the effect could be obtained, but at the concentration of HMPC
required to achieve the effect, the background flavour notes
were too obvious, meaning that they would have to be masked
with very strong flavours, limiting its application in this setting
(Edwards-​Stuart, 2009).
Hydroxypropylcellulose
Like MC and HPMC, hydroxypropylcellulose (HPC) is a non-​
ionic water-​soluble cellulose derivative. Cellulose is derivatised
with propylene oxide to make HPC, but as well as substituting
604
with available hydroxyl groups on the AGU, the substitutions
also occur on the side chain, where secondary hydroxyl groups
are present, resulting in branched side chains. The molar substi-
tution of a typical HPC is 3.0–​4.5.
Solutions of HPC are exceptionally clear and smooth flowing.
HPC solutions show pseudoplastic or shear thinning behaviour,
with the original viscosity returning once the shear is removed;
however, they display almost no thixotropy. The viscosity of
HPC solutions is affected by the DP. HPC is also thermoplastic
and so is often used in injection moulding (Imeson, 2010).
Thermoplastics are a class of polymers that can be softened and
melted by the application of heat and can be processed either in
the heat-​softened state or in the liquid state (Mallik, 2010).
Unlike MC and HPMC, HPC does not have a gelled state when
hot (Hoefler, 2002). HPC is insoluble in water above 45 °C; pre-
cipitation will occur in water solutions raised to temperatures
between 40 and 45 °C, but the compound will re-​dissolve on
cooling. The precipitation temperature is reduced by the presence
of solutes; 10% sucrose will lower the precipitation temperature
by 5–​8 °C, and 5% salt will lower it by 10 °C (Imeson, 2010).
The HPC polymer is soluble in many organic solvents, including
ethanol and propylene glycol. To prepare a solution of HPC, a
similar method is used as for the preparation of MC and HPMC;
the powder should first be dispersed in hot water and then cooled
under agitation by adding the balance of cold water or by cooling.
HPC may also be dispersed in glycerine (Imeson, 2010).
The key function of HPC in the food industry is to provide
thickening/​viscosity when cold. It is also a good film former
(Hoefler, 2002). HPC also acts as a good foam promoter/​whipping
agent due to the presence of both hydrophobic and hydrophilic
groups on its chain. It can also act as an emulsifier to stabilise
oil–​water mixtures (Imeson, 2010; Hoefler, 2002).
Ethylcellulose
Ethylcellulose (EC), like MC, HPMC and HPC, is a non-​ionic
cellulose derivative, soluble in a range of organic solvents but not
water. Since it is completely hydrophobic, in food applications it
is solvated in ethanol. It is resistant to changes in pH, with a par-
ticular resistance to alkalinity. EC is used particularly as a mois-
ture barrier, preventing moisture migration in foods prepared in
two stages, such as pizzas and snacks. It is also an excellent film
former (Imeson, 2010).
It has other applications in food; e.g., it stabilises flavours
and protects nutritional ingredients against active interactions,
Rachel Edwards-Stuart
hydrolysis and oxidation. It can also be used to slow down the
release of active ingredients by encapsulation.
The grades of EC available differ in molecular weight and
substitution; changes in molecular weight will affect properties
such as the film tensile strength, elongation and flexibility, while
changes in DS affect properties such as softening point and solu-
bility in ethanol (Imeson, 2010).
REFERENCES
Belitz H-​D, Grosch W, Schieberle P. 2004. Food Chemistry. Springer.
Coultate TP. 2009. Food: The Chemistry of Its Components. RSC
Publishing.
Cullen PJ, Brijesh KT, Valdramidis VP. (eds). 2012. Novel Thermal
and Non-​Thermal Technologies for Fluid Foods. Academic
Press.
Cummings JH. 1984. Cellulose and the human gut. Gut, 25(8),
805–​810.
Du B, Li J, Zhang H, Chen P, Huang L, Zhou J. 2007. The sta-
bilization mechanism of acidified milk drinks induced by
carboxymethylcellulose. Lait, 87, 287–​300.
Edwards-​Stuart R. 2009. Creating Innovative Flavour and Texture
Experiences. PhD Dissertation. http://​ eprints.nottingham.
ac.uk/​27644/​
Granstrom M. 2009. Cellulose derivatives: synthesis, properties and
applications. Helsingin yliopisto, Helsinki. https://​core.ac.uk/​
download/​pdf/​14916693.pdf
Hamad A, Ates S, Durmaz E. 2016. Evaluation of the possibilities
for cellulose derivatives in food products. Kastamonu Uni.,
Orman Fakultesi Dergisi, 16 (2), 383–​400.
Haque A, Morris ER. 1993. Thermogelation of methylcellulose.
1. Molecular-​structures & processes. Carbohydrate Polymers,
22(3), 161–​173.
Hoefler AC. 2002. The cellulose derivatives. Cellulose gum, methyl­
cellulose, methylhydroxypropylcellulose, and hydroxy­
propylcellulose. Chemistry, functionality and applications.
www.scribd.com/document/398430475/juimo6
Imeson A (ed.). 2010. Food stabilisers, Thickeners and Gelling
Agents. Blackwell Publishing Ltd.
McGee H. 2004. Food & Cooking. An Encyclopedia of Kitchen
Science, History and Culture. Hodder & Stoughton.
Mallik PK. 2010. Materials, Design and Manufacturing for
Lightweight Vehicles, Woodhead Publishing.
Medronho B, Romano A, Garça Miguel M, Stigsson L, Lindman B.
2012. Rationalizing cellulose (in)solubility: reviewing basic
physicochemical aspects and role of hydrophobic interactions.
Cellulose, 19, 581–​587.
Xu Y, Li L, LamYC, Hu X. 2004. Controllable gelation of
methylcellulose by a salt mixture. Langmuir, 20, 6134–​6138.
3D Printing of Food

Megan M. Ross1, Róisín M. Burke2 and Alan L. Kelly1

1 School of Food and Nutritional Sciences, University College Cork, Cork, T12 YN60, Ireland
2 School of Culinary Arts and Food Technology, Technological University Dublin, City Campus, Dublin, Ireland

Introduction and History have the potential to construct novel food forms and customise
Three-​dimensional (3D) printing is a process that creates 3D food textures, colours and flavours (Wegrzyn et al., 2012).
objects from a virtual model, usually designed through com- The broad objective of 3D food printing is to create universal
puter aided design (CAD) software. The principle of a 3D printer printers that will allow consumers to freely shape their food. This
is analogous to that of a regular office printer, to which has would offer the possibility to browse and download recipes from
been added a third dimension, giving the ability to print three-​ a mobile phone or any device and then order their 3D printer
dimensional objects from a range of materials. to deliver a proper meal. 3D food printing offers the possibility
Charles W. Hull of 3D Systems Corporation created the of customising food to suit one’s dietary needs or health condi-
first working 3D printer in 1984 (Hull, 1986). Since then, tion, but it also has the ability to change the shape and enhance
advancements in printing technology have widened the possibil- the visual aspect of food. Ingredients like seaweed or insects
ities in terms of 3D-​printed objects. A decade ago, 3D printers can be incorporated in recipes and made to look appealing
were an expensive hobby, costing about the same as a family car, (Molitch-​ H ou, 2014).
but in more recent times, a basic printer is roughly the same cost
as a low-​end laptop (Miller, 2016).
3D printing has rapidly become a disruptive technology in Possible End-​Use Scenarios
many industrial areas around the globe, as seen in Figure 91.1
3D food printing has been seen to be used in a variety of
(Morris, 2014). Automotive, building construction, electronic
applications, ranging from home-​cooking with families to pro-
manufacture, biomedical and aerospace are just a number of
fessional chefs at Michelin star restaurants (Table 91.1). This
sectors where 3D printing has been implemented in order to
section explores how 3D printing can benefit each of these key
increase automation and also decrease waste (Ngo et al., 2018).
food-​related areas and why 3D printing is set to become the tech-
However, in more recent times, 3D printing has progressively
nology of the future.
appeared in the realm of food manufacture. Confectionary, choc-
olate and even pizza can now be printed from a machine. 3D
food printing is beginning to emerge into society, and the idea
of having a 3D food printer at home as a method of food prepar- Food Applications
ation doesn’t seem to be so unrealistic. Since 1988, various rapid Currently, 3D food printing has been established in food areas as
prototyping techniques have emerged, which can be classified as diverse as the military, meals for the elderly, confectionery and
liquid-​based, solid-​based and powder-​based (Chua et al., 2010). food designed for space missions. The list of food materials that
can be utilised in such applications using the 3D food printing
process is growing at an incredible rate and includes ingredients
Principle of 3D Printing of Food such as chocolate, sugar, fruit and vegetables. In this section, rele-
vant applications for everyday food ingredients that can be 3D
3D food printing is the action of mechanically layering food
printed using commercially available printers will be discussed
material in such a way that a 3D shape is created. The process
in detail.
involves computer technology, which converts the blueprint
of the shape and structure required into coordinates where the
printer head must travel and deposit food material. A complex Chocolate
and unique 3D shape is created layer by layer, applying phase Chocolate extrusion has been by far the most common material
transitions (i.e., liquid to solid) or chemical or enzymatic reactions to be used in creating 3D food structures. The Choc Creator V1
to fuse layers together. 3D food printing technologies have the and V2 (Plus) became the first commercially available 3D choc-
potential to meet the growing demand for customised food. They olate printer in 2012 (Choc Edge, 2018). The Choc Creator is

605
606 Megan M. Ross et al.

FIGURE 91.1 Timeline of companies involved in the 3D food printing sphere.

(Adapted from Van der Linden, 2015)

TABLE 91.1
Examples of 3D Food Printing Applications in Different Use Scenarios
Scenario Examples of applications
Domestic kitchen • Enables creative, consistent designs to be created in bakery and/​or decoration applications
• Allows mass production for parties and large family events
• Reduces food wastage by combining commonly discarded, but nutritious, parts of fruits and vegetables (e.g., stalks, skins, etc.) into
purée to be used as sauce, topping, plate decoration, etc.
• Allows control over portion size
• Can encourage children to develop an interest in their own nutrition if allowed to create their own design
• Can facilitate personal, precise and consistent nutrition
Restaurant kitchen • Increases chef efficiency by delegating time-​consuming and repetitive tasks to the printer
• Produces consistent meals in terms of shape, portion size and precision placement
• Creates shapes and structures that may be difficult or impossible to do by hand
• Incorporates elements of 3D-​printed foods into traditional dishes for a modern twist
• Produces personalised cake toppers for special events
• Creates unique and personalised culinary experiences (e.g., printing customers’ name or custom designs on the plate or dessert)
Vending machines • Potentially offers healthier snacking option
• Offers convenience; order in advance online or through phone application and collect when completed
• Allows customised nutrition, flavours and texture to suit each individual user
• Can be placed in high-​flow areas such as airports, train stations, shopping centres, cinemas, universities/​colleges, work places,
fitness centres, etc., for healthy, personalised convenience options
Nursing homes and • Shearing effect of printing creates softer food material, which can be ingested more easily for individuals with swallowing
hospitals difficulties (i.e., dysphagia)
• Creates nutritious and attractive meals compared with traditional bland high-​calorie pastes typically found in nursing homes to
maintain weight
• Nursing home residents can look forward to the nutritious and tasty meals that they used to enjoy pre-​admittance. Encourages
healthy appetite, which helps in maintaining muscle mass and weight
• 3D-​printed meals in hospitals might benefit patients who, due to surgery or treatment, might have a small appetite, be consuming
strong medicines, and/​or require high-​calorie and nutrient-​dense foods
Food processors • Novel shapes can be created and changed as desired, with little or no changeover in equipment required
• Only one production line is required for multiple products; changes to product design or size can be made immediately on the same
production line, meaning that fewer lines are required, thereby saving factory floor space and capital cost on individual machines
for individual products
• Enables rapid in-​house prototyping with minimal cost
• Promotes production of mass customised products (i.e., personalised items have added value)
3D Printing of Food 607

a three-​axis Cartesian computer numerical control printer with Confectionery

a chocolate extruder head (Godoi et al., 2016). The innovation
In 2014, Print2Taste GmbH was established as a spin-​off com-
relies on an extrusion system capable of handling chocolate,
pany from the University of Weihenstephan-​Triesdorf in Freising
which requires accurate control of viscosity and temperature
(Germany). Procusini 3.0 was developed as an extrusion-​based
conditions. The printer is composed of three components: the
3D printer that would be suitable for professionals in catering,
first is a heat-​controlled chamber, and the second is a pump that
event gastronomy and hotels, as well as bakery and confectionery.
pumps the melted chocolate into the third component, which is
The company claims that the Procusini halves production time
the extrusion head and nozzle (Hao et al., 2010).
and can create 84 small structures in just 93 minutes, allowing
Recently, on 7 July 2019, Cadbury launched the world’s first
professionals to complete other tasks whilst printing is in oper-
Cadbury Dairy Milk 3D printer in Melbourne, Australia, for
ation. The Procusini includes a temperature-​controlled cartridge
World Chocolate Day. According to Kopp (2019), chocolate-​
holder to maintain the temperature of food materials contained
lovers were offered the chance to be among the first to take away
within. The printer can also operate through a Wi-​Fi connection
their very own personal selection of 3D-​printed chocolate pieces
from an electronic device such as a tablet or phone. Procusini is
built from layer upon layer of delicious Cadbury Dairy Milk
a mobile printer with dimensions of 60 × 60 × 65 cm and weighs
chocolate.
only 9 kg.
Food materials that have been printed using the Procusini
Pasta include pasta (four colours available), chocolate, marzipan
Extrusion-​ based processes have been employed by TNO (Figure 91.2), fondant (five colours), goat’s cheese, sausage meat,
(Netherlands Organisation for Applied Scientific Research) mashed potato and cassis (i.e., blackcurrant puree) (Procusini,
researchers to print a large variety of foods using sugars, proteins, 2018). In 2019, Procusini launced the Mycusini, which is a small,
meat purees and other nutrients extracted from alternative compact chocolate printer starting at €198; it will fit in any kit-
sources, such as algae and insects (Van der Linden, 2015). chen at 19 × 19.5 × 27 cm, which is described as being ‘smaller
In 2012, TNO researcher Kjeld van Bommel announced that, than most coffee machines’.
in partnership with Barilla (an Italian pasta company), they had In 2014, the American-​ based manufacturer 3D Systems
developed a 3D pasta printer that can print customised pasta developed Chefjet and Chefjet Pro, two NSF (National Sanitation
shapes and colour. Most recently, TNO and Barilla created 3D-​ Foundation) and UL (Underwriters Laboratory) certified kitchen-​
printed pasta in the shape of roses, using classical pasta recipes ready 3D food printers. The printer operates using liquid binding
(ingredients: durum wheat semolina and water, without additives) (LB) technology, which requires water and powder-​ based
(Van der Linden, 2015). ingredients. A roller-​
c omponent lays down a thin layer of powder
Massachusetts Institute of Technology (MIT) has brought (i.e., sugar) on the printing surface. A print head then sprays water
3D food printing another step further in terms of innovation by or liquid binder in a pattern that draws a shape on the powder sur-
introducing the concept of ‘4D printing’. Wang et al. (2017) face. The powder recrystallizes when in contact with the water
introduced this term through their experiments with pasta, where or binder, which encourages a solidifying and binding process
the initially flat-​shaped pasta will twist and curl up in a unique through the joining of adjacent particles. The piston supporting
pattern when hydrated and/​or heated. The 4D effect is caused by the print bed is progressively lowered at the completion of
the addition of carefully placed edible films made of common each layer until the object is completely formed. The process is
food materials such as protein, cellulose or starch, which contract repeated several times until the 3D structure is built (Figure 91.3)
when water and/​or heat is incorporated. by cross-​linking of the particle surfaces (Godoi et al., 2016).

FIGURE 91.2 Procusini Printer printing chocolate pieces and an example for a marzipan-​printed car cake topping.
(Procusini, 2018)
608 Megan M. Ross et al.

FIGURE 91.3 ChefJet and sugar sculpture made by ChefJet Pro.

(Brill 3D Culinary Studio –​Powered by 3D Systems, 2020)

swirling sugar sculpture (Figure 91.4) was 3D printed to balance

across the rim of a cocktail glass for the 50th birthday celebration
of renowned Spanish chef Ferran Adrià. The specially designed
perforated shape of the sculpture allowed the absinthe to flow
through the Gaudi-​style structure while absorbing the sweetness
of the sugar, creating a colourful, theatrical and sweet cocktail
experience.

Meals
Natural Machines, a Barcelona-​based start-​up company, launched
the Foodini food printer prototype in 2013. Foodini is a syringe-​
based deposition printer with a tactile user interface. The printer
(10 kg) is a 43.8 × 43 × 43 cm cube with a touchscreen display
powered by Android OS with Wi-​Fi connection. The printer has a
capsule receiver at the top where users can insert a cartridge with
fresh pureed ingredients (Natural Machines, 2018).
The process proposed by Foodini is quite simple; users have
to cook and prepare liquid-​ based food by mixing different
kinds of ingredients. Then, they load the liquid-​based mixture
into capsules (five maximum). Each capsule can hold 123 mL
FIGURE 91.4 Example of 3D-​printed sugar sculpture topping a custom-​
of liquid, and the five combined can hold 615 mL. Users insert
made cocktail. capsules in the Foodini and select printing programs through the
touchscreen. The printer will then start extruding food one cap-
(The Sugar Lab by 3D Systems, 2020)
sule at a time onto the print bed (Sevenson, 2014). Not all foods
can be printed; however, the Foodini allows the user to print intri-
The basic ChefJet printer can only print monochrome food cate designs that may not be possible to achieve by hand. This
decoration, whilst the ChefJet Pro prints in full colour and in a is accomplished through the use of a smaller nozzle diameter
multitude of flavours, including sour cherry, chocolate, vanilla, (0.5 mm) and computer-​controlled print head. Different nozzles
mint and watermelon (Ngo, 2015). The ChefJet printer can print are also available to accommodate textural differences between
edible sculptures and confections of all shapes and sizes which food types (Chadwick, 2017).
were previously impossible to create by hand. An example of The machine has successfully printed different kinds of food
these complex and creative designs can be found in the collab- products, such as crackers, individually shaped cereal, gnocchi,
oration between The Sugar Lab by 3D Systems and Modernist ravioli, bread sticks, butter, pizza, cereal bars and biscuits
Cuisine, whereby, using the Chefjet Pro, a colourful and intricate (Figure 91.5). The machine is equipped with a heating element
3D Printing of Food
FIGURE 91.5 Foodini printer and example of 3D-​printed biscuit spoons.
(Natural Machines, 2018)
FIGURE 91.6 Examples of Focus printing capabilities.
(byFlow, 2020; Food Ink, 2016)
with a maximum temperature of 100 °C. The heating function
can only keep food materials melted (like chocolate) or keep food
warm. In some cases, where meals require cooking, users will
have to cook the printed food in a separate process before eating
(Hoopes, 2013).
The Foodini contains built-​in 3D-​scanner hardware capable of
scanning an object, digitally saving the shape and then printing it.
It can also scan an object such as a plate or a cupcake and print on
top of this (Sevenson, 2014). The Wi-​Fi connection allows users
to share their recipes with the Foodini community. In addition,
the founders of Foodini also propose to produce pre-​prepared
capsules and make them available in supermarkets. These
capsules will contain all the ingredients required for a recipe pre-​
programmed by the printer (Natural Machines, 2018).
Netherlands-​based company byFlow has specialised in 3D
printing since 2009. The Portable ‘Focus’ 3D food printer,
609
created by byFlow, is currently sold as a business-​to-​business
(B2B) product, optimised for desserts, including chocolates and
meringues (Figure 91.6). byFlow claims that chefs and patissiers
use ‘Focus’ on a daily basis ‘to experiment with textures and
shapes, save time and money, create new designs, and amaze their
customers’ (Chadwick, 2017).
byFlow’s clients in the restaurants space so far include La
Boscana in Barcelona. This restaurant uses both Focus and
Natural Machines’ Foodini. The printer is not only based in the
kitchen but placed in front of the customer’s table so that they
can see their dishes coming to life. Focus printers also appeared
at Food Ink’s 3D printing project in 2016 (Figure 91.7), a pop-​up
restaurant in London. Currently, the Focus printer has been used
in printing applications such as marzipan, ganache, hummus,
fondant, mango caviar, avocado, meringue, pureed tomato and
mozzarella, and even pureed chicken and beef.
610
FIGURE 91.7 byFlow’s Focus Printer.
(byFlow, 2020; Food Ink, 2016)
Specialist Nutritional Foods
One of the unique applications of 3D-​printed food is producing
foods with a soft texture for elderly customers who have diffi-
culty swallowing (Deloitte, 2015). One in 25 adults is affected by
chewing and swallowing difficulties, otherwise called dysphagia
(Bhattacharyya, 2014).
The German company Biozoon launched the ‘3D Smoothfood
Project’ in 2014, which aims to design and create individualised
nutritious meals. The meals will have a jelly-​like texture that
resembles solid food when printed but dissolves easily when
ingested and swallowed by the consumer (Deloitte, 2015).
Biozoon originally specialised in a range of texturisers that
change the consistency of food. Their SeneoPro powder is
combined with pureed ingredients to form a smooth paste or
gel. The mixture is then inserted into a cartridge in the printer
and printed to resemble the solid shape of that ingredient
(Figure 91.8).
Mathias Kück, owner of Biozoon, claims that the look and
taste of the end product match the original food item: ‘When
eaten by a patient, the food can be destroyed without using the
teeth and flow like a gel through the throat’ (Chadwick, 2017).
Fruit
Dovetailed is a design studio and innovation lab founded in 2011
in Cambridge, UK. In 2014, Dovetailed created their first 3D food
printer, ‘nūfood’, which is controlled using an app on the user’s
phone, allowing the user to design their own unique flavoured
shape, which can be added to beverages, breakfasts and salads to
provide added bursts of flavour. According to Vaiva Kalnikaitè,
creative director and founder, the printing process is based on
the spherification technique, a well-​known molecular-​cooking
method (Figure 91.9).
Megan M. Ross et al.
This method involves combining liquids such as fruit purée
or juices with sodium alginate. The combination is then dropped
in a controlled fashion by the 3D printer into a bowl of cold
calcium-​based food-​grade liquid, where it forms tiny caviar-​like
spheres. Fruits and other shapes are formed by layering the tiny
spheres of juice until the desired shape is built. For example,
nūfood explored unique flavour pairings such as balsamic and
raspberries, which they combined in a 3D-​printed cube and added
to a whiskey sour cocktail.
Culinary Applications
Bakeries
The adoption of 3D printing is evolving rapidly in the baking
scene (Duchêne et al., 2016; Deloitte, 2018). For example, a large
international company, CSM Bakery Solutions, which produces a
broad range of products for customers in more than 100 countries
(CSM Bakery Solutions, 2018), has entered into partnership with
The Sugar Lab by 3D Systems to enable collaborative research
and development, engineering, design and printer development,
which will be focused on specific sourcing, food product devel-
opment and go-​to-​market plans (Unrein, 2017). Initially, CSM
Bakery Solutions will target how 3D printing may assist chefs
creating high culinary art in venues such as casinos and cruise
liners (CSM Bakery Solutions, 2018).
The artisanal field of confectionery that produces decorative
bakery products (Figure 91.10), chocolates and other types of
sweets is one of the food industry fields that can benefit from
the dimensions that 3D printing can offer. However, 3D-​printed
chocolates and other decorations should not be confused with
moulding, which has already been developed in the confectionery
sector (Duchêne et al., 2016).
3D Printing of Food 611

FIGURE 91.8 Examples of entire meals printed by Biozoon.

(Biozoon, 2018)

FIGURE 91.9 Dovetailed’s 3D printer nūfood –​printing a strawberry.

(nūfood, 2014)
612
Restaurants and Chefs Serving 3D-​Printed Foods
Many smaller R&D service-​focused companies are developing
innovative concepts, such as 3D-​ printed dinner at pop-​ up
restaurants. Additionally, pop-​up fine dining concepts have been
tested in Europe and the US (Duchêne et al., 2016). In 2016,
diners at the pop-​up restaurant Food Ink were served a nine-​
course meal in which all the food was fabricated using 3D
printing technology (Lupton, 2017). The dishes were made from
pizza dough, hummus, mushy peas, chocolate mousse and goat’s
cheese (Figure 91.11). Two chefs of elBulli and La Boscana, Joel
Castanye and Mateu Blanch, were behind the menu. The utensils
and even the chairs were 3D printed (Hartman, 2016).
FIGURE 91.10 A ChefJet Pro 3-​D printer from 3D Systems uses powdered
sugar and hydrated food colouring to create products like cake decorations.
(The Sugar Lab by 3D Systems, 2020)
FIGURE 91.11 3D-​printed meals served at the pop-​up restaurant Food Ink in London in 2016.
(Mendoza, 2016)
Megan M. Ross et al.
Figure 91.11 shows, on the left, ‘Caesar’s Flower of Life’
(seasoned bread, assorted flowers and vegetables) and, on the
right, ‘3D Boscana’ (nocilla, chocolate cream, hazelnut polvoron,
milk ice cream, Dutch chocolate leaves and 3D-​printed centre
spiral.
3D Printing and Fine Dining
One of the emerging regional areas for 3D printing of food
is Barcelona. Here, restaurants such as Dos Cielos and La
Boscana have been actively involved in experimenting (Duchêne
et al., 2016; Ahmed, 2017). At La Enoteca in the Hotel Arts in
Barcelona, chef Paco Perez, who has won several Michelin stars
for his restaurants, has created a new dish entitled ‘Sea Coral’,
which is a 3D-​ printed ‘coral’ shape made of seafood purée
(Figure 91.12). A report for the BBC website showed how Perez
and his co-​workers constructed this dish in one of his restaurants
(Koenig, 2016).
Also in Catalonia is Reimagine Food, a research and design
group who develop digital gastronomy ecosystems and robots in
cooking (Duchêne et al., 2016) and use disruptive technology,
such as drones, artificial intelligence, robotics, wearable devices
and big data, adapting them to the needs of consumers and the
food industry (Reimagine Food, 2014).
In 2018, chef Jan Smink opened the first full-​time 3D-​printed
food restaurant in the town of Wolvega in the Netherlands. Smink
previously worked for De Librije, a triple Michelin star-​winning
restaurant. Now, after creating new recipes (Figure 91.13) and
aiding in R&D, he is ready to provide a unique dining experience
that takes advantage of 3D printing technology. Chef Jan Smink
has stated: ‘By using the Focus 3D Printer I’m able to make forms
and shapes that would otherwise not be possible. I can surprise
3D Printing of Food
FIGURE 91.12 A 3D-​printed coral.
(Koenig, 2016; Natural Machines, 2018)
FIGURE 91.13 3D-​printed hazelnut, celery and cream cheese by Jan Smink.
(byFlow, 2020)
my guests with a unique experience that is very tasty as well. 3D
Printing is the future!’ (Smink, 2018).
In the United States, Melisse is a Michelin 2-​star restaurant in
Santa Monica, run by chef Josiah Citrin. The restaurant partnered
with 3D systems and has, for example, developed a dish where
the 3D printer crafted a fresh crouton using an aromatic onion
powder, as seen in Figure 91.14 (Ahmed, 2017).
Collaborative Projects and Applications
The 3DS Culinary Lab in Los Angeles brings together 3D food
printing technology developers with chefs and food industry
613
representatives (Lupton, 2017). One such example is the dish
created by Mei Lin, the winner of Bravo’s Top Chef Season 12,
which was based entirely on the flavours of Hawaii and inspired
by the beauty of a passion fruit flower (Duchêne et al., 2016;
Lupton, 2017; The Sugar Lab by 3D Systems, 2020). The dish
contained cylinders of passion fruit curd, caramelised banana
crème anglaise chilled to a gelato consistency by liquid nitrogen,
freeze-​dried strawberry powder, bee pollen crumble, toasted
yogurt, and sliced fresh bananas and strawberries (Figure 91.15a).
A delicate, perforated 3D-​printed interpretation of the passion
fruit flower, flavoured with actual passion fruit, crowned each
plating. Mei invited guests to shatter the delicate sugar passion
614
FIGURE 91.14 3D Systems partnered with Josiah Citrin of Melisse to create a 3D-​printed crouton.
(The Sugar Lab by 3D Systems, 2020)
fruit flower with their spoons (Figure 91.15b) to incorporate the
flavour and texture of the piece into each bite (The Sugar Lab by
3D Systems, 2020).
TNO has been involved in an interesting gastronomy pro-
ject with food designer Marijn Roovers and chef Wouter van
Laarhoven, who together created a chocolate shell just 0.8
millimetres thick using the Focus Printer. The chocolate’s con-
tinent of origin is embossed in gold, whilst inside, it holds delica-
cies that symbolise that region (Figure 91.16). Roovers mentions
that the chocolate globes take an hour to print and claims that
they have the texture of aerated chocolate bars as a result of
printing the globe in 200 layers of chocolate. The North America
globe contains segments of crème of sweetened corn and bourbon
whiskey. Meanwhile, the South America globe is filled with choc-
olate with allspice and popped corn. The African globe consists
of portions of ras el hanout with cumin and yoghurt (TNO, 2015).
In Silicon Valley, a team of over 30 people, including 3D
artists, designers, food scientists, chefs and engineers, created a
way for people to immerse themselves in a new reality in which
they enjoy different meals (The Project Nourished Initiative). All
participants need is a VR headset (for stimulating vision), an aro-
matic diffuser (for smell), 3D-​printed cubes (for texture), a bone
conduction transducer (for chewing), a gyroscopic utensil (fork
for the virtual and physical food), and a virtual cocktail glass (for
intoxication). The 3D-​printed cube is made of algae, which adds
taste and texture. The texture of the cube, the aromatic diffuser
and the sounds produced by the bone conduction transducer trick
the consumer’s mind into believing they are eating actual sushi
(The Sensorama, 2017).
In 2014, Dutch industrial design student Chloé Rutzerveld
presented her Edible Growth project in collaboration with
Eindhoven University of Technology and TNO. For this project,
she 3D printed an edible ecosystem made up of a carbohydrate
matrix containing miniature plants and mushrooms. At the start,
Chloé created the design for the carbohydrate structural matrix
on a 3D modelling program, printed it out and then printed a
combination of seeds, spores and yeast embedded in agar-​agar
Megan M. Ross et al.
into it. Agar-​agar is a gelatinous paste, which functioned as a soil
for the seeds and yeast to grow and feed on. After about five days,
the plants had grown sufficiently, and the product was ready to be
eaten (Figure 91.17). However, waiting a little longer gives the
plants some more time to grow, and the taste of the food intensi-
fies (Rutzerveld, 2014).
Applications of Molecular Gastronomy
In recent years, applications of molecular gastronomy such as
‘molecular cooking’ and ‘note-​by-​note cooking’ have emerged
(This, 2008, 2013, 2014). Molecular cooking is defined as produ-
cing food in kitchens using ‘new’ tools, ingredients and methods
(ttz-​Bremerhaven, 2018); examples of this could include use of
equipment such as siphons, ingredients such as sodium alginate,
and methods such as sous-​vide cooking. In the case of ‘note by
note cooking’, meat, fish, vegetables or fruits are not used to
make dishes, but instead compounds, either pure or in mixtures,
are assembled by the chef to design the shapes, colours, tastes,
odours, temperatures, trigeminal stimulation, textures, nutritional
aspects and more of the desired dish (This, 2013).
In many instances, the ingredients used in these molecular
gastronomy applications, e.g., hydrocolloids, including gelling
agents such as agar-​agar and gelatine, are suitable for 3D printing.
This will allow freedom of design in terms of not only compos-
ition, structure and texture but also taste (Duchêne et al., 2016).
In particular, it is possible to design and print customised note-
by-note foods. An example of a 3D-​printed note-by-note proto-
type recipe can be found in Part III of this book. It is suitable for
vegans and/​or those who are lactose-​intolerant.
Specialised Stores and Personalised Products
Specialised shops that offer 3D-​printed products include the
Amsterdam-​based MELT Ice pops, which has offered custom-​
made ice pops since 2012 (Meltpops, 2018) and uses 3D-​printed
moulds that are printed using domestic 3D printer manufac-
turer Ultimaker’s printers (Duchêne et al., 2016). The Magic
3D Printing of Food
FIGURE 91.15 The Flavours of Hawaii.
(The Sugar Lab by 3D Systems, 2020)
Candy Factory also specialises in personalised products and is
the world’s first 3D gummy candy printer, allowing anyone to
create shapes, write messages and draw their own custom candies
(Magic Candy Factory, 2017).
Limitations
Major challenges still to be solved relate to the need for multi-​
material printing systems and integration with traditional cooking
processes, like baking or boiling. One of the major hindrances is
the low printing speed (Duchêne et al., 2016)
The cost of home food-​printing machines has also been noted
in some media reports as a potential barrier to consumer interest,
as was the problem that these devices were still in development,
so that people who might be interested in purchasing them would
have to wait until they came onto the market (Lupton, 2017).
The expensive nature of the gourmet printed food featured in
restaurants has also received attention, as in the following head-
line in an online newspaper article about the Food Ink pop-​up
restaurant: ‘Is this the future of fine dining? Restaurant where all
615
the food and even the table is 3D printed –​but it will cost you
£250 a head’ (Best, 2016).
Some media reports also quoted experts in the food industry as
voicing some reservations about whether 3D-​printed food would
be widely accepted by consumers. This was usually in response
to the more speculative uses of the technology, as in the virtual
reality meals served through the Project Nourished initiative or
the use of cultured meat (Lupton, 2017).
Future Speculations
The global 3D printing food market is expected to expand
during the period 2017–​2024 and to reach USD 400 million
by 2024. The market growth is driven by factors such as the
growing demand for customised food (Researchnester, 2018).
Richard Watson, futurist, writer and founder, predicted in The
Telegraph’s feature on ‘back to the future’ that ‘By 2045 many
kitchens will feature a 3D Printer … a fun kitchen gadget to
sit alongside the Soda stream and waffle maker’ (Titcomb and
Murgia, 2015).
616
FIGURE 91.16 TNO Michelin star chocolate dessert.
(TNO, 2015)
FIGURE 91.17 Chloé Rutzerveld’s award-​winning Edible Growth Project.
(Rutzerveld, 2014)
Companies like Natural Machines see the future of 3D printing
food as a faster and more precise operation, possibly even
including more textures. A US startup called Modern Meadow
is working on a technique to 3D print meat without having to
slaughter an animal at all. The process includes using stem cells
Megan M. Ross et al.
to create what they call ‘bio-​ink’, which is then inserted into a
nozzle similar to that of a 2D inkjet printer. The ‘live’ bio-​ink is
then extruded into an agarose gel mould. After that, it matures in
a ‘bioreactor’, resulting in organ tissue. The agarose gel is then
removed, leaving the end product (Houser, 2017).
Production capacity can be increased through the use of a well-​
managed print farm or print cell of multiple 3D printers, allowing
simultaneous production in multiple materials (Formlabs, 2017).
Conclusions
In recent times, the users of 3D-​printed foods have become many
and varied, ranging from domestic to professional kitchens to
retailers and large food manufacturers. The use of this technology
has opened up an array of applications allowing personalised
food design and nutritional meal formulation. In the last few
years, more and more companies have been developing 3D
food printers that have specific design features, e.g., for printing
doughs or even complete meals. As a result, it has become pos-
sible to diversify ingredients beyond those used by the pioneering
confectioners. A number of collaborative 3D food printing
projects are currently underway, involving multi-​ disciplinary
teams (e.g., chefs, scientists, engineers and artists), who aim to
3D Printing of Food 617

further improve our eating experiences. While there are some Hoopes H. 2013. Foodini 3D food printer customizes and automates
limitations to 3D printing technology, the pace of developments, your kitchen prep, New Atlas. https://​newatlas.com/​food-​prep-​
together with the economic forecast for 3D food printing, signals 3d-​printer-​foodini/​30079/​
Houser F. 2017. 3D Printed Food: A Culinary Guide to 3D Printing
a positive future ahead.
Food, All 3D P. https://​all3dp.com/​1/​3d-​printed-food-3d-
printing-​food/​
REFERENCES Hull CS. 1986. Apparatus for production of three-​ dimensional
Ahmed R. 2017. 4 Famous Restaurants that Use 3D Printers, objects by stereolithography, US 4575330 A.
3D Printing.com. https://​3dprinting.com/​food/​4-​famous- Koenig N. 2016. How 3D printing is shaking up high end dining.
restaurants-that-​use-​3d-​printers/​ Available at: www.bbc.com/​news/​business-​35631265
Best S. 2016. Is this the future of fine dining? Restaurant where all Kopp C. 2019. Cadbury Launches World’s First Cadbury Dairy Milk
the food and even the table is 3D printed –​but it will cost you 3D Printer for World Chocolate Day. Available at: www.news.
£250 a head, DailyMail. www.dailymail.co.uk/​sciencetech/​ com.au/​technology/​innovation/​inventions/​cadbury-​launches-​
article-​3685998/​Is-​future-​fine-​dining-​Restaurant-​food-​table-​ worlds-​ f irst-​ c adbury-​ d airy-​ m ilk-​ 3 d-​ p rinter-​ f or-​ w orld-​
3D-​printed-​cost-​250-​head.html chocolate-​day/​news-​story/​90e9ee3a32e65218c9b3e8f692c6
Bhattacharyya N. 2014. The prevalence of dysphagia among adults bfa2
in the United States. Otolaryngology–​Head and Neck Surgery, Lupton D. 2017. ‘Download to Delicious’: Promissory Themes
151:765–​769. and Sociotechnical Imaginaries in Coverage of 3D Printed
Biozoon. 2018. SeneoPro, Biozoon. http://​biozoon.de/​en/​products/​ Food in Online News Sources. https://​simplysociology.files.
seneopro/​(accessed 2 May 2020). wordpress.com/ ​ 2 017/ ​ 0 8/ ​ a rticle- ​ d ownload- ​ t o- ​ d elicious-​
Brill 3D Culinary Studio –​Powered by 3D Systems. 2020. www. postprint.pdf
the-​sugar-​lab.com/​ Magic Candy Factory. 2017. The Magic Candy Factory Family!
byFlow. 2020. www.3dbyflow.com/​ http://​magiccandyfactory.com/​our-​vision/​
Chadwick J. 2017. Here’s how 3D food printers are changing Meltpops. 2018. Custom 3D Icepops. Melt Ice-​ Pops. http://​
what we eat, Tech Republic. www.techrepublic.com/​article/​ melticepops.com/​
heres-​how-​3d-​food-​printers-​are-​changing-​the-​way-​we-​cook/​ Mendoza HR. 2016. Everything but the People is 3D Printed at
Choc Edge. 2018. Choc Edge Limited -​3D ALM Chocolate Printing. Pop-​Up Restaurant Food Ink, 3D Print. https://​3dprint.com/​
http://​chocedge.com/​ 141563/​3d-​printing-​popup-​food-​ink/​
Chua CK, Leong KF, Lim CS. 2010. Rapid Prototyping: Principles Miller A. 2016. The Evolution of 3D printing: Past, present and
and Applications, World Scientific Publishing. future, 3D Printing Industry. https://​3dprintingindustry.com/​
CSM Bakery Solutions. 2018. CSM Bakery Solutions. www. news/​evolution-​3Dprinting-​past-​present-​future-​90605/​
csmbakerysolutions.com/​en/​about-​us/​our-​company.html Morris JA. 2014. A Taste of the Future, Consumer Market Report
Deloitte. 2015. 3D opportunity serves it up: Additive manufac- (June 2014).
turing and food, Options for deploying additive manufac- Molitch-​Hou M. 2014. Print pasta fazul right on your plate, 3D
turing. www2.deloitte.com/​content/​dam/​insights/​us/​articles/​ Printing Industry. https://​3dprintingindustry.com/​news/​print-
3d-​printing-​in-​the-​food-​industry/​DUP_​1147-​3D-​opportunity-​ pasta-fazul-​right-​plate-​21968/​
food_​MASTER1.pdf Natural Machines. 2018. Frequently asked questions. www.
Deloitte. 2018. 3D Printed Food –​Just Because We Can, Doesn’t naturalmachines.com/​
Always Mean We Should. www.forbes.com/​ sites/​ deloitte/​ Ngo D. 2015. 3D Systems unveils ChefJet 3D printers for those with
2018/ ​ 0 5/ ​ 2 9/ ​ 3 d- ​ p rinted- ​ f ood- ​ j ust- ​ b ecause- ​ w e- ​ c an- ​ d oesnt-​ a sweet tooth, CNET. www.cnet.com/​ products/​ 3d-​ systems-​
always-​mean-​we-​should/​#3a7331962e93 chefjet-​3d-​printer/​preview/​
Duchêne V, Padilla P, Van de Velde E, Wastyn A, Nuñez L, Knotter Ngo T, Kashani A, Imbalzano G, Nguyen K, Hui D. 2018. Additive
S, Magistrelli G, Nieminen M, Puukko P, Eklund P, Rilla N, manufacturing (3D printing): A review of materials, methods,
Deschyvere M, Mäntylä M, Kasztler A, Leitner K, Schiebel E, applications and challenges. Composites Part B: Engineering,
Wepne B. 2016. Final Report: Identifying current and future 143, 172–​196.
application areas, existing industrial value chains and missing Procusini. 2018. Passion in Food Creation. www.procusini.com/​
competences in the EU, in the area of additive manufacturing Reimagine Food. 2014. Reimagine Food. Reimagine Food to serve
(3D-​printing). http://​ec.europa.eu/​DocsRoom/​documents/​ diners 3D-​printed meals. newatlas.com
18741/​attachments/​1/​translations/​en/​renditions/​native Researchnester. 2018. 3D Food Printing Market: Global Demand
Food Ink. 2016. Food Ink at Villa Flora, Venlo, The Netherlands. Analysis & Opportunity Outlook 2024, Researchnester. www.
http://​foodink.io/​venlo/​ researchnester.com/​reports/​3d-​food-​printing-​market-​global-​
Formlabs (2017). How a 3D Printer Farm Can Help Your Business. demand-​analysis-​opportunity-​outlook-​2024/​272
https://​formlabs.com/​blog/​3d-​print-​farm-​business-​benefits/​ Rutzerveld C. 2014. Edible Growth –​The use of additive manu-
Godoi FC, Prakash S, Bhandari BR. 2016. 3D printing technologies facturing technologies to create an edible ecosystem. www.
applied for food design: Status and prospects. Journal of Food chloerutzerveld.com/​edible-​growth
Engineering, 179, 44–​54. Sevenson B. 2014. Foodini 3D Printer Seeks $100,000 on
Hao L, Mellor S, Seaman O, Henderson J, Sewell N, Sloan M. Kickstarter, 3D Print. https://​3dprint.com/​1803/​foodini-​3d-
2010. Material characterisation and process development for printer-​seeks-​100000-​on-​kickstarter/​
chocolate additive layer manufacturing. Virtual and Physical Smink, J. 2018. 3D Food Printing. www.jansmink.com/​ 3d-​food-
Prototyping, 5, 57–​64. printing/​
Hartman S. 2016. The world’s first 3D-​printing restaurant is coming to The Sensorama. 2017. Project Nourished: Eating Food Without
London, Timeout. www.timeout.com/​london/​blog/​the-​worlds-​ Consequences. Project Nourished. foodinspiration.com
first-​3d-​printing-​restaurant-​is-​coming-​to-​london-​071416 The Sugar Lab by 3D Systems. 2020. www.the-​sugar-​lab.com/​
618
This H. 2008. Molecular gastronomy vs molecular cooking, Food
Technology, 62, 108.
This H. 2013. Molecular gastronomy is a scientific discipline and
note-​by-​note cuisine is the next culinary trend. Flavour, 2, 1.
This H. 2014. Note by Note Cooking: The Future of Food (MB
DeBevoise, Trans.). Columbia University Press, New York.
Titcomb J, Murgia M. 2015. Back to the Future Day: Six experts
predict life in 2045, The Telegraph. www.telegraph.co.uk/​
technology/​news/​11943575/​Back-​to-​the-​Future-​Day-​Five-​
experts-​predict-​life-​in-​2045.html
TNO. 2015. 3D Food Printing. www.tno.nl/​ media/​6574/​
3d-​food-​
printing.pdf
ttz-​Bremerhaven. 2018. Introduction of Innovative Technologies in
Modern Gastronomy for Modernisation of Cooking. www.
ttz-​bremerhaven.de/​en/​research/​food/​research-​projects/​893-​
inicon.html
Megan M. Ross et al.
Unrein J. 2017. 3D Printing Comes to the Baking Industry. www.
bakemag.com/​Trends/​Bread-​Winners/​2017/​9/​3D-​Printing-​
Comes-​to-​the-​Baking-​Industry.aspx?cck=1
Van der Linden D. 2015. 3D Food Printing Creating Shapes and
Textures. www.tno.nl/​media/​5517/​3d_​food_​printing_​march_​
2015.pdf
Wang W, Yao L, Cheng C, Zhang T, Levine D, Ishii H. 2017.
Transformative Appetite, Tangible Media Group. https://​tan-
gible.media.mit.edu/​project/​transformative-​appetite/​
Wegrzyn TF, Golding M, Archer RH. 2012. Food Layered
Manufacture: A new process for constructing solid foods.
Trends in Food Science & Technology, 267, 66–​72.
Umami: The Molecular Science of Umami Synergy

Ole G. Mouritsen
Department of Food Science, Taste for Life, Design and Consumer Behavior, University of Copenhagen,
26 Rolighedsvej, DK-​1958 Frederiksberg C, Denmark

Umami is a basic taste that is elicited by the stimulation of umami Molecular Mechanism of Umami Synergy
receptors in the taste buds by the interaction with some free
Umami was proposed as a basic taste based on an analysis of the
amino acids, in particular glutamate, often in synergy with free
components in the classical Japanese soup stock, dashi, which
5′-​ribonucleotides such as inosinate, adenylate, and guanylate.
is made of an aqueous extract of a dried brown seaweed, konbu
Deliciousness can be obtained by pairing of appropriate foods
(Saccharina japonica) (Ikeda, 2002; Japanese Culinary Academy,
with free glutamate and nucleotides as is well known, e.g., in the
2016), and a highly processed fish product, katsuobushi.
Japanese kitchen when pairing konbu with katsuobushi (dashi)
Vegetarians and vegans would use dried shiitake mushrooms
and in the Western cuisine when pairing eggs with bacon, cheese
instead of katsuobushi. Ikeda discovered that konbu is very rich in
with ham, or tomatoes with meat.
free glutamate, and he ascribed the delicious (umai) taste of dashi
Although the concept is also becoming widely recognized in
to glutamate, as also described in detail in another article in the
the Western world, it is difficult for many people, even chefs,
present volume (see chapter on seaweed). Later, Kodama (1913)
to come to grips with a fifth taste, umami, as a basic taste sen-
discovered that katsuobushi contains significant amounts of the
sation. Part of the difficulty derives from the fact that the term
free 5′-​ribonucleotide inosinate (IMP), and in 1957, Kuninaka
‘umami’ is not yet firmly encoded in the Western languages
(1960) demonstrated that dried shiitake mushrooms contain high
(O’Mahony and Ishii, 1986) in the same way as sweet, bitter,
amounts of another free nucleotide, guanylate (GMP). However,
sour, and salty are. As already pointed out in the pioneering
the most important observation made from Kuninaka’s work
work by Ikeda (2002), a main reason behind this is that this taste
was that it is the combination of free glutamate and free
quality is often hidden under the other basic tastes, and that it
nucleotides that leads to maximal umami sensation (Mouritsen
has a time course in the mouth, which often overlaps with that of
and Styrbæk, 2014). This finding is at the very core of the umami
mouthfeel, mouthfullness, kokumi, complexity, etc. (Nishimura
synergy, which was quantitatively investigated by sensory ana-
and Kuroda, 2019). Another reason is perhaps that most world
lysis in a pioneering work by Yamaguchi (1967), whose dis-
cuisines, apart from the classical Japanese cuisine (which has
covery is illustrated in Figure 92.1.
the soup stock dashi), do not have a unique, single ingredient in
Umami as a basic taste was not immediately accepted, and
the kitchen that is very pure in umami, except for MSG (mono-
it took almost 100 years after Ikeda’s discovery before it was
sodium glutamate).
commonly accepted among scientists. It happened only after the
Foods that are associated with strong umami taste are as
identification of certain umami taste receptors (Chaudhari et al.,
common and diverse as matured and fermented cheeses (particu-
2000; Kunishima et al., 2000; Li et al., 2002; Nelson et al., 2002;
larly Parmesan cheese), cured ham, sardine paste, fish sauce, soy
San Gabriel et al., 2005). It is interesting to note that Ikeda’s
sauce, sun ripe tomatoes, mushrooms, ketchup, etc. (Maga, 1983;
landmark paper on umami from 1909 was translated into English
Ninomiya, 1998; Mouritsen and Styrbæk, 2014). A particular
only in 2002 (Ikeda, 2002), i.e., after the discovery of the umami
feature of the umami taste sensation is its mouthfullness and
receptors. The molecular mechanisms behind the umami synergy
lingering, which is long compared with the short time duration of
have been worked out in the case of the most prominent umami
basic tastes like salty, sour, and sweet. However, the most pecu-
receptor, T1R1/​T1R3, which is a so-​called G-​protein coupled
liar and almost magic aspect of umami is that it can be enhanced
heterodimer receptor embedded in the taste cell membranes
manifold by particular combinations of food ingredients that
(Zhang et al., 2008; Mouritsen and Khandelia, 2012).
contain some well-​defined molecular substances (Schmidt et al.,
The power of umami synergy becomes apparent from
2020). In contrast to other food-​pairing theories (Ahn et al., 2011),
Figure 92.2, which demonstrates that it is the joined event of sim-
there is actually some molecular science behind umami synergy.
ultaneous binding to the receptor of a free glutamate ion and a

619
620 Ole G. Mouritsen

0
0 10 20 30 40 50 60 70 80 90 100%
Pure glutamate (MSG) Mixture Pure inosinate (IMP)

FIGURE 92.1 The synergy principle in umami sensation, where the combination of free glutamate and free inosinate leads to a highly non-​linear enhance-
ment of the perceived taste intensity on a scale 0–​10.
(From Yamaguchi, 1967)

Nucleotide

Glutamate

FIGURE 92.2 The part of the umami receptor (T1R1/​T1R3) that can bind free glutamate and free nucleotides, here seen in three stages from the left:
unbound, binding of glutamate, and simultaneous binding of glutamate and the free nucleotide guanylate. The top panel shows a molecular representation,
while the bottom panel shows a corresponding Pacman analogy.
(Courtesy of JD Mouritsen)

free nucleotide ion that leads to a so-​called allosteric enhance-

ment of the binding and hence, an enhancement of the electrical TABLE 92.1
signal to the taste centre in the brain. It is the same fundamental Pairing of Ingredients Providing Umami Synergy
principle that is at work when pairing eggs with bacon, cheese
Ingredients with free glutamate Ingredients with free nucleotides
with ham, and konbu seaweed with the fish product katsuobushi
in a classical Japanese dashi (Table 92.1). Konbu Katsuobushi/​Niboshi Shiitake
Eggs Bacon
Cheese Ham
Tomatoes Meat
Umami in the Kitchen Green peas Scallops
Once we have established the principle of umami synergy, it is Champagne Oysters
possible to both rationalize as well as design particular delicious
Umami: Molecular Science of Umami Synergy 621

25
Basal umami: glutamate 23 26

21 24
17
19 22
9 11 13
15 18 20
16
1 2 3 4 5 6 7 8 10
12 14
14
12
13
4 11
2 6 8
9 10
1 3
5 7
Synergistic umami: nucleotides (IMP+GMP+AMP)

FIGURE 92.3 Selection of raw ingredients and processed food products that contain umami taste substances, ranging from very small quantities (on the left)
to an abundance (on the right). The products in the top panel have basal umami (from glutamate), while those on the bottom have synergistic umami (from the
nucleotides inosinate (IMP), guanylate (GMP), and adenylate (AMP)). The horizontal axes are not linear, and the position of a given product on the axis does
not correspond to its absolute content of umami substances. However, the individual products on each axis are placed in the correct relationship to each other.
Basal umami: 1: cow’s milk, 2: apple, 3: carrots, 4: egg, 5: pork, 6: Worcestershire sauce, 7: mackerel, 8: chicken, 9: green asparagus, 10: caviar, 11: green peas,
12: oysters, 13: potatoes, 14: ketchup, 15: air-​dried ham, 16: miso paste, 17: sun-​dried tomatoes, 18: walnuts, 19: soy sauce, 20: dried shiitake mushrooms, 21:
anchovies in brine, 22: blue cheese, 23: Parmesan cheese, 24: fish sauce, 25: Marmite, 26: dried seaweeds (konbu). Synergistic umami: 1: green asparagus, 2:
oyster mushrooms, 3: sun-​ripened tomatoes, 4: crab, 5: beef, 6: lobster, 7: dried shiitake mushrooms, 8: scallop, 9: shrimp, 10: pork, 11: chicken, 12: mackerel,
13: anchovy paste, 14: katsuobushi.
(Adapted from Mouritsen and Styrbæk, 2014)

combinations of foodstuffs and prepared dishes in terms of

umami optimization. Figure 92.3 shows a series of foodstuffs, raw
ingredients, and prepared food, listed in the order of increasing
amounts of free glutamate and free nucleotides. To optimize
umami, one would need to combine items from both series. This
approach is true molecular cooking based on a well-​established
scientific and molecular principle. As an example, Figure 92.4
shows a dish with high levels of umami because it combines free
glutamate (from green peas) and the free nucleotide adenylate
(AMP) from scallops. Adding a bit of seaweed may impart add-
itional glutamate.
Because the umami synergy is highly non-​linear (Figure 92.3), FIGURE 92.4 Green pea soup with scallops and seaweed.
it is possible to enhance umami in a dish by adding very little
of a specific ingredient. Well-​known examples include adding a Styrbæk, 2014). Poultry meat, particularly from chicken, duck,
drop of soy sauce or fish sauce, a bit of an anchovy or Parmesan and turkey, has a high concentration of glutamate, and the same
cheese, or some tomato paste to a gravy. Many types of (fast) food is true for game and veal. Pork and chicken are rich in inosinate,
also use umami synergy, e.g., pasta dishes, pizza, and burgers. while beef has a reasonable amount. As lamb is a poor source of
A flavourful fish stock can be made using only fish and shell- umami, it benefits from being prepared in duck stock, which has
fish, as they contain both the glutamate and the nucleotides needed a great deal of it. The amount of umami in meat is proportional
to impart umami. The combination of shellfish with vegetables to how long it has been aged: the longer, the more umami. That
that have a good quantity of glutamate yields a particularly robust is why beef is much tastier after it has been aged than when it is
stock with intense umami. A good example is the green pea soup freshly slaughtered. It is also true that older animals are better
with scallops shown in Figure 93.4. sources of umami than younger ones. Furthermore, cooking,
Fresh meat is a major source of umami, as it contains both ageing, curing, drying, salting, smoking, and fermenting,
glutamate and the nucleotides that create synergy (Mouritsen and either alone or in combination, can in some cases appreciably
622
intensify the umami taste in a meat product. Fermenting is the
most powerful method to induce umami (Maga, 1983), whereas
cooking, e.g., by sous vide, does not necessarily produce more
free glutamate in meat (Clausen et al., 2018).
Of all the vegetables, tomatoes have the greatest content of
free glutamate, and it is enhanced when they are prepared in con-
junction with other ingredients that contribute synergistic umami,
for example, fish and shellfish (Mouritsen and Styrbæk, 2014).
This is undoubtedly the reason why tomatoes are so popular
and are used in a wide range of dishes in kitchens all around the
world. In fact, tomatoes also have a free nucleotide of their own,
adenylate, rendering this popular food ingredient one of the few
that can contribute umami synergy on its own. Only a few kinds
of raw foodstuff contain appreciable amounts of both free gluta-
mate and nucleotides. The tomato is one; nori from the red sea-
weed Pyropia spp. is another (Ninomiya, 2002), which explains
why nori is so important in sushi based on cooked white rice that
has no umami of its own.
Research has shown that levels of glutamate and adenylate are
three times as high in the pulp of the tomato compared with the
flesh (Oruna-​Concha et al., 2007). Powerful umami synergy with
tomatoes arises when combined with mackerel or beef, which are
both high in inosinate. This is the molecular science behind an
Italian sauce bolognese, combining tomatoes and minced meat,
which does wonders for a simple pasta dish.
Purely vegetarian soups can derive glutamate from a variety
of sources, such as cooked potatoes, green peas, soybeans, corn,
green asparagus, cauliflower, mushrooms, and of course, toma-
toes. Fungi, especially dried shiitake mushrooms, have good
quantities of the nucleotides that can interact synergistically, but
there are very few vegetables that can play this role.
Umami and Healthy Eating
Apart from the umami synergy involving free glutamate and free
nucleotides, there are other powerful interactions between umami
and the other basic tastes (Mouritsen and Styrbæk, 2014). The
molecular mechanisms behind these interactions are not well
understood. However, it is known that by administering umami,
it is possible to cut down on salt (sodium chloride) by up to 50%
without compromising the perception of salty. Umami appears
also to be capable of enhancing sweetness without adding sugar
(sucrose), e.g., in a dessert (Mouritsen and Styrbæk, 2014). In
addition, umami is capable of cutting the top of bitterness, e.g., of
vegetable dishes (Mouritsen, 2018). Umami and umami synergy
also contribute to the koku attribute by enhancing continuity and
mouthfullness of the overall taste experience (Mouritsen, 2019).
If one adds to these phenomenological findings that by using
the principle of umami and dashi it is possible to produce deli-
cious meals without adding fats and oils to, e.g., dressings, there
are reasons to believe that a more informed use of the principle
behind umami and umami synergy may not only lead to delicious
food but may also be a means to improve human health and regu-
late food intake (Mouritsen, 2016).
Ole G. Mouritsen
Acknowledgements
This work was supported in part by a grant to Smag for Livet
(Taste for Life) from Nordea-​fonden.
REFERENCES
Ahn YY. Ahnert SE, Bagrow JP, Barabási AL. 2011. Flavor net-
work and the principles of food pairing. Scientific Reports,
1,196.
Chaudhari N, Landin AM, Roper SD. 2000. A novel metabotropic
glutamate receptor functions as a taste receptor. Nature
Neuroscience, 3, 113–​119.
Clausen MP, Christensen M, Djurhuus TH, Duelund L, Mouritsen
OG. 2018. The quest for umami: can sous vide contribute?
International Journal of Gastronomy and Food Science, 13,
129–​133.
Ikeda I. 2002. New seasonings. Chemical Senses, 27, 847–​ 849.
[Translation from the original article in Journal of the Chemical
Society of Japan, 30, 820–​836 (1909)].
Japanese Culinary Academy. 2016. Flavour and Seasonings: Dashi,
Umami, and fermented Foods. Shuhari Initiative Ltd, Tokyo.
Kodama S. 1913. On a procedure for separating inosinic acid.
Journal of Tokyo Chemical Society, 34, 751–​757.
Kuninaka A. 1960. Studies on taste of ribonucleic acid derivatives.
Journal of Agricultural and Chemical Society Japan, 34,
487–​492.
Kunishima N, Shimada Y, Tsuji Y. 2000. Structural basis of glutamate
recognition by a dimeric metabotrobic glutamate receptor.
Nature, 407, 971–​977.
Li X, Staszewski L, Xu H, Durick K, Zoller M, Adler E. 2002.
Human receptors for sweet and umami taste. Proceedings
of the National Academy of Science of the United States of
America, 99, 4692–​4696.
Maga JA. 1983. Flavor potentiators. CRC Critical Review in Food
Science and Nutrition, 18, 231–​312.
Mouritsen OG. 2016. Deliciousness of food and a proper balance in
fatty-​acid composition as means to improve human health and
regulate food intake. Flavour, 5, 1.
Mouritsen OG. 2018. Tsukemono –​crunchy pickled foods from
Japan: a case study of food design by gastrophysics and nature.
International Journal of Food Design, 3, 103–​124.
Mouritsen OG. 2019. The quest for umami. In Nishimura T, Kuroda
M (eds.) Koku Attribute; Food Science and Biology. Springer,
Singapore, 33–​45.
Mouritsen OG, Khandelia H. 2012. Molecular mechanism of the
allosteric enhancement of the umami taste sensation. FEBS
Journal, 279, 3112–​3120.
Mouritsen OG, Styrbæk K. 2014. Umami: Unlocking the Secrets of
the Fifth Taste. Columbia University Press, New York.
Nelson G, Chandrashekar J, Moon MA, Feng L, Zhao G, Ryba NJ,
Zucker CS. 2002. An amino acid taste receptor. Nature, 416,
199–​202.
Ninomiya K. 1998. Natural occurrence. Food Review International,
14, 177–​211.
Ninomiya K. 2002. Umami: a universal taste. Food Review
International, 18, 23–​38.
Nishimura T, Kuroda M (eds.). 2019. Koku in Food Science and
Physiology. Springer, Singapore.
O’Mahony M, Ishii R. 1986. A comparison of English and Japanese
taste languages: taste descriptive methodology, codability and
the umami taste. British Journal of Psychology, 77, 161–​174.
Umami: Molecular Science of Umami Synergy
Oruna-​Concha MJ, Methven L, Blumenthal H, Young C, Mottram
DS. 2007. Differences in glutamic acid and 5′-​ribonucleotide
contents between flesh and pulp of tomatoes and the rela-
tionship with umami taste. Journal of Agricultural and Food
Chemistry, 55, 5776–​5780.
San Gabriel A, Uneyama H, Yoshie Y, Torii K. 2005. Cloning and
characterization of a novel mGluR1 variant from vallate
papillae that functions as a receptor for L-​glutamate stimuli.
Chemical Senses, 30, i25–​i26.
623
Schmidt CV, Olsen K, Mouritsen OG. 2020. Umami synergy as
the scientific principle behind taste-pairing champagne and
oysters. Nature Scientific Reports, 10, 20077.
Yamaguchi S. 1967. The synergistic taste effect of monosodium glu-
tamate and disodium 5′-​inosinate. Journal of Food Science, 32,
473–​478.
Zhang FB, Klebansky B, Fine RM, Xu H, Pronin A, Liu H, Tachdjian
C, Li X. 2008. Molecular mechanism for the umami taste syn-
ergism. Proceedings of the National Academy of Sciences of
the United States of America, 105, 20930–​20934.
 Part II​

Educational Practices
The Right Words for Improving Communication in Food Science,
Food Technology, and between Food Science and Technology
and a Broader Audience
Hervé This vo Kientza1,2
1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
Attending scientific lectures in food science and technology or
reading scientific and technological articles reveals contradictions
between scientists, sometimes using different words for the
same object and even changing words in their own talks or art-
icles; such contradictions can even be observed within the same
research institution or within the same team. Here, we analyse
some examples that are frequent in our discipline, promoting the
idea that our community would benefit from a more consistent
use of some words.
Correct wording is so important that Antoine Laurent de
Lavoisier (1789) discussed the issue in the introduction to his
Elementary Treaty of Chemistry:
When I began the following Work, my only object was to
extend and explain more fully the Memoir which I read at the
public meeting of the Academy of Sciences in the month of
April 1787, on the necessity of reforming and completing the
Nomenclature of Chemistry. While engaged in this employ-
ment, I perceived, better than I had ever done before, the
justice of the following maxims of the Abbé de Condillac, in
his System of Logic, and some other of his works.
We think only through the medium of words. –​Languages
are true analytical methods. –​Algebra, which is adapted to its
purpose in every species of expression, in the most simple,
most exact, and best manner possible, is at the same time a
language and an analytical method. –​The art of reasoning is
nothing more than a language well arranged.
Thus, while I thought myself employed only in forming a
Nomenclature, and while I proposed to myself nothing more
than to improve the chemical language, my work transformed
itself by degrees, without my being able to prevent it, into a
treatise upon the Elements of Chemistry.
The impossibility of separating the nomenclature of a
science from the science itself, is owing to this, that every
branch of physical science must consist of three things; the
series of facts which are the objects of the science, the ideas
which represent these facts, and the words by which these
ideas are expressed. Like three impressions of the same seal,
the word ought to produce the idea, and the idea to be a pic-
ture of the fact. And, as ideas are preserved and communicated
by means of words, it necessarily follows that we cannot
improve the language of any science without at the same
time improving the science itself; neither can we, on the other
hand, improve a science, without improving the language or
nomenclature which belongs to it. However certain the facts
of any science may be, and, however just the ideas we may
have formed of these facts, we can only communicate false
impressions to others, while we want words by which these
may be properly expressed.
Lavoisier is certainly not the only scientist of the past to have
insisted on the clarity of scientific language. Since 1919, the
International Union of Pure and Applied Chemistry (IUPAC)
has been contributing to international standardization in chem-
istry (IUPAC, 2019), after the Geneva Nomenclature of 1892,
based on a series of meetings organized by the German chemist
August Kekulé. For food science and food technology, and for
molecular and physical gastronomy, chemistry is much involved,
so that many terms are clearly defined internationally, but in other
fields, practitioners still suffer contradicting terminologies. In
what follows, we examine this issue, moving from physiology
to chemistry, and we conclude with the name of this handbook.
About Flavour (and Flavor)
Because (1) my country is Alsace, with a distinctive language that
is not French or German, and (2) the international language for
scientific communication (English) is not my native tongue, it is
easier for me to be surprised when language inconsistencies arise
in French or in English.
In French, for example, food generates a perception called
“goût”, which is the result of many stimuli plus some brain pro-
cessing; we shall see that, for some English-​speaking scientists
(only for some), this corresponds to “flavour” (in what follows,
627
628
I am using the spelling “flavour”, except when authors using
“flavor” are quoted). It is based on visual and auditory clues,
antenasal odour (“odeur”), taste (“saveur”), retronasal odour,
and also many other clues, such as those given by stimulation of
the trigeminal receptors and others (e.g., unsaturated long-​chain
fatty acids (LCFA) have specific receptors). At the perception
level, memory is important, as is demonstrated by the experiment
of repeating many times the word “lemon”: saliva flows in the
mouth, so that its buffering capacity can change the perception
of sourness.
It is said in French that the sensory system for the “goût”, i.e.,
the whole synthetic sensation, is “gustation”, and the French
scientific community seems to accept that the word “saveur”
(taste) is associated with the stimulation of receptors of taste
buds. However, no name is given to the specific activation of
this sensory pathway, so that the word “sapiction” was proposed
(This, 2009).
Regarding olfactory stimuli, French flavour scientists and
technologists speak of antenasal and retronasal odour (“odeur”),
and some (only some: there is no consensus, as was demonstrated
repeatedly during scientific meetings of the “Club ECRIN arôme”)
use the word “arôme” (aroma would be the equivalent in English)
for the “sum” of taste and retronasal odour, while the French dic-
tionary (TLFi, 2019) defines it contradictorily as the odour of aro-
matic plants, also called “aromates”. For some scientists, “arôme”
designates both the perception and the set of compounds respon-
sible for it, which is confusing (Richard, 2014). Finally, a French
regulation makes the picture worse, accepting the word “arôme”
for mixtures of extracts and pure compounds (isolated from essen-
tial oils or synthesized), which are named “flavourings” in English
(but fortunately, the word “aromatisant” seems to appear in French
nowadays as a replacement for “arôme”).
Last but not least, in the 1950s, a group of French physiologists
insisted on naming “flaveur” the “whole set of gustatory (taste),
olfactory and trigeminal perceptions” (Canon, 2019). But in the
same scientific community –​and within the same research insti-
tution –​others defined it differently as the “sum” of “saveur”
(taste) and “arome” (aroma) (Guichard, 2012). About this “sum”,
one should observe that the sciences of nature call for measure-
ment and refutability, so that an object is “scientifically sound”
when it can be measured: how could we make the “sum” of olfac-
tory and taste signals (and these two only, without the effect of
vision, noise, tactile perceptions and others)?
In English, also, definitions do not converge. For the
Cambridge Dictionary (2019), “flavour” is the “particular way
a substance, esp. food or drink, is recognized from its taste and
smell”. For Merriam Webster (2019), it is “the blend of taste
and smell sensations evoked by a substance in the mouth”. For
scientists, it is different: “The sensation of flavour occurs when
certain flavour-​active molecules are released from food and are
transported to stimulate sensors in the mouth and nose. The human
brain integrates all the signals from the sensors and produces an
overall flavour perception” (Taylor, 1998): let us observe that this
definition includes only the molecular stimulation, not taking into
account any visual or auditory clues, for example, and it does
not consider the fact that “thermal tastes” are not based on the
Hervé This vo Kientza
binding of “flavour-​active molecules” on receptors (Cruz and
Green, 2000).
Others scientists (Auvray and Spence, 2008) give different
definitions, such as: “Flavor is a multisensory perception, and
this decribes a combination of tastes, smells, trigeminal, tactile
sensations, visual and auditory cues, temperature”. And finally,
we comment (in square brackets) on the definition given by a
reference textbook (published by German scientists for whom
English is not the native tongue):
When food is consumed, the interaction of taste, odor and tex-
tural feeling [this word is strange, in this context] provides an
overall sensation which is best defined by the English word
“flavor” [here the authors stick to a particular definition that
does not include only molecular stimulation of receptors,
and they don’t explain their choice]. German and some other
languages do not have an adequate expression for such a broad
and comprehensive term [which is not true, as the French word
“goût” can be defined as the same as flavour]. Flavor results
from compounds that are divided into two broad classes: those
responsible for taste and those responsible for odors, the latter
often designated as aroma substances. However, there are
compounds which provide both sensations.
(Belitz et al., 2004)
Another question would be whether flavor is the same as
flavour, but in different versions of the English/​American lan-
guage (Caul, 1957). ASTM International (ASTM, 2019) answers
negatively, giving two definitions: “(1) perception resulting
from stimulating a combination of the taste buds, the olfactory
organs, and chemesthetic receptors within the oral cavity; (2) the
combined effect of taste sensations, aromatics, and chemical
feeling factors evoked by a substance in the oral cavity.” Here,
“chemical feeling” corresponds to chemesthesis, i.e., stimulation
of the trigeminal receptors.
And finally, one also finds the word “flavors” for compounds
that can stimulate a receptor (Running, 2018) in the same text:
“Olfaction, the sense of smell, contributes enormously to defining
and identifying food flavors.” Let us observe that the same text
includes “Another vital component of food flavor is texture”,
showing that for this author, flavor is more than simply the taste
and odour.
Now, regarding “taste”, the English word (for the French
“saveur”) seems to be well accepted for the detection of
hydrosoluble compounds that stimulate receptors in taste buds.
But regarding “Gustation is the sense of taste” (Running, 2018),
there would be a mismatch between French and English, as what
is relevant to flavour in the first language is about only taste in
the second.
On the contrary, there is no consensus about the words “smell”,
“odour” and “aroma”. About this last question, for sure Wikidiff
(2019) is not an official source, but it is interesting to observe that
the confusion is generalized: “As nouns the difference between
odor and aroma is that odor is any smell, whether fragrant or
offensive; scent; perfume while aroma is a smell; especially a
pleasant spicy or fragrant one.” Indeed, this occurs as well in the
scientific literature: “Many of these compounds have been related
Improving Communication in Food Science
to particular odor and aroma notes” (Arques et al., 2007). And
finally, for some, the odour is divided into recognizable “aromas”.
How can we interpret titles such as “Factors contributing to
interindividual variation in retronasal odor perception from aroma
glycosides: the role of odorant sensory detection threshold, oral
microbiota and hydrolysis in saliva” (Parker et al., 2019)? This
article goes on to say that that glycosides are “non volatile odor
precursors that have no odor themselves”. Later, the same text
speaks of “odorants”, and we can observe that the (welcome) use
of this word is spreading in the scientific literature.
All this being said, there is the question of the stimula-
tion of receptors by unsaturated LCFA, which has been named
“oleogustation” by some: “a new term, ‘oleogustus,’ has been
proposed to isolate the gustatory experience of fatty acids”
(Running et al., 2015). An attraction to palatable foods rich in
lipids is shared by rodents and humans. Over the last decade, the
mechanisms responsible for this specific eating behaviour have
been actively studied, and compelling evidence implicates a taste
component in the oro-​sensory detection of dietary lipids (i.e.,
LCFA), in addition to textural, olfactory and post-​ingestive cues.
The interactions between LCFA and specific receptors in taste
bud cells (TBC) elicit physiological changes that affect both food
intake and digestive functions.
Finally, how could we improve our scientific communication?
Firstly, a possibility would be to keep the word “flavour” (and
the equivalent “goût” in French) for the perception created by
eating one particular food. For example, a raspberry would have
a flavour of raspberry, and a particular wine would have the fla-
vour of this particular wine. When flavours reminiscent of other
food or non-​food items are distinguished during consumption,
the word “flavours” would be more appropriate than aromas,
because indeed, what is recognized is the synthetic perception,
and not only an odour: one can recognize a flavour of rose in a
gewurtztraminer, and also a flavour of chestnut in an oyster, etc.
If we accept the comprehensive definition of flavour, one has
to admit as well that it is impossible to get a particular, isolated
perception of taste, odour (antenasal or retronasal), consist-
ency, trigeminal stimulation, or visual and auditory stimulations,
because flavour is a multisensory perception. The easiest way is
to recognize that some compounds are “odorant” (when they can
bind to receptors in the nose), or tasty (when they have receptors
in the mouth), or that they can stimulate trigeminal receptors, etc.
The word “aroma” has no real reason to be used, except for
the odour (ante-​or retronasal) of aromatic plants, and finally,
it would probably be convenient to keep the word “texture” for
the perceived consistency (equivalent to microstructure) of some
food, not to be confused with its actual physical state (when one
dives into a swimming pool, the water can appear liquid or solid
depending on how the dive is made, but it remains a liquid phys-
ical system).
About Colloids
Regarding flavour, part of the difficulty lies in the fact (well
recognized by ASTM) that the sensory scientific community is
not the same as the chemist’s one. For colloids, one could imagine
629
that the issue would be less serious, because (1) the scientific
community is more homogeneous and (2) definitions were issued
by IUPAC. However, observation of the scientific literature also
reveals discrepancies. Let us begin with the definitions:
• Colloid: a short synonym for colloidal system (IUPAC,
1972a).
• Colloidal system: the term refers to a state of subdiv-
ision, implying that the molecules or polymolecular
particles dispersed in a medium have at least in one dir-
ection a dimension roughly between 1 nm and 1 μm, or
that in a system discontinuities are found at distances of
that order (IUPAC, 1972a).
Here, there does not seem to be much debate, but when it comes
to particular colloidal systems, the matter gets more complex.
Emulsions and Emulsified Systems
Let us begin by emulsions, which were defined by IUPAC
(1972b):
A fluid colloidal system in which liquid droplets and/​or liquid
crystals are dispersed in a liquid. The droplets often exceed the
usual limits for colloids in size. An emulsion is denoted by the
symbol O/​W if the continuous phase is an aqueous solution
and by W/​O if the continuous phase is an organic liquid (an
“oil”). More complicated emulsions such as O/​W/​O (i.e., oil
droplets contained within aqueous droplets dispersed in a con-
tinuous oil phase) are also possible. Photographic emulsions,
although colloidal systems, are not emulsions in the sense of
this nomenclature.
Here, there are three questions, which we will discuss in the
following:
(1) is it consistent to use the word “emulsion” for
Ramsden (also called Pickering) systems?
(2) are the words “nanoemulsions” and “microemulsions”
correctly given?
(3) is it proper to say that systems such as cream, butter
or ice creams are emulsions?
Regarding the first question, it has to be observed first that it
was very explicitly recognized by W. Ramsden in 1903 that oil
droplets or bubbles can be permanently stabilized by means of
fine solids or highly viscous particles placed at the interface of
two liquids (Ramsden, 1903). Later, it was observed that the
colloidal systems that have sometimes been called Pickering
emulsions are emulsions of any type, either oil-​ in-​
water (O/​
W), water-​in-​oil (W/​O), or even multiple, stabilized by solid
particles in place of surfactants (Binks, 2002). However Spencer
Umfreville Pickering himself (1907) recognized that “The sub-
ject had already been investigated by Ramsden (Proc. Roy. Soc.,
1903, 72, 156)”.
Here, we can come back to the first question, and we have to
recognize that Ramsden emulsions are fairly called emulsions, as
they are indeed “fluid colloidal systems in which liquid droplets
630
and/​or liquid crystals are dispersed in a liquid”. Now, a pro-
posal is to call them Ramsden emulsions rather than Pickering
emulsions, as is often the case (Nicolai and Murray, 2017).
Regarding nanoemulsions and microemulsions, confusion
often arises because nanometres are smaller than micrometres.
However, nanoemulsions and microemulsions do not describe –​
today, but the scientific community could decide to change
this –​emulsions having dispersed droplets in the nanometre
and micrometre range, respectively. It is a fact that “currently,
there is considerable confusion about the use of the terms
‘microemulsions’ and ‘nanoemulsions’ in the scientific litera-
ture. However, these are distinctly different types of colloidal
dispersions: a microemulsion is thermodynamically stable,
whereas a nanoemulsion is not” (McClements, 2012).
More precisely, an oil-​in-​water microemulsion is a thermo-
dynamically stable colloidal dispersion consisting of small
spheroid particles (comprised of oil, surfactant and possibly
co-​surfactant) dispersed within an aqueous medium, whereas
nanoemulsions can be considered to be conventional emulsions
that contain very small particles; in particular, an oil-​in-​water
nanoemulsion is defined as a thermodynamically unstable col-
loidal dispersion consisting of two immiscible liquids, with one
of the liquids being dispersed as small spherical droplets (r <
100 nm) in the other liquid.
This being observed, it is clear that the word “microemulsion”
is at the root of the confusion, and this is why a new name could
be usefully given, considering that such systems can be very
different from emulsions, as defined by IUPAC (with cylinder-​like
structures, lamellar structures, sponge-​like structures, etc., instead
of the droplets evoked in the IUPAC definition). Considering, on
the one hand, that size is not the distinctive characteristic of the
systems today called microemulsions, and on the other hand, that
microemulsions are not emulsions as defined by IUPAC, the two
parts of the name should be replaced with a more descriptive
name, or even an acronym, such as TSDP, for “thermodynamic-
ally stable dispersion of two phases”, for example.
Finally, are cream, butter and ice creams emulsions? Let us
start the analysis with milk, which is often said (and taught) to
be a O/​W emulsion: the milk fat globules, about 4 micrometres
in diameter, are made of a core rich in triacylglycerols (TAGs)
enveloped by the milk fat globule membrane (Lopez, 2011).
Milk fat has a TAG composition that changes with bovine
species, animal nutrition, gestational age of cows, lactation
stage, etc. However, concerning the question discussed here, we
have primarily to concentrate on the thermal behaviour of TAGs,
which depends on their fatty acid residues. About 400 fatty acid
residues have been identified, so that the melting temperature of
milk fat spreads over a temperature interval of −10 °C to 60 °C
(Jensen, 2002; Bouteille et al., 2013). For milk at 20 °C, the fat
solid content is about 40%: is it fair to speak of an emulsion,
when a colloidal dispersion contains about 50% of the dispersed
material in a solid state? The IUPAC definition allows for a posi-
tive answer.
Now, cream is made from creaming of fat globules, and even if
the fat concentration is higher, the same answer can be given. But
what about butter, of which it is often repeated (with no analysis)
that it is a W/​O emulsion? Now, the question of temperature is
Hervé This vo Kientza
very important, because it is clear that when all fat is crystallized,
e.g., at temperatures below −20 °C, butter is a multiphasic solid.
But at room temperature?
The question has been studied extensively (Rønholt, 2014).
Butter, spreads and margarines are all multiphase water-​in-​oil
dispersions, consisting of liquid fat, crystallized fat and water
droplets (Juriaanse and Heertje,1988). The fat crystal network
strongly contributes to product stability by physical stabiliza-
tion of the water droplets dispersed within the fat phase, hence
preventing microstructural changes (Rousseau et al., 2003).
Indeed, the observation of a solid network in which one can find
liquid fat and liquid water (this can occur, as the solid content
is more than about 5%) should make us consider that butter is a
gel, rather than an emulsion. But what about the liquid fat and
liquid water? If water were dispersed in the liquid fat phase, then
the system would be a complex oleogel, and the system could
be considered as an emulsified system, but not an emulsion. Of
course, when the solid fat is no longer making a continuous net-
work, now a complex emulsion can be obtained, with liquid water
and fat crystals dispersed in the liquid oil phase. The use of dis-
perse system formalism (DSF) makes all this very simple and
clear (see the chapter on DSF in this book).
Regarding ice cream, as well, it depends, of course, on the par-
ticular recipe being used, but for many of them, the fat content
is low, and even if some fat remains liquid, most of it is solid,
and we should consider, rather, that the system is made of a con-
tinuous aqueous liquid phase (because of the high sugar concen-
tration) in which are dispersed ice crystals, fat crystals, a small
quantity of liquid fat, and other ingredients such as vegetable
solid particles. This certainly does not correspond to the defin-
ition of an emulsion but rather, of a suspension.
Gels
For the question of gels, another chapter of this book is already
considering the question, and we simply give here the IUPAC
(2007) definition:
Non-​ fluid colloidal network or polymer network that is
expanded throughout its whole volume by a fluid. Notes: (1)
A gel has a finite, usually rather small, yield stress; (2) A gel
can contain: (2.1.) a covalent polymer network, e.g., a net-
work formed by crosslinking polymer chains or by non linear
polymerization; (2.2.) a polymer network formed through the
physical aggregation of polymer chains, caused by hydrogen
bonds, crystallization, helix formation, complexation, etc,
that results in regions of local order acting as the network
junction points. The resulting swollen network may be termed
a thermoreversible gel if the regions of local order are ther-
mally reversible; (2.3.) a polymer network formed through
glassy junction points, e.g., one based on block copolymers.
If the junction points are thermally reversible glassy
domains, the resulting swollen network may also be termed
a thermoreversible gel; (3) lamellar structures including
mesophases, e.g., soap gels, phospholipids and clays; (4) par-
ticulate disordered structures, e.g., a flocculent precipitate
usually consisting of particles with large geometrical anisot-
ropy, such as in V2O5 gels and globular or fibrillar protein gels.
Improving Communication in Food Science
With this definition, it can be observed that emulsions (e.g.,
mayonnaise) are not gels, even if their rheological behaviour
mimics that of real gels: not all that glitters is gold, as the saying
has it.
Foams and Aerated Systems
Regarding foams, IUPAC (1972b) opens the door to discussion
by recognizing that they are
dispersions in which a large proportion of gas by volume
in the form of gas bubbles, is dispersed in a liquid, solid or
gel. The diameter of the bubbles is usually larger than 1 μm,
but the thickness of the lamellae between the bubbles is often
in the usual colloidal size range. The term froth has been
used interchangeably with foam. In particular cases froth
may be distinguished from foam by the fact that the former is
stabilized by solid particles (as in froth flotation q.v.) and the
latter by soluble substances.
Here, we have the same discussion as when a small quantity of
oil is dispersed in a complex colloidal system: in the second case,
it is more appropriate to speak of an emulsified system, and here,
of an aerated system.
About Chemistry in General
Concerning chemical words, it is a sad observation that the words
“compounds” (molecular species) and “molecules” are some-
times confused. For example (personal translation of a text, of
which the author is not given because it is a colleague): “The
second composant is aroma. Now, there are volatile molecules
that, during mastication, are released by food and go up toward
the olfactory system. This is called retronasal odour. The percep-
tion of aroma is more complex, because an aroma can be a mix-
ture of many molecules. For example, the aroma of pomegranate
is made of six volatile molecules.”
Not only do we observe here the previous confusion between
odorants and aroma, about the stimulus and its result, but
also, the author is giving the name “molecule” to a molecular
species. Hopefully, the difference is understood by scientists
and technologists, but it is my personal experience that “science
journalists” suffer this confusion, and part of the public certainly
does: I had to explain to a TV journalist that wine does not con-
tain only 350 molecules, but 350 compounds, with very large
numbers of molecules for each compound.
In this field of chemistry, one of the most important confusions
that could be usefully corrected in view of science improvement
is the confusion of amino acids and fatty acids, respectively,
with amino acid residues (in proteins) and fatty acid residues (in
triglycerides). In some scientific articles, it is absolutely not clear
what the authors have in mind (residues or free forms) when they
speak of amino acids and fatty acids, and even reading their text
carefully gives no clue.
For saccharides, sugars, “carbohydrates”, -​ oses and their
chemical cousins, the question is different, but it is worth refer-
ring to the history of science in order to observe that the word
631
“carbohydrate” could be usefully dropped, because it is a remnant
of an obsolete state of knowledge when it was thought, through
elementary analysis (chemical analysis by which the proportions
of the various chemical elements are determined), that a “water
moiety” was linked to a carbon one. Indeed, in some polyols from
the mono-​, oligo-​and polysaccharide families, for sure there are
as many oxygen atoms as carbon ones, and double the number of
hydrogen atoms, but no water molecule is present, and they are
certainly not “hydrates” (with loosely bound water molecules).
Here, the IUPAC definitions are:
Saccharides: The monosaccharides and di-​ , oligo-​and
polysaccharides, which are made up of n monosaccharide
units linked to each other by a glycosidic bond. Considered by
some to be synonymous with carbohydrates (IUPAC, 1995a).
Sugars (IUPAC, 1995b): A loose term applied to
monosaccharides and lower oligosaccharides.
Should We Speak of Maillard Reactions?
Probably Not
We finish with the often-​used expression “Maillard reaction”, for
which we shall see that it is often used in very strange ways,
so we propose that the name “glycation reaction” should be
used instead … when it does indeed correspond to a glycation.
Moreover, we observe that the Maillard reaction is often said to
be a “complex group of reactions”, and we propose here thinking
that when the word “complex” is used, it sometimes means that
scientific work remains to be done.
In This (2016), the scientific contribution of the French chemist
Louis Camille Maillard (1878–​1936) was discussed, and it was
shown that the terminology “Maillard reactions” is spreading in
spite of the fact that it is certainly not legitimate for this name
to be given to all non-​enzymatic browning processes. Here, we
focus on this question of the name given to molecular transform-
ations of food, because they are too often wrongly associated
with the name of Maillard. We also give some hints for the inter-
pretation of browning of organic systems being thermally treated.
Looking at the scientific literature about “Maillard reactions”
reveals a muddled state of the art; the analysis of some examples
demonstrates the need for stricter naming of molecular transform-
ations during food production. First, it can be observed that the
name of Maillard is sometimes given to “reactions” or to “a reac-
tion”, which shows a lack of coherence; an explanation for this
can be acquired through sentences such as “The Maillard reac-
tion is a very complex series of reactions” (Gobert and Glomb,
2009), and it is certainly true that most classical reactions (Würtz
condensation, Grignard reaction, Diels–​Alder reaction, etc.) are
the result of many steps. However, if a reaction is to be given the
name of Maillard, the singular should be preferred for “reaction”.
But the main inconsistency is not here: it is the confusion of
“Maillard reaction” or “Maillard reactions” with “non-​enzymatic
browning reactions” and even caramelization (Rao et al., 2018).
Let us analyse a sentence such as “Naturally occurring reac-
tion during food processing, glycation, also known as non-​
enzymatic browning or Maillard reaction, can improve food
632
protein physiochemical properties and functionality.” This sounds
strange, first, precisely because food processing is not natural
(dictionaries define something as natural when it is not the result
of human action), and second, because such a text assimilates
all non-​enzymatic browning to Maillard reaction: caramelization
and pyrolysis are certainly non-​enzymatic browning reactions,
and the reactions creating brown colours are in huge numbers,
but Maillard did not study them, and they often have very little
in common with glycation reactions between sugars and amines.
Phrases such as “Non-​enzymatic Maillard reactions between
reducing sugars and proteins, lipids, or nucleic acids” (Zhu
et al., 2018) are also disputable, because it is a pleonasm to write
“non enzymatic Maillard reactions”, as “Maillard reactions” are
always non-​enzymatic. But the question is also whether Maillard
ever considered reactions between reducing sugars and lipids, or
with nucleic acids (the answer is no). Moreover, another main
point is whether Maillard reactions can occur between reducing
sugars and proteins, or between reducing sugars and amino acids.
Finally, with sentences such as “the Maillard reaction is one of
the most important chemical reactions occurring during heating
or storage of food” (Hellwig et al., 2019), the question is how
important Maillard reactions are: indeed, the “importance”
should be measured and expressed quantitatively before one is
able to write such a sentence (in our laboratory, adjectives and
adverbs are “forbidden” and have always to be replaced by the
answer to the question “How much?”).
When Maillard reactions are considered, most authors
acknowledge that initially, amino compounds, such as free
amino acids, N-​terminal amino groups of peptides or proteins,
or side chains of peptide-​bound lysine residues, react with redu-
cing sugars, such as glucose, maltose, or lactose, to generate
so-​called Amadori products (aminoketoses). During prolonged
heating or storage, Amadori compounds are degraded to form
1,2-​dicarbonyl compounds (“advanced stages of the Maillard
reaction”), which may subsequently react with free amino acids
via Strecker degradation to form flavour compounds.
The term “glycation” was introduced by IUPAC as a general
term for sugar−protein adducts, with “glycosylation” being a
subtype of glycation, i.e., resulting from an enzyme-​catalysed
reaction (Sharon, 1985). However, this distinction has never been
widely recognized; today, glycation is used synonymously with
the Maillard reaction. The term “glycation” is preferred in bio-
medical research, whereas the “Maillard reaction” prevails in food
chemical research. The expression “AGEs”, denoting “advanced
glycation end products”, is a historically grown collective term
for products of the Maillard reaction formed in vivo. Initially, the
term was introduced for “brown fluorescent pigments that cross-​
link proteins”, i.e., irreversibly bound sugar adducts on proteins,
in contrast to reversibly bound adducts, such as Schiff bases and
Amadori products. In recent years, the term AGEs has increas-
ingly been used to summarize all Maillard reaction products,
including amino acid derivatives, dicarbonyl compounds (e.g.,
3-​deoxyglucosone and methylglyoxal) and stable end products,
such as 5-​hydroxymethylfurfural (HMF).
All this being said, an analytical history of the “Maillard reac-
tion” shows clearly that chemistry and food science should be
Hervé This vo Kientza
more careful about denominations. We start from the observa-
tion that too many food scientists and technologists speak of the
Maillard reaction each time that there is some non-​enzymatic
browning of food tissues, but the thermal treatment (200 °C,
20 min) of separated samples of pure starch (polysaccharides),
flour (polysaccharides and proteins, plus other compounds), a
mixture of starch and egg white proteins, sucrose, and sucrose
and proteins shows very clearly that the brown colour can be
the result of more than glycation reactions. The question of how
much brown colour is produced during glycation reactions (This,
2016) seems to be clearer using this result. Now, it appears that the
browning of a mixture of proteins and saccharides was observed
long before the work of Louis-​Camille Maillard (1878–​1936).
A short history of glycation reactions begins in 1866, before
Maillard was born: Hugo Schiff (1834–​1915) discovered that
aldehydes (including monosaccharides) can react with amines,
and in particular with amino acids, and also that sugars react with
amines, forming dark compounds (Schiff, 1866; Qin, 2013). Schiff
proposed the formation of secondary imines from aldehydes and
aromatic amines (“Schiff bases”). In his article, he discussed the
reaction of D-​glucose, aniline and p-​toluidine, and he proposed
the formation of secondary imines (today called Schiff’s bases)
from aldehydes and aromatic amines. And in 1871 (again before
Maillard was born), R. Sachsse observed the reaction between
lactone and aniline, or benzenamine (Sachsse, 1871).
In 1884 and 1886, the great German chemist Emil Fischer
(1852–​1919) explored the reactions of saccharides and amino
compounds, first D-​glucose or D-​fructose and phenylhydrazine,
and then also sucrose and phenylhydrazine, obtaining 1-​amino-​
1-​deoxyfructose (Fischer, 1884). Fischer noted that D-​glucose
and D-​fructose lead to the same compound, and he identified
that the reaction of sucrose with phenylhydrazine was producing
1-​amino-​1-​deoxyfructose. This would hint at the name “Fischer
reactions” for what are sometimes called Maillard reactions.
In 1888, B. Sorokin studied the reaction between D-​glucose and
aniline (Sorokin, 1888), and in 1898, Lobry de Bruyn obtained
D-​glucosamine from D-​fructose and ammonia (crystallized by
Breuer), a compound important in the reaction scheme published
by Hodge in 1948 (Hodge and Rist, 1952). Contrary to what
some sources wrongly indicate, the work by AR Ling came later:
he observed during beer making that colour and odour appear
when the temperature of the malt is between 120 °C and 150 °C.
Using all the information already published, he assumed without
demonstration that amino acids react with sugars, but this guess
was easy to build on the work by Fischer and others. By the way,
it is interesting to observe that articles dealing with the history of
Maillard reactions attribute this quotation to two authors, “Ling
and Malting”; however, Malting was not the co-​author of Ling,
but only the title of the article by Ling: “Ling AR. 1908. Malting,
J. Inst. Brew., 14:494–​521”. And in the same year, JC Irvine and
R Gilmour explored the reaction of D-​glucose and p-​toluidine.
Finally, it was only in 1913 that Maillard published his “Genèse
des matières protéiques et des matières humiques”, with a subtitle
stating clearly that he was considering the action of glycerol and
saccharides on amino acids. Now, it has to be observed that the
title of the work does not say anything new. And indeed, Maillard
Improving Communication in Food Science
himself recognized that the work was part of what chemists had
been producing for a century on the question of “albuminoïds”, after
Braconnot, who treated gelatine with sulphuric acid, producing a
“sugar of gelatin”, glycocolle. Maillard’s goal was the synthesis of
proteins, and it was in the fourth part of his text that he considered
the action of saccharides on alpha amino acids, without giving credit
to previous researchers; the only new information is an analysis of
the reactions in more depth than Ling. Maillard did not explain the
reaction, and indeed, the mechanism was only investigated later by
Mario Amadori (1886–​1941), i.e., reacting hydroxyaldehydes with
amines to make alpha-​aminoketones (Wrodnigg and Eder, 2001),
before K Heyns explored alpha-​hydroxyketones to make 2-​amino-​
2-​deoxyaldoses (Kawamura, 1983).
Finally, what is a Maillard reaction? It is certainly a glycation
one, so that the IUPAC proposal is to be promoted. By the way, it
is not true that Maillard discovered melanoidins, as is sometimes
published: they were known and named by M Schmiedeberg as
early as 1897 (Schmiedeberg, 1897).
Why was Maillard credited for something that he did not dis-
cover? Certainly, science is universal, but one should remember
that there was strong opposition between France and Germany at
that time, with much nationalism, to the point that the Alsatian
chemist Charles Adolphe Würtz declared in 1874 that “chemistry
was a French science” (Würtz, 1874).
Conclusion
If we want to make better science and technology, the right words
have to be used, and this also applies to the whole of scientific or
technological activities. In another chapter of this book, it is well
explained that “molecular and physical gastronomy” (sometimes
shortened to “molecular gastronomy”) is the part of food science
that looks for the mechanisms of phenomena occurring during
culinary processes, not to be confused with “molecular cooking”
or “molecular cuisine”. But what about “culinary science”, often
associated with quotations from the cook Auguste Escoffier, who
(wrongly, in our opinion) predicted that cooking would become
“scientific”?
The main issue while discussing this question is that the word
“science” is used both for knowledge in general and for “natural
sciences”. Although it is certain that cooks have some knowledge
(about their technical activity), they don’t –​and will never –​use
scientific methods for their work, because their goal is to produce
“good dishes” and not new knowledge on the mechanisms of phe-
nomena. By the way, it is perhaps safe to repeat here that the
method of the natural sciences goes through (1) identification
of a phenomenon, (2) quantitative characterization of the phe-
nomenon, (3) grouping the data in synthetic laws, i.e., equations,
(4) inducing a theory by the possible introduction of new
concepts quantitatively compatible with the equations and data,
(5) looking for theoretical refutable consequences of the theory,
and (6) testing such consequences experimentally (This, 2017).
In the 19th century, the French chef Marie-​Antoine Carême
discussed the “science of cooking”: “Cooking also wants to be a
science”. What does it mean? Certainly, Carême is proposing that
there is a theoretical knowledge, and not only a practice (which
633
is why he wrote books). And the same meaning occurs with “The
culinary science is more safe for human health than all the know-
ledgeable prescriptions of those who make the diseases easier by
prescription” (attacking the medicine of his time).
Carême’s students Urbain Dubois, Emile Bernard, Jules
Gouffé and Joseph Favre used the same idea, but when they said
that one has to use precise measurements, this did not mean nat-
ural science, but simply more precise technique. In particular,
Favre discussed a “scientific culinary activity”, which would
mean “the art of preparing food well”: if the meaning of science
is here about knowledge, then it is almost a tautology, but if it is
natural science, then Favre is mistaken.
Finally, we come to Escoffier: “Cooking, remaining an art, will
become scientific, and will have to submit its empirical formula to
a method and a precision that will leave nothing at random.” This
is again either a tautology or a mistake. More precisely, Escoffier
was wrong to think that precision in an activity is enough to trans-
form this activity into a natural science. It is not true that being
precise makes someone into a practitioner of the natural sciences.
Cooks are cooks, and scientists are scientists. They are different
… even if they can –​and are invited to –​discuss!
REFERENCES
Arques L, Garde S, Fernández-​Garcıa E, Gaya P, Nunez M. 2007.
Volatile compounds, odor, and aroma of La Serena cheese
high-​pressure treated at two different stages of ripening, Dairy
Science, 90, 3627–​3639.
ASTM International. 2019. Standard Terminology Relating to
Sensory Evaluations of Materials and Products, E253-​ 09a.
ASTM International, West Conshohocken, PA. E253-​09a.
Auvray M, Spence C. 2008. The multisensory perception of flavor,
Consciousness and Cognition, 17, 1016–​1031.
Belitz HD, Grosch W, Schieberle P. 2004. Aroma Compounds. In
Food Chemistry. Springer, Berlin, Heidelberg, 342–​408.
Binks BP. 2002. Particles as surfactants–​similarities and differences,
Current Opinion in Colloid and Interface Science, 7, 21–​41.
Bouteille R, Perez J, Khifer F, Jouan-​ Rimbaud-​ Bouveresse D,
Lecanu B, This H. 2013. Influence of the colloidal structure
of dairy gels on milk fat fusion behavior: Quantification of
the liquid fat content by In Situ Quantitative Proton Nuclear
Magnetic Resonance Spectroscopy (isq 1 H NMR) , Journal of
Food Science, 78(4), E535–​E541.
Cambridge Dictionary. 2019. https://​ dictionary.cambridge.org/​
fr/​
dictionnaire/​anglais/​flavor
Canon F. 2019. www.inra.fr/​Grand-​public/​Alimentation-​et-​sante/​
Toutes-​les-​actualites/​Gout-​et-​flaveur
Caul JF. 1957. The profile method of flavor analysis. In Mrak EM,
Stewards M (eds.) Advances in Food Research. Academic
Press, New York, vol. 7, 1.
Cruz A, Green BG. 2000. Thermal stimulation of taste, Nature, 403,
889–​892.
Fischer E. 1884. Verbindunger des Phenyldydrazins mit den
Zuckerarten, Berichte der Deutschen Chemischen Gesellschaft,
17, 579–​584.
Gobert J, Glomb M. 2009. Degradation of glucose: Reinvestigation
of reactive α-​dicarbonyl compounds, Journal of Agricultural
and Food Chemistry, 57(18), 8591–​8597.
Guichard E, Genot C. 2012. Texture et flaveur des aliments. In
Elisabeth Guichard (ed.) Vers une conception maîtrisée,
Educagri editions
634
Hellwig M, Humpf HU, Hengstler J, Mally M, Vieths S, Henle T.
2019. From the Maillard reactions to advanced glycation
end products (AGES): Chemical considerations, Journal of
Agricultural and Food Chemistry, 67, 11307−11311.
Hodge JE, Rist CE. 1952. N-​ glycosyl derivatives of secondary
amines, Journal of the American Chemical Society, 74,
1494–​1497.
IUPAC. 1972a. Manual of Symbols and Terminology for
Physicochemical Quantities and Units, Appendix II:
Definitions, Terminology and Symbols in Colloid and Surface
Chemistry, Pure and Applied Chemistry, 31, 605.
IUPAC. 1972b. Manual of Symbols and Terminology for
Physicochemical Quantities and Units, Appendix II:
Definitions, Terminology and Symbols in Colloid and Surface
Chemistry, Pure and Applied Chemistry, 31, 606.
IUPAC. 1995a. Glossary of class names of organic compounds
and reactivity intermediates based on structure (IUPAC
Recommendations 1995), Pure and Applied Chemistry, 67,
1364.
IUPAC. 1995b. Glossary of class names of organic compounds
and reactivity intermediates based on structure (IUPAC
Recommendations 1995), Pure and Applied Chemistry, 67,
1367.
IUPAC. 2007. Definitions of terms relating to the structure and pro-
cessing of sols, gels, networks, and inorganic-​organic hybrid
materials (IUPAC Recommendations 2007), Pure and Applied
Chemistry, 79, 1806.
IUPAC. 2019. https://​iupac.org/​who-​we-​are/​our-​history/​
Jensen RG. 2002. The composition of bovine milk lipids, Journal of
Dairy Science, 85(2), 295–​350.
Juriaanse AC, Heertje I. 1988. Microstructure of shortenings, mar-
garine and butter: A review. Food Microstructure, 7, 181–​188.
Kawamura S. 1983. Seventy Years of the Maillard Reaction. In
Waller GR and Feather MS (eds.) The Maillard Reaction in
Foods and Nutrition. ACS Symposium Series. American
Chemical Society, Washington DC.
Lavoisier A. 1789. Elements of Chemistry, In a New Systematic
Order, Containing all the Modern Discoveries (transla-
tion Robert Kerr), www.gutenberg.org/​cache/​epub/​30775/​
pg30775.txt
Ling AR. 1908. Malting, Journal of the Institute of Brewery, 14,
494–​521.
Lopez C. 2011. Milk fat globules enveloped by their biological
membrane: Unique colloidal assemblies with a specific com-
position and structure, Current Opinion in Colloid & Interface
Science, 16, 391–​404.
McClements DJ. 2012. Nanoemulsions versus microemulsions:
Terminology, differences, and similarities, Soft Matter, 8,
1719–​1729.
Merriam Webster Dictionary. 2019. www.merriam-​webster.com/​dic-
tionary/​flavor
Nicolai T, Murray B. 2017. Particle stabilized water in water
emulsions, Food Hydrocolloids, 68, 157–​163.
Parker M, Onetto C, Hixson J, Bilogrevic E, Schueth L, Pisaniello
L, Borneman A, Herderich M, De Barros Lopes M, Leigh F.
2019. Factors contributing to interindividual variation in
retronasal odor perception from aroma glycosides: The role
of odorant sensory detection threshold, oral microbiota,
and hydrolysis in saliva, Journal of Agricultural and Food
Chemistry, https://​doi.org/​10.1021/​acs.jafc.9b05450
Pickering SU. 1907. Emulsions, Journal of the Chemical Society:
Transactions, 91, 2001–​2021.
Hervé This vo Kientza
Qin W, Long S, Panunzio M, Biondi S. 2013. Schiff Bases: A
short survey on an evergreen chemistry tool, Molecules, 18,
12264–​12289.
Ramsden W. 1903. Separation of solids in the surface-​ layers
of solutions and “suspensions” (observations on surface-​
membranes, bubbles, emulsions, and mechanical coagula-
tion). Preliminary account, Proceedings of the Royal Society,
72, 156.
Rao Q, Jiang X, Li Y, Samiwala M, Labuza TP. 2018. Can glycation
reduce food allergenicity? Journal of Agricultural and Food
Chemistry, 66(17), 4295–4299.
Richard H. 2014. Les composés odorants. In Lavelle C (ed.) Science
culinaire, Belin, Paris, 314–​348.
Rønholt S, Kirkensgaard JJK, Mortensen K, Knudsen JC. 2014.
Effect of cream cooling rate and water content on butter micro-
structure during four weeks of storage, Food Hydrocolloids,
34, 169–​176.
Rousseau D, Zilnik L, Khan R, Hodge S. 2003. Dispersed phase
destabilization in table spreads, Journal of the American Oil
Chemists’ Society, 80, 957–​961.
Running CA. 2018. Oral sensations and secretions, Physiology &
Behavior, 193, 234–​237.
Running CA, Craig BA, Mattes RD. 2015. Oleogustus: The unique
taste of fat, Chemical Senses, 40(7), 507–​516.
Sachsse R. 1871. Ueber einige stickstoffhaltige Verbindungen des
Milchzuckers von Anilin, Berichte der Deutschen Chemischen
Gesellschaft, 4, 834.
Schiff H. 1866. Eine neue Reihe organischer Diamine [in German],
Justus Liebigs Annalen der Chemie, 140, 92–​137.
Schmiedeberg O. 1897. Ueber die Elementarformeln einiger
Eiweisskörper und über die Zusammenselzung un die Natur
der Melanine, Archiven für expermental Pathologie und
Pharmakologie, 39, 1.
Sharon NB. 1985. Nomenclature of glycoproteins, glycopeptides
and peptidoglycans, IUPAC-​ IUB Joint Commission on
Biochemical Nomenclature (JCBN), recommendations 1985.
Sorokin B. 1888. Ueber Anilide der Glycose, Berichte der Deutschen
Chemischen Gesellschaft, 19, 513.
Taylor AJ. 1998. Physical chemistry of flavour, International Journal
of Food Science and Technology, 33, 53–​62.
This H. 2009. Goût, odeur, saveur, arôme? L’Actualité chimique,
332, 9–​11.
This H. 2016. “Maillard products” and “Maillard reactions” are much
discussed in food science and technology, but do such products
and reactions deserve their name? Notes Académiques de
l’Académie d’agriculture de France /​Academic Notes from the
French Academy of Agriculture, 3, 1–​10.
This H. 2017. Molecular gastronomy, as a prototype for discussing
the issue of scientific strategy, Irish Chemical News, 5, 1–17
TLFi. 2019. http://​stella.atilf.fr/​Dendien/​scripts/​tlfiv5/​advanced.
exe?8;s=3760138455
Wikidiff. 2019. https://​wikidiff.com/​odor/​aroma
Wrodnigg TM, Eder B. 2001.The Amadori and Heyns rearrangements:
Landmarks in the history of carbohydrate chemistry or unrec-
ognized synthetic opportunities? Topics in Current Chemistry,
215, 115–​152.
Würtz CA. 1874. Dictionnaire de chimie pure et appliquée, https://​
gallica.bnf.fr/​ark:/​12148/​bpt6k3383f.image
Zhu Y, Snooks H, Sang S. 2018. Complexity of advanced glycation
end products in foods: Where are we now? Journal of
Agricultural and Food Chemistry, 66, 1325−1329.
Experimental Flavour Workshops
Hervé This vo Kientza1,2
1 INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
In France, culinary lessons were once given at primary school
and in high schools for girls (only), but these courses disappeared
from the French curriculum in the 1970s, which was unfortu-
nate, because, if it is true that the “technology classes” that were
introduced instead were useful (one is happy to be able to change
an electric bulb, for example), it is also true that the young citi-
zens were left without the important knowledge of how to prepare
their food (instead of relying on the food industry), and had to
rely on private books and journals, in spite of much enthusiasm
for food in that country for many centuries (Montaigne, 1581).
The development of scientific activities based on cooking,
which came to be called molecular and physical gastronomy in
1988, and the development of molecular cooking (after 1980, but
with this name given in 1999 only), were considered with much
interest after 1980, with lectures, seminars and courses in uni-
versities, for example, but the development in primary schools,
and more generally in the various instructional centres run by
the French public Department of Education, occurred only after
the publication of a book entitled La casserole des enfants
(Children’s pan) in 1997 (This, 1997).
This book, like the later instructional programmes that we dis-
cuss here, is based on the wonderful remark by Michael Faraday:
“It is not enough to know the great principles; one should manipu-
late” (Faraday, 1842).
Developments since 2000 in Primary Schools
Upon the invitation of the French minister of public education,
in 1998, the experiments proposed in the book (This, 1997) were
transformed into class activities that were officially introduced
in all French primary schools in September 2000 under the
name “Ateliers expérimentaux du goût” (Flavour Experimental
Workshops), linking cooking to science and other cultural activ-
ities: following the examples in the book, and with minimal hard-
ware, children whip eggs until they get huge volumes of foam
(but they primarily understand the chemical composition of eggs
and the importance of surfactants), change the colour of tea (and
they understand why lemon juice makes such a change), look for
the parameters of good bread, and even make wine or cheese. The
prototypes of the activities were, of course, educationally tested
and validated in 2000, and after one year of successful implemen-
tation, the initial 20 activities were increased to 35.
How were these activities designed? The answer to the
minister’s demand had to be analysed carefully before an appro-
priate proposal was made, and it was also considered why
culinary classes had disappeared from schools in order to avoid
a new disappearance. For sure, culinary activities need special
hardware and relatively costly products, and expose children to
danger (burning, cutting, etc.), but also, traditional recipes were
not intended to invite critical thinking, and their methodological
content was low. In France, the culinary activities had been based
on former publications by Edouard de Pomiane, which contained
many errors (Pomiane, 1922; 1931; 1948; This, 2006), and they
had often been restricted to technique instead of technology (crit-
ical thinking on technique). Molecular and physical gastronomy
was useful in order to circumvent these difficulties, because it
offered a link between cooking and science, the latter being a
goal for public education, and also because it was investigating
the roots of cooking, allowing innovation through technology.
Moreover, it was a way to add a chemical component to the activ-
ities of the “Main à la pâte” initiatives (2019), a French develop-
ment similar to “Hands on” (2019) in the USA.
In brief, the Flavour Experimental Workshops were intended
to introduce both enlightening cooking activities and experi-
mental science.
Departing from “Instinct”
There are two great ancestors of this project. On the one hand,
Jean-​Anthelme Brillat-​Savarin (born Bellay, 1755; died Paris,
1826) placed among the introductory aphorisms of his Physiology
of taste: “The animals feed themselves; the man eats; the man
of spirit alone knows how to eat” (Brillat-​Savarin, 1825). At a
time when obesity strikes, children have to learn how to eat,
and cooking is important for this skill. As said before, it is also
important to escape the mere execution of technical gestures that
are not understood.
The second godfather of these activities is Antoine Laurent de
Lavoisier (born Paris, 1743 –​died Paris, 1794), who wrote in the
introduction to his Elements of Chemistry: “The impossibility of
635
636 Hervé This vo Kientza

isolating the nomenclature of science, and the science of nomen-

clature, holds that all physical science is necessarily founded on
three things: the series of facts that constitute science, the ideas
that recall them, the words that express them. […] As words
retain ideas and transmit them, it follows that languages cannot
be perfected without perfecting science or science without lan-
guage” (Lavoisier, 1798). From this perspective, science, espe-
cially chemistry, finds a perfect instructional justification; it
invites the control of language and ideas, while revealing the
beauties of the world where the children live; it combines experi-
mental practice with theoretical interpretation, and makes the
hand skilful when it is commanded by critical thinking.

Protocols and Their Instructional Support

The experimental protocols and instructional materials of the
document entitled “Flavour Experimental Workshops” (down-
loadable from the French website for instructional resources,
as well as video clips; Ministère de l’education nationale,
2000) describe experiments that can be performed by children at
school under the guidance of any adult. They offer an exploration
of heritage products (e.g., bread, cheese, butter, milk, vinegar and
wine) from the technical, technological and scientific point of
view. The proposed instructional method is active.
For example, the first activity proposed is the study of the
question: how much whipped egg white can be prepared from a
single egg? (See the full protocol later.) These experiments are
cheap and safe, and can be done anywhere. Initially proposed for
second year elementary classes, they were also tested in nursery
schools.
FIGURE 94.1 A simple calculation for determining the quantity of butter
The proposed activities are initially experimental, as the
and flour needed for making a dough. Imagine that dough is made of starch
name of these activities underlines, but they can be consider-
granules (white discs in top picture) that are packed in butter (in yellow).
ably enriched by including elements of writing, reading, history, The overall quantity of starch can be interpreted as the sum of the areas for
geography and calculation. For example, in high schools, one can all the white discs, and the overall quantity of butter can be interpreted as
show how knowing the formulas that give the surface of a disk the sum of the areas for all the yellow parts. The students can understand
and a square helps to predict the proportions of flour and butter in that dealing with only one disc and one square can give the answer, as the
a pie dough (Figure 94.1). result will be simply multiplied by the number of elementary cells. For the
These workshops relate to an artistic activity, through which simple case of a disk in a square, it can be calculated (assuming radius 1) that
they reveal the technical aspect. They frame the space of artistic the area of the disk is about 3, whereas the area of the yellow part is (2 ×
freedom, but do not limit it; on the contrary, they reveal the pos- 2) −3. This simplistic calculation leads to a 1:3 ratio, which can be tested. Of
sibilities of innovation. Thus, after the workshop shown later, it course, when the children are taught the formula for the volume of a sphere,
they determine a more realistic 1:1 ratio. But primarily, this activity can help
is possible to modify the texture of an egg white foam and, there-
them to understand what modelling can be, and how formal activities can
fore, of any preparation including such a foam. In addition, con-
help cooking, instead of following recipes (for the technical component of
stant links with aesthetic considerations can be made. culinary activities).
Each activity includes instructional objectives, a list of neces-
sary materials and the experimental protocol, plus a detailed
instructional discussion, which gives information to organize the Why is the egg white called “white” when it is transparent
workshops and to answer children’s questions. It should be noted and yellowish? Why is it liquid? Why do we observe areas of
that gestures that are difficult to describe are shown on down- different thickness? Why does the liquid egg white become solid
loadable films at the same time as the documents describing the when heated, and why does it become white and opaque? Why
activities. are the air bubbles that are often present when you break an egg
Let us consider the active aspects of these activities. Their white into a pan on the surface?
main objective is the development of the sense of observation Such observations and questions are valuable. The teacher will
and questioning. For example, when you break an egg and put not be able to answer them all, far from it, but he/she can arouse
the white in a pan, a thousand questions arise to those who know curiosity through observation and manipulation. Better still, a
how to look, such as: careful observation of the culinary world will lead to verbalization
Experimental Flavour Workshops
(a)
FIGURE 94.2 (a) One of the many description of eggs that can be found online. Sometimes, it is argued that the chalaza keeps the yolk in the middle of the
egg, but storing an egg in vinegar does not disrupt the chalaza and reveals that the yolk is indeed floating in the white (b). This is not surprising, as the egg white
is a 10% protein solution, whereas the yolk contains about 50% water, 15% proteins and 35% lipids.
activities; let’s not forget the idea of Lavoisier. Moreover, teachers
should not be afraid of questions that they cannot answer: this is
a possibility for group research, and one example for this is the
question “Where is the yolk inside an egg?” Here, many people
have the wrong “intuition” that the yolk is in the centre of the
egg (probably because this is what is shown in textbooks; see
Figure 94.2), or because this is what is said in some culinary books.
To extend the oral descriptions, children may be invited to have
a notebook where they can record the observations made during
these experimental activities. It is recommended that this “exer-
cise book” should be “private”, and that children do not write
during the workshops, but only afterwards: on the one hand, these
activities aim to emphasize the experimental, which is too often
excluded by an excessive formalism of instructional systems;
on the other hand, the daily experience of a laboratory shows
that keeping a laboratory notebook is difficult when conducting
experiments, even for science professionals; one can hardly
observe, experiment, think and write at the same time. Indeed, it
would be a pity to occupy experimental time with writing time. To
avoid this pitfall, it was instead proposed that synthesis sessions
take place after the sessions of experimentation; they can consist
of drawings, free editorials and discussion sessions directed by
the observations made.
For each activity, some extensions are described, but the possi-
bilities are innumerable, because the kitchen, with its utensils and
its ingredients, is the occasion for numerous simple experiments,
within reach of the children and at a reduced cost.
The use of a microscope, for example, reveals an unsuspected
world, which explains the macroscopic world. Obviously, chil-
dren who use such equipment will have to learn not only to
use it but also to “see” with it. Some activities are much more
interesting when such an instrument is available, but it is never
essential. Similarly, a thermometer and a scale considerably
extend the possibilities of technical, technological or scientific
exploration of the kitchen.
637
(b)
Instructional Efforts
An interest in cooking clearly acts as an invitation to experimen-
tation and exploration of the physical and chemical world, on the
ethics of such an approach (do we have the right to “spoil” eggs
when people are dying of hunger?), on the relationship between
science and culinary culture, and more, but all these consider-
ations are beyond the scope of this discussion.
One commentary is needed, however, on the general approach
of the various projects mentioned. They all have in common that
they are aiming for a method rather than knowledge. The activ-
ities aim to put the children on the path of questioning, to arouse
curiosity about everyday objects. The world is not obvious for
those who know how to see it, and the activities described here
aim, beyond the culinary culture they propagate, to show children
that a keen eye can be used to understand a lot.
Can the concept of “molecules” and “compounds”, so powerful
for these explorations, be discussed as early as in primary schools?
There was a fierce debate about this question, but one can observe
that children hear about these concepts daily, and that they are
useful for the interpretation of many phenomena, including phe-
nomena occurring during culinary processes. For sure, there is
an act of faith in admitting the existence of molecules, but the
activities proposed in the Flavour Experimental Workshops give
the possibility of facilitating the necessary conceptual leap; this
means analogy and identification. For example, if we compare
running sand and flowing water, we can give children the idea
that water may be made up of objects (too small to be seen by the
naked eye) that have a certain independence, analogous to that of
grains of sand. Those who fail to make the necessary conceptual
effort will gain from an idea formerly used by Albert Einstein,
and which receives the assent of children’s psychoanalysis: the
proposed “dance of molecules” (see this activity in the commen-
taries of Ministère de l’education nationale, 2000) invites chil-
dren to move like moving marbles, at different speeds. Having
638
thus brought abstract ideas to their bodies, the children better
understand the idea of a molecule; at least, it opens up in them an
intellectual “file”, which is no more “false” than the one that will
be opened, later, at high school level.
An Example Activity: The Whipped Egg White
Contest
Instructional Commentary
The Instructional Objectives
Understand what a foam is. Explore the colour of egg white.
Understand why egg white foams, but pure water does not.
Introduce the notions of liquid, gases, molecules and proteins.
Introduce the notion of reflection and absorption of light.
Experimental Sheet
Inquiry: egg whites are often used in the kitchen (e.g., in soufflés
or chocolate mousse); in these cases, we want to obtain a large
volume of firm foam. What is the maximum volume you can get
with an egg white? Why is it white? Why is it firm? We propose
that the children of the class, in pairs, have a competition to get
the biggest volume of egg whites.
Material for a class of 30 children:
18 eggs
15 large bowls (one in two children is asked to bring one
from home)
15 hand whips (one in two children is asked to bring one
from home)
Water (a jug, available to the teacher)
A large vessel (to collect the egg yolks)
A trash receptacle (for shells, etc.)
Protocol
1. The children are asked (one after the other and in front
of the group, telling the group what they are doing) to
break the eggs and to put the egg whites in their bowl.
The yolks are put in the communal vessel.
2. Common observation of the egg white, considerations
on the egg (structure, production …).
3. Children are asked to whip only once or twice in order
to get some large bubbles. Observation of the gesture:
one wonders what is the objective of the process of
whipping, and how, knowing this objective, one can per-
form it better. The bubbles are observed and described
(colour, shape, position …).
4. Comparison with pure water that is whipped (using a
clean whisk).
5. Whipping the egg white again until there are sev-
eral layers of bubbles big enough to be seen with the
unaided eye. Observation of the bubbles, of the general
colour of the system.
6. Whipping further by groups of two (the two members
of each group alternate). Observation of the foams
obtained.
Hervé This vo Kientza
7. Comparison of foams obtained by a jury composed of
the whole class.
8. Reflection on the method used to whip egg whites.
Consideration of limitations.
9. Add a tablespoon of water to the foams and whip. The
operation is repeated, and the class tries to make as
much foam as possible, collectively.
1. The children are asked to break the eggs and to put the
egg whites in their bowl. The yolks are put in the com-
munal vessel.
• The gesture of breaking an egg and separating the
yolk from the white is often known to school chil-
dren, but few children have practised it by them-
selves. The teacher has to organize this group by
group, in order to avoid all the eggs being poorly
broken, with the yolk mixed with the white. It is best
to invite a first child to do it (after he or she has said
what he or she is going to do), while the other chil-
dren analyse and comment. Finally the class will
understand that the shell can be broken by tapping
the centre of the egg against the rim of the bowl
(also, the method called Hazard Analysis Critical
Control Point –​HACCP, which proposes instead to
knock the egg on the table, can be discussed).
• The group will have to understand that it is then
necessary to turn the egg (above the bowl) when it
is broken, to put the break upward, to place the two
thumbs in the break and open, keeping the yellow in
one of the half shells. Then, the yolk has to be trans-
ferred from one half shell to the other by pouring the
white into the bowl.
• Children will be reminded to recover as much white
as possible for maximum foam volume. To make
them understand this point, it is often useful to
remind them that one gets more foam in the kitchen
with two egg whites than with only one.
• It will be pointed out that “to break an egg” means
generally separating the yolk from the white rather
than simply breaking the egg: therefore, when the
egg is passed from one half shell to the other, care
must be taken not to break the yolk; otherwise, it
will mix with white, and it will be difficult to sep-
arate them. A group can be allowed to keep some
yolk, and the effect on the foam will be observed.
• Note that for organizational reasons, it is better
to have the eggs broken successively, the whole
class commenting on the gesture made by one
child performing the separation of the egg white
and the yolk. This procedure also gives the oppor-
tunity to verbalize the gesture, which often leads
to a gradual improvement. The succession of
observations finally allows the constitution of a
collective experience.
• The yolks are together. If the “whipped egg white”
activity is done in the morning, the afternoon can be
used for experiments on yolks or shells.
Experimental Flavour Workshops
2. Observation of the egg white, considerations on the egg
(structure, production …)
• The class is invited to observe the egg white: it is
not white, but transparent, and the colour is yellow-​
green. Discussion of the name: the children will
probably say that cooked egg white is white. It will
be noted that this (familiar) phenomenon of egg
becoming white when cooked is quite extraordinary,
and this can lead to a discussion: do children know
other examples of products that harden on heating?
To continue, we can illuminate the cooked egg white
with coloured light (desk lamp hidden by a coloured
plastic interlayer) and observe that the white is then
no longer white, but coloured according to the colour
of the light it receives.
• Comment on the colour: what is the colour of egg
whites lit by coloured light (if possible, experi-
ment by lighting with a flashlight, in the dark, with
transparent coloured plastic in front of the lamp)?
Possibility of commenting on the colour of the
objects: absorption, emission.
• Discussion on the structure of the egg: yolk, white.
Observation of different areas in the white (by pla-
cing an egg white in a flat container, we see parts of
different thicknesses).
• Discussion on the production of the egg: presenta-
tion of documents showing the successive stages
of the constitution of an egg in the chicken, pres-
entation of documents on the industrial production
of eggs.
3. Children are asked to whip for some seconds in order to
get only a few bubbles.
  Observation of the gesture: one wonders what is
the objective of the operation and how, knowing this
objective, one can perform it better. The bubbles are
observed and described (colour, shape, position …).
• To make more understandable the role of whipping
in the formation of a foam (bubbles dispersed in
a liquid, here air dispersed in water), one moves
a pencil or a small wire in an egg white perpen-
dicularly to the surface: this does not introduce air
bubbles. On the other hand, if the same pencil or
wire is inclined and plunged, keeping a constant
angle with the horizontal, it pushes air into the liquid
and forms bubbles.
  It is deduced that to make a foam, composed of air
bubbles in the liquid that the egg white is, it will be
necessary to push air into the liquid with sufficient
whipping velocity. Also, the process will be more
efficient if the whisk has many wires: each wire
pushes air into the liquid. It should be added that
the handle must be big enough, otherwise one gets
cramps.
• Comments on the whipping action: if the shoulder
and arm are contracted, the children will get tired
quickly. Hence the need to whip only with the
wrist.
• The bubbles will be observed individually. Children
will be asked to describe them, and they will finally
639
be able to say that bubbles are transparent (they are
indeed composed of air), and the better observers
will be able to say, rightly, that only reflections on
the upper part of the bubbles are seen. Depending
on the observation conditions, the children can
count the number of reflections on each bubble and
see that it corresponds to the number of powerful
lamps in the room. In daylight, the reflection of the
light passing through the windows will be observed.
These reflections will often be white.
  Then, the bubbles will be illuminated with col-
oured light (as previously described, using coloured
plastic placed in front of a desk lamp), and it will be
observed that reflections of colour appear.
• It can also be observed that the bubbles formed are
on the surface of the liquid. Density considerations
(air less dense than water). Analysis of each bubble,
and diagram on the board. Each bubble is covered
with a thin film of liquid. One can try to pierce the
bubbles with a pencil point.
  Possible groupings of bubbles will be observed.
With the help of a pencil, we will try to ungroup
them, to move them.
• The class will conclude this part by asking why
bubbles are stable (relatively) in egg whites and not
in water. Hence, considerations on the constitution
of water and egg white.
  Note that these studies would be more difficult
with soap bubbles because they are more fragile.
• Water: without dwelling on this theme, children
will be asked what water is (observation of a glass
of water). They generally answer that it is a trans-
parent liquid. What is it made of, and why is the
water liquid? To introduce the constitution of water
by molecules (the word will be written on the board,
with etymology and history, as given by the dic-
tionary), we will take small balls that we put in a
glass, and then pour them into a salad bowl. It will
be explained that the molecules are like invisible
marbles (we can experiment, several times in a row,
with smaller and smaller beads, transparent if pos-
sible). However, the difference between beads and
molecules is their movement: the molecules are in
permanent motion even when the water is still, while
the balls do not move if the glass is still.
  Note that children –​even very young children –​
understand this idea perfectly. Those who have the
most difficulty in imagining the molecules will be
asked to close their eyes and to add balls after balls
in their mind.
• Egg white will be examined to find the molecular
differences that explain the differences in
foamability between egg white and pure water.
For example, one can gently heat an egg white in
a pan, and observe a release of white smoke; we
can explain the difference between vapour (invis-
ible) and smoke (visible because it is composed of
condensed droplets, so small that they are individu-
ally invisible).
640
Then the smoke will be condensed on a cold transparent sur-
face (a salad bowl or canteen bowl, for example). With a lid then
placed on the pan, we can get enough water on the lid to taste
it and conclude that the egg white contains water. At the end
of the operation, only a thin, yellow-​brown, transparent leaf is
left, resembling a sheet of gelatine. It will be explained that it
is composed of molecules that, in a whipped egg white, coat the
air bubbles and stabilize them in the water. These molecules are
called “proteins”.
4. Comparison with whipping pure water.
Here, bubbles are formed, but they are not stable. It is concluded
that the bubbles are not stabilized, so that they explode in the air.
We corroborate the previous explanation.
5. We resume the whipping until there are several layers
of big enough bubbles in the white. Observation of the
bubbles, of the general colour.
• It will be observed that the bubbles become more and
more numerous, and smaller and smaller. The chil-
dren will be questioned in order to be able to say that
the whipping repeatedly divides the already formed
bubbles while introducing new ones. Constitution of
several layers of bubbles.
• Then, we will wonder about the white colour that
appears gradually (from about three layers of
bubbles: we can manipulate the bubbles with a
pencil to see from what number of layers the white
colour appears).
  We will observe each bubble in search of the lumi-
nous reflections already observed. We conclude that
each bubble still has reflections, that the bubbles
become almost imperceptible to the naked eye, but
that we continue to see reflections, more and more
numerous. Finally, we only see the reflections, and
more bubbles. Because the reflections are white, the
whipped egg white appears white.
• Then, we will observe the foam using coloured
light, and we will see that it is the colour of the light.
Interpretation in terms of reflections.
This can lead to a discussion on the naming of “whipped egg
white” in relation to colour. Eventually, comparison of the whisk
with the whip used for horses.
6. We continue the whipping (the two students of each
group alternate). Observation of the foams obtained.
• Now, the question is to know when whipped egg
whites are sufficiently beaten. What is a “firm”
whipped egg white, and is it the same as having
the maximum volume of foam? Generally, children
think that whipped egg whites are firm when they
do not sink if the bowl is turned over. It should be
noted that the professionals beat until the egg whites
support an entire egg, in its shell, without it sinking.
  Observation of the foam at this stage: the bubbles
are no longer visible to the naked eye, but a magni-
fying glass still allows them to be seen.
Hervé This vo Kientza
• Colour: we will continue the observations of (5),
looking at the bubbles with a magnifying glass and
seeing the reflections. We conclude that whipped
egg whites are white because bubbles, which are
invisible to the naked eye, reflect white light.
Hence the question that will be introduced: if the light is reflected
by the bubbles, it does not enter the whites; therefore, what is the
colour inside a whipped egg white? Discussion to organize, to
conclude that it is probably black (no light).
To check this, the participants could put a transparent plate
diagonally in a salad bowl and look at a light through layers of
increasing thickness of whipped egg white. We will see that from
15 cm, the light no longer reaches the eye. On the other hand, if we
place ourselves on the side of the light, we will see that the white
colour is constituted progressively, from the thinnest zones (very
few reflections) to the thickest zones (almost total reflection).
We conclude that a whipped egg white of more than 15 cm
radius would have a dark interior.
* Firmness of the formed foams: students might be encouraged
to wonder why the whipped egg white is firm, since it is composed
of egg white, which is liquid, and air, which is gaseous.
To explain it, we will again take the preceding observation,
where the few bubbles could move. In the well-​beaten white, the
bubbles are packed against each other, and they do not individu-
ally move easily. If none can move, the set does not move easily,
and it does not flow. We can introduce the term “foam”, which is
the one by which physicists describe such a system.
7. Comparison of foams obtained by a jury composed of
the entire class.
• Note that it is difficult to measure the volume of the
foams. Can we make a regular upper surface to better
appreciate the height? (Answer: yes). By moving the
whites, do not we risk making them fall? (Answer:
yes, but only a little).
• This evaluation must show that apart from outlier
cases (presence of yolk, loss of white that would
have overflowed during whipping …), all foams
have a volume of the same order of magnitude.
The jury might thus disagree with the competitors
without an objective measure.
  Discussion of better ways to judge, focussing
(under discussion) on comparison or measurement
methods.
8. Reflection on the method used to whip egg whites, and
consideration of limitations.
• The question is why the volume of whipped egg
white is limited to that which is finally observed. By
changing the method, would you get more volume?
Here, we whipped, but we could also make bubbles
come from the bottom, just like bubbles are formed
when we blow with a straw in a glass. What volume
would we get then? What would be the size of the
bubbles? Participants should experiment with this.
• Why is the volume limited? The analysis of this
question will be conducted by having the children
list the components of the system: essentially water,
proteins and air.
Experimental Flavour Workshops
We will go on to say that we get more whipped egg whites when
we use more egg white. For a quantity of egg white limited to
one, it will be understood that we lack either water, protein or air.
Then, children will be told that the air is not lacking. So, we
lack either water or protein. How can we know which of the two
is the limiting element? A list of conjectures provided by the chil-
dren will be listed.
Then we’ll say we can experiment to find out: add water, or add
protein. Students will be asked what, in their opinion, is the sim-
plest experiment. And we will conclude that we must add water.
9. Add water to the foams and whip. We are thus looking
to see what maximum volume of foam can be achieved.
• We will make a collective experiment: in a first
bowl, we will put a spoonful of water; in a second,
two spoonfuls, etc. We will resume the whipping,
and we will observe that the volume obtained
increases with the amount of water added (for the
larger amounts of water added, we will add the
water gradually).
  From this experiment, it will be deduced that
water was the limiting factor for obtaining a higher
volume of foam. It will be observed that the foam
is, however, more fragile than with the pure egg
white.
• Hence the question: what is the maximum volume
that can be achieved? If one is content to add
water, this volume will ultimately be limited by the
amount of protein (about 10% of an egg white, by
mass). We will show a whipped egg white, with
one or two layers of protein around each bubble,
and give the method of determining the maximum
volume of whipped egg white: gradually add water
while whipping. It should be noted that experience
shows that it is easy to obtain several litres of foam
from a single egg white (our record was over 40
litres).
• Conclusion on the experimental method: we will
point out the method that has been used. We started
from an observation, and we tried to understand,
by assuming the existence of proteins, molecules
with surfactant properties (that is to say here,
“foaming”). Then, we analysed the formation
of whipped egg whites, and it was expected that
the water was missing, which was confirmed by
experiment.
Possible Extensions
• To continue the experiment on the resistance of the egg-
shell, one can perform an experiment, which is to lay
a wooden plate on rolls of toilet paper and to mount
several on the plate without the rolls being crushed.
Experience shows that plate-​like materials generally
resist well in compression.
• Another experiment can be usefully done with shells:
this consists of covering shells with white alcohol
vinegar (also called crystal vinegar). Bubbles appear
641
immediately. We can then explore this reaction (later
sheet).
• With the yolks, it will be possible to test for the presence
of water in the yolks by heating them gently and then
examining the condensation of steam on a glass placed
above the pan.
• We can continue the study of the colours of the materials
by illuminating various objects coloured by various
colours, always using the system of coloured dividers,
or by using tops made of a cardboard disk fixed on a
pencil and coloured in various ways.
• With the most disciplined classes, we can “play
molecules”: some students will move straight on, boun-
cing on obstacles (walls or other students) with an angle
of incidence equal to the angle of reflection.
  Three situations will be repeated: liquid at room tem-
perature (the children will have constant, noticeable and
different speeds); solid (they will hold hands, vibrating
on the spot, like water molecules in ice); and steam
(they will move at high speed, occupying all available
space).
• Presentation of proteins: we will explain that they are
molecules. We will present a simple model: the proteins
are like pearl necklaces packed on themselves and
composed of beads of 20 kinds (we can make such a
necklace, with pearls in blue tones and pearls in red
tones), and explain that some pearls (the red pearls) are
hydrophobic (like oil, they do not mix well with water:
experiment with pouring oil on water, and whipping to
observe the phenomena that occur), while other pearls
(blue pearls) are hydrophilic (like alcohol at 90°, they
mix perfectly with water). To refine the protein model,
we will put the necklace in a configuration where the
red beads are at the heart and the blue beads outside
the cluster.
• For children who ask, we could explain why the egg
white proteins stabilize the air bubbles: the whipping
unfolds them, so that they spontaneously place their
hydrophobic part against the air bubbles, which are
also hydrophobic. The air bubbles, thus covered with
unwrapped proteins, are stabilized in the water.
• You can explore the cooking of the hard-​boiled egg:
recipe collection, experimental analysis of procedures,
rational methods to obtain a yolk centred in the white,
conservation, recognition of a fresh egg and a hard-​
boiled egg, and concepts such as buoyancy.
• Study of the eggshell: observation of the membrane
(“chorion”), study of the resistance of the chorion (it
is dried), measurement of the mass of the shell of sev-
eral eggs (weighing), and comparison of the masses of
various parts of eggs (yolk, white and shell).
• This study can be extended by comparing egg whites
beaten with a little salt or a little lemon juice. We could
also compare with whites where we put a drop of oil,
and whites where we put some egg yolk.
• Geometric tracing of the egg. We work on a wooden
board, on which we fix a sheet, attached by pins. One
end of a string is tied to a thumbtack, and two other
thumbtacks are planted on the board. The free end of
the string is attached to a pencil. Then, we stretch the
642
string and draw the curve that is established when we
rotate the string around its axis.
• Different results are obtained according to the relative
positions of the three bugs. When the three pins are
aligned, we obtain a symmetrical curved line: that of the
egg. The gap between the two free pins varies the con-
trast between the two tops of the egg. When the three
bugs are not aligned, we get an asymmetrical egg.
Conclusion
It is proposed that schools can be the place where chemistry is
introduced, if it is shown in an experimental form and without
formalism. How will children develop if they understand, from
school on, why whipped egg whites are white (while they are
a foam, that is to say, a dispersion of transparent air bubbles
in transparent and yellowish egg white) and why they are firm
(while they are composed of air bubbles in a liquid)? This is an
exciting question, isn’t it?
REFERENCES
Brillat-​Savarin JA. 1825. La physiologie du goût, chez l’auteur,
Paris, France (English translation https://​ebooks.adelaide.edu.
au/​b/​brillat/​savarin/​b85p/​”https://​ebooks.adelaide.edu.au/​b/​
brillat/​savarin/​b85p/)​
Hervé This vo Kientza
Faraday M. 1842. Chemical manipulations. https://​ archive.org/​
details/​chemicalmanipul04faragoog/​page/​n9
Hands on. 2019. https://​ed.fnal.gov/​lsc/​ed_​lsc_​facts.shtml
Lavoisier A. 1789. Elements of Chemistry, In a New Systematic
Order, Containing all the Modern Discoveries (Translator:
Robert Kerr), www.gutenberg.org/​cache/​epub/​30775/​pg30
775.txt
Main à la pâte. 2019. www.fondation-​lamap.org/​
Ministère de l’éducation nationale. 2000. www.ac-​paris.fr/​portail/​
jcms/​p1_​80293/​ateliers-​science-​cuisine
Montaigne M. 1581. Journal de voyage en Alsace et en Suisse.
https://​data.bnf.fr/​11970594/​michel_​de_​montaigne_​journal_​
de_​voyage/​
Pomiane E. 1922. Bien manger pour bien vivre. Le code de la bonne
chère, Albin Michel, Paris.
Pomiane E. 1931. La cuisine et le raisonnement, Société du gaz de
Paris, Paris.
Pomiane E.1948. Gastrotechnie. La cuisine est un laboratoire,
Atomes, 29, 255–​258.
This H. 1992. Les secrets de la casserole, Belin, Paris.
This H. 1993. Révélations gastronomiques, Belin, Paris.
This H. 1995. Science et gastronomie, Pour la Science, Paris.
This H. 1997. La casserole des enfants, Belin, Paris.
This H. 2002. Traité élémentaire de cuisine, éditions Belin, Paris.
This H. 2006. Pourquoi la cuisine n’est pas une science? Sciences des
aliments, 26 (3), 201–​210.
Teaching Argumentation and Inquiry through Culinary Claims

Erik Fooladi
Department of Science and Mathematics, Faculty of Humanities and Education, Volda University College,
P.O. Box 500 Volda, Norway

Introduction be experienced as relevant to learners’ own lives (Sadler, 2009;

Aikenhead, 2006). Consequently, we need some rationalization
The present chapter is an adapted version of a previously
for how science relates to society, and one model was described
published article (Fooladi, 2013a) and places itself in the field of
by Roberts and Gott (2010), as illustrated in Figure 95.1. The
science-​and cross-​curricular education research, also including
model consists of three segments: design and conduct of
culinary education. A research-​based teaching approach has been
experiments, or second-​hand inquiry into findings by others, to
developed, tested and refined over a five-​year period in tertiary
produce evidence (left-​hand side); argumentation to generate a
education in a mixed group of 20–​35 undergraduate students
claim based on this evidence (middle); and the claim as a factor
(each year) from science teacher and home economics courses.
in a socio-​scientific issue or decision to be made in an everyday
Taking its departure from perspectives in molecular gastronomy,
situation (right-​hand side).
it has sought and met some major challenges in (science and
According to the same authors, one might envision that a
cross-curricular) education. These challenges include argumenta-
researcher would mainly work from left to right in this model,
tion skills, inquiry in scientific and everyday contexts, and source
“looking forwards”, whereas the perspective of a scientific-
awareness and critical thinking. This is predominantly framed in
ally literate person/​citizen would be to work from right to left
theoretical and empirical perspectives from science education
(“looking back”), to retrace the evidence for a given claim related
research. Thus, it is sought to draw benefits from both science
to a societal or “everyday” issue. In the work described herein,
education research, a field where culinary practices are not vis-
this model is employed to demonstrate that starting from claims/​
ibly present, and molecular gastronomy, where educational
specifications about cooking (culinary claims; see later), one may
research has thus far not been strongly present.
deal with all these three aspects of science in society in a context
that is close to learners’ everyday lives (on the issue of “context”
in science education, see e.g. Gilbert et al., 2011). Herein, this
Background context is given through relevant content from culinary practices,
Argumentation, Inquiry and Socio-​Scientific Issues in namely, claims and specifications about cooking. Culinary
practices can be seen to draw their knowledge, procedures and
Education
ways of thinking from two different spheres: on the one hand,
Within the international science education community and among natural sciences (e.g., food science, health and nutrition) and on
policy makers, the development of quality teaching methods to the other hand, the practical craftsmanship of cooking, which is
promote scientific literacy, focusing not only on what we know characterized by experiential knowledge communicated orally or
but also on how we know and why we do so, has been defined as through written recipes (Sutton, 2006). As a result, and as will
a major challenge (e.g., Driver et al., 2000; Osborne and Millar, be demonstrated below, culinary practices may offer fruitful
1998; Rocard et al., 2007). Consequently, recent years have contexts for teaching, concerning not only declarative content
seen an increasing amount of research and educational material/​ knowledge (a subject’s “facts”) but also scientific methods and
resources focusing on the promotion of various cognitive skills ways of thinking.
in science education, such as talking, reading and writing science
(Fang et al., 2010; Wellington and Osborne, 2001) as well as
Toulmin’s Argumentation Pattern (TAP) and Education
the development of students’ competencies in reasoning and
argumentation (Driver et al., 2000; Duschl and Osborne, 2002; Throughout the last three to four decades, science education
Erduran and Jiménez-​Aleixandre, 2008; Osborne et al., 2004). researchers have seen the importance of discourse in science
Science content knowledge, argumentation and inquiry do not learning, and there appears to be consensus that argumentation
exist in a vacuum but are inherent parts of society. Thus, science is a form of discourse that needs to be taught explicitly through
teaching that takes into consideration science as a societal phe- suitable instruction (Erduran et al., 2004). Simply observing
nomenon has been promoted for achieving an education that can or being part of a discussion is not sufficient, and the learner

643
644 Erik Fooladi

FIGURE 95.1 Model for practical work, declarative/​content knowledge, argumentation and science in a societal context according to Roberts and Gott (2010).
(Figure used with permission)

needs to be engaged in a meta-​discussion about argumentative

practices for such instruction to be effective. Consequently,
researchers have sought heuristics and tools to deal with
these issues. In educational research, Toulmin’s argumenta-
tion pattern (“TAP”) developed by the British philosopher
Stephen Toulmin as a tool/​heuristic within informal reasoning
(Toulmin, 1958/​2003), has gained a foothold as an analytical
tool for researchers studying argumentative discourse among
learners (Erduran et al., 2004; Jimenez-​Aleixandre et al., 2000;
Zohar and Nemet, 2002). The essential parts of an argument
according to Toulmin are often displayed schematically, as in
Figure 95.2, and are:

Claim – the assertion/​specification put forth.

Data/​facts – facts given in support of the claim.
Warrant – to provide the connection between data and FIGURE 95.2 Toulmin’s argumentation pattern.
claim. (Toulmin, 1958/​2003)
Backing – assumptions commonly agreed upon which
support the warrant.
Qualifier – special conditions under which the claim 2004) and the epistemic nature of their own discipline (Simon
holds true. et al., 2006). It has also been introduced directly to 12–​13-​year-​
Rebuttal – statements that contradict either of the other old school pupils as part of nature-​of-​science and science literacy
elements. topics (Simon et al., 2006). In Osborne and coworkers’ words:
“the use of these features of TAP offers teachers a richer metalan-
Revisiting Figure 95.1 (middle part), it is seen that this is guage for talking about science and for understanding the nature
exactly what is used as the link between the left-​hand and right-​ of their own discipline” (Osborne et al., 2004). Here, the authors
hand sides in the model by Roberts and Gott (2010). TAP can form refer to the detailed elements of Toulmin’s pattern (see Figure
a bridge between, on the one hand, science content knowledge 95.2) as opposed to the more general/less specific terms “ideas”
and experiment, and on the other hand, a socio-​scientific issue. (claims) and “evidence” (data, warrants, etc.). Hence, TAP allows
In some cases, TAP has also been used in the explicit teaching the instruction of argumentation in a way that explicates the rhet-
of argumentation, as a tool for teachers to gain an increased orical elements of an argument as it is commonly used in natural
understanding of discourse in their own classroom (Osborne et al., settings (Driver et al., 2000; Toulmin, 1958/​2003).
Teaching Argumentation and Inquiry 645

The “Kitchen Stories” Project to apply it when structuring their argument. The constructed argu-
In 2009, a project was started at Volda University College to ment need not be true per se, but it should be coherent and plaus-
approach inquiry and argumentation in a more explicit manner ible when placing oneself in the shoes of the source of the claim.
than previously for undergraduate pre-​service teacher students Indeed, retracing someone’s argument to uncover erroneous facts
in science and home economics. The project, named “Kitchen would be to uncover a myth, and in those cases, students would
stories”, drew inspiration from molecular gastronomy and used be asked to give a colour code to the element in the TAP that was
what This (2005) has termed “culinary precisions” as an approach (expectedly) not correct.
to teaching argumentation and inquiry. Such culinary precisions Second phase (2–​3 weeks’ duration, part-​time)
can, in the context of informal reasoning, be considered as claims,
Step 3 –​Test/​experiment analysed claims
and thus, the term “culinary claims” is used herein. Culinary
claims can be defined as “the technical or procedural information Each group selects one or two of the analysed claims for experi-
present in a recipe (oral or written), which provides added value mental testing. Based on their prior analysis, the experiment must
in terms of improved quality and greater chance of a successful be designed in detail and carried out accordingly.
product” (Fooladi and Hopia, 2013, 2). Examples are: Step 4 –​Record and publish results, documentation of
cultural heritage
• You can’t make jelly containing fresh kiwi because then
The students must record and publish the culinary claims, ana-
the jelly will not set.
• Cucumbers decay/​rot more quickly if stored together lyses and experimental results, thus disseminating knowledge of
with tomatoes. scientific and/or food cultural nature. A culinary claim handed
• Leavened bakery will rise more if baked at high (flow) down orally can in some cases be experiential knowledge that
tide compared with ebb tide. would otherwise have been lost unless the students had collected
it. This was done by publication in a wiki.
The research design of the project draws inspiration from design-​
based research in which a research-​based teaching sequence is Case Example: You Can’t Make Jelly Containing
designed, implemented and analysed (Cobb, 2001; Edelson, Fresh Kiwi
2002; Juuti and Lavonen, 2006). Throughout the process, data is
Below follows a description of how a group of four pre-​service
collected by various means such as observation, interview, stu-
science teacher students carried out part of their project based
dent text analysis and tests/​questionnaires. The teaching sequence
on the structure given in the previous section. The sample is
is then revised (re-​designed), another cycle is conducted, and
taken from the first design cycle, which was carried out in 2009.
so forth. In the process, data is also collected to shed light on
Note that this might be adapted to different educational levels
selected educational issues, thus at the same time producing
as long as modifications are made to suit age group, availability
research-​based knowledge about teaching and learning.
of equipment and so forth, one example being Vartiainen et al.
(2011), who have described the use of culinary claims as an
Practical Approach in the “Kitchen Stories” Project approach to inquiry in lower secondary school.
The concrete approach to culinary claims used in teaching pre-​
service teacher students in this project is as follows: First phase (3 weeks)
Step 1 –​Collection and documentation: The group collected 18
First phase (2–​3 weeks’ duration, part-​time) culinary claims from various sources. The claims were listed in
Step 1 –​Collect and document culinary claims a document together with their respective sources. Among these,
The students can find culinary claims by interviewing family, four were selected as promising candidates for close analysis
professionals (e.g. chef, supermarket employee, artisan, etc.) or (claims that are difficult to test due to methodological obstacles
others; they can search in literature and cookbooks, food pages or lack of resources are usually ruled out, such as health claims,
on the internet and so forth. The source of each culinary claim traditional rules for slaughtering animals, time-​of-​year-​dependent
must be documented and situated in space and time: where it was claims and so forth).
found, when it was published or stated. Each student must collect Step 2 –​ Analysis: The students attended lectures about argu-
at least four culinary claims. The students must also decide mentation, including an introduction to Toulmin’s argumentation
whether it would be possible to test the claim through experiment. pattern. The students had to collect data/​facts, warrants/​reasons,
rebuttals, etc. and construct a complete and plausible, though not
Step 2 –​Analyse and construct plausible arguments necessarily correct, argument; see Figure 95.3.
for a few selected claims
The group selects one claim for each group member for closer ana- Second phase (3 weeks)
lysis: What is the actual claim? What subject matter or evidence Step 3 –​Experiment: Through the data collection process and
is hidden behind this claim? Are there facts and justifications that argumentation, the students decided that the claim about kiwi in
support, weaken or contradict the claim? Here, TAP is introduced jelly was also suitable for experimental testing. The hypothesis
for explicit teaching of argumentation, and students are required was: the jelly does not set because protease enzymes in fresh kiwi
646
FIGURE 95.3 Example of TAP applied to a culinary claim about making jelly with fresh kiwi. The TAP was completed, starting from only the claim, using
inference and scientific content knowledge derived through second-​hand inquiry.
FIGURE 95.4 Jellies with kiwi.
(Photo by Dagfinn Bakklund; used with permission)
break down the gelatine proteins in jelly. Since enzymes lose their
functionality when heated, blanching the kiwi might render the
protease enzymes inactive, and hence, the jelly would set properly.
Furthermore, there were at the time two sorts of jelly available
in grocery shops: standard jelly based on gelatine (a protein) and
“fast-​setting jelly” based on locust bean gum (a carbohydrate). If
the problem is due to proteases in kiwi, which are specific for
degrading proteins, the jelly should set properly when using the
carbohydrate-​based jelly. Figure 95.4 elegantly demonstrates how
the different parallel experiments shed light upon this argument.
Step 4 –​Documentation and publication: The students
were introduced to the IMRaD structure of writing scientific
papers (IMRaD is an abbreviation for Introduction, Methods
(and materials), Results and Discussion), a structure that has
Erik Fooladi
dominated as the template for writing scientific papers in several
sciences (Sollaci and Pereira, 2004), and their report had to be
written following this structure.
Results and Discussion
Research Questions, Observations, Preliminary Results
and Discussion
Some research questions from this educational project have been:
• To what extent is it possible to consider culinary claims
as parts of formal arguments, seen from an epistemic or
argumentation theoretical perspective?
• Can culinary claims be used as an approach to teaching
about argumentation?
• To what extent is TAP a suitable tool in dealing with
culinary claims, be it educational or as a heuristic in
structuring kitchen stories as arguments?
• In what sense can culinary claims be used to promote
inquiry activities in science and culinary contexts
(herein, the school subject home economics)?
• Which other relevant topics, if any, appear in the pro-
cess of collecting and exploring culinary claims?
Initial results and the accompanying discussion are based on sys-
tematic observations, evaluation of student assignments, and field
notes during design, implementation and revision of the teaching
sequences.
Culinary Claims, TAP and Argumentation Instruction
Through the project, it is evident that it is possible to use TAP
in the exploration of culinary claims, consequently supporting
the hypothesis that it is possible to construct formal arguments
Teaching Argumentation and Inquiry 647

based on culinary claims. Although the students reported it to Epistemic Status: Source Awareness, Sourcing Skills
be intellectually demanding, they were able to use this tool and Critical Thinking
for structuring arguments, and all groups were ultimately
In the last decades, awareness and evaluation skills related to
successful in producing coherent arguments, even though TAP
trustworthiness and credibility of information sources have
is said to be difficult to apply in analysis of real-​life verbal data
become increasingly important among both experts and the non-​
(Erduran et al., 2004). However, as opposed to verbal data,
expert members of society (e.g. Bråten et al., 2011; Norris and
“Kitchen stories” deals with arguments of a different kind
Phillips, 1994; Wellington and Osborne, 2001). We are constantly
and might thus be an arena where TAP is easier to apply, with
bombarded with large amounts of information and statements,
a positive contribution to argument analysis. In the study of
often of competing nature, which standard textbooks seldom pro-
existing verbal data, TAP is used to deconstruct an argument
vide help in dealing with. In the “Kitchen stories” project, the
(Erduran et al., 2004), whereas the role of TAP in “Kitchen
students naturally used a broad selection of information sources
stories” is to structure or construct an argument. The stu-
in addition to course textbooks. After the first round of implemen-
dent groups involved in this project have been mixed groups
tation, it became evident that source awareness was an important
with respect to subject as well as academic achievement
matter to consider, and this was one of two major revisions in
levels. However, all groups were able to construct TAPs, but
the design between the first and second cycles. Hence, a source
lower-​achieving students required more guidance than high-​
credibility step was introduced, in which the students were to rate
achieving students (groups received one or two sessions of
every source they used on one of two scales (Fooladi, 2013b).
supervision for each phase). A challenge arising in this con-
If the students considered a source to be of scientific nature,
text is that one single culinary claim might have different sets
they were to rate it on a six-​point scale between 1 (lowest, e.g.
of facts, warrants, rebuttals and so forth. Consequently, to dig
“internet with no other references”, “otherwise undocumented
deep into a culinary claim, one might either build several TAPs
old wives’ tale”) and 6 (highest, e.g. “scientific literature on inter-
or produce one rather complex TAP.
national level”). However, the source could also be a (more or
less experienced) craftsperson, in which case rating on a scale
Promoting Minds-​On as Well as Hands-​On of scientific credibility might not be appropriate. These sources
Practical Work were rated on a scale from A (lowest, no relevant experience) to F
Often, practical work (e.g. lab work) is carried out simply to con- (highest, e.g. an expert chef or artisan); see Figure 95.5.
firm textbook content rather than inquire into authentic questions. In evaluating the sources, the students themselves had to
Furthermore, a common problem is lack of coherence between, choose whether a source should be rated on the scientific or the
on the one hand, the practical work/​experiments and, on the other craftsmanship scale, or both. The students also had to rate their
hand, content knowledge, scientific methods and scientific ways own experiments on one of these scales, since their experimental
of thinking (Abrahams and Millar, 2008). However, achieving results often constituted a part of the final argument. Hence, a
high-​quality minds-​ on practical instruction is not straightfor- byproduct of the “Kitchen stories” project has been not only
ward. In the “Kitchen stories” project, the traditional pattern is an increased focus on a wider selection of information sources,
turned upside down, because the students spend a lot of time but also a strong emphasis on epistemic status of information,
asking questions, discussing, searching and conducting second-​ an increasingly important topic in education in recent decades.
hand inquiry (Palincsar and Magnusson, 2001) before the actual The context afforded by cooking is unique in this respect, since
experiment is planned and conducted. Hence, the occurrence of it represents a meeting point between science and craftsman-
genuine “minds-​on experimenting” seems to be predominant. No ship. As students were asked to rate their own experiments,
group (a total of 26 groups throughout four implementations) self-​monitoring and self-​evaluation became a natural part of the
had problems finding researchable questions among the culinary project, which is considered a high-​hanging goal in education
claims they collected. This may be a result of the requirement for (for a seminal article, see Sadler, 1989).
each group to collect a large number of culinary claims to select
from in later stages.
Using TAP seems to scaffold the students’ inquiry in a positive Kitchen Stories and Declarative Knowledge
manner, and they apparently adopt a shared lexicon and structure One potential problem in this project is that the teacher has limited
of argumentative reasoning, which was one of the main aims of control over which declarative knowledge (factual/​ textbook-​
the project. Notably, when students were offered the hypothetico-​ type knowledge) is covered, at least if the students select freely
deductive method as alternative support, they reported that TAP among the collected claims. However, Vartiainen et al. (2011)
and the IMRaD structure combined was a sufficient scaffold have shown that relevant chemistry topics do arise naturally in
to carry out their project. (The hypothetico-deductive method, the process, as was also observed in this project. Hence, “Kitchen
HDM, is a model much used to display the process of scientific stories” affords ample opportunities for teaching declarative
inquiry, occasionally described as “the scientific method”. It is knowledge in addition to the mentioned procedural knowledge
also commonly used to scaffold inquiry teaching.) The products and reasoning, albeit less control over the actual content covered.
of the students’ work, as assessed by the lecturer, support this If high control over declarative knowledge is required, working
notion. from predefined claims or delimiting the areas from which
648 Erik Fooladi

FIGURE 95.5 Scales for students’ judgements of source credibility.

(Fooladi, 2013b)

FIGURE 95.6 The process in a “Kitchen stories” project, with participants starting from a culinary claim “looking back” along the line of plausible argument
and evidence. This is followed by the opposite approach “looking forward” as the students assume the role of a researcher.
(Figure adapted from Roberts and Gott, 2010. Used with permission)

claims may be collected (e.g., only claims about meat or proteins
allowed) would be options to consider.
Conclusions
Reconsidering Roberts and Gott’s (2010) model of science in
society (Figure 95.1), in “Kitchen stories”, the participants start
out by looking back (from the societal perspective), retracing, or
“unpicking”, possible arguments for a certain claim, taking the
role of an inquisitive cook or food professional. The students must
then assume the role of the researcher and start “looking forward”:
based on their constructed argument, they must draw up an experi-
mental design, carry out the experiment and move all the way to
the right-​hand side through a coherent argument, ending up with
a publication related to the societal issue at hand (Figure 95.6).
Teaching Argumentation and Inquiry
Through this process, the students have encountered real-​
world questions, dealt with argumentation in an explicit manner,
designed and carried out one or more experiments, documented
results based on evidence from their own experiment as well as
second-​hand sources in an argumentative manner, and finally
conveyed their findings to the public. Furthermore, the students
are part of a project in which data is collected for the common
good of society by documenting culinary knowledge/​heritage; in
a sense, they assume the role of true researchers and not only
students.
Outlook
One matter that would benefit from further studies is the epistemic
aspects of this multidisciplinary setting. The possible tension
between, and the possibilities afforded by, scientific knowledge
and methods, on the one hand, and the epistemic characteristics
and values of experience-​based food and cooking (craftsman-
ship), on the other hand, represents a highly relevant issue when
seen in the context of science in society. The fact that science is
in constant change and development is often overlooked in school
science (Abd-​El-​Khalick and Lederman, 2000; Vesterinen et al.,
2009). In dealing with culinary claims, science does not always
have the one true answer or may not yet have investigated all
relevant issues. Thus, the “Kitchen stories” approach, or more
generally using culinary claims at various educational levels as
well as in informal contexts, is one avenue to approach this. Other
examples of efforts to use culinary claims for learning and devel-
opment exist, from compulsory education to life-​long learning in
non-​formal contexts, such as in France (This, 2019), in Finland
(Vartiainen et al., 2011) and in Finland/​Norway (Fooladi and
Hopia, 2014). However, educational research on the matter is
still scarce. Furthermore, it would be no exaggeration to say that
culinary claims constitute a promising arena for the interaction
between science and society (Fooladi and Hopia, 2013), and
future efforts should be welcomed.
REFERENCES
Abd-​El-​Khalick F, Lederman NG. 2000. Improving science teachers’
conceptions of nature of science: a critical review of the lit-
erature. International Journal of Science Education 22 (7),
665–​701.
Abrahams I, Millar R. 2008. Does practical work really work?
A study of the effectiveness of practical work as a teaching
and learning method in school science. International Journal
of Science Education 30 (14), 1945–​1969.
Aikenhead GS. 2006. Science Education for Everyday Life: Evidence-​
Based Practice, Ways of knowing in science and mathematics
series. Teachers College Press, New York, NY.
Bråten I, Strømsø HI, Salmerón L. 2011. Trust and mistrust when
students read multiple information sources about climate
change. Learning and Instruction, 21 (2), 180–​192.
Cobb P. 2001. Supporting the improvement of learning and
teaching in social and institutional context. In Cognition and
Instruction: Twenty-​Five Years of Progress (Carver SM, Klahr
DK eds.), Lawrence Erlbaum Associates, Mahwah, New
Jersey.
649
Driver R, Newton P, Osborne J. 2000. Establishing the norms of sci-
entific argumentation in classrooms. Science Education, 84
(3), 287–​312.
Duschl RA, Osborne J. 2002. Supporting and promoting argu-
mentation discourse in science education. Studies in Science
Education, 38 (1), 39–​72.
Edelson DC. 2002. Design research: What we learn when we
engage in design. Journal of the Learning Sciences, 11 (1),
105–​121.
Erduran S, Jiménez-​Aleixandre MPS (eds.). 2008. Argumentation
in Science Education: Perspectives from Classroom-​ Based
Research. Springer Science, Dordrecht, Germany.
Erduran S, Simon S, Osborne J. 2004. TAPping into argumenta-
tion: Developments in the application of Toulmin’s Argument
Pattern for studying science discourse. Science Education, 88
(6), 915–​933.
Fang Z, Abell SK, Lamme LL, Pringle RM. 2010. Language and
Literacy in Inquiry-​Based Science Classrooms, Grades 3–​8.
Corwin Press, Thousand Oaks, California.
Fooladi E. 2013a. “Kitchen stories”: Argumentation and inquiry
in the context of food and cooking In ESERA 2013 confer-
ence. Nicosia, Cyprus, European Science Education Research
Association (ESERA).
Fooladi E. 2013b. Molecular gastronomy in science and cross-​
curricular education –​The case of “Kitchen stories”. LUMAT –​
Research and Practice in Math, Science and Technology
Education, 1 (2), 17–​30.
Fooladi E, Hopia A. 2013. Culinary precisions as a platform for
interdisciplinary dialogue. Flavour, 2 (6), www.flavourjournal.
com/​content/​2/​1/​6
Fooladi E, Hopia A. 2014. Culinary claims as entry to the “how we
know”s of science in an informal learning environment. In
NFSUN 2014 conference. University of Helsinki, Finland.
Gilbert JK, Bulte AMW, Pilot A. 2011. Concept development and
transfer in context-​ based science education. International
Journal of Science Education, 33 (6), 817–​837.
Jimenez-​Aleixandre MP, Rodriguez AB, Duschl RA. 2000. “Doing
the lesson” or “Doing science”: Argument in high school gen-
etics. Science Education, 84 (6), 757–​792.
Juuti K, Lavonen J. 2006. Design-​based research in science educa-
tion: One step towards methodology. NorDiNa, 2 (4), 54–​68.
Norris SP, Phillips LM. 1994. Interpreting pragmatic meaning when
reading popular reports of science. Journal of Research in
Science Teaching, 31 (9), 947–​967.
Osborne J, Erduran S, Simon S. 2004. Enhancing the quality of argu-
mentation in school science. Journal of Research in Science
Teaching, 41 (10), 994–​1020.
Osborne J, Millar R. 1998. Beyond 2000: Science Education for the
Future. King’s College, London, UK.
Palincsar AS, Magnusson S. 2001. The interplay of first-​hand and
second-​hand investigations to model and support the devel-
opment of scientific knowledge and reasoning. In Cognition
and Instruction: Twenty-​Five Years of Progress (Carver SM,
Klahr DK eds.), Lawrence Erlbaum Associates, Mahwah, New
Jersey.
Roberts R, Gott R. 2010. Questioning the evidence for a claim in a
socio-​scientific issue: An aspect of scientific literacy. Research
in Science & Technological Education, 28 (3), 203–​226.
Rocard M, Hemmo V, Csermely P, Jorde D, Lenzen D, Walberg-​
Henriksson H. 2007. Science Education NOW: A Renewed
Pedagogy for the Future of Europe. Office for Official
Publications of the European Communities, Luxembourg.
Sadler DR. 1989. Formative assessment and the design of instruc-
tional systems. Instructional Science, 18 (2), 119–​144.
650
Sadler TD. 2009. Situated learning in science education: Socio-​
scientific issues as contexts for practice. Studies in Science
Education, 45 (1), 1–​42.
Simon S, Erduran S, Osborne J. 2006. Learning to teach argu-
mentation: Research and development in the science class-
room. International Journal of Science Education, 28 (2–​3),
235–​260.
Sollaci LB, Pereira MG. 2004. The introduction, methods, results,
and discussion (IMRAD) structure: A fifty-​ year survey.
Journal of the Medical Library Association, 92 (3), 364–​367.
Sutton D. 2006. Cooking skill, the senses, and memory: The fate
of practical knowledge. In Sensible Objects: Colonialism,
Museums and Material Culture (Edwards E, Gosden C,
Phillips RB eds.), Berg, Oxford, UK.
This H. 2005. Modelling dishes and exploring culinary precisions:
The two issues of molecular gastronomy. British Journal of
Nutrition, 93, S139–​S146.
Erik Fooladi
This H. 2019. Les Ateliers expérimentaux du goût. AgroParisTech
s.a. Available from http://​www2.agroparistech.fr/​Les-​Ateliers-​
experimentaux-​du-​gout.html
Toulmin S. 1958/​2003. The Uses of Argument. Cambridge University
Press, Cambridge, UK.
Vartiainen J, Hopia A, Aksela A. 2011. Using kitchen stories as
starting point for chemical instruction in high school. In
ESERA 2011 conference (Bruguière C, Tiberghien A, Clément
P eds.). European Science Education Research Association
(ESERA), Lyon, France.
Vesterinen VM, Aksela M, Sundberg MR. 2009. Nature of chemistry
in the national frame curricula for upper secondary education
in Finland, Norway and Sweden. NorDiNa, 5 (2), 200–​212.
Wellington JJ, Osborne J. 2001. Language and Literacy in Science
Education. Open University Press, Buckingham, UK.
Zohar A, Nemet F. 2002. Fostering students’ knowledge and argu-
mentation skills through dilemmas in human genetics. Journal
of Research in Science Teaching, 39 (1), 35–​62.
Cooking and Science Workshops: The “Soft of the World”,
Gelling Agents

Pere Castells
Science and Cooking World Congress, Barcelona, Spain

The creative revolution undergone in the last decades by cooking
and the worldwide diffusion of gastronomical knowledge have
been key to the development of modern cooking. This blossoming
in the field of gastronomy has contributed to making cooking a
driving force for economic growth and territorial development.
Using culinary practice as a teaching tool to explain science is
the primary objective of this activity. This workshop is presented
as an investigation game. The aim is to make the participants
manipulate gelling agents and gels.
The objectives of the activity are to:
• observe the changes produced in the liquids when
gelling agents are dispersed;
• learn about different kinds of gelling agents;
• learn about the application techniques;
• study gelling agents’ properties concerning heat, salts,
etc.
Gelling Agents: Agar-​Agar and Gelatine
As a first part of the educational activity, definitions are given
(Alícia Foundation and elBullitaller, 2010):
ACTIVITY 1. MAKING JELLIFIED PRODUCTS
AND APPLICATION OF HEAT
Recipes to prepare:
Agar-​Agar
200 g water or fruit juice + 2 g agar-​agar
Mix both products and bring them to boil. Remove from
the heat and allow to cool.
Gelatine
200 g water or fruit juice + 2 gelatine sheets (2 g for unit)
Firstly, dip the gelatine sheets in cold water. When the gel-
atine sheets are hydrated, dry them and add them to the
water or fruit juice. Bring to the boil, remove from the heat
and allow to cool.
Once both preparations have jellified, apply some heat,
around 70 ºC.

Gelling agents are products that form gels when added to Question
liquids. Many of them belong to the hydrocolloid family.
Gelatine: a mixture formed by proteins that are soluble in Describe the differences that you observe between the two
water that is traditionally used as a gelling agent. It has the products.
properties of a hydrocolloid. It is obtained by physicochemical
separation of bone or skin collagen, mainly from pork but also
from veal and fish.
ACTIVITY 2. JELLIFIED PRODUCTS SOFT/​HARD
Agar-​agar: a fibrous polysaccharide that is used as a gelling
agent. It has the properties of a hydrocolloid. It forms thermo-​ Recipes to prepare:
reversible gels. It is extracted by physicochemical treatments
from red algae Gelidium and Gracilaria.
Hydrocolloid: a protein or polysaccharide that has the Agar-​Agar
capacity to attract water, causing the formation of gels, or to 500 g water + 13 g agar-​agar
thicken a blended product or a liquid.
Mix the ingredients together and bring to boil. Remove
Then, the participants are invited to perform activities along with
from the heat and allow to cool.
answering questions.

651
652 Pere Castells

Here, the main objectives are:

Gelatine
• creativity in the kitchen;
500 g water + 10 gelatine sheets (2 g for unit)
• observe and learn about new gelling agents;
• learn about the application techniques;
Firstly, dip the gelatine sheets in cold water. When the gel- • learn the “spherification” technique.
atine sheets are hydrated, dry them and add them to the
water. Heat until dispersed, remove from the heat and allow Here again, definitions are given first:
to cool in a refrigerator. Sodium alginate: an organic salt derived from fibrous polysac-
charide used as a gelling and thickening agent and as a stabilizer.
It has the properties of a hydrocolloid.
Questions
It is extracted by physicochemical treatments of brown algae
Which gel is harder? (Macrocystis, Fucus, Laminaria, Ascophyllum, etc.), which are
You can also taste them; what are the differences? found in cold-​water seas and oceans.
What are the differences between a gel made from agar-​ Calcium chloride: calcium salt extracted from mineral
agar and one from gelatine? products.
Calcium lactate: calcium salt extracted from milk.
Spherification technique: a culinary technique consisting of
the controlled gelification of a liquid, which, when submerged in
ACTIVITY 3. APPLICATIONS OF AGAR-​AGAR
a bath, forms blobs. There are two kinds of spherification process:
Cabbage Jelly in the “direct” one, a liquid in which sodium alginate is dissolved
Ingredients is dipped in a bath containing calcium ions; in the reverse pro-
cess, the liquid containing calcium ions is dipped in an aqueous
200 g red cabbage juice solution of sodium alginate. These techniques can obtain spheres
2 g agar-​agar of different sizes, ressembling caviar, salmon eggs, gnocchi and
ravioli.
Method The process is then explained to the workshop participants
before they move on to the practical activities.
Red cabbage juice:

Crush 50 g of red cabbage with 200 g of water and strain.

Mix the red cabbage juice and agar-​ agar with a stick ACTIVITY 1. DIRECT SPHERIFICATION
blender. Fruit or Vegetable Spheres
Start boiling and place into moulds.
Ingredients
200 g fruit or vegetable juice
Purée of peach
5 g sodium alginate
Ingredients 6.5 g calcium chloride in 1 litre of water (for the bath)

200 g peach juice

3 g agar-​agar
Method
Mix peach juice and agar-​agar. Bring to boil.
Leave in the refrigerator to gel.
Blend it until you get a soft purée.
More Gelling Agents: Sodium Alginate
After this first series of activities has been performed, more
experiments are proposed, using gelling agents that have been
used in the retail food industry for a long time.
Method
Mix the sodium alginate with the fruit or vegetable juice.
It can be left to disperse slowly, or you can use a stick
blender. If left to disperse, the natural dispersion process
takes at least 24 hours. If you use a stick blender, dis-
persion is immediate, but residual air will be produced,
which can then be removed with a vacuum machine or
left to rest for a minimum of 4 hours.
Wait until the dispersion has stabilized and then apply the
spherification technique via immersion into the calcium
chloride bath.
Wait until a gelatinous texture has been achieved.
Cooking and Science Workshops 653

ACTIVITY 2. REVERSE SPHERIFICATION These workshops help the participants to understand the
Ravioli Yoghurt relationship between science and cooking. Gels have been
selected to represent one of the most important research
Ingredients fields in modern cooking.
More about all these projects can be found in Koppmann
Yoghurt
(2009), Rowat et al. (2013), and Vega and Ubbink (2008).
5 g sodium alginate for each litre of water (for the bath)

Method
REFERENCES
As yoghurt is rich in calcium, no calcium lactate is required. Alícia Foundation and elBullitaller. 2010. Modern Gastronomy
Prepare the sodium alginate bath. A to Z. CRC Press Taylor and Francis Group, Boca Raton,
Apply the spherification technique by immersion into the Florida.
sodium alginate bath. Koppmann M. 2009. Manual de gastronomía molecular. Siglo XXI,
Buenos Aires, Argentina.
Rowat AC, Sinha NN, Sorensen MP, Brenner MP, Weitz D. 2013.
Questions The Kitchen as a Physics Classroom. School of Engineering
and Applied Sciences, Harvard University, Cambridge, USA.
What do you observe? Vega C, Ubbink J. 2008. Molecular gastronomy: a food fad or science
Why is it called spherification? supporting innovative cuisine? Trends in Food Science &
Why do you think they are called ravioli? Technology, 19, 372–​382.
Culinary Sciences for the Enhancement of the Public Understanding
of Science
Ole G. Mouritsen
Department of Food Science, Taste for Life, Design and Consumer Behavior, University of Copenhagen,
26 Rolighedsvej, DK-​1958 Frederiksberg C, Denmark
Food and cooking provide a unique setting for public outreach
to all age groups with a focus on the enhancement of the public
understanding of the importance of science for society and the
way we as humans perceive each other and the world at large.
Using examples from a major Danish center for public com-
munication about taste, Taste for Life, we demonstrate how the
culinary sciences can help to promote the public understanding
of science.
In many countries in the Western world, the general public
understanding of science, in particular the natural and tech-
nical sciences, appears to be rather poor. Moreover, it appears
that the emergence of the postfactual society, the spreading of
‘fake news’, and the hesitance of some political systems to make
decisions based on scientific facts rather than public polls and
opinions appears to be spreading, with the risk of undermining
the foundations of a democratic society. At the same time, there is
a general trend in many affluent societies that fewer young people
choose to study hard sciences but rather focus on training for
professions that traditionally lead to high status and high income
for life. The culinary sciences may serve as a means to counteract
this unfortunate development by taking as a starting point some-
thing we all share, food, cooking, flavor, and eating, and then
appealing to a general human curiosity for an understanding of
how ‘things work’.
Science in Formal Educational Settings
Elementary and high-​school students are exposed to the sciences
in the classroom, but there is no need to hide the fact that some
find science classes boring and in many cases, more difficult than
other parts of the school curriculum. Girls and boys, depending
on their age, react somewhat differently to science concepts;
some have a preference for the more theoretical concepts, while
others excel in practical exercises and experimentation. Even
if science is recognized as an important part of an individual’s
‘Bildung’ (education, formation), the science labs at schools are
typically characterized by an artificial atmosphere in a setting
that can appear remote from daily life circumstances.
This is where the culinary sciences can come into play by
taking as a starting point the observation that the kitchen is a
lab (Parkers, 2004; Vega et al., 2012; Rowat, 2013; Rowat et al.,
2014), and most homes have in this sense a laboratory. Kitchens
deal with materials (food) that is of biological origin, and the
experimentation (cooking) involves using methodologies and
concepts via protocols (recipes, culinary precisions) and by
means of experimental equipment (kitchen tools). All cooking
procedures have analogies in more or less well-​defined processes
controlled by the laws of physics and chemistry. There is no
reason why one should not be able to teach and learn science
in a kitchen, even in a quantitative setting if the arsenal of kit-
chen tools contains proper measuring devices. In Denmark, we
have recently, within the framework of the massively interdis-
ciplinary national center Taste for Life, developed a communica-
tion platform (Hedegaard and Leer, 2017; Schneider et al., 2018;
Sörensen and Mouritsen, 2018) that, based on food and flavor
among other things, offers free-​ of-​
charge teaching materials
in all school disciplines, including the sciences, aimed at both
elementary and secondary schools (see www.taste-​for-​life.org)
(Figure 97.1). This platform is currently being expanded to
include formal and inspiratory teaching material for secondary
schools and chef schools.
In chef schools and at culinary academies, as reviewed recently
by Christensen and Edwards Stuart (2018), the situation with
regard to science ‘Bildung’ is somewhat different. Although sci-
entific principles and scientific thinking are considered valuable
tools for chefs, the science of cooking as such is only taught in
a few chef colleges and culinary schools around the world. The
challenge appears to be how to teach an academic topic in a voca-
tional setting and to develop a standardized syllabus for science
classes directed toward chef students. Another issue is related to
the still limited understanding of the importance of science in the
various workplaces where new chefs get jobs. However, recent
years have witnessed an enhanced interaction between chefs and
scientists (Sörensen and Mouritsen, 2018) that holds promise for
a trickling-​down effect on the formal educational system. The
growing interest in molecular gastronomy and gastrophysics
among chefs in the top tiers is also likely to have an effect on the
655
656
FIGURE 97.1 High-​school students in a gastrolab in the process of pre-
paring a cheese curd under accurate thermal control.
(Courtesy of Casper Thrane)
motivation for chef students to pay more attention to the scientific
underpinnings of cooking and gastronomy.
In several countries, it is difficult to recruit new and qualified
students to chef schools for a variety of reasons, including pos-
sible lack of status and recognition of the ‘ordinary’ chef, low
pay, and long working hours. It is possible that by increasing the
knowledge base of chefs’ training, e.g., by putting more emphasis
on scientifically inspired cooking and scientific reasoning and
innovation, it will be more attractive to students who can com-
bine the work of the hand and the mind. More skilled workers
and masters of craft are likely to be important for the New Urban
Economy (Ocejo, 2017), where chefs are not ‘just chefs’ but also
knowledge workers who deal with innovation, entrepreneurship,
communication, creative business, and the arts.
Finally, it should be mentioned that the introduction of some
culinary topics into the curriculum may also be a key to revitalize
hard-​core science education at college and university level (Barham
et al., 2010; Brenner and Sörensen, 2015; Brenner et al., 2015). An
example is the Science and Cooking Program at Harvard, which
is a wide-​ranging, interdisciplinary program that aims to use food
and cooking as a way to teach basic chemistry, physics, and biology
(Brenner et al., 2020). The program includes online courses, public
lectures, YouTube/​ iTunes channels, and educational programs
from pre-​school to college (Sörensen and Mouritsen, 2018).
Science Outreach on Informal Platforms
In this section, we use our experiences from the Danish commu-
nication center Taste for Life to demonstrate how one can conduct
science outreach on a number of informal platforms addressing
different age groups and different sectors of the public audi-
ence (Sörensen and Mouritsen, 2018). The center’s mission is to
convey knowledge about flavor to the general public, in particular
children and young people, in order to improve the quality of life
in general (Figure 97.2). Flavor offers a powerful platform for
research, teaching, and learning, not least in relation to science
(Schneider et al., 2018). While addressing children, students, and
Ole G. Mouritsen
FIGURE 97.2 Outreach to the general public at a folk rally (Folkemødet på
Bornholm, Denmark) about the multisensory perception of taste explored via
the so-​called jelly-​bean test.
(Courtesy of Mikael Schneider)
the broader public at outreach events and at festivals through and
about flavor, it is also possible to study their use of taste, flavor
preferences, and learning processes by gathering empirical data
for anthropological, sensory, and pedagogical research. One of
our projects studies the senses as a facilitator in breaking through
disgust barriers by teaching school children about fish through
arts (gyotaku), science, and cooking –​in parallel with quantita-
tive and qualitative studies of the children’s learning and changes
in taste preferences.
Working with a variety of stakeholders in a massively inter-
disciplinary mode across different disciplines, Taste for Life
engages scholars and researchers from the humanities (phil-
osophy, literature, pedagogy, and didactics), the natural sciences
(sensory science and gastrophysics), and the social sciences
(anthropology, media, and cultural science) as well as educators,
chefs, food innovators, and food entrepreneurs. By integrating
research on flavor, learning, didactics, and communication, the
center produces new knowledge on taste in three main areas: sen-
sory sciences and didactics; gastrophysics and interdisciplinarity;
and innovation and honing of culinary skills. The result of the
center’s work is a wide-​ ranging outreach program spanning
from natural sciences to human sciences, and spanning from
sciences to craft, design, and cooking. A unique initiative was
the international symposium from 2017, Creative Tastebuds, that
took place in a theatre with scientists and practitioners on stage
(Mouritsen and Frøst, 2018) in front of a diverse audience of
scholars, chefs, food innovators, educators, designers, politicians,
architects, professionals from the cultural sector, researchers,
food writers, food journalists, etc. The symposium explored how
different people understand and work with food and taste, e.g.,
from the points of view of physiology, neurology, habits, places,
memories, traditions, and culture.
We have discovered that the emerging field of gastrophysics
and its take on food and flavor (Mouritsen, 2012; Mouritsen and
Styrbæk, 2017; Spence, 2017) has been especially fruitful when
it comes to communicating science and engaging children and
young people, as well as the general public. One of the reasons
The Enhancement of Public Understanding
FIGURE 97.3 Chefs and chef school students doing a cooking class for
seventh-​graders at a traditional cattle show (Odense, Denmark).
(Courtesy of Styrbæks)
for this, as mentioned earlier, is to be found in the concept of
the kitchen as a laboratory, where one can experiment and tinker,
combining practical hands-​on work and design with intellectual
exercises and challenges. Being a creative and curious amateur
chef blends in with being an inquisitive and exploring amateur
researcher and scientist.
When addressing school children and the general public to
encourage them to reflect about flavor and the science behind
flavor, it can sometimes be advantageous to introduce foodstuffs
that are less common in the childrens’ sphere of food culture.
In Taste for Life, we have been using a range of foodstuffs that
are strange to Westerners, such as insects (Evans et al., 2015),
jellyfish (Pedersen et al., 2017), seaweeds (Mouritsen, 2017),
and cephalopods (squid, cuttlefish, and octopus) (Mouritsen and
Styrbæk, 2018; Faxholm et al., 2018). Focusing on these little-​
used food sources, at least in the Western world, also furnished a
welcome opportunity to address how science can be used to show
ways to exploit the world’s resources in a more sustainable way.
The question of healthy, sustainable, and delicious future food
has a certain appeal.
One of the more powerful modes of carrying out science out-
reach is by implementing various models of so-​called ‘bridge
building’. The bridge building can be between different age groups,
different generations, different disciplines, different levels of the
educational system, different ethnicities, and between craft and
mind, between theory and practice, etc. A bridge is founded on
food and taste and elements of a scientific understanding thereof.
Multiple bridges can be constructed by combining different
bridges. As an example, people from Taste for Life engage with
a group of chef school students and their teachers, focusing on a
specific topic, e.g., cheese making, combining the craft and art of
making cheese with the scientific principles behind it. The chef
students then act as tutors and supervisors at a cooking school
event for children at a food fair or a folk festival (Figure 97.3).
By instructing the children, the chef students gain further own- culinary transformations of a particular kind of foodstuff by
ership of their own learning by teaching the children. Further schematic illustrations of the involved molecules and molecular
657
bridge building can involve extension to children in different age
brackets, or having a starred chef come on the scene together with
the chef students and the children, who see the starred chef as a
role model one can aspire to become oneself with appropriate
training and motivation.
Some Examples Using the Science of Food and
Taste in Outreach Activities
In the following, we list a number of examples that we have
found particularly powerful for communication of science, be it
chemistry, physics, or biology, to the general public using food
and taste as a starting point.
1. Chocolate: Craft: tempering and mouthfeel. Science:
phase transitions and melting, non-​equilibrium phe-
nomena, viscosity, emulsions, crystal formation, fatty
acids, sugars, lipids, and triglycerides.
2. Cheese: Craft: making cheese. Science: structure of
milk, whey, casein, fats, micelles, aggregation, salts
and ions, acids and pH, thermal denaturation, micro-
biology of fermentation, enzymes.
3. Butter: Craft: making butter. Science: fats, water,
proteins, acids and pH, emulsions, crystal formation,
fermentation.
4. Bread: Craft: making and baking bread. Science:
proteins, yeast fermentation and other means of
leverage, carbon dioxide, foams, carbohydrates, gel-
atinization, retrogradation.
5. Pickles: Craft: making crunchy pickles. Science:
pectin, calcium binding, acids, salts and ions, fermen-
tation, osmosis, turgor.
6. Hydrogels: Craft: forming gels; spherification.
Science: water binding, polysaccharides, ions, cal-
cium binding, acids, thermal stability.
7. Mayonnaise: Craft: making mayonnaise. Science:
emulsion.
8. Chili: Craft: making hot food. Science: non-​taste and
chemesthesis, capsaicin, false heat.
9. Tomatoes: Craft: making delicious food. Science:
umami and umami synergy, calcification (cf. Figure
97.4), mouthfeel.
10. Apples: Craft: making cider and vinegar. Science: fer-
mentation processes, polysaccharides, sugar, alcohol,
acids.
11. Potatoes: Craft: preparing different kinds of pota-
toes. Science: polysaccharides, starches and starch
granules, gelatinization, texture.
12. Seaweeds: Craft: making dashi. Science: umami, tex-
ture versus structure.
13. Squid: Craft: caring for mouthfeel. Science: muscle
structure, texture, collagen, thermal processes.
In several cases, we have found it instructive to illustrate the
658
FIGURE 97.4 Chef students prepare calcified tomatoes by injecting a
herbal-​flavored tomato jus into the pulp of peeled calcified tomatoes.
(Courtesy of Eva Rymann)
FIGURE 97.5 Two young girls and a scientist use a microscope at a food
festival (Kulinarisk Sydfyn, Svendborg, Denmark) to explore how a mayon-
naise looks inside.
(Courtesy of Mikael Schneider)
assemblies in addition to images obtained by different kinds
of microscopy. Such illustrations can highlight the possible
relationships between the perceived mouthfeel (texture) and the
structure of the food (cf. Figure 97.5). Bringing a simple light
microscope into a chef school class to image a mayonnaise
(emulsion) or making it available to the audience at a public
event gives testimony to the old proverb that a picture is worth
more than a thousand words.
Acknowledgments
This work was supported in part by a grant to Smag for Livet
(Taste for Life) from Nordea-​fonden.
Ole G. Mouritsen
REFERENCES
Barham P, Skibsted LH, Bredie WL, Frøst MB, Møller P, Risbo J,
Snitkjaer P, Mortensen LM. 2010. Molecular gastronomy:
a new emerging scientific discipline. Chem. Rev., 110,
2313–​2365.
Brenner MP, Sörensen PM. 2015. Biophysics of molecular gas-
tronomy. Cell, 161, 5–​8.
Brenner MP, Sörensen PM, Weitz DA. 2015. Science and Cooking:
A Companion to the Harvard Course. Interactive e-​ book,
self-​published.
Brenner M, Sörensen P, Weitz D. 2020. Science and Cooking. WW
Norton & Co, New York.
Christensen M, Edwards Stuart R. 2018. Teaching science to chefs:
The benefits, challenges and opportunities. Int. J. Gast. Food
Sci. 16 [100133]. https://​doi.org/​10.1016/​j.ijgfs.2019.01.001
Evans J, Flore R, Pedersen JA, Frøst MB. 2015. Place-​based taste:
geography as a starting point for deliciousness. Flavour, 4, 7.
Faxholm PL, Schmidt CV, Brønnum LB, Sun YT, Clausen MP, Flore
R, Olsen K, Mouritsen OG. 2018. Squids of the North: gas-
tronomy and gastrophysics of Danish squid. Int. J. Gast. Food.
Sci., 14, 66–​76.
Hedegaard L, Leer J. 2017. Perspektiver på smag. SMAG #06,
University College Lillebælt og Aarhus Universitet, Odense,
Denmark. www.smagforlivet.dk/​sites/​default/​files/​documents/​
SMAG06%20-​%20Perspektiver%20p%C3%A5%20smag.pdf
Mouritsen OG. 2012. The emerging science of gastrophysics and its
application to the algal cuisine. Flavour, 1, 6.
Mouritsen OG. 2017. Those tasty weeds. J. Appl. Phycol., 29,
2159–​2164.
Mouritsen OG, Frøst MB (eds.). 2018. Creative tastebuds. Int.
J. Food Design, 3, 79–​168.
Mouritsen OG, Styrbæk K. 2017. Mouthfeel: How Texture Makes
Taste. Columbia University Press, New York.
Mouritsen OG, Styrbæk K. 2018. Cephalopod gastronomy –​a promise
for the future. Front. Comm. Sci. Environ. Comm., 3, 38.
Ocejo RE. 2017. Masters of Craft: Old Jobs in the New Urban
Economy. Princeton University Press, Princeton, New Jersey.
Parkers K. 2004. Recipe for success: teachers get inspiration from
‘gastrophysics’. Phys. Educ., 39, 19.
Pedersen MT, Brewer J, Duelund L, Hansen PL. 2017. On the
gastrophysics of jellyfish preparation. Int. J. Gast. Food Sci.,
9, 34–​38.
Rowat AC. 2013. The molecules we eat: Food as a medium to com-
municate science. Flavour, 2, 10.
Rowat AC, Sinha NN, Sörensen PM, Campas O, Castells P,
Rosenberg D, Brenner MP, Weitz DA. 2014. The kitchen as a
physics classroom. Phys. Educ., 49, 512–​522.
Schneider M, Kamuk A, Wistoft K. Frøst MB, Olsen A, Hedegaard
L, Mouritsen OG, 2018. Taste for Life: an exemplary case
for interdisciplinary collaboration between scientists and
practitioners on taste research and communication. Int. J. Food
Design, 3, 166.
Sörensen PM, Mouritsen OG. 2018. Science education and public
understanding of science via food, cooking and flavor. Int.
J. Gast. Food. Sci., 15, 36–​47.
Spence C. 2017. Gastrophysics: The New Science of Eating. Penguin,
New York, NY.
Vega C, Ubbink J, van der Linden E. 2012. The Kitchen as
Laboratory: Reflections on the Science of Food and Cooking.
Columbia University Press, New York, NY.
“Science and Cooking Activities” for Secondary School Students

Marie-​Claude Feore1,2, Laure Fort1,2, Marie-​Blanche Mauhourat3 and Hervé This vo Kientza1,2
1 UMR 0782 SayFood, AgroParisTech (INRAE), F-​75005 Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech, International Centre for Molecular Gastronomy,

F-​75005 Paris, France

3 Education nationale, Inspection générale, 110 rue de Grenelle, 75007 Paris, France

The “Science and Cooking Activities” that are discussed in this

chapter describe a series of practical classroom activities that are
connected to national curricula of lower and upper secondary
schools. They combine elements of science, technology, art, cul-
ture and cooking. They have been produced since 2004 upon the
request of the French school inspectorate of education, building
on the “Flavour Experimental Activities” introduced after 2001
in all French primary schools (see the corresponding chapter)
(AgroParisTech, 2019a).
These kinds of school activities are steadily increasing in
number in response to the demand by students, who are becoming
more and more fond of “cooking”, inspired, for example, by
television programs. The “Science and Cooking Activities” are
presented online in two parts: one for the students and the other
for the teachers. They are downloadable at various websites
(Academie de Paris, 2019; AgroParisTech, 2019b).

Objectives and Methods
The main goal of the “Science and Cooking Activities” is to
approach physics, chemistry and biology through everyday life
experiences, especially by activities and observations from the
kitchen. The activities are based on some basic ideas:
1. daily life observations are used to get or increase new
knowledge;
2. all ideas or beliefs must be tested by an experimental
approach;
3. the classical steps of scientific process must be followed:
observation of a phenomenon –​precise description with
scientific vocabulary –​proposition of a hypothesis –​
experiment to test the hypothesis –​return to hypothesis
formulation –​ conclusion;
4. as far as possible, the knowledge involved in these activ-
ities belongs to the official lower and upper secondary
school curriculum (physics, chemistry and biology);
5. articles such as scientific reviews and doctoral or
master’s theses may be used to provide more advanced
information about the latest research in some cases.
FIGURE 98.1 The extraction of a mixture of chlorophylls and carotenoids.
The experimental procedures can be performed by students in
lower and upper secondary school at a very low cost and with
safety. They propose an exploration of everyday culinary trans-
formations from technical, technological and scientific points of
view and also, more generally, from the cultural point of view
(artistic, literary …). The proposed pedagogy is active, and its
purpose is the promotion of knowledge.
Today, the list of the available documents online is:
1. How to prepare a green colored mayonnaise
(Figure 98.1).
2. How to extract the coloring from a foodstuff.
3a. How to transform milk from the liquid state to the
solid state of yoghurt.
3b. How to prepare yoghurt with cows’ milk.
4. What are the constituents of bread?
5. Why are there holes in bread?
6. Where does the smell of roast beef come from?

659
660 Marie-Claude Feore et al.

7. How are proteins digested? • how to use the laboratory scale to measure the mass and
8. What is the difference between short pastry crust graduated glassware to measure volumes of liquids;
and sugar pastry crust? • how to follow a scientific method to determine the
9a. How fast do apples turn brown when they are cut? density of a liquid.
9b. What role does vitamin C play?
10. How to make wine. The different syrups prepared (mint, lemon and strawberry) are
11a. What can we extract from tea? ranked in order of increasing density. Then, students introduce
11b. Determination of pigments in tea. the syrups into a glass in order of decreasing density, and this
12a. What are color pH indicators in the kitchen, in our should give a successful outcome.
plates or glasses? For advanced students:
12b. Where do the colors of red cabbage come from? Here, the objectives and requirements are more numerous: to
13. Alginate gelling. make as many layers as possible, with different colors, the best
14. How to make quince jelly (plasmolysis). flavors possible, and the use of vegetable or fruit juices.
15a. How do we go from cocoa beans (or pods) to It can be discussed how buoyancy forces (Archimedes’ law)
cocoa dough? are important for the superposition of layers. Natural (E140
15b. How do we get chocolate from cocoa dough? chlorophyll, E150 caramel) and artificial (E102 tartrazine, E122
16. How do we model plate tectonics in our kitchen? azorubine) dye molecules can be used to create strong, contrasting
17. How do we use sugar syrup to create multilayered colors. Also, when an ultraviolet lamp is available, the fluores-
cocktails? cence of quinine contained in the drink Schweppes® can also
18. How can we statistically approach the volatile reac- amaze the students by adding a fluorescent layer; this enables the
tion between Coca-​Cola and Mentos sweets? tutors to introduce the notion of excitation of the molecules and
19. Where does the color of green beans come from their return to the fundamental level (Valeur, 2004).
after cooking? Also, preliminary experiments can show that diffusion (of
20. Why does crustaceans’ color change after cooking? colorants, for example) between adjacent fluid layers depends
21a. Sugars, sweets and honeys: use 3D-​glasses to iden- on their viscosity. Based on this, the students can develop their
tify various sugars. own experimental protocol and make some very nice and good
21b. Sugars, sweets and honeys: create pulled sugar. cocktails.
21c. Sugars, sweets and honeys: use nuclear magnetic
resonance (NMR) to study acacia honey.
22. How do we prepare butter from milk? Workshop 19: Where Does the Color of Green
23. What do we use different salts in the kitchen for?
Beans Come from after Cooking?
Each activity is closely related to the French national school cur- The color of food depends not only on the color of the food
riculum (physics, chemistry and biology), and the connection is ingredients but also on the cooking method, the introduction of air
presented in the teacher’s part of each document. into the food, the environmental pH, the oxidation, and the color
Let us now describe some examples. of the light. There are color changes when cooking green beans
(immature pods of Phaseolus vulgaris L.), eggs, shellfish, etc.
In particular, many recipes for green beans offer indications to
prevent color change (Valverde and This, 2008), such as cooking
Workshop 17: How Do We Use Sugar Syrup to with the pan uncovered, adding sodium bicarbonate, using a large
Create Multilayered Cocktails? volume of water, or cooking in a copper pan with no tin lining.
This question was based on a demonstration by one of us of a For this activity, the following questions are to be asked first:
cocktail with ten layers, as discussed in the chapter “Molecular Where does the green color come from? How can we extract it?
Mixology: Welcome Coffee, a Cocktail with 10 Layers” in Part How do we identify the pigments?
III of this handbook (see also This, 2005). Here, it is proposed These questions are given to high school students, who already
that children from 12 to 16 years old reproduce the experiment. know methods for the extraction and separation of compounds,
Often, the students try to make the multilayered cocktail in empir- such as thin layer chromatography (TLC) (Valverde et al., 2007).
ical ways without success. The activity calls for discussions with Using such methods, students can recognize such pigments as
the teacher to guide them toward the successful making of layers. chlorophylls (green), carotenoids (yellow, orange) and their
Depending on their age, students are given different targets. derivatives, including pheophytins, which impart a darker color
For the youngest: to plant tissues.
Objective: create a cocktail consisting of three different col- What happens when green beans are thermally processed in
ored layers. The different layers are not allowed to be mixed. water? When the temperature of the plant tissues (such as green
Knowledge used for that purpose: beans) increases, some of these cells rupture, releasing organic
acids. The H+ ions of these acids react with the chlorophyll
• notions of absolute and relative densities; molecules, replacing the magnesium ion at the center of the
• notions of solvents, solutes and dilution; tetrapyrrolic ring. A new complex is formed: pheophytin. Instead
“Science and Cooking Activities” in Schools
of retaining all light radiation except green, pheophytin reflects
many wavelengths, so a browner color is observed.
But then, what happens to magnesium ions? After cooking
green beans for different durations in very pure water (MilliQ
water), a search for magnesium by atomic absorption spectrom-
etry was carried out. This method confirmed that the magnesium
had migrated into the cooking water, was in ionic form, and was
in sufficient quantity to be measured by titration with EDTA, a
method that is known to scientific students.
Of course, ordinary culinary processes do not use MilliQ
water, and the pH of the water in which the beans are cooked
is important. Indeed, the same magnesium determination can be
done with water to which sodium bicarbonate has been added,
for example. The results of the titrations obtained are consistent
with the spectrometric method. The study of one of the cooking
tips, using sodium bicarbonate, showed that the green color is
better preserved; in fact, the H+ ions of the released acids react
with the basic ions HCO3− of the sodium hydrogenocarbonate,
neutralizing the acid attack (Besançon et al., 2016).
661
REFERENCES
Academie de Paris. 2019. www.ac-​paris.fr/​portail/​jcms/​p1_​80293/​
ateliers-​science-​cuisine
AgroParisTech. 2019a. www2.agroparistech.fr/​-​Les-​Ateliers-
Science-Cuisine-colleges-​lycees-​
AgroParisTech. 2019b. www2.agroparistech.fr/​Des-​documents-​
generaux.html
Besançon S, Feore MC, Fort L, This H. 2016. La détermination de
la quantité de magnésium dans l’eau de cuisson des haricots
verts, L’Actualité chimique, 412, 32–​35.
This H. 2005. Welcome coffee, www.pierregagnaire.com/​ pierre_​
gagnaire/​travaux_​detail/​47.
Valeur B. 2004. Invitation à la fluorescence moléculaire, De Boeck
Supérieur, Bruxelles, Belgium. ISBN 9782804145972.
Valverde J, Vignolle M, This H. 2007. Quantitative determination of
photosynthetic pigments in green beans using thin-​layer chro-
matography and flatbed scanner as densitometer. Journal of
Chemical Education, 84, 1505–​1507.
Valverde J, This H. 2008. 1H NMR quantitative determination of
photosynthetic pigments from green beans (Phaseolus vul-
garis L.). Journal of Agricultural and Food Chemistry, 56 (2),
314–​320.
How to Reduce Oil in French Fries: A Student Experiment
Hervé This vo Kientza1,2
1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
It appears that the rise in obesity worldwide is linked to
overconsumption of energy-​dense nutrients, as well as reduced
physical exercise, especially in cities. In particular, consuming
too much fried food can be unhealthy, not only because of the
richness of such products in saccharides but also because of oil
consumption. Moreover, poorly fried food can result in the con-
sumption of dangerous compounds such as acrylamide (see the
chapter on frying by Pedreschi).
Here, we propose a simple demonstrative experiment (This,
1999) that can both show how a good frying technique can avoid
oil consumption and also indicate that, in contrast to what has been
published for a long time, the roughness of fried products is less
important for oil uptake during frying than water condensation.
Protocol
The experiment is simple, cheap, and easy to implement: only
a balance is needed in addition to the usual culinary tools. Here
it is:
1. Peel a potato.
2. Cut it into parallelepipeds about 10 × 1 × 1 cm.
3. Heat oil (e.g., in a pan, or in a special frying system, or
in a beaker on lab heating equipment).
4. When the temperature is about 180 °C, select two
potato parallelepipeds, and weigh them; cut the heaviest
so that the masses are about the same (a precision of
0.1 g is enough, as we shall see), and record the final
measurements.
5. Prepare a large quantity (about 20 sheets) of absorbing
paper.
6. At the same time, put the two parallelepipeds into the
hot oil and record the phenomena: bubbles flowing out
of the potato tissue; this steam condensing on the wall
of a cold glass that you put over the oil bath; potato
parallelepipeds first sinking, then floating; progressive
disappearance of the many jets of bubbles; with bubbles
flowing only from one to ten places.
7. When the two French fries are golden in colour,
take them out of the oil at exactly the same time and
immediately sponge the first one carefully, while you
put the other on a plate.
8. After 2 minutes, sponge the second French fry as care-
fully as the first.
9. Measure the mass of the two French fries.
10. Repeat the experiments for other parallelepipeds, and
calculate the standard deviation for all recorded masses.
Results
The last step is included for best practice, but indeed, the result
does not need it; for a 10 g initial mass of potato, there is between
0.5 and 1 g difference in mass between the two potato sticks,
with a lower mass for the French fry that is carefully sponged
immediately after frying. As the two French fries are otherwise
submitted to the same process, it means that this is a difference
in oil absorption, and indeed, another experiment, measuring the
pressure inside the potato parallelepipeds, shows that the pressure
increases with time inside the French fries, which explains the
flow of steam during frying, preventing oil from getting inside.
In order to make such measurements, ready-​to-​use equipment
is available, certainly, but a good idea of the pressure increase
can be obtained by connecting a Pasteur pipette to a U-​tube filled
with oil. The difference in levels can lead to the determination of
pressure. Using such equipment, it can be observed (Figure 99.1)
that the pressure increases regularly at the centre of the French
fries, but after the potato sample is out of the bath, the pressure
stops increasing and decreases after about 2 minutes. This can
be correlated with the recording of the temperature of the inside
tissue of potatoes, which can be made by inserting a thermo-
couple into the parallelepipeds at various places; in this way, it
can be observed that the temperature inside the potatoes remains
below 100 °C except at the limit of the crust (Figure 99.2).
The interpretation of all these phenomena is clear: with the
potato tissue (80% water) in the oil at a temperature higher than
100 °C, its water boils, and this creates a large volume of steam
(assuming that 18 g of water generates 24 L of steam, it can be
calculated from the mass variation of French fries that about 30 L
of steam is produced for one parallelepiped: this prevents oil
from coming inside the fries).
663
664 Hervé This vo Kientza

Pressure (mm of oil)

40

35

30

Pressure (mm of oil) 25

20

15

10

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Time (min)

FIGURE 99.1 The pressure increases inside potato parallelepipeds (side 1 x 1 cm) during frying. Here, the oil temperature was 180 °C.

100

90

80

70

60
Temperature (°c)

50

40

30

20

10

0
0 20 40 70 100 130 160
Time (s)

FIGURE 99.2 The temperature inside a potato parallelepiped during frying (the uncertainty is smaller than the size of the dots).

For the fry that is not sponged immediately, the condensation were able to recognize the French fries that had been sponged,
of the steam inside the fried system (cutting a French fry trans- but preferred the more oily fries!
versely shows a lot of space filled with gas) sucks the oil inside,
and this can be very important for public health. Now, one should REFERENCES
also know that a triangle test organized during one of the monthly This H. 1999. Experiment for the Science TV Programme of NHK.
seminars on molecular gastronomy in Paris showed that the jury Japan. 25 November 2019.
An Educational Satellite Project around the Scientific Elucidation
of Culinary Precisions in Lebanon and in the Middle East

Reine Barbar1,2, Jean-​Marie Malbec3, Christophe Lavelle4 and Hervé This vo Kientza5,6
1 Department of Agriculture and Food Engineering, School of Engineering, Holy Spirit University of Kaslik,

P.O. Box 446 Jounieh, Mount Lebanon, Lebanon

2 UMR IATE, Univ. Montpellier, INRAE, Institut Agro, 2, place Pierre Viala, 34 060 MONTPELLIER Cedex 02, France
3 Conseiller pédagogique sur la zone MOPI, Lycée Bonaparte, Doha, Qatar
4 National Museum of Natural History, CNRS UMR7196 /​INSERM U1154, Sorbonne University, 43 rue Cuvier,

75005 Paris, France

5 Groupe de gastronomie moléculaire, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,

F-​75005, Paris, France

6 UMR 0782 SayFood, AgroParisTech, INRAE, Université Paris-​Saclay, 91300 Massy, France

Introduction “Traditional means proven usage in the community market for a

time period showing transmission between generations; this time
Food and cooking habits are important features of Lebanese
period should be the one generally ascribed as one human gener-
society, as they link with the social life in Lebanon. Lebanese
ation, at least 25 years” (EU, 2006). In this regard, the convention
culinary culture and terroir are rich in traditional recipes and
for the safeguarding of intangible cultural heritage was passed by
ingredients that are in high demand and appreciated by affluent
the UNESCO General Conference held in 2003. At that time, the
as well as poor people. Tradition dominates the food system in
international community recognized the need to raise awareness
rural and urban areas and is demonstrated at home, on the street
about cultural manifestations and expressions that until then had
and in restaurants (Zurayk and Abu Ghyda, 2009).
had no legal or programmatic framework to protect them. The
The term “tradition” derives from the Latin word “tradere”,
know-​how linked to traditional crafts and traditions is part of the
which essentially means “to transmit” or “to send”. Tradition thus
intangible cultural heritage.
represents a form of transmission, or an information flow through
In support of this convention, UNESCO recognized this
time, of ideas, praxes, habits, methods, etc. Through such con-
culinary heritage by integrating the Lebanese region of Zahle
tinuity, traditions not only have a link to the past but also push
into its Creative Cities Network in the gastronomy category
towards the future (Cannarella and Piccioni, 2011). Similarly,
(Ilkincas, 2013). Zahle is the first city in the Arab world to join
“heritage”, from Latin patrimonium, refers etymologically to
the UNESCO network in the food category. The Creative Cities
the legacy of the father, that is, by extension, to all properties
Network is currently formed of 116 members from 54 countries
inherited by the family. Accordingly, the cultural heritage is an
covering seven creative fields: Crafts and Folk Art, Design, Film,
essential part of the content of the daily life of a community,
Gastronomy, Literature, Music and Media Arts (http://​en.unesco.
but it remains linked with its past and its future. Also, tradition
org/​creative-​cities/​).
and modernity are not necessarily conflicting terms: they collide
Lebanese cuisine is mainly based on rural dishes. Until the pol-
when one of these human expressions becomes an absolute value.
itical independence of the country in November 1943, Lebanon
For this reason, progress and modernity are opposed to tradition
was under the rule of several major foreign powers: the Persians
when modernization tends to interrupt the links with the past
(6th–​4th century BC), the Assyrians (9th–​6th century BC), the
(Cannarella and Piccioni, 2011).
Greeks and Seleucids (4th–​1st century BC), the Romans and
Traditional foods are considered a long-​ lasting legacy
Byzantine Greeks (64 BC –​AD 636), the Arabs (AD 636–​1099),
throughout the centuries, passed on to us by our ancestors. Due
the Crusaders (AD 1099–​ 1291), the Mamelukes (AD 1291–​
to altered lifestyles and international exchange of information
1516), the Ottoman Empire (AD 1516–​1918) and the French
and goods, this national legacy is slowly vanishing, and future
mandate (AD 1918–​1941) (Gabriel, 1978). This has certainly left
generations may be totally deprived of it (Trichopoulou et al.,
an impact on the national cuisine and led to its current richness
2006). However, ongoing research initiatives and funding are
and diversity.
taking place to counteract this. In 2006, the European Commission
Among the specific customs in Lebanese culinary preparations,
gave the following definition of “traditional” related to foods:
we can cite the mezze, from the Arabic word maza, originating

665
666
from the Persian word maza (taste) (which is an array of starters
or snacks), the mouneh (provisions made in autumn in prepar-
ation for winter to transform seasonal perishable food into dur-
able preserved food), and the slikké (edible plants picked by
hand) (Yazbeck, 2009). Mezze, for example, while being com-
monly inherited by the whole Lebanese population, is at the same
time diverse. Each region in Lebanon will have its own variation
of some mezze items, depending on its climate, its geographical
location and thus, its natural resources.
Although Lebanese cuisine is not as precisely codified as
French cuisine, we should find a rigorous way of preserving the
recipes. One way to preserve the culinary heritage is to have a
closer look at the technical steps described in a recipe leading
to the final result. While several books list typical traditional
Lebanese recipes, no link has been made between academia and
rural communities to bridge the gap between the two and build
other ways to preserve culinary heritage.
Molecular gastronomy (MG) is the scientific activity of
looking for the mechanisms of phenomena that occur during
culinary processes (preparation and consumption) (This, 2011;
Barham et al., 2010). The word “gastronomy” is derived from
the greek γαστήρ (gaster, which means “stomach”) and νόμος
(nómos, which refers to “laws”); it is therefore the “laws of the
stomach” and became “the knowledge and understanding of all
that relates to man, as he eats” under the pen of Brillat-​Savarin
in his masterpiece Physiology of taste (Brillat-​Savarin, 1825).
One could therefore paraphrase Brillat-​ Savarin to state that
molecular gastronomy is the physical and chemical knowledge
and understanding of all that relates to man as he eats.
In 2011, the Faculty of Agricultural and Food Sciences at the
Holy Spirit University of Kaslik organized a series of lectures
and workshops addressed to different audiences. A project
funded by the Agence Universitaire de la Francophonie (AUF)
was undertaken by an academic team in Lebanon under the theme
“Preservation and development of Lebanese culinary heritage
through molecular gastronomy” in order to better understand sci-
entifically the particularity of Lebanon’s traditional recipes. The
project targets several actions in teaching, research, social out-
reach and regional expansion (Figure 100.1). These actions aim
to orchestrate interdisciplinary work about Lebanese culinary
precisions for transfer in the future to the Middle East region. The
term “culinary precisions” describes the technical or procedural
information present in a recipe (oral or written), which provides
added value in terms of improved quality and a greater chance of
a successful product (Fooladi and Hopia, 2013). Precisions lead
to a better understanding of traditional culinary culture while
providing opportunities for innovation. Current actions aim for a
better understanding of Lebanese culinary heritage and the pres-
ervation, conservation and development of the latter (Barbar and
This, 2012).
In this context, the focus in this project is on developing the
following objectives (Barbar and This, 2012):
• preservation of Lebanon’s culinary heritage;
• help for rural development by the valorization of
culinary ancestral know-​how in Lebanese villages;
• helping sustainable development through the develop-
ment of local products;
Reine Barbar et al.
FIGURE 100.1 The different components targeted by the actions of the
project “Preservation and development of the Lebanese culinary heritage by
using molecular gastronomy”.
• helping the development of agro-​tourism in Lebanon by
ensuring better visibility of Lebanese rural regions and
their food as well as culinary resources;
• helping sustainable food production;
• valorization of natural resources found in Lebanon, like
plant exudates, medicinal herbs, typical spices, etc.
Research Activities
The information given by recipes can be challenged through
scientific investigation; at the same time, these investigations
provide abundant pedagogical supports for scientific classes,
since cooking is familiar to most human beings (Fooladi and
Hopia, 2013). Beside scientific and pedagogical opportunities,
molecular gastronomy can not only help us to better understand
(hence potentially improve) the way we cook, but also provide
inspiration for new techniques and dishes (Arboleya et al., 2008;
Barbot et al., 2014).
In this instance, several projects were conducted. Examples are
listed in Table 100.1. These projects were carried out by food
engineering students, who scientifically explored culinary tech-
nical proposals (“culinary precisions”) in traditional recipes by
means of molecular gastronomy as well as helping small-​scale
food producers to develop and stabilize their food products. Each
specific research task fits a broader objective set at the begin-
ning of the project. These objectives could be transferred to other
countries of the Middle East in order to better link science to trad-
itional food at both industrial and consumption scales.
One example in Table 100.1 is concerned with the analytical
exploration of elementary production steps of Lebanese hommos
bi tahine. Different types of hommos bi tahine were studied: indus-
trial and ready-​to-​eat hommos bi tahine, and traditional hommos
bi tahine (from the traditional restaurant Akra in Tripoli). The
main ingredients are the seeds of Cicer aretinum L. (chick peas)
and “sesame cream” or tahine (a suspension obtained by mech-
anical treatment of the seeds of Sesamum indicum), which were
Culinary Precisions in the Middle East 667

TABLE 100.1
Research Topics Addressed in the Scope of the Project
Research topic
Optimization of gluten-​free kneffe bread
Study of traditional marzipan from Zouk-​Mikael
Exploration and colloidal stability development of fermented
milk drinks (ayran)
Scientific exploration of culinary precisions of different
preparation modes of hommos bi tahine
Development of a new Lebanese menu: explorative study
Objective
Strengthening links between academia and the private sector through research and
development and innovation
Developing specific protection labels for local traditional products
Supporting rural development and agri-​tourism by promoting Lebanese culinary
ancestral know-​how
Exploring by molecular gastronomy the molecular reasons for instability of certain
Lebanese food products
Strengthening links between academia and the private sector through research and
development and innovation
Promoting traditional Lebanese products and their scientific development by molecular
gastronomy
Understanding consumers’ perception of traditional and new Lebanese recipes
Linking academia to food production enterprises at industrial and restaurant scales

also explored. The results of experimental and rheological tests

demonstrated the influence of a flow chart (and thus, the steps of
the production process) on the composition and microstructural
and textural quality of the end product (Barakat, 2014).
The literature data on consumer perception of Lebanese
cuisine is sparse. This is why one study conducted by our team
aimed to explore and understand the relationship between the
Lebanese population and Lebanese cuisine (Mitri, 2014). The
study, conducted on 506 Lebanese consumers (between 18 and
60 years old), revealed that Lebanese cuisine represents for them
habits, culture and heritage (95% totally agreed or completely
agreed). Moreover, all agreed that this cuisine had its origin in
rural areas. Despite their emotional attachment to familiar food,
68% of Lebanese participants accepted the introduction of new
flavours into their dishes. According to the study, consumer curi-
osity is explicit from the fact that 65% were ready to examine FIGURE 100.2 Renovated kebbe dish made by chefs Marijana Pedovic and
Sasa Hasic (Lebanon, November 2015).
a reworked menu. Analysis of the results showed that 90% of
participants were ready to eat dishes presented differently, while
68% were willing to try dishes fused with other cuisines, and
Among some examples of topics discussed during these
73% of consumers were interested in discovering new textures
monthly seminars are the following: mixing order of arak (alco-
in dishes.
holic beverage obtained from fermented grape juice and flavoured
Today, leading worldwide chefs are competing for innovation.
by adding aniseed and distilling again) with water, whitening
Many of them are looking closely at science and often work dir-
effect on baba ghanoush or moutabal (which is a dish made
ectly with scientists to help create new textures, shapes and tastes
by mixing grilled fruits of Solanum melongena (aubergines)
(Arboleya et al., 2008). While some customers seek reassurance
and sesame cream), salting aubergines before frying, marin-
by consuming well-​known traditional dishes, others are open to
ating chick peas with sodium bicarbonate, browning of different
being surprised. Balance is thus sought in some new Lebanese
fat sources used in Lebanese villages, cooking acid yogurt for
restaurants between traditional dishes and the introduction of
some Lebanese dishes, and the effect of different ingredients on
some reworked Lebanese items. An example by the chefs Pedovic
maintaining the stability of heated yogurt.
and Hasic is illustrated in Figure 100.2.
A collaboration took place between the Department of
Agriculture and Food Engineering and the laboratory IDEES
(Identités et Développement –​Espaces de l’Environnement
Social Outreach et des Sociétés) of the social sciences department at USEK
Similarly to France and other countries in the world, Lebanon (Université Saint-​Esprit de Kaslik). The laboratory is dedicated
organizes monthly seminars on molecular gastronomy to scientific and technical experiments as well as targeted pro-
(Figure 100.3). Anyone interested in culinary activity is invited fessional studies, and promotes the emergence of a working
to join these monthly meetings (cooks, scientists, teachers, engin- attitude starting from a problem identified in Lebanese society.
eers, food writers, etc.) in order to consider open questions about It collaborates with a number of municipalities, NGOs (non-​
culinary transformations at both home and restaurant scales. governmental organizations), and local and international
668 Reine Barbar et al.

FIGURE 100.3 First monthly seminar of Lebanon, organized with Hervé This in 2011.

organizations, in particular with UNESCO National Commission School Level

for Education and Culture. It manages its projects in a multidis-
The action entitled “Between Science and Cooking” initiated
ciplinary way, and for this purpose, it uses the expertise of other
through this project is focused on the idea of recovering and
faculties at USEK.
preserving the culinary heritage of our parents (in Arabic and
The collaboration took place in the scope of a main axis for this
French) by means of molecular gastronomy. This transversal
laboratory, which is the rehabilitation of the Lebanese cultural
research work helps to develop intercultural collaboration among
heritage. Through this broad theme, the aim is to raise awareness
science teachers, on the one side, and language teachers, on the
of the importance of the intangible heritage, to foster recogni-
other. It explores culinary transformations from the perspective
tion of its content and to design a strategy for action involving
of technique, technology and science and also from a cultural
schools, academia and municipalities.
perspective (art, literature, history, society and languages). The
A major part of this project included visits to different rural
model can be applied in schools as well as at university level.
regions in Lebanon conducted by students enrolled in a food
To create a multidisciplinary collaboration around taste, the
engineering programme. These field trips initiated work with
current project suggests the study of culinary sayings regarding
women’s associations as part of different municipalities involved
dishes from all over the world with a focus on French and Arab
in the project with IDEES laboratory on intangible cultural heri-
countries’ cuisines. This action is already being implemented in
tage. Scientific exploration followed in order to study the scien-
Doha (Qatar) with the objective of extending this pilot project
tific validity of culinary sayings in traditional recipes in Lebanon.
to the surrounding countries of the Middle East. It involves 40
With molecular gastronomy, it is possible to track the technical
teachers in primary schools.
components of the recipe through history and understand how
It is therefore important to initiate the new generation, starting
they occurred, how they evolved, and why they are true or false.
in school, on how to re-​appropriate the old recipes and culinary
At various events, the students explained to visitors, mainly
sayings but also how to modernize them (when relevant) using
composed of school students and their families, the work done
molecular gastronomy without altering the recipe and its spirit.
on traditional Lebanese recipes explored by means of molecular
The students are asked to collect recipes from their parents
gastronomy (Figure 100.4).
and grandparents; these recipes should be regional. In Doha,
the recipes are of Arab tradition. The recipe is then written, and
sometimes translated from Arabic to French, and contains all the
Teaching Structures technical steps for its realization.
In France, educational efforts towards better knowledge and Each group of students works on a specific recipe, and the
understanding of the culinary heritage have been carried out in following should be investigated for each recipe:
schools at both primary and secondary level, as well as univer-
• a complete procedure (e.g., mass –​temperature –​dur-
sities (This 2007; This, 2013). At the same time, professional
ation of cooking);
teaching is also evolving towards an “investigation-​ based”
• utensils to use (robot, whisk, grinder, traditional
approach rather than an “imitation-​based” approach, as testified
utensils);
by the new programme of the cooking technology Baccalaureate
• foods defined (amount, mass, volume);
(Cardinale et al., 2015). • condiments.
Culinary Precisions in the Middle East 669

FIGURE 100.4 Social outreach actions implemented by the project. (a) Food engineering students exploring recipes of Tannourine region; (b) collaboration
with WADA Women’s association of Deir El-​Ahmar; (c) Food engineering students at the Beirut Science Days fair; (d) Interaction between students and women
of Ehmej having ancestral culinary know-​how.

The recipe is translated into French and explained to everyone. Bachelor Degree Level
The teacher presents more details about the ingredients, and in
Molecular gastronomy courses were also developed at bach-
doing so, he/​she will explain the main food components and
elor and master levels of the food engineering programme at the
their composition. The recipe is then distributed to students
Faculty of Agricultural and Food Sciences of USEK (Lebanon).
and colour-​coded; blue is used for the culinary precisions,
Visits took place in several regions (in collaboration with
while red is used for the definitions section, and colourless
municipalities) to meet women working on traditional recipes
sections indicate the unnecessary parts from the technical
and investigate, as part of field trips, the details of recipes that
point of view.
have been prepared on site.
A small statistical calculation on the three colours of the
Students’ work includes the scientific study of culinary
recipe allows us to highlight the fact that the precision compo-
precisions of each recipe as well as the results of tests done at the
nent outweighs the other two parts. We can therefore conclude
laboratory to check their accuracy. The presentation of students’
that the technical precision part of a recipe is always the most
work includes an oral presentation, a scientific poster prepared
important part.
by the students and a tasting of an improved version of the trad-
Then, a list of culinary precisions for each recipe is prepared.
itional recipe.
Each group of students works on a chosen culinary precision.
Several culinary precisions can then be treated in the same class.
An experiment is then done in order to conclude the validity, or Master Degree Level
otherwise, of each culinary precision. Revision of the recipe can Students of the Master’s in Food Engineering have a stand
be done with the help of a chef. each year at the Science Days Fair that takes place at the Beirut
The experiments are a very efficient way to encourage students Hippodrome (Figure 100.5).
to use language abilities. The curiosity aroused by the proposed By addressing the general public and school students as well as
experiments requires the use of a specific vocabulary, and peri- families, the master students have highlighted several activities:
odic times of brainstorming and summaries must be facilitated awareness of molecular gastronomy through simple experiments,
with the help of the teacher. explaining gelification, and stability of sauces observed under a
670 Reine Barbar et al.

FIGURE 100.5 Food engineering students at Beirut Science Days fair. Interaction around culinary sayings in traditional Lebanese recipes with students
coming from schools.

microscope. In addition to this, the students demonstrated sim-

plified scientific elucidation of culinary precisions in traditional
Lebanese recipes collected during their visits to the Lebanese
villages; examples of such culinary sayings include enzymatic
browning of aubergines during their cooking, acidification of
kechek (which is a dried Lebanese cheese made from fermented
milk and wheat), frike cooking (which is made from green wheat
that goes through a roasting process) and water cooking of
aubergines while covering with a textile to preserve the colour.
The ancestral know-​how is thus best described by the scien-
tific mechanisms involved. The university students at both bach-
elor and master levels were able to discover a region and its
culinary resources with which they were unfamiliar at first. They
were able to understand them, to explore them and to modernize
them by working with a chef. This collaborative work between
FIGURE 100.6 Workshop with Cordon Bleu and food engineering students
science and social aspects aims to provide a better understanding
held at USEK (Lebanon) in October 2014 hosted by Christophe Lavelle.
of the traditional culinary culture, while it allows innovation and
promotes the domestic economy. Traditional crafts related to
cooking are reliable growth factors for economic development of
the country, especially for a small country like Lebanon. They introduced their methodology to work on traditional
recipes of their countries while making sure to preserve the taste.
They also conducted sessions during the Beirut Cooking Festival,
which is a culinary event open to the public.
Continuing Education and Knowledge Transfer
In the scope of the project, regular workshops with international
experts are planned for food science students, culinary school
students and professional cooks to update their knowledge and Regional Expansion
give them information on the latest developments in the field The project initiated in Lebanon in 2011 aimed to expand its
(Figure 100.6). In addition to professional workshops, round activities to surrounding regions of the Middle East having in
tables are regularly organized with NGOs and municipalities common a rich food heritage and culinary similarities in some
in order to better link academia and science in the service of recipes.
small producers and associations implicated in traditional food An educational project around culinary precisions at both uni-
production. versity and school levels could be one way to unite the efforts
In November 2015, chefs Marijana Pedovic (Serbia) and of academia, the private and public sectors, and communities in
Sasa Hasic (Croatia) gave several lectures to food engineering each country towards the common target of exploration of food
and culinary school students as well as professional cooks heritage and culinary practices (Table 100.2). A pilot project
(Figure 100.7). was therefore suggested in order to orchestrate multidisciplinary
Culinary Precisions in the Middle East
FIGURE 100.7 Workshops in Lebanon in November 2015 by chefs Marijana Pedovic and Sasa Hasic (a) as well as their participation in the Fifth Beirut
Cooking Festival (b and c).
efforts and complementary work to be done by all disciplines
related to science, food production and food heritage.
Conclusion
A specific Unit of Molecular Gastronomy has been built in
Lebanon with the scientific findings shared in the scope of this
chapter, with a main academic core interacting with multi-
disciplinary partners in the field: municipalities, universities,
cooks, schools, industries, restaurants and research centres. The
collection and study of culinary precisions has the potential to
create a framework for research, not only in food science and
molecular gastronomy, but also in other disciplines such as social
sciences and humanities, allowing multidisciplinary approaches
and cross-​fertilization among a broad range of sciences (Fooladi
and Hopia, 2013). In order to innovate, one must learn more
about culinary traditions and precisions.
“The discovery of a new dish does more for the happiness of
mankind than the discovery of a star”, Brillat-​Savarin (1825) once
said. Obviously, the (re)discovery of an old dish also matters.
A balance is sought through this project between the meaning of
671
traditional cooking and innovation through the knowledge gained
by food science. This is how we keep cuisine as both a social
activity and a living art.
Acknowledgements
This work was made possible by the financial support of the
Agence Universitaire de la Francophonie (for the project
“Preservation and development of Lebanese cultural heritage
through molecular gastronomy”). The help of the hotel manage-
ment department at USEK is also highly appreciated.
REFERENCES
Arboleya J, Olabarrieta I, Luis-​Aduriz A, Lasa D, Vergara J, Sanmartín
E, Iturriaga L, Duch A, Martínez de Marañón I. 2008. From the
chef’s mind to the dish: How scientific approaches facilitate
the creative process. Food Biophysics, 3, 261–​268.
Barakat A-M. 2014. Analytical Exploration of the Elementary Stages
of Hommos bi Tahiné Production. Master’s thesis in food
science, Holy Spirit University of Kaslik.
Barbar R, This H. 2012. Molecular gastronomy in Lebanon. Journal
of Culinary Science and Technology, 10, 277–​293.
672 Reine Barbar et al.

TABLE 100.2
Pilot Educational Project around Culinary Precisions
Step Disciplines involved Persons involved Objective
Choice of a recipe –​ Students and their parents/​grandparents Promote local recipes of the country/​
Teachers region
Food heritage associations
Explore the regional variations of History Students Promote the culinary diversity of regions
the recipe Anthropology Teachers Develop agro-​tourism around culinary
Database of special culinary Geography NGOs in the field of rural development traditions
gestures (manual techniques) Municipalities
Translation to French/​English Arabic Students Study the culinary vocabulary of a recipe
French Language teachers
English
Describe the ingredients/​ Mathematics Students Identify the culinary precisions of the
technological/​technical Chemistry Science teachers recipe and confirm their predominance
components of the recipe Physics in the recipe
Food processing
Experimental study of culinary Scientific disciplines Students working in groups Prove scientifically the validity or
precisions Science teachers otherwise of culinary precisions by
studying their variations
Create a database of ancestral culinary
precisions validated by experiment
Renovation proposals Cooking Professional cooks Let students explore their innovation
Science Students and their families potential to re-​appropriate traditional
Teachers recipes according to their consumption
Restaurants trends
Animation and presentation of All disciplines listed Audio-​visual media Celebrate knowledge around food and
project results Students/​teachers/​families promote students’ work
Municipalities and NGOs
Wider public
International associations: UNESCO, FAO
(Food and Agriculture Organization)
International Center of Molecular
Gastronomy

Barbot P, Gutrin N, Martin D, Lavelle C. 2014. De la science à l’art
culinaire. In Science culinaire: matière, procédés, dégustation
(Lavelle C, ed.). Belin, Paris.
Barham P, Skibsted LH, Bredie WLP, Bom Frøst M, Møller P, Risbo
J, Snitkjær P, Mortensen LM. 2010. Molecular gastronomy:
A new emerging scientific discipline. Chemical Reviews, 110,
2313–​2365.
Brillat-​Savarin JA. 1825. La Physiologie du goût, ou Méditations
de gastronomie transcendante. Gabriel de Gonet, Paris,
France.
Cannarella C, Piccioni V. 2011. Traditiovations: Creating innov-
ation from the past and antique techniques for rural areas.
Technovation, 31, 689–​699.
Cardinale B, Lavelle C, Lemagner O, Van Sevenant R. 2015. Sciences
et Technologies Culinaires. Delagrave, Paris, France.
EU. 2006. Council Regulation (EC) No 509/​2006 of 20 March
2006 on agricultural products and foodstuffs as traditional
specialities guaranteed. Official Journal of the European
Union L 93/​1.
Fooladi E, Hopia A. 2013. Culinary precisions as a platform for inter-
disciplinary dialogue. Flavour, 2, 6–​10.
Gabriel PL. 1978. In the Ashes: A Story of Lebanon. Whitmore
Publishing Company, Pennsylvania.
Ilkincas A. 2013. Le patrimoine culinaire de Zahlé reconnu par
l’Unesco, L’Orient-​le-​jour: www.lorientlejour.com/​article/
839176/ ​ l e- ​ p atrimoine- ​ c ulinaire- ​ d e- ​ z ahle- ​ r econnu- ​ p ar-​
lunesco.html.
Mitri J. 2014. Development of a new Lebanese menu: explorative
study. Master’s thesis in human nutrition and dietetics, Holy
Spirit University of Kaslik (USEK).
This H. 2011. Dix ans de gastronomie moléculaire [Ten years of
molecular gastronomy]. L’actualité Chimique, 2, 353–​354.
This H. 2007. Les Ateliers expérimentaux du goût: www2.
agroparistech.fr/​-​Les-​applications-​pedagogiques-​de-​.html.
This H. 2013. Programme “Dictons et plats patrimoniaux”. www2.
agroparistech.fr/​Le-​projet-​Dictons-​e t-​Plats-​Patrimoniaux.
html
Trichopoulou A, Vasilopoulou E, Georga K, Soukara S, Dilis V.
2006. Traditional foods: Why and how to sustain them. Trends
in Food Science & Technology, 17, 498–​504.
Yazbeck C. 2009. The Rural Taste of Lebanon: A Food Heritage
Trail. Raidy, Beirut, Lebanon.
Zurayk R, Abu Ghyda T. 2009. The Lebanese Terroir: a challenge
of quality. In Localizing Products: A Sustainable Approach
for Natural and Cultural Diversity in the South? International
Symposium, Paris, 9–​11 June 2009, UNESCO, Paris, France.
Bon Appétit, Marie Curie! A Stanford University Introductory
Science of Cooking Course

Markus W. Covert1 and Imanol Arrieta-​Ibarra2

1 Department of Bioengineering, Stanford University, Stanford, CA, United States
2 Department of Management Science and Engineering, Stanford University, Stanford, CA, United States

Introduction and Overview engaging with food. Third, students are required to analyze a
At Stanford, we have noticed a demand for more content about particular food experience scientifically and in more depth. The
cooking and its interaction with other fields. As a response to this latter two requirements are implemented differently as the course
demand, we developed a course to teach scientific principles in is held in different locations, as will be discussed later.
the context of high-​end cuisine. This 10-​week course has now
been taught eight times, including five times on the Stanford
campus, twice in Paris, and once online (due to the 2020 cor- Details of the Class Sessions
onavirus pandemic). Here, we discuss the goals and outline of
As noted, the course meets once per week in a three-​hour session,
the course, together with the ways in which it has been func-
which includes “lecture, lab, and lunch.” Table 101.1 lists the
tionally implemented in each of its three locations. We conclude
fundamental course lectures, which are taught in every location,
with a discussion of the course’s impact and our aspirations for
together with the labs and cooking that accompany them and the
the future.
scientific disciplines evoked. We describe each class session in
As stated in the syllabus,
more detail in the following.
this seminar is for anyone who loves food, cooking or
science! We will focus on the science and biology behind the Session 1: Taste and Flavor
techniques and the taste buds. Not a single lecture will pass by The lecture for this session introduces students to the passion for
without a delicious opportunity –​each weekly meeting will food and cuisine. We start by giving a small synthesis of the his-
include not only a lecture, but also a lab demonstration and a tory of haute cuisine and how French cuisine was popularized
chance to prepare classic dishes that illustrate that day’s sci- in the United States. We emphasize the idea of the importance
entific concepts.
of trying new things and how science can enrich our aesthetic
experience of the world. We then cover the genetic, develop-
The key goals of the course are that students (1) learn key
mental and environmental components of taste. Finally, we pre-
scientific concepts associated with cooking, (2) apply these
sent the categorization of taste and dismiss the myth that different
principles to give them intuition and insight while preparing
parts of the tongue react to different flavors.
meals, (3) enjoy a variety of new food experiences, and (4) be
Session 1 includes four experiments. The first explores the
able to scientifically analyze the meals they eat in depth.
genetic basis of taste, using phenylthiocarbamide (PTC) strips,
To support these goals, and in contrast to other courses or
whose perceived bitterness varies between individuals. We then
books, the course draws on an extensive variety of scientific dis-
explain how this variability depends on the individual’s genetic
ciplines beyond chemistry and physics –​from developmental and
code and is due to a change in the amino acid composition of the
evolutionary biology to statistics and artificial intelligence (some
protein present in the taste receptor TAS2R38 (taste receptor 2
of the materials consulted in developing the course are provided
member 38). In the second experiment, students determine the
in the references; this includes Barham (2000), McGee (2004),
number of taste buds per area near the front of the tongue, using
This (2008), Potter (2010), Myhrvold et al. (2011), Pollan (2014),
food coloring to dye a small area and then examining that area
López-​Alt (2015) and Wrangham (2010).
with a strong magnifying glass to count the taste buds. We then use
The class grades are determined by student performance in
a histogram to analyze the variability in the class. A third experi-
three areas. First, students are expected to attend classes and par-
ment concerns the taste perception of very low concentrations of
ticipate actively in experiments and cooking. Second, students
sugar, using a blind taste test. Finally, we demonstrate how taste
are encouraged to explore actively and push their boundaries in
sensations can be chemically altered at the molecular level using

673
674
miracle berries, which induce a sweet taste sensation at low pH
(created by ingesting sour food, such as lemon juice) due to the
glycoprotein miraculin.
For the lunch component of the session, we prepare complex
salads. Our purpose with this component is to introduce knife
skills and kitchen safety, and also to show how a complex flavor
profile can be constructed based on a variety of ingredients.
Examples of salads we have used include a tomato salsa, a
Tuscan-​style tuna salad and an ingredient-​heavy tabouleh. As
the students prepare these salads, we emphasize that they should
taste the different ingredients and figure out how many of these
they can identify in the finished product.
Session 2: Why Cook Food?
This lecture starts by explaining why cooked food is important and
beneficial for human beings, from the perspective of evolutionary
biology. We consider the risks and drawbacks that arise from
following fad diets, particularly those that eschew cooked foods.
Then, we discuss how cooking food may have given humans an
advantage over other species –​an advantage that may have given
rise to our enhanced cognitive abilities. We then shift to a molecular
and physics-​ based explanation of heat transfer, presenting the
thermodynamic equations that relate to the well-​known chef’s rule:
if the steak is twice as thick, it needs to be cooked four times as
long. Finally, we talk about different means of heat transfer –​for
example, convection, conduction and radiation –​with some extra
detail on how microwave ovens work.
Three lab experiences support this material. First, students
use a molecular dynamics simulation of water (PhET Interactive
Simulations, 2019) to investigate the impact of heat on intra-
molecular collisions, phase changes and temperature. Our next
experiment is to measure the gelatinization of potatoes boiled
in water over time. Students submerge potatoes in boiling water
and remove one potato at 60-​second intervals. The starches in
the potato gelatinize at 60 °C, a transition that is readily visible
as a “ring” on the outside of the potato. Students can measure
the width of these rings and plot it against the time elapsed,
which results in the identification of a square root relationship.
Third, students watch a video showing that heating grapes in a
microwave oven results in the generation of plasma. The scien-
tific explanation of this phenomenon (Veritasium, 2019) helps
students to better understand microwaves.
Lunch for this session is a roasted chicken on a bed of roasted
root vegetables, the preparation of which illustrates several of the
concepts described. We also present a video about mass market
chicken farming, which shows the different practices that are
used in raising and killing these animals, catalyzing an oppor-
tunity for students to reflect on the ethics and tradeoffs between
our need to eat and animal welfare.
Section 3: The Chemistry of Cooking
We continue the discussion of heat, but now considering the
effects of heat on muscles and plants. This lecture concentrates
on changes that food goes through by the act of cooking. We
frame the subject in the context of baking chocolate chip cookies
Markus W. Covert, Imanol Arrieta-Ibarra
(Warren, 2013). The video gives way to a thorough explanation
of the different reactions that occur at different temperatures. We
teach the processes of protein denaturation (with an emphasis
on collagen, myosin, actin, myoglobin, ovotransferrin and
ovalbumin) as well as the Maillard and caramelization reactions.
We then comment on how different cooking techniques (e.g.,
baking, boiling, steaming) and equipment (e.g., wok, pressure
cooker, sous-​vide) lead to different final outcomes.
The cooking for this class session is extensive, so we typic-
ally run only one experiment: we either boil eggs or fry small
and similarly cut pieces of steak, and remove them from the heat
at one-​minute time intervals. For the eggs, students compare
doneness of the white, doneness of the yolk, taste and smell. For
the steak, students measure the contraction along the length and
width dimensions, as well as taste and tenderness.
For lunch, we cook a sous-​vide steak, which clearly shows how
the sous-​vide approach to cooking achieves an even and controlled
temperature distribution inside the steak, which means that the
outside can be briefly seared without the steak being overcooked.
As part of the steak explanation, we present students with the
work of Temple Grandin (2006) in renovating slaughterhouses
to make them more humane. We then use a technique involving
sodium bicarbonate and a pressure cooker to caramelize carrots
efficiently as part of a caramelized carrot soup. Finally, we bake
the cookies that were described earlier in the lecture.
Session 4: Fermentation
We introduce the class to the reality that a great part of what we
eat is a direct consequence of microbial activity. We then present
the Malthusian growth equations and a typical metabolic network
for a microorganism, which leads us to reflect on the drawbacks
(i.e., infectious disease) and opportunities (e.g., fermented foods,
production of pharmaceuticals) presented by microbial growth.
We then focus on the baker’s yeast Saccharomyces cerevisiae and
its role in producing alcohol and bread. We explain what wheat
contributes to bread-​making, in particular the process by which
gluten is formed and its purpose (America’s Test Kitchen, 2013).
We then turn to the production of cheese, focusing on the micro-
bial aspects (NakedScientists, 2011). We end with a discussion of
recent research on the human microbiome, especially as it relates
to human health and disease.
Two experiments support this material. First, we use micros-
copy to examine the microbes in samples of common fermented
items such as kombucha and blue cheese. Then, to illustrate how
rapidly genes can spread by simple human contact, we apply a
UV reflective powder to one participant’s hands and have students
shake hands with each other in a line to see how far down the
line the powder will spread. We can also apply this powder to
students’ hands for them to test their handwashing abilities.
Our lunch is homemade and involves dough made from scratch
and also mozzarella cheese, both products of fermentation.
Session 5: Sauces as Complex Mixtures
To set the scene for the lecture on Modernism, this lecture
covers sauces –​and how they have often remained essentially
Bon Appétit, Marie Curie!
unchanged –​over centuries of time. We then describe how sauces
can be emulsions, clear dispersions (gels) and/​or cloudy, col-
loidal suspensions, and comment on the chemical characteristics
of each. We then show a video demonstration of a beurre blanc
sauce made in two ways, the first traditional and the second using
modern ingredients and equipment (Stella Culinary with Chef
Jacob, 2011).
Our experiment for this session is a comparison of different
methods to create emulsions. Specifically, we combine water and
oil in three ways: shaken by hand with no additions, shaken by
hand in the presence of emulsifier, and thoroughly mixed using
a blender. Then, we observe that the blended one separates most
slowly, followed by the emulsifier one and then by the simple
hand-​shaken one.
We have tried several different lunch possibilities to corres-
pond to this lecture, including hollandaise sauce with asparagus,
a rustic French bread salad with an interesting vinaigrette, and a
honey citrus chicken dish, in which we use the gelatin rendered
from the chicken to thicken the sauce. Probably the most popular
has been our “pasta bar”, in which we make a pesto sauce, a gar-
licky tomato sauce and a cheese sauce based on smoked pork and
tomme de savoie cheese.
Session 6: Modernist
This session continues the historical approach from Session 5,
starting at the beginning of the 20th century. We discuss how
modernist principles in other fields (art and music), together
with technical developments inspired by industrialization, all
played a significant role in the modernist cuisine movement in
the later part of the century. We then introduce the class to sev-
eral key players in the movement, including Ferran Adrià, Hervé
This, Dominique Crenn, Harold McGee and others. Finally, we
describe many of the different instruments and ingredients used
in molecular gastronomy, as well as newer initiatives such as the
use of artificial intelligence to guide the creation of new recipes
(a Silicon Valley flourish).
In this lecture, we perform the lab work and make lunch sim-
ultaneously. We teach the students how to use gelification and
spherification to cook modernist versions of classical dishes. One
example is a caprese salad where the balsamic vinegar is gellified
into pearls and basil is pureed to create a juice, which is then
gellified into noodles. We also use spherification to create a mod-
ernist version of a PBJ –​using brioche crackers for the bread,
speculatis butter and spherified caviar made from blueberry juice.
Finally, using the reverse frozen spherification technique, we
make beet juice salad spheres with goat cheese shavings.
Session 7: Chocolate
This session is a deep dive into the botany, ecology, ethics and tech-
nology connected with chocolate manufacture. We explain how
a chocolate bar and chocolate bon bons are created (Discovery
UK, 2018) and then describe where the Theobroma cacao tree
grows and how it is threatened. This is followed by an explan-
ation of the different varietals of cocoa beans and the importance
of terroir in chocolate flavor, including recent thinking on the role
675
of microbes in this process. We continue with an explanation of
the ethics and societal problems derived from the cacao industry,
using the manufacturer Momotombo in Nicaragua as a case study
(de Tourreil, 2016). The lecture continues with the explanation
of the physics of tempering and conching chocolate (HarvardX,
2017). Chocolate judge and advocate –​and molecular biology
expert –​Sunita de Tourreil is an extremely popular guest lecturer
for this session of the course.
To support this material, we perform a tasting of several bars
from different manufacturers and countries from around the
world. Students are introduced to the triangle test in order to make
statistical statements in real time, determining whether the class
can distinguish between chocolates that differ only in the type of
bean used, or else whether or not conching occurred, for example.
We conclude with an exploration of what makes chocolate deli-
cious and high-​quality, comparing mass market products with
smaller-​batch, artisanal products.
Teaching the Course on Stanford Campus
Although the class originated to support a study-​abroad quarter
for Stanford students in Paris, it quickly found an enthusiastic
home on campus in California. In its current form, we are able
to admit 14–​16 students. The Residential and Dining Enterprises
Teaching Kitchen @ Stanford (TK@S) was launched in January
2015 as a learning and educational resource for the Stanford com-
munity, which was an incredible enabling event for this class. The
founding belief in the R&DE Teaching Kitchen @ Stanford is
that food knowledge and cooking skills are essential to health and
wellness. TK@S fundamentally believes in educating students in
the use of healthy cooking techniques and sustainable ingredients
and passionately believes that this knowledge can help them live
healthier lives both during and after school (Malan et al., 2019).
Our class depends on the TK@S and gratefully acknowledges
David Iott, Amanda Zeitlin and Eric Montell for their support in
making this course a reality.
As noted earlier, the experiences we provide in order for
students to fulfil their grade requirements vary from site to
site. On the main campus, to encourage students to push their
boundaries in engaging with food, we have two requirements.
The first requirement is to engage in one food adventure and to
submit a brief write-​up including one or two paragraphs of text
(what they decided to do and why, and what the experience was)
and a picture of them at the site or performing the described
activity. The food adventures are self-​guided and highly flex-
ible, with the only real guidance being that the students should
try to expand their horizons. The most common example of a
food adventure is simply to visit a restaurant that represents a
cuisine that has not been experienced before; others include
shopping at ethnic markets or visiting a farm. The students uni-
formly love these adventures, almost invariably finding that
their initial apprehensions lead to a new appreciation of a food
or culture.
The second requirement that helps students to push their
boundaries is the student cooking challenge. Here, students
676
form groups of three to five in order to plan and prepare a three-​
course meal in class. Recipes are identified by the students and
submitted three weeks before the challenge, so that instructors
can identify and discuss any safety or resource limitations and
suggest modifications as needed. Students then purchase needed
materials (with a budget allocated by the class) and prepare the
meal in the Teaching Kitchen during a highly animated three-​
hour class session. The challenge has seen spectacular triumphs
as well as epic failures and is considered an unforgettable student
favorite for both reasons.
For the requirement that students analyze a particular food
experience scientifically and in more depth, the class meets at
a restaurant for a “final exam”. Students are expected to take
detailed notes during the entire meal and then to write a final
paper that analyzes the meal comprehensively, placing every
possible aspect of the meal –​surroundings, ingredients, prep-
aration, tastes, etc. –​in a scientific context, using the concepts
and material presented in each of the previous sessions of class.
This experience thus not only creates community and a sense of
completion, but also serves as a thorough review of the course
material. This assignment always receives extremely positive
feedback –​for many students, it becomes one of the best, most
memorable meals of their lives. Several students also mention
that they felt they would not have truly appreciated what they ate
if not for the concepts they learned in class.
Teaching in Paris
The course was taught in Paris in 2014 and in 2020. Teaching
the material as a study-​abroad course is significantly different
from teaching on campus for several reasons. First, the students
enrolling in the course abroad typically have less interest in and
experience with cooking than those enrolled on campus. This
is a result of the fact that the course on campus is extremely
selective due to limited space and high demand. Students are
evaluated for admission based on an essay submitted with their
application, and successful applicants typically already have
worked in a professional kitchen, written a successful food
blog, or similar. In contrast, the Paris course is open to all
study-​abroad students, and while this fact has motivated some
food enthusiasts who were unable to enroll in the course on
campus to study abroad, that number is small. Although there
is uniformly high enthusiasm for the Paris course, we are con-
stantly aware that we are primarily serving students for whom
Paris is their source of main focus and greatest excitement (and
rightly so). A second difference concerns the venue for the
course. Although Stanford has ready access to lecture space, a
large kitchen space is not readily available. Fortunately, Paris
is home to a number of rentable teaching kitchens, some of
which also have conference space that can be used for a lec-
ture. However, using these can add significantly to the cost of
running the course.
As a result, the Paris course is modified to include many more
out-​of-​class experiences and fewer joint cooking components.
Specifically, the sessions on taste, microbiology and chocolate
can be modified for presentation in a classroom –​replacing the
Markus W. Covert, Imanol Arrieta-Ibarra
preparation of complex salads with a “guess the ingredients”
challenge for salads obtained from noted Parisian caterers such as
Maisons Verot or Mulot, or bringing in an assortment of cheeses
from famous fromageries, such as that of Marie Quatrehomme.
The cooking challenge and the “final exam” lunch can also be
replaced by interesting field trips. Over the two times teaching
the course, we have visited Louis Pasteur’s home, laboratory and
crypt; the laboratory of molecular gastronomy pioneer Hervé
This; a museum exhibit at the Musée de l’Homme entitled “Je
mange donc je suis” (our tour was guided by the scientific curator
of the exhibit, Christophe Lavelle); the Musée du Chocolat; and
L’Atelier du Sens, a school restaurant in Antony (also generously
guided by Lavelle, who works as a teacher there). Students report
that these off-​the-​beaten-​path experiences contributed greatly to
their enjoyment of Paris itself as well as to the course material.
We also modify the assignments in order to give the students
as much exposure as possible to Parisian food. Instead of a single
food adventure (as required on campus), the Paris students are
expected to write about five adventures and are given an inter-
active Google map, which points them to fun recommendations.
These assignments are not only a way to push the students to
explore the city, but as we have since heard, have also become
a treasured journal reminding them of these experiences once
they arrive home. This was particularly meaningful in 2020, as
students were required to return home early as a result of the
worldwide pandemic. The “final exam” remains essentially the
same requirement as on campus; however, the meal is not held
in class. Instead, students find a restaurant to visit on their own.
Again, the students are excited by this opportunity and note it as
one of the most formative meals they have experienced.
Teaching during a Global Pandemic
Together with teachers the world over, we were forced to adapt
the course in response to the coronavirus pandemic earlier this
year. The requirements of our university were such that no
sessions could be held in person; that all content, assignments
and exams had to be available to students who were sheltered in
place across the globe; and that the final grade would be reported
as credit/​no credit in the place of a letter grade. Our recognition
that the student body would be largely in a state of shock, iso-
lation, depression and/​or anxiety also motivated us to consider
ways to bring joy and connection to our students’ lives in addition
to simply conveying the material.
Thus, we set out to create an online version of the class (cur-
rently only available to the Stanford community) that took these
considerations into account. After discussions with the students,
we decided to create short videos, about five per class session
(15–​20 minutes each), which contained a narrated slide deck,
performance of the labs described earlier, and finally a cooking
demonstration, also as described earlier. The students preferred
these videos to Zoom lectures because they were more flexible
for them to access, and in terms of presentation, easier for them
to digest. To supplement these videos, we also scheduled weekly
Zoom meetings. Sometimes these meetings were short and
simply involved asking questions about the video material, but
Bon Appétit, Marie Curie!
others involved more complex shared experiences. In two cases,
we shared a recipe in advance and asked students to be ready
to prepare the food in real time over Zoom. The first case was
a baked omelet from a recipe by Napa chef Thomas Keller, and
the second was an olive oil birdseed cake from a recipe by the
San Francisco restaurant State Bird Provisions. We also created
a chocolate tasting experience, in which ten different chocolate
samples were packaged together and shipped to the students’
locations. We then tasted these chocolates together and discussed
them in the context of the chocolate video lectures.
Assignments and grading were also different. We required a
report summarizing what was learned from the assigned video
material each week, attendance at the Zoom meetings and a final
“Challenge Project”. The latter was left open-​ended in order for
the students to identify a project that would work for them in their
unique circumstances. To date, some students have chosen to pre-
pare new and exciting meals, while others are performing novel
experiments (one involves optimizing a cupcake recipe) and still
others will be writing reports on food-​related topics.
Although changing this course so dramatically involved a great
deal of work, it also led to several novel insights. First, a major
question about online coursework is how to approach lab work.
Simply showing a video of a trained professional performing a
protocol is unlikely to translate well to student knowledge. In our
course, we solved this by having MWC perform labs with his
children. Because these labs were totally novel to them, there was
a greater possibility of failure –​all of which created an enhanced
sense of excitement and immediacy during the filming. The chil-
dren also asked basic questions, which were often resonant to
the student observers. Another insight regarded connection.
Having the students watch the videos freed us to engage more
deeply with them during the Zoom sessions. Another connection
was unexpected. We discovered that the parents, families and
roommates of many students had become engaged in the course,
not only by watching the videos but also by helping to prepare
and taste food, and even performing experiments. Thus, while
TABLE 101.1
Description of the Fundamental Sessions Covered in the Class
# Lecture Labs
1 Taste and flavor Genetic taste strips; taste bud count;
sugar dilution; miracle berries
2 Why cook food? Heat transfer simulation
Boiling potatoes
3 Chemistry of cooking Boiling eggs
Frying steak
4 Fermentation Microscopy
Disease transmission (UV powder)
5 Sauces as complex Emulsions
mixtures
6 Modernist cuisine See “Lunch”
7 Chocolate Triangle tests
677
our efforts to create community between students taking the class
were restricted, the connectivity between families and communi-
ties sheltered-​in-​place was enhanced. We also believe that these
families feel a deeper connection to Stanford, and their student’s
education, as a result of this class.
Conclusions and Future Directions
In summary, we have worked to create and expand a course that
integrates cooking with a large number of scientific disciplines.
Students are very enthusiastic about the chance to combine so
many interests and disciplines, as well as the hands-​on and highly
communal nature of the course. Moreover, several participants in
the class have gone on to successfully apply for other exciting
food-​related opportunities, including an appearance on a cooking
competition show and an internship at a three-​star Michelin res-
taurant in San Francisco. Thus, the course is evaluated highly and
oversubscribed by a large margin every year.
We also believe that the course would be highly portable and
could be adopted for use by any place of higher education that has
access to teaching kitchen facilities. Additionally, one version of
the course is optimized for a study-​abroad experience in France,
and it is easy to envision further versions being adapted to other
centers of cuisine (Tokyo, Mexico City, Copenhagen, Barcelona,
New York City, etc.). We have also now seen how the course could
potentially be offered to students online, and even how it can help
create a measure of connection and community –​even within indi-
vidual households –​under our current conditions of isolation.
All of this relates to the foremost criticism of the class: that it
reaches too few students. In fact, this was the subject of an opinion
article written in the Stanford student newspaper (Rizkalla,
2018). Three factors have prevented the class from getting much
bigger. The first is the size of the venue, which can only accom-
modate a maximum of 16 students safely. The second is the time
constraint of the teaching staff. Although one could imagine
Lunch Scientific topics
Complex salads (e.g., tabouleh, salsa, Genetics; developmental, molecular
tuna cannellini) and cellular biology; psychology
Roast chicken on a bed of root Evolutionary biology; molecular
vegetables dynamics; materials science; heat
transfer
Sous-​vide steak; caramelized carrot Biochemistry; physiology; animal
soup; chocolate chip cookies science; physics
Pizza Microbiology; infectious disease
Pasta bar; honey citrus chicken; Materials science; chemistry
asparagus with hollandaise;
vinaigrettes
Modernist caprese salad; PBJ; beets Chemistry; physics; artificial
with goat cheese intelligence
Chocolate tasting Botany; ecology; climate science;
chemistry; statistics
678
teaching several sections of this course, it is more logistically
challenging than most classes, meaning that the time outlay for
even one section is significant. Finally, there appears to be an
inherent tradeoff between the class size and the ability to forge
a deep sense of connection within the class. The latter relates to
Stanford’s fundamental new initiatives to create a greater sense
of belonging and inclusion within an increasingly diverse student
population. Any efforts to increase the breadth of this course will
have to address these important factors.
Finally, another obvious way to improve the class would be
to include more diversity. One consequence of its history is
that the course is largely focused on a Western culinary trad-
ition, and French cuisine in particular. However, the majority
of the students who enroll in the class have significant roots in
other traditions. One strength of the San Francisco Bay Area is
that so many cuisines are readily available, not only in terms
of restaurants but in local markets and expertise. As a result,
our primary goal moving forward with the class is to bring in
Asian, Central and South American, and African cuisines and
perspectives. We anticipate that such an expansion will make the
class more interesting, more inclusive and even more delicious to
our incoming students.
REFERENCES
America’s Test Kitchen. 2013. Science: What is gluten? Here’s how
to see and feel gluten.
www.youtube.com/​watch?v=zDEcvSc2UKA
Barham P. 2000. The Science of Cooking. Springer Verlag,
Heidelberg, Germany.
de Tourreil S. 2016. The Chocolate Garage: in Nicaragua! www.
youtube.com/​watch?v=xU5dBXCwuJY
Discovery UK. 2018. The Whole Process of Making Chocolate | How
do they do it? www.youtube.com/​watch?v=P_​JuQCiKWUc
Markus W. Covert, Imanol Arrieta-Ibarra
Grandin T. 2006. Thinking in Pictures: My Life with Autism. Vintage
Books, New York.
HarvardX. 2017. Different Phases of Chocolate. www.youtube.com/​
watch?v=3ODAI2gZyFw
López-​Alt JK. 2015. The Food Lab: Better Home Cooking through
Science. W. W. Norton & Company, Reno, NV.
Malan J, Watson TD, Slusser W, Gliik D, Rowat AC, Prelip M. 2020.
Challenges, opportunities, and motivators for developing and
applying food literacy in a university setting: A qualitative
study. Acad. Nutr. Diet. 120(1), 33–​44.
McGee H. 2004. On Food and Cooking: The Science and Lore of The
Kitchen. Scribner, New York, NY.
Myhrvold N, Young C, Bilet M. 2011. Modernist Cuisine: The
Art and Science of Cooking. The Cooking Lab, Bellevue,
Washington.
NakedScientists. 2011. How is Cheese Made? Naked Science
Scrapbook. www.youtube.com/​watch?v=Pnw-​XwCctYY
PhET Interactive Simulations. 2019. States of Matter: Basics. phet.
colorado.edu/​sims/​html/​states-​of-​matter-​basics/​latest/​states-​
of-​matter-​basics_​en.html
Pollan M. 2014. Cooked: A Natural History of Transformation.
Penguin Books, London, UK.
Potter J. 2010. Cooking for Geeks: Real Science, Great Hacks, and
Good Food. O’Reilly, Newton, Massachusetts.
Rizkalla A. 2018. The value of cooking classes. The Stanford Daily.
www.stanforddaily.com/​2018/​11/​26/​the-​value-​of-cooking-
classes
Stella Culinary with Chef Jacob. 2011. Sauces & Soups. https://​
stellaculinary.com/​cooking-​videos/​sauces-​and-​soups
This H. 2008. Molecular Gastronomy: Exploring the Science of
Flavor. Columbia University Press, New York, NY.
Veritasium. 2019. How Microwaving Grapes Makes Plasma. www.
youtube.com/​watch?v=wCrtk-​pyP0I
Warren S. 2013. TED-​Ed. The Chemistry of Cookies. www.ted.com/​
talks/​stephanie_​warren_​the_​chemistry_​of_​cookies
Wrangham R. 2010. Catching Fire: How Cooking Made Us Human.
Basic Books, New York, NY.
Molecular Gastronomy in Science Education and Science
Communication at the National University of Singapore

Linda Sellou and Lau Shi Yun

National University of Singapore, 21 Lower Kent Ridge road, Singapore 119077, Singapore

In this chapter, we will describe our efforts to use molecular gas-
tronomy as a vehicle for communicating and teaching science in
Singapore. We will share how to set up molecular gastronomy
workshops and also the impact of these efforts on the people
involved. It turns out that the benefits go beyond just sharing and
teaching science.
and their parents to investigate and reinvent local favourites
(Figure 102.1).
Some local delicacies we included were ice kacang (shaved
ice topped up with red beans, grass jelly, agar agar, sweet corn,
evaporated milk and syrups), cendol (pandan-​ infused dessert
made of mung bean flour, palm sugar syrup and coconut milk) and
bubble tea (hot or cold tea with milk and tapioca pearls). In these

The Project: Reaching Out with Molecular

Gastronomy in Singapore
One of the characteristics of Singaporean society is its obsession
with food. There are many food centres (locally referred to as
‘Hawker Centres’) scattered all over the country. These centres
offer a variety (e.g., Chinese, Indian, Malay, Thai) of affordable
food through small stalls. It is not uncommon for the denizens of
Singapore to stay in long queues just to treat themselves to local
favourites (e.g., chicken rice) at these hawker centres. In this
sense, food also plays an essential role in gathering and bonding
people.

Science Education and Science Communication

The National University of Singapore (NUS) strongly encourages
efforts to improve the learning, understanding and appreciation
of science both within and outside the campus. Promoting the
science of local food and engaging the public through Molecular
Gastronomy (This, 2009) was, therefore, an obvious choice.
Molecular Gastronomy allows us to use a topic, food, that is close
to everyone’s heart to demonstrate how art and science can work
together (Symcox, 2013). The fact that food is relevant, access-
ible and adaptable to all audiences makes it an ideal vehicle for
outreach. We carried out workshops centred around the science
of cooking.

An Example of a Typical Workshop

To give an idea of what a typical workshop entails, we will briefly
describe an event organised in 2015 to rediscover local favourites FIGURE 102.1 Poster for advertising our workshop celebrating the 50th
called “More Rojak, Less Rocket”. This was part of the 50th birthday of Singapore.
birthday celebration (SG50) of Singapore. We invited children (Hua Qian Hui, 2015)

679
680
desserts, colours, consistencies and flavours paint a delicious and
intricate visual and gustatory picture for the consumer. We also
discussed science! We talked about polymers, hydrocolloids, and
the possibilities and results of substituting starch hydrocolloids
with non-​starch hydrocolloids. The ingredients and utensils used
were very simple and mostly inexpensive, so that the participants
could pursue their experimentations back at home. A few parents
asked us for some ingredients to take back home!
People and Planning
People
A key feature of the molecular gastronomy workshops was that
students undertook most of the organising and planning of these
efforts under the guidance of members of the academic staff.
These students were volunteers from different parts of the Faculty
of Science of NUS: undergraduates from various science discip-
lines (chemistry, life sciences, mathematics and physics), from
the Special Programme in Science (SPS), and graduate students
from the NUS-Australian National University (ANU) Science
Communication programme (Figure 102.2). Another important
partner was the Young Educators in Science (YES) programme,
which organises many other outreach events throughout the
year with undergraduate volunteers. This diversity of individ-
uals added significant interdisciplinary strength to the planning
and execution of our molecular gastronomy workshops. We will
refer to all these individuals (students and faculty academic staff)
involved in the planning as ‘facilitators’ and those attending the
workshops as ‘attendees’.
Roles, Tasks and Teaching Material
The student facilitators contributed in three general ways:
(1) Designing molecular gastronomy activities and
experiments
FIGURE 102.2 NUS undergraduates experimenting with molecular gastronomy activities. From left to right: Srishti Arora (Life Sci.), Hua Qian Hui (Life
Sci.), Aparajita Hariharan (Engineering), Aditya Nair (Life Sci.), Megha Verma (Environmental Sci.) and Kristy Chang (Life Sci.).
Linda Sellou, Lau Shi Yun
(2) Creating teaching material
(3) Public engagement.
The student facilitators met regularly to explore and brainstorm
on what type of cooking experiments they could undertake and
share with their audience. They would record their reflections in
a journal. Some focused on the design of hands-​on practicals,
others on the creation of teaching material.
Some teaching materials included the design of videos or web
pages. In Figure 102.3, Yu Yuebo, a graduate student from Science
Communication, used analogies and visualisations to explain the
concept of states of matter involved in cooking. Another student,
Lim Vee Heng (from the same programme), who is also a high
school chemistry teacher, went further by doing his final science
communication project on how the systematic approach of
science can encourage and help people cook better. For example,
he designed graphics and a recipe explaining pan-​fried salmon
with snow peas and sweet and sour pork. He highlighted the sci-
entific concepts behind fish sticking to the pan and how to prevent
it, and the role of baking soda in the frying batter.
He took great care to focus on traditional Asian cuisine in
order to be more relatable to the local audiences.
Once the student facilitators came up with molecular gas-
tronomy experiments and teaching topics, the faculty advised on
appropriate pedagogical strategies to target the content to specific
audiences (e.g., young school children vs. adults) in a better way.
Most students were comfortable with the scientific aspects of the
culinary experiments; however, they found it more challenging to
explain the culinary aspects of each experiment.
Choice of Topics Related to Molecular Gastronomy
We tried to relate the topics we chose for our workshops to
specific audiences. For school children, we connected with the
public school syllabus. When possible, we extended beyond the
curriculum: for instance, the properties, structure and interactions
Molecular Gastronomy in Science Education
FIGURE 102.3 States of matter by Science Communication MSc, Yu Yuebo (2019).
of ingredients (water, proteins, fats and saccharides), states of
matter and phase transition during cooking.
We also included sensory effects and nutrition. For example,
we talked about ‘chocolate chantilly’ (This, 2006), ice cream and
sorbet with dry ice. We also talked about the colour, taste and
smell of teas. Another topic was the properties of hydrocolloids
like agar-​agar (a local favourite), which is used ubiquitously in
Asian cuisine. The experiments with bubble tea (another local
favourite) were probably the most popular topic, especially with
young audiences.
Benefits and Impacts
Teaching Science to Workshop Attendees
We ran our molecular gastronomy workshops intending to share
science with several different audiences. The participants were
either school students, families (i.e., young children together with
their parents) or only adults.
For school students, the impact of our workshops is similar to
most informal science education activities. They get to connect
science with authentic and relatable experiences (food in our
case). This alignment between the science they learn at school
and the workshop activities allows deeper engagement, leading
to better learning and further questioning (Bennet et al., 2007;
Vartiainen et al., 2013).
An interesting observation with some students was that
they seem to have forgotten that you can go back to basics and
experiment, discover, and learn with simple apparatus. Who
knew you could learn so much from whisking chocolate and
water? Anecdotally, from discussions with students, the use of
simple tools sounds too trivial to them. This may be because the
Singapore education system has long been adopting digital tech-
nologies in the classroom with great benefits (Tan, 2017). Many
Singapore schools have access to the latest technologies (e.g.,
personal learning devices such as iPads) (Ang, 2020; Ministry
of Education, n.d.). In such a setting, it was gratifying to the
facilitators to see students not only connecting with science
through cooking exploration but also engaging all their senses.
For example, while making the chocolate chantilly, they had to
engage their sight to observe the gradual change in appearance
681
of the chocolate emulsion being whisked. At the same time,
they could taste the changes in the mousse. Similarly, when
experimenting with bubble tea, their sense of smell was engaged
when tea was brewed, and the alginate sphere topping engaged
their senses of sight, taste and touch.
From the perspective of parents, these workshops allow them
to discover or rediscover science together with their family. This
opportunity to experiment and learn science using cooking can
continue even when they are back home. For the organisers,
what is essential here is to use the workshops for kindling the
attendees’ interest in science through food. This can empower
them to embark on a self-​regulated, life-​long journey of learning
by exploring cooking.
Empowering Undergraduates
It might appear that those attending the workshops will gain the
most from these workshops. However, it turns out that the student
facilitators who plan, organise and run these events reap unin-
tended and sometimes life-​altering benefits. The motivations of
the student facilitators to take part in this molecular gastronomy
initiative vary from sheer curiosity to being avid cooks or to
wanting to extend their educational portfolios. Regardless of their
primary motivations, we noted that the student facilitators gained
as much as, if not more than, the workshop attendees.
Basing our analysis on comments shared with the organisers,
we can break down the gains into three different categories: aca-
demic, cultural and professional.
For one attendee, thinking of cooking allowed them to see
science come alive, connecting some of what they have learned
through their academic journey to real life. This is what one of
them said:
How we can apply scientific concepts to food, to make it more
interesting and exciting and delicious was something I came
to appreciate as well. And what I also came to understand was
how cooking and the execution of the scientific method were
so similar. Precision, technique, and protocol are all required
but innovation is not limited. The balance between freedom
and method are reflected both in the scientific method and in
cooking. Those have been the key personal impacts from this
educational experience.
682
Another significant benefit was that the student facilitators began
to see molecular gastronomy as having a local identity relevant to
their own home culture. Here is what one of them said:
It doesn’t have to be limited to the exotic. And somehow,
while using local culture, the scientific gastronomy project felt
more applicable, personally. It felt more real, and felt more
understandable, and much, much less foreign. And I was then
able to see how I could use it with my own culture’s cooking
and really think about the flavours and textures that went into
it, and appreciate it.
This sentiment can be better understood by highlighting that
molecular gastronomy as a discipline emerged and has developed
much more in Western societies. Most literature (research or edu-
cational) is based on “Western” tastes and ingredients (Ivanovic
et al., 2011; Roosth, 2013; Wang and Wang, 2016). So, although
the chocolate chantilly is a very popular experiment, even in Asia,
students could relate more, culturally, to experiments with bubble
tea or pandan-​infused dishes. One of our future objectives is to
develop further educational materials with local flavours to make
molecular gastronomy more relatable to a wider local community.
Another benefit to the student facilitators was related to inter-
personal skills. The fact that molecular gastronomy is so interdis-
ciplinary provided an opportunity for teaching and learning from
each other. Students from different disciplines such as chemistry,
life sciences, mathematics and physics had to work together to
design experiments, create material, and develop communication
and educational strategies around a topic on which they had no or
little prior background. They had to work in a team and practise
excellent communication skills to reach their objectives. They
also had to become innovative and creative when engaging the
different workshop attendees (e.g., children, adults).
Unfortunately, we have not yet had an opportunity to survey the
workshop attendees about the impact of the workshops on their
views of food and molecular gastronomy, science and culture.
Plans for the Future
The universal attraction of food, together with the versatility and
interdisciplinarity of molecular gastronomy, makes it an ideal
topic for science outreach and education. However, our experi-
ence with molecular gastronomy workshops has indicated that
we can use the discipline not only to promote science but also
to instil critical thinking in students from schools to higher edu-
cation and the wider public. Critical thinking skills are used
and developed when multidisciplinary students with different
viewpoints work together to come up with the final food product
(Ivanitskaya et al., 2002; Burke and Danaher, 2018). The work-
shop provides a familiar platform for students or people in gen-
eral to combine culture and science.
We have only just started our initiatives, and we see many more
opportunities to move forward. More university students across
many faculties (e.g., humanities and engineering) have shown
interest in exploring the scientific sides of cooking. Recently,
some undergraduates have also proposed launching a course on
Linda Sellou, Lau Shi Yun
molecular gastronomy under a new “Design Your Own Module”
(DYOM) scheme of NUS.
Outside the university, we are also planning to connect with
local communities to extend the cultural aspect of our workshops
and offer a better link between culture, cooking and science.
Acknowledgements
Many kind individuals, both faculty and students, were involved in
successfully organising these molecular gastronomy workshops.
Unfortunately, we don’t have room to mention all of them.
However, we would like to thank, at least, the following individuals,
departments and programmes from NUS for their kind support:
Dr Lim Kim Yong, Dr Liu Mei Hui, Mr Kuang Jianhong, Mr
Poon Jun Xian, Cyrus; NUS Science Demo Lab, NUS Young
Educators in Science (YES) programme; NUS Department of
Chemistry, NUS Department of Food Science & Technology,
Special Programme in Science (SPS) and NUS FoS Science
Communication Programme.
REFERENCES
Ang J. 2020. Parliament: All secondary school students to have
personal digital devices by 2028, $200 Edusave top-​up to
support purchase. The Straits Times. www.straitstimes.com/​
politics/​parliament-​all-​secondary-​school-​students-​to-​have-​
personal-​digital-​devices-​by-​2028–​200.
Bennett J, Lubben F, Hogarth S. 2007. Bringing science to life: A syn-
thesis of the research evidence on the effects of context-​based
and STS approaches to science teaching. Science Education,
91(3), 347–​370.
Burke R, Danaher P. 2018. Interdisciplinary teaching and learning
within molecular gastronomy education: Does it benefit
students? International Journal of Molecular Gastronomy,
1, 1–​12.
Ivanitskaya L, Clark D, Montgomery G, Primeau R. 2002.
Interdisciplinary learning: Process and outcomes. Innovative
Higher Education, 27(2), 95–​111.
Ivanovic S, Mikinac K, Perman L. 2011. Molecular gastronomy
in function of scientific implementation in practice. UTMS
Journal of Economics, 2(2), 139–​150.
Ministry of Education. (n.d.). Masterplan 4 Key Features. Retrieved
from www.ura.gov.sg/​Corporate/​Planning/​Master-​Plan
Roosth S. 2013. Of foams and formalisms: Scientific expertise
and craft practice in molecular gastronomy. American
Anthropologist, 115(1), 4–​16.
Symcox K (Ed.). 2013. Using Food to Stimulate Interest in
the Chemistry Classroom. American Chemical Society,
Washington, DC.
Tan SC, Cheah HM, Chen W , Choy D. 2017. ICT Environments
in Singapore. In Pushing the Frontier (pp. 45–​55). Springer,
Singapore.
This H. 2006. Molecular Gastronomy: Exploring the Science of
Flavor. Columbia University Press, New York, NY.
This H. 2009. Molecular gastronomy, a scientific look at cooking.
Accounts of Chemical Research, 42(5), 575–​583.
Vartiainen J, Aksela M, Hopia A. 2013. Introduction to molecular
gastronomy and to its applications in science education.
LUMAT (2013–​2015 Issues), 1(2), 143–​150.
Wang H, Wang J. 2016. An analysis on the influence of the molecular
gastronomy on the Chinese cooking development. Journal of
Culinary Science & Technology, 14(3), 191–​197.
Molecular Gastronomy: A Universal Portal to the Molecular Sciences
Patricia B. O’Hara
Amherst College, Amherst, MA 01002, United States
By providing answers to simple questions, such as why certain
apples turn brown once bitten, to more complex questions, such
as the advantages of sous vide cooking, molecular gastronomy
provides a motivating reason for learning the underlying chem-
ical, biological, and physical properties. In this chapter, we show
how foundational topics in acid–​base chemistry, redox chemistry,
enzyme kinetics, and elasticity can be explored through topics in
food science. Each topic is developed alongside lab exercises in
which students test hypotheses about molecular behavior using
the scientific method. When students can observe, taste, feel,
and smell differences for themselves, molecular changes become
more meaningful.
Acids and Bases
The first example is drawn from acid–​base behavior: one of the
foundational concepts of chemistry. Most of us might be familiar
with acids and bases in our culinary world through topics such
as the acidity of a wine, the need to add vinegar or lemon to
“brighten” up a soup or sauce, and the bases that are present in
bubbles or soaps.
However, students of chemistry will say that quantifying these
concepts and using them in problems can be confusing. Acidity
is measured by the pH of a solution, which is determined from
the concentration of the H+ ion, [H+], but not in a straightforward
or intuitive way. Rather, the pH = −log([H+]), a negative loga-
rithmic scale. This means that as the acidity of a solution goes up,
[H+] increases (so far so good), but the pH decreases. A ten-​fold
increase in the [H+] drops the pH by one unit (say from two to
one). Add to that the fact that different materials yield up their
protons to different extents (strong and weak acids), that some
prefer to extract protons from other materials (bases), and that
these properties depend upon temperature and the nature of other
compounds in solution, and you can understand the novice’s
frustration.
Exploring methods of preparing food with acid helps students
to grab hold of these ideas for themselves. When we learn that
the fundamental structures of the proteins in our foods are
transformed in the presence of acid in the same way as they
unfold when exposed to heat as in traditional cooking, many
new cooking processes suddenly make sense. Lime juice, which
naturally contains citric acid, can unfold or denature the proteins
in a fish or shrimp after marinading for 15–​45 minutes, a South
American method of cooking known as ceviche (McGee, 2004).
In the Philippino dish kinilaw, the process is similar, but the acid
used to denature the fish is made from a locally brewed vinegar
made from coconut, palm, or sugarcane with spices added.
Thus, food is “cold cooked” in both a ceviche and a kinilaw pro-
cess. In many fermentation processes, lactic acid, a byproduct of
the transformation of lactose sugars in dairy products by the bac-
teria Streptococcus thermophilus and Lactobacillus bulgaricus,
denatures or curdles the milk proteins and forms yogurt (McGee,
2004). Acids are useful as palate cleansers, and in many New
England pancake houses, pickles are served to cleanse the palate
between mouthfuls of sweet maple syrup-​coated pancakes. Taste
receptors for acid exist on our tongue and are commonly referred
to as “sour taste.” These taste receptors are often clustered in
combination with other taste receptors and are in communica-
tion with them, thus influencing our perception of sweet or sour
(Stuckey, 2012).
A logical place to start a discussion of pH and cooking is to
consider a kitchen staple, vinegar, or from the French, vin aigre
(sour wine). Chemically, vinegars are dilute solutions that con-
tain more than 5% of acetic acid, a weak organic acid, in 100 mL
water. Typical pHs for vinegar are about 2.4, but they can vary
widely. Biologically, vinegars are produced through the action
of the bacterium Acetobacter aceti on fermented solutions
containing alcohol, and are “living, breathing organisms that
should be treated with the same respect you would any fruit tree
or vegetable plant. While normally made from wine (grapes),
vinegars can also be produced from most grains including
rice vinegar (rice vinegar), apple cider vinegar (apples) and honey
vinegar (mead)” (Turkell, 2017).
Organic solutions can also be distilled to produce a clear dis-
tilled vinegar (just acetic acid and water), which lacks any of the
individual tastes and smells that accompany the natural products
mentioned. One good experimental way to explore acid–​base
properties is to begin with a sampling of vinegars from dis-
tilled vinegar through to as many of the plant-​and honey-​based
vinegars as can be found. Students can taste a ten-​fold dilution
(one in ten), a twofold dilution (one in two) and an undiluted
sample of the vinegars and be asked to record the “sourness”
of the solutions and the variety of flavors that accompanied the
683
684
FIGURE 103.1 Roll of pH paper with a color scale to indicate pH.
sourness. This “organoleptic” analysis can go along with simple
pH measurements using test paper similar to the paper shown
in Figure 103.1, in which the acidity of a solution changes the
color of a dye on a strip of filter paper, and the particular color of
your test sample can be matched up against the standards. How
do the diluted samples compare with the undiluted samples? Do
the twofold diluted samples record a pH twice as high as the
undiluted sample? What about the ten-​fold diluted sample? How
do the recorded pH values relate to the perceptions of sourness?
Oxidoreduction
While changes in molecular shapes are at the heart of hot and
cold cooking, another aspect of the transformations that happen
during cooking involves changing the number of electrons in
some of the molecules in the food. Gaining an electron reduces
a molecule, but no molecule can gain an electron without another
molecule losing one. Having donated an electron, this second
molecule is described as a reducing agent, an antioxidant, and
has itself been oxidized.
Confused yet? Once again, novices can find redox chemistry
confusing. Understanding that an oxidant can turn into a reduc-
tant depending on its partner might make better sense if we can
illustrate it with food. Evaluating redox states of individual atoms
in a complex molecule can help one follow the flow of electrons.
Browning reactions are of particular interest here. Students
are familiar with apples or avocados turning brown after being
bitten or sliced. But why don’t all fruits turn brown? What is the
molecular basis for these discoloration reactions, and can we
slow them down or even stop them? Molecular oxygen, present
as 20% of the air around us, is the ultimate oxidizing agent. It is at
the heart of many browning reactions. In an unbitten apple or an
unsliced avocado, molecules in the fruit are compartmentalized
and sheltered from molecular oxygen. Once sliced or bitten, poly-
phenol molecules in the fruit are exposed to the air and oxidized.
This oxidation turns them brown and breaks down some of the
Patricia B. O’Hara
cellular structure responsible for the crispness. This reaction is
accelerated by an enzyme, phenolase; avocados and apples have
high concentrations of this enzyme.
Students might be aware that sprinkling apples with lemon
juice can slow down the reaction. This happens because, as we
said before, the protein that makes up the phenolase enzyme
needs to be folded properly to do its job. Once the acid denatures
the protein, it can no longer act as an enzyme to accelerate the
rate of the oxidation reaction, and we stop it dead in its tracks.
Sliced fruit can also be wrapped in plastic wrap, which prevents
access of oxygen. Cut fruit in fast food restaurants has citric acid
added to prevent browning, and the fruit is packaged in such a
way so as to exclude oxygen from the package containing the
fruit. Some companies are exploring new types of packaging
material that inhibit browning of sliced fruit by a combination of
oxygen exclusion and enzyme inhibition.
Another way of stopping the action of this enzyme and the
oxidation of the fruit is to give the fruit a brief irradiation in
the microwave. The heat produced in the microwave can also
denature the enzyme and slow down the oxidation. Students can
set up their own experiments in which they explore how acid,
brief heat, and physical exclusion of oxygen affect the browning
time for a sample of cut apples.
Glycation/​Maillard Reaction
Another type of browning, the glycation reactions (“Maillard
reaction”), also involves an oxidation/​reduction reaction. The
glycation reactions lie at the heart of so much flavor and color
development in grilled meats, baked bread, and melted cheeses.
Simple reducing sugars such as glucose react with and reduce
nitrogen-​containing amino acids to produce a whole array of
intermediate compounds that further react in somewhat unpredict-
able ways depending on pH, time, and temperature. The blend of
products, many of which are caramel colored and smell heavenly,
produce the sear on a steak that many of us crave. These reactions
transform the molecules in uncooked food to produce smaller
molecules that are volatile with succulent flavors and fragrances.
The reactions taking place are complex, and scientists have used
the field of electrochemistry (looking at the flow of electrons) to
learn about intermediates and discover ways that one might be
able to steer these complex reactions to the most desirable and
predictable outcomes. The action of that first reducing sugar with
that first amino group yields a group of intermediate compounds,
“reductones”, known for their exquisite ability to go on and react
with and reduce other molecules.
The general Amadori reaction shown in Figure 103.2
(Biosynthesis, 2019), illustrates the reactants: the six-​ carbon
glucose-​ related molecule reduces the nitrogen-​ containing
amino acid-​related molecule. The product, the now condensed
carbanolimine, easily loses a water molecule to become
what is known as a Schiff base, which itself is quite reactive
(Figure 103.3).
Many food scientists focus on how to predict, measure, and
control the flavor development paths and consider the molecular
transformations, the redox chemistry, that transforms an odorless
Molecular Gastronomy: A Universal Portal
and rather tasteless mixture into a griller’s delight (Martins
et al., 2001).
Enzyme Kinetics
Catalysts cause reactions to go faster, and biological catalysts
are most often proteins called enzymes. Chefs sometimes take
advantage of enzymes to accelerate the rate of a reaction, as when
tenderizing meat with the enzyme papain, or in some cases, slow
it down, such as using lemon juice to slow the browning of fruit
by the enzyme phenolase.
Students in biology and chemistry learn about the field of
enzyme kinetics –​the study of how enzymes work to change the
speed of reactions. Enzymes bind to the reactants, or substrates,
in a manner that is exquisitely selective, picking out a particu-
larly shaped molecule and excluding other very similarly shaped
molecules.
This property is described by a term, Km, the Michaelis con-
stant, and in the peculiar way of scientists’ choice of scales
(remember pH), the higher the Km, the weaker the binding.
A second parameter that describes enzymes’ capacity to cata-
lyze reactions is kcat, the catalytic rate constant. It is critical for
students to master understanding of these terms in the study of
FIGURE 103.2 Glucose (C6H12O6, left) is a reducing sugar that reacts
with an amino acid (H2N-​R) to produce an oxidized intermediate known as
a carbanolimine.
FIGURE 103.3 The carbanolimine loses a water molecule to become a
reactive Schiff base.
FIGURE 103.4 Cleavage of the glycosidic bond of sucrose by the enzyme invertase results in the inversion of the optical rotation.
685
biochemistry. This understanding of enzyme kinetics was laid out
by a team of scientists –​the German Leonore Michaelis and the
Canadian Maud Menten –​over 100 years ago. The enzyme they
were studying was invertase (Science Borealis, 2019).
Students often become interested when they learn that invertase
is used in the confectionery industry to speed up the rate at which
the disaccharide sucrose is broken down into the monosaccharides
glucose and fructose (Figure 103.4). Uncatalyzed, this reac-
tion takes place over years, if not decades. When invertase
is added to sucrose, the reaction is complete in just hours or
days. Confectioners use this trick in creating sweets with liquid
centers, such as chocolate-​ covered cherries. What better way
to introduce these foundational concepts in enzymology than
by having students create for themselves a chocolate-​covered
cherry with a sweet liquid center: one that starts out as a solid
paste that surrounds the fruit but is transformed into a liquid gel
by an enzyme? Maud Menten knew that the enzyme would bind
to the sucrose substrate (as measured by 1/Km) and then accel-
erate the velocity of the reaction: the addition of water to the bond
connecting the two monosaccharides, as measured by kcat, or Vmax
(maximum velocity). The appearance of the product molecules
can be detected using light by measuring the “optical rotation” of
the sample, which goes from positive to negative as the sucrose
is converted to product; thus the name invertase (Figure 103.5).
Elasticity
A Middle Eastern frozen sweet dairy treat sold by street-​side
vendors, dondurma, is a relative of ice cream. It can be used as
a delicious example of elasticity in foods. Many food products
include thickeners such as carrageenan or xanthum gum to
augment the mouthfeel of our food, making it more palatable and
its appearance more pleasing. Dondurma includes a plant product,
salep or Sahlab, which is a powder made from orchid roots and
mastic, which comes from a tree resin and adds to the chewiness.
The salep powder is hydrated when added to the warmed milk
products and stirred with repeated pouring from a ladle held high
above the pot. As the plant starches are heated and poured, they
begin to align in the hot liquid and stay aligned when the cream is
frozen. The result is an exquisite treat that is almost as much fun
to eat as it is to make and doesn’t drip or melt easily. The exercise
allows a discussion of biopolymers and their properties, and one
can measure the elastic modulus of the ice cream with various
amounts of the orchid root powder added. Various thickeners can
686 Patricia B. O’Hara

FIGURE 103.5 The Michaelis–​Menten saturation curve: a generic plot of

the rate at which an enzyme’s product is produced, d[P]‌/​dt, with respect to the
concentration of the starting material (substrate), [S]. The Michaelis constant
KM is determined by the rates of the three key processes (enzyme–​substrate
binding, chemical conversion of substrate to product, and release of the final
product from the enzyme.)

be compared, and students can be challenged to imagine different

methods for measuring the elasticity (Figure 103.6).
Connecting scientific principles to culinary practices is an
effective way to interest students by making the abstract con-
crete. The examples here range from those that address basic
concepts, such as acidity and basicity and oxidation and reduction
reactions, to more complex topics such as enzyme kinetics and
elasticity. The appeal of this approach can be measured in many
ways, but one concrete example is the chronic oversubscription FIGURE 103.6 Elasticity of dondurma, a frozen dairy dessert sold in
and long wait list for a course, Molecular Gastronomy, offered at Turkey, is due to the molecular properties of two ingredients, salep (lower
my institution. right) and mastic (upper left), which also provide its unique mouthfeel and
texture.

REFERENCES
McGee H. 2004. On Food and Cooking: The Science and Lore of The Martins SIFS, Jongen WMF, van Boekel WJJS. 2001. A review of
Kitchen, Scribner, New York, NY. Maillard reaction in food and implications to kinetic modeling,
Stuckey B. 2012. Taste: Surprising Stories and Science about Why Trends in Food Science and Technology, 11, 364–​373.
Food Tastes Good, Atria, New York, NY. Science Borealis. 2019. https://​blog.scienceborealis.ca/​maud-​leonora-
Turkell MH. 2017. Acid Trip: Travels in the World of Vinegar, menten.
Abrams, New York, NY.
Biosynthesis. 2019. www.biosyn.com/​tew/​The-​Maillard-​reaction-​
and-​Amadori-​rearrangement.aspx
Heat Transfer in the Kitchen –​Exercises
Manuel Combes
Teacher of Physics in Prep Class, La Pérouse-​Kerichen, Brest, France
Research Associate in Laboratoire Géosciences Océan (UMR 6538), Brest, France
Cooking Pasta
You are cooking spaghetti in boiling water (Figure 104.1). Two
processes control the quality of the pasta: wheat starch gelatiniza-
tion, which occurs at 60 °C, and protein coagulation (necessary to
avoid disintegrating spaghetti) at temperatures above 80 °C. The
pasta is cooked when 95% of the volume is gelatinized, lasting
around 9 minutes.
Let us consider the parameters for this operation.
Geometry: spaghetti of diameter d = 1 mm; pot of radius
r = 10 cm and height H = 10 cm.
Pasta parameters: m = 250 g; k = 0.45 W.m−1.K−1;
ρ = 1500 kg.m−3; c = 1800 J.kg−1.K−1
Water parameters: mw = 1 kg; cw = 4180 J.kg−1.K−1;
Lvap(100 °C) = 2260 kJ.kg−1
Electric hob power: P = 1200 W
Air convection intensity is characterized by a Newton coeffi-
cient h = 20 W.m−2.K−1
Questions:
(1) Estimate the temperature decrease when the spa-
ghetti (at room temperature) is put into the boiling
water. How much time will it take to bring it again to
the boil?
(2) How much time does it take to start gelatinization
and coagulation in the core of the spaghetti?
(3) Apart from starch gelatinization and protein coagu-
lation, how much energy is used for cooking pasta
here, considering initial water temperature elevation,
FIGURE 104.1 The setup.
heat loss at the pot surface, thermal radiation, and
vapor exhaust? Estimate the mass of evacuated vapor.
Answers:
(1) Considering an adiabatic transformation of the
isolated system {water + spaghetti}, its enthalpy is
conserved:
ΔH = 0 = mwcw (Tf − Tboil) + mc (Tf − Tamb)
where Tboil = 100 °C and Tamb = 20 °C lead to a final
temperature Tf = 92 °C. To bring it back to 100 °C
using an electric hob power P, we need a time
Δt = (mc + mwcw)(Tboil − Tf)/​P ~ 30 s.
(2) One can extract from the heat equation a character-
istic diffusion time t ~ ρc(d/​2)² /​k < 3 s. The cooking
time for spaghetti is then controlled by gelatinization.
(3) The cost for the temperature elevation of the water
from 20 to 100 °C is:
ΔHw, elev = mwcw (Tboil − Tamb) = 334 kJ.
Taking into account the hob (reaching 200 °C, with a thermal cap-
acity Kh ~ 500 J.kg−1.K−1) and the pot (reaching 100°C, thermal
capacity Kp ~ 100 J.kg−1.K−1), we can add a supplementary cost:
ΔH′elev = Kh (100 + Tboil − Tamb) + Kp (Tboil − Tamb) ~ 100 kJ.
During the cooking time (t = 9 min), heat loss at the pot surface
(S = 2πrH + 2πr²) is:
ΔHloss = hS (Tboil − Tamb) t ~ 110 kJ.
Thermal radiation can be estimated with Stefan’s law:
ΔHrad = σTboil4 S t ~ 75 kJ.
In the meantime, the electric hob yields ΔHpower = P t = 648 kJ,
the difference being mainly used to evaporate water. This corres-
ponds to a mass of vapor exhausted during 9 minutes of:
Δm ~ (ΔHpower –​ ΔHw, elev –​ ΔH′elev –​ ΔHloss –​ ΔHrad)/​Lvap ~ 15 g.
687
688
Freezing an Apple Sauce
Consider a food tray filled with apple sauce, placed in a freezer
at temperature Tf = −18 °C. In a first approach, the edge effects
are neglected, considering only thermal transfer in the direction
of thickness e = 4 cm.
Geometry: food tray dimensions are 30 × 20 × 4 cm.
Apple sauce parameters: k = 1.3 W.m−1.K−1, ρ = 1500 kg.m−3,
cliq = 3500 J.kg−1.K−1, csol = 1500 J.kg−1.K−1, Lfus(0 °C) =
334 kJ.kg−1.
Freezer parameters: h = 5 W.m−2.K−1 (free convection) or
h = 20 W.m−2.K−1 (forced convection).
Questions:
(1) Compare the sensible and latent heats lost by the
food tray during its cooling from the liquid phase at
5 °C to the solid phase at 0 °C.
(2) Under which temporal conditions can we use
thermal resistances to describe heat transfer through
the food tray?
(3) Considering that both sensible and latent heats are
evacuated simultaneously when the freezing front
reaches a given depth, estimate the time necessary to
get a solid apple sauce in the freezer.
Answers:
(1) Latent heat L = Lfus(0 °C) = 334 kJ.kg−1 and sensible
heat: q = cliq [T5 °C –​ T0 °C] = 17.5 kJ.kg−1.
(2) A quasi-​ stationary thermal regime is needed to
use thermal resistances. Here, the characteristic
time for thermal diffusion over a depth e is td ~
ρcsole 2
= 45 min . The migration of the freezing
k = exp   with τ =
front must be much slower than this to introduce the
thermal resistance of the solid phase.
(3) The external and internal resistances are connected
in series, leading to a global resistance, Rtot, equal to
the sum of the two.
The heat flow through a solid phase of surface area
S and thickness x (position of the freezing front at
T = 0 °C), with external temperature Tf = −18 °C, is
j(x).S = (Tf –​ T)/​Rtot = (Tf –​ T)S /​ [x/​k + 1/​h].
During a short time, dt, a new mass dm = ρ.S.dx
is frozen, and the amount of evacuated heat is:
j(x).S.dt = ρ.S.dx (L+q).
This allows us to integrate the total duration Δt of the
freezing front migration:
dt = ρ (L + q) dx/​j(x)
Manuel Combes
leading to: Δt = [e²/​8k +e/​2h]. ρ (L+q)/​(T –​ Tf) ~ 29 h
for free convection, and 8 h for forced convection.
Refrigerating a Soup Tray
The growth of microbial populations is reduced below 10 °C.
Consider the refrigeration of a food tray filled with hot soup. In
a first approach, the edge effects are neglected, considering only
thermal transfers in the direction of thickness e = 4 cm.
Geometry: food tray dimensions are 30 × 20 × 4 cm.
Soup parameters: k = 0.5 W.m−1.K−1, ρ = 1500 kg.m−3,
c = 3500 J.kg−1.K−1, T0 = 50°C.
Refrigerator parameters: h = 5 W.m−2.K−1 (free convection) or
h = 20 W.m−2.K−1 (forced convection).
Questions:
(1) What is the time to reach T = 10 °C (in order to limit
microbial growth) in a refrigerator at Tf = 4 °C?
(2) What if we consider a bigger container of thickness
e′ = 20 cm?
Answers:
(1) The Biot number, Bi, compares the internal con-
ductive resistance with the external convective
resistance. Here, considering a distance e/​2 from
surface to core, Bi = he/​2k = 0.2 with free convec-
tion, and Bi = 0.8 with forced convection. In a first
approximation, the external transfer is limiting only
for free convection. A global heat balance gives in
this case:
T − Tf  −t  mc ρ(e / 2)c
= = 21000 s.
T0 − Tf  τ hA h
The duration to reach T = 10 °C is then:
 T − Tf 
t = τ ln  0 = 12 h.
 T − T 
f
Taking an internal diffusion delay into account, the
duration would be slightly increased; the characteristic
ρc (e / 2 )
2
diffusion time is: t′ = = 1h10 .
k
(2) For a thickness e′ = 20 cm instead of 4 cm, edge
effects are no longer negligible. In addition, Bi = 1
for free convection (and Bi = 4 in a ventilated fridge);
in this case, both heat transfer modes must be taken
into account (numerical resolution).
Ionic Diffusion in Spherified Calcium Alginate Gels: A Laboratory
Experiment

Lorenzo Soprani, Lara Querciagrossa, Silvia Cristofaro, Luca Muccioli, Silvia Orlandi, Elena Strocchi,
Alberto Arcioni and Roberto Berardi
Dipartimento di Chimica Industriale “Toso Montanari”, Università di Bologna, Viale del Risorgimento 4, I-​40136, Bologna, Italy

Introduction are composed of three polymers: cellulose (long and stiff fibres,
giving mechanical endurance), hemicellulose (a branched sugar-​
Dilute water solutions of sodium alginate are used in molecular
based polymer) and alginic acid (Cosgrove, 2005). These last two
cooking and the dairy industry as thickening and gelling agents
materials act as a “glue”, giving flexibility to the stiff cellulose
(Barham et al., 2010) and also as edible films. Typically, they are
fibre matrix. Alginic acid is typically present as sodium, calcium
obtained from brown algae or seaweeds and form stable gels in
or magnesium alginate. Terrestrial plants, and in particular their
the presence of calcium ions.
fruits, contain large amounts of pectin instead of alginic acid as
Calcium alginate gels can be easily prepared in the form of
a second cellulose glue: this material can be used as a gelling
small solid spheres by dropping, with a syringe or a pipette, the
agent in the preparation of fruit preserves, conferring on them
sodium alginate solution into a water bath of calcium ions (“direct
their jelly-​like consistency, aspect, and mechanical and textural
method”). After rinsing the calcium alginate spheres, fully or just
properties, as well as increasing their resistance to bacterial
partially jellified, their external pH can be easily regulated using
biodegradation.
ordinary acid and/​or basic solutions. The difference of the pH
The major industrial source of alginic acid is cultivated marine
value between the inside and the outside of the spheres drives the
brown weeds of the Laminaria family. The process of sodium
diffusion of H3O+ ions from the solution to the liquid phase, which
alginate production is schematically shown in Figure 105.1
is trapped in the porous, sponge-​like, calcium alginate gel. If a suit-
(McHugh, 1987). It shows how pH, as well as Na+ and Ca2+
able pH indicator is added to the initial sodium alginate solution,
concentrations, can be used to separate and purify water-​soluble
the time-​dependent pH variation inside the sphere can be easily
sodium alginate from chopped seaweeds following two different
detected as a gradual change in its colour, and also by an evolving
pathways.
colour gradient. This colour change is observed moving from
From the chemical point of view (Figure 105.2), alginic acid
the surface towards the centre of the solid sphere. As one would
and its salts are copolymers made of random blocks of β-​D-​
expect for a solid, no change of shape and/​or size of the sphere
mannuronate (M) and α-​L-​guluronate (G) monomers (Draget
occurs, whereas the fast and complete diffusion from the outside
et al., 2005).
to the core of the sphere represents the typical behaviour of liquid
As shown in Figure 105.3a, the glucosidic bond between two
systems. This experiment clearly shows how calcium alginate gels
neighbouring G units leads to the formation of a cavity, which
surprisingly share some physical properties of both states of matter.
can be filled by positive ions, e.g., Ca2+, attracted by the negative
The dual solid/​liquid nature of calcium alginate gels, strikingly
charges of the hydroxyl and carboxyl groups of α-​L-​guluronate.
highlighted by the preceding example, has been extensively used
Alginic acid at basic pH and its sodium salts are soluble in
over many decades in several fields, from dentistry to the food and
water thanks to the great number of strong hydrogen bonds that
dairy industry, molecular gastronomy and, more recently, regen-
can be formed with the solvent and to the electrostatic repulsion
erative medicine, lithium-​ ion batteries, self-​
healing materials,
between negatively charged carboxylate groups, which does not
and biodegradable disposable water pouches. Further details and
favour the aggregation of polymer chains. Conversely, at low
references are reported in the section “Uses of Sodium Alginate”.
pH, alginic acid is protonated and as such, not soluble in water
(Draget et al., 2005).
If calcium ions are present in a sodium alginate solution,
Sources and Chemistry of Alginic Acid and they fill the available cavities formed by the GG pairs and bind
Sodium Alginate strongly to them (Figure 105.3a). At the same time, calcium ions
Plant (eukaryotic) cells are characterized by a semi-​rigid cell wall are large enough to protrude from each cavity and coordinate a
enveloping the phospholipidic membrane. Brown algae cell walls still free GG pair cavity in a different position or belonging to

689
690
FIGURE 105.1 Flow chart for the production of sodium alginate from seaweeds.
(After McHugh, 1987)
another nearby polymer chain. These opposite pairs of GG cav-
ities filled by the same calcium ion are the elementary units of the
so-​called “egg box model” (Figure 105.3b) (Grant et al., 1973;
Li et al., 2007).
The appearance of these casual 3D junctions and linkages is
responsible for the insolubility of calcium alginate, since they
readily transform the sodium alginate liquid solution into a
sponge-​like gel (Figure 105.4).
The formation of a calcium alginate gel is not thermoreversible:
when heated, water is permanently lost and spheres become
dry. Nonetheless, a calcium alginate gel can still be restored to
its constituents through approaches aimed at weakening and
eventually destroying the gel structure by reducing either the
number of calcium ions or the number of GG cavities avail-
able to form the basic egg box structures. Three approaches are
easy to perform with students in a laboratory experiment on a
water solution containing alginate spheres: (1) to lower the solu-
tion pH below the pKa of alginic acid in water, so that the poly-
meric chains become insoluble; (2) to increase the pH and use
ethylenediaminetetraacetic acid (EDTA), which strongly binds
calcium ions and subtracts them from the egg box formation
equilibrium; (3) to increase the pH and use an excess of Na2CO3,
Lorenzo Soprani et al.
which acts as a source of Na+ ions and removes Ca2+ by forming
solid insoluble CaCO3. The third is the most common industrial
process used to prepare the sodium alginate salt (Figure 105.1).
Uses of Sodium Alginate
Medicine and Dentistry
Stiff and strong calcium alginate gels have been used for a long
time to prepare dental casts, and more recently, other innovative
applications have been reported. For instance, reconstruction of
damaged cardiac tissues after a stroke can be helped using a 3D
matrix gel with trapped gold nanofilaments (Dvir, 2011).
Moreover, calcium alginate spheres can be used as vehicles
for drug delivery. Alginate capsules are used to cover healthy
cells that can be grafted into patients to repair damaged tissues
and/​or to produce active ingredients without causing adverse
reactions to the immune system (Dettmar et al., 2011; Lee and
Mooney, 2012). For instance, drugs, like insulin, can be provided
to a patient by inserting a healthy cell that produces insulin in
an alginate capsule. Insulin diffuses through the wall of the cap-
sule into the patient’s bloodstream, while patient antibodies are
Ionic Diffusion: A Laboratory Experiment 691

(a)

H H H OH
– –
HO O COO OOC O H
H HO
H OH H OH
HO H
HO H HO H

β-D-mannuronate (M) α-L-guluronate (G)

(b)

G G M M G
– – –
OOC OH OOC OOC
O O O HO O
OH
OH
OH HO O OH
O – O
O OH OOC O

O O
OH –
OOC OH

(c)

MMMMGMGGGGGMGMGGGGGGGGMMGMGMGGM
M-block G-block G-block MG-block

FIGURE 105.2 Alginate monomers (a), polymer 3D shape (b) and example of M-​and G-​blocks in a random copolymer (c).
(Adapted from Draget et al., 2005)

-
OOC OH
Ca2+
O O

OH O
O OH

O
-
OH OOC

(a) (b)

FIGURE 105.3 (a) A cavity formed by two neighbouring G units, which bind a calcium ion through electrostatic interactions with oxygen-​containing groups.
(b) Two parallel polymeric chains bound by means of electrostatic interactions with calcium ions, forming the so-​called egg box junction.
((a) Adapted from Draget et al., 2005; (b) adapted from McHugh, 1987)

unable to reach the host cell inside the capsule because their size
is larger than the calcium alginate porosity (Tan and Takeuchi,
2007; Martinez et al., 2012; Tahtat et al., 2013).
In a similar way, alginate salts can be used for cartilage and
bone tissue engineering (Liu et al., 2017).
Innovative Materials
Alginic acid and its derivatives have been used as components
of highly stretchable and tough hydrogels, as reported by
Niekraszewicz and Niekraszewicz (2009) and Sun et al.
(2012). High-​capacity Li-​ion batteries can be produced with Si-​
nanoparticles embedded in alginate gels (Kovalenko et al., 2011;
Gu et al., 2014; Yoon et al., 2014).
Alginate gels are employed also as biodegradable blob-​like
water containers called “Ooho”, produced by the Skipping
Rocks Lab. The “Ooho” spheres are made by dropping ice into
solutions of calcium salt and sodium alginate, leading to small
water volumes enclosed in membranes. Besides degrading in
four to six weeks, the capsules are edible, since they are made of
100% brown algae. The manufacturing process is easy and cheap,
692
Ca2+
Ca2+
FIGURE 105.4 Effect of increasing Ca2+ concentration on alginate
chain aggregation: 3D junctions and crosslinks progressively appear in
random positions and transform the sodium alginate liquid solution into a
sponge-​like gel.
(Adapted from McHugh, 1987)
representing a step towards plastic-​free packaging (www.notpla.
com/​).
Art
The calcium alginates used in dentistry are also used as the main
materials for a sculpture technique called life-​casting. It consists
of recreating a 3D copy of a living human body, or part of it, with
the aid of moulding and casting. Details as small as fingerprints
and pores can be reproduced.
Molecular Cooking
In the food industry, calcium alginate has been used for about
50 years as a thickening and gelling agent (e.g., in ice creams
and yoghurt). Nowadays, the spherification process of calcium
alginate is commonly used in the preparation of molecular
Lorenzo Soprani et al.
cocktails and salads to create small caviar-​like spheres, which
release a flavour burst in the mouth when they are crushed by
the teeth. Sodium alginate is first mixed with the tasty/​flavoured
liquid to be spherified. This liquid is then dropped into a solu-
tion with calcium chloride (CaCl2), calcium lactate or calcium
lactate gluconate, generating a sealing membrane, which forms
the spheres. After this treatment, the spheres can be removed and
rinsed in distilled water, ready to be served in cocktails.
Alginate gels are also used for low-​cost production of lactose-​
free milk by room temperature enzymatic degradation with lac-
tase. The lactase enzyme is first immobilized inside the calcium
alginate spheres during the spherification. These spheres are then
employed as a solid-​state catalyst in a continuous flow reactor,
and the raw milk is percolated through them, thus allowing the
disaccharide lactose molecules diffusing inside the spheres to be
split into their glucose and galactose monomeric units, which then
do not cause allergic reactions in humans (Woodward, 1985).
Physical Properties of Calcium Alginate
In the direct method, the thickness of the external skin of cal-
cium alginate spheres can be optimized by choosing the time
they remain in the calcium bath: shorter times lead to a thin skin,
longer ones to a thicker one and eventually to a completely jelli-
fied sphere.
In addition, the liquid nature of the gel can be probed by
modifying external conditions and observing how they influence
the inner part of the sphere and its shape (Chuang et al., 2017).
Alginate gels resemble wet sponges: they are characterized
by liquid-​filled nano or micro canals and cavities, able to let
small molecules flow through them (Shapiro and Cohen, 1997;
Boontheekul et al., 2005). Therefore, by modifying the pH value
of the medium outside and using a pH indicator, one can verify
that the pH of the inner part of the sphere changes. The colour
change does not occur instantaneously, just as when a tea bag
slowly modifies the colour of warm water during the infusion
period of tea, even though, for the jellified spheres, the indicator
starts changing from the outside and slowly reaches the inside.
The underlying physical process is diffusion, an entropy-​
driven net mass movement of a chemical species from a region
of higher concentration to a region of lower concentration, which
in a liquid is assisted by thermal random motion of solute and
solvent particles. Diffusion takes place also in the absence of
concentration gradients, and in this case, the disordered motion
produced by random collisions is termed Brownian or random-​
walk. In both cases, the distance travelled by a given particle, or
by the concentration front, is proportional to the square root of
the elapsed time.
Experimental Protocol
In this section, we describe in detail the various steps of the pro-
cedure adopted for the preparation of calcium alginate spheres
and for detecting the diffusion of protonated water inside them by
observing the colour change of a suitable pH indicator contained
in the spheres, as schematically depicted in Figure 105.5.
Ionic Diffusion: A Laboratory Experiment 693

Prepare sodium Wait for complete Place drop in Measure turning

alginate solution spherification solution at low pH times and drop
dimensions

FIGURE 105.5 Flowchart for calcium alginate droplets’ preparation and diffusion experiment.
(Photos by Studio Fotografico Marco Fortini, Bologna, Italy)

Preparation of Sodium Alginate Solution

The first step consisted in preparing a sodium alginate solution at
an adequate concentration.
In general, the smaller the sodium alginate concentration, the
higher the gel porosity and the faster the diffusion process. After
testing several concentration values, we decided to use a solu-
tion at 1% w/​w, even though this is still fairly concentrated with
respect to culinary concentrations (~0.5% w/​w): this is a good
compromise between mechanical resistance of the spheres and
diffusion time of H3O+ ions inside them.
The pH of this sodium alginate solution was measured with a
universal indicator as almost neutral (Figure 105.6).
Sodium alginate is characterized by slow solubilization kin-
etics and presents technical drawbacks. In order to obtain a uni-
form solution, clots should be broken, first with a mortar and
then with an ultrasonic bath. In molecular cooking, this is not
a viable option for chefs, so alginate is usually prepared 24 h FIGURE 105.6 Neutrality of the initial sodium alginate solution, measured
with a universal pH indicator.
before usage by mixing water –​possibly salty –​and sodium
alginate with a blender. The solution we obtained in this way,
which incorporated air bubbles, was put overnight in a fridge:
even though the low temperature increases the solution viscosity,
it also reduces gas solubility and allows air bubbles to escape
the mixture, yielding a clear and transparent liquid. Some chefs
also use vacuum packaging machines, when available, to help to
remove the air bubbles.
Once we obtained a homogeneous solution, we divided it
into two batches and added, as pH indicators, phenolphthalein
(P) in the first one and bromothymol blue (B) in the second batch
(Figure 105.7). The indicator amount was one drop for each gram
of alginate solution. During this step, it is important to take care
to mix slowly: this avoids incorporating air bubbles again into
the alginate. Note that phenolphthalein in these amounts is not
FIGURE 105.7 Sodium alginate solution at almost neutral pH with two
completely soluble in alginate solution and partially precipitates,
different pH indicators: phenolphthalein (left) and bromothymol blue (right).
making the solution slightly opaque.
(Photo by Studio Fotografico Marco Fortini, Bologna, Italy)

Jellification and Conservation Bath

In order to obtain a more uniform jellification, the Ca2+ ions the sodium alginate solution is dropped with calcium carbonate
should be introduced into sodium alginate solution through a fine into an acidic solution, the H3O+ ions diffuse into the sphere and
powder of CaCO3, so that the dispersion is homogeneous. When induce CaCO3 solubilization, generating Ca2+ ions in situ. Since
694 Lorenzo Soprani et al.

for the scope of this work this would have been an unnecessary sufficient. We suggest letting the droplets rest in the last bath for
expedient, we decided to use a CaCl2 solution as a source of at least 15 min.
Ca2+ ions. Two consecutive pH measurements, one performed at the end
We prepared a solution of CaCl2 at 2% w/​w. As before, we of this procedure and the other one on the following day, gave
divided it into two batches and added, as pH indicators, P to the nearly identical results (pH ≈ 10.5).
first batch and B to the second, in order to have the same indicator
concentration as in the alginate solution (one drop of indicator for Preparation of Acidic CaCl2 Solutions and
each gram of CaCl2 solution). We then brought the pH value to
Measurements of Turning Times and Diameter of
10.0–​10.5 by adding NaOH. The pH was measured with an Amel
338 pH meter. Spheres
Note that the indicator was added both to the CaCl2 and to After the spherification was complete and the system had reached
the sodium alginate solution before dropping the latter into the equilibrium in the original bath conditions, we took some spheres
former, in order to avoid a net flow of the colourant from the of similar size from the solution and placed them into three
CaCl2 solution to the inside of the droplet and vice versa: in this
way, we do not have to take care of pH indicator diffusion, which
could be slower compared with that of H3O+ and Ca2+.
The use of a pH indicator should allow a clear distinction
between the basic and acidic colours. For this reason, we chose
to analyse both phenolphthalein and bromothymol blue. The first
indicator is magenta in basic solution and colourless at acid pH,
while the second one turns from blue to yellow for the same pH
variation.
Once the jellification bath was prepared, we dropped sodium
alginate 1% w/​w solution into it with a plastic pipette from a
suitable and constant height (~5 cm from the surface of the jel-
lification bath) in order to obtain spherical droplets of uniform
dimensions (Figure 105.8, where we used a solution without
pH indicator just to make the droplets more visible). Indeed, the
impact with the liquid from an excessive height would deform
the droplet, while dropping from an insufficient distance from the
solution surface would produce droplets with a “tail”. Other pos-
sible deformation effects are shown in Figure 105.9.
We let the drops rest inside the solution for 20–​30 min to allow
the diffusion and jellification processes to conclude. After this
period, one can reasonably assume that ions (H3O+ and Ca2+) and
indicator are uniformly distributed inside the spheres.
We then transferred the droplets into a second conservation
bath (Figure 105.10), with the same composition as the first one,
since the pH and the Ca2+ concentration of the original bath were
slightly decreased after the diffusion inside the droplets. This FIGURE 105.8 Dropping sodium alginate into CaCl2 solution (the solution
transfer procedure must be repeated until the pH of the bath in the picture is without pH indicator to make droplets more visible).
stays constant: we observed that one additional step is typically (Photo by Studio Fotografico Marco Fortini, Bologna, Italy)

FIGURE 105.9 Effect of inclusion of air in a droplet (a) and formation of two conjugated drops (b).
Ionic Diffusion: A Laboratory Experiment
different 2% w/​w CaCl2 solutions (the same as used for the jel-
lification process) at pH values of 2.0, 3.0 and 3.5, respectively.
The higher H3O+ concentration pushes ions to diffuse from the
bulk of the solution into the droplet core: this provokes a gradual
variation of pH inside the spheres, detected thanks to a change in
the indicator colour, starting from the external skin and moving
towards the centre. In Figure 105.11 we show some photos of two
calcium alginate droplets containing the pH indicators P and B,
respectively, which change their colour as a consequence of the
H3O+ diffusion: in (a) phenolphthalein turns from magenta to col-
ourless, and in (b) bromothymol blue turns from blue to yellow.
We measured the turning time (τ, i.e., the time required by the
pH indicator inside a droplet to completely change its colour due
to pH variation) for a total of five droplets for each value of pH.
Timings were recorded with a stopwatch, and at the end of the
diffusion process, we measured the diameter of the spheres with
FIGURE 105.10 Conservation bath with P indicator.
(Photo by Studio Fotografico Marco Fortini, Bologna, Italy)
FIGURE 105.11 Effect of H3O+ diffusion inside two calcium alginate spheres as a function of time. (a) Phenolphthalein changing from magenta to colourless
and (b) bromothymol blue changing from blue to yellow.
(Photo by Studio Fotografico Marco Fortini, Bologna, Italy)
695
a calliper (Figure 105.12). We also took time-​lapse videos of the
turnings to analyse them with a “simulated” digitalized spectro-
photometric procedure, as explained in the next section.
As a summary of the experimental protocol, there follow step-​
by-​step instructions, also illustrated in Figure 105.13.
1. Preparation of 500 g of 1% w/​w sodium alginate solu-
tion. (We suggest the preparation of at least 250 g of
sodium alginate solution, since it must be blended
with a mixer, which is difficult with small volumes of
mixture.)
1.1. Weigh 5 g of sodium alginate and add this to
495 g of distilled water.
1.2. Blend with a mixer and let it de-​gas overnight in a
fridge (this can take some hours).
1.3. Put 10 g of sodium alginate solution into a vessel,
add 10 drops of 1% w/​w phenolphthalein (P) (1:1
FIGURE 105.12 Measurement of sphere diameter with a calliper.
696
FIGURE 105.13 Graphic illustration of the experimental procedure.
water-​ ethanol solution) and mix carefully,
avoiding air incorporation.
1.4. Put 10 g of sodium alginate solution into another
vessel, add 10 drops of 0.04 % w/​w bromothymol
blue (B) (water solution) and mix carefully,
avoiding air incorporation.
1.5. Measure the solution pH with a universal indi-
cator and take note of the value.
2. Preparation of 2% w/​w CaCl2 solution.
2.1. Weigh 21 g of CaCl2 and dissolve in 1030 g of
distilled water.
3. Jellification and conservation bath.
3.1. Calibrate the pH meter for basic pH
measurements.
3.2. Put 550 mL of CaCl2 solution prepared as in point
2.1 into a beaker with a magnetic stirring bar,
place it on a magnetic stirrer and switch it on.
3.3. Using the pH meter, bring the pH value to 10.0–​
10.5 by adding 0.5 M NaOH solution (take note of
the final value).
Lorenzo Soprani et al.
3.4. Transfer 200 mL of CaCl2 solution into another
beaker and add 20 drops of P indicator (this will
be the jellification bath).
3.5. Transfer 30 mL of CaCl2 solution into another
beaker and add 30 drops of P indicator (this will
be the conservation bath).
3.6. Check the pH value of the conservation bath
(point 3.5) and, if necessary, adjust it to the value
measured at point 3.3.
3.7. By means of a plastic pipette cut at a suitable internal
diameter (2–​4 mm), drop the sodium alginate solu-
tion with P indicator (point 1.3) into the jellification
bath (point 3.4) so as to obtain about 20 droplets
with a good spherical shape (as described in section
“Jellification and conservation bath”; to obtain
droplets with a good spherical shape, the sodium
alginate solution should be dropped at a distance of
about 5 cm from the CaCl2 solution).
Let them rest inside the jellification bath for
20–​30 min.
Ionic Diffusion: A Laboratory Experiment
3.8. Transfer the droplets with a good spherical shape
into the conservation bath (point 3.5) and let them
rest there for at least 15 min.
3.9. Check the pH value of the conservation bath
(point 3.5) and, if necessary, adjust it to the value
measured at point 3.3. If the pH differs more than
0.5 pH units from that value, we suggest pre-
paring another conservation bath, transferring
the droplets into it and letting them rest there for
another 15 min.
3.10. Repeat the procedure from point 3.4 to 3.9 with
the B indicator.
4. Preparation of acidic CaCl2 solutions and measurements
of turning times and sphere diameter.
4.1. Calibrate the pH meter for acidic pH
measurements.
4.2. Put 150 mL of the CaCl2 solution prepared at
point 2.1 in a beaker with a magnetic stirring
bar, place it on a magnetic stirrer and switch
it on.
4.3. Using the pH meter, bring the pH value to 3.5 by
adding 0.5 M HCl solution (take note of the final
value).
4.4. Put 25 mL of the solution obtained at point 4.3
in a beaker, put a sphere with P indicator into the
solution and measure the time it takes to com-
pletely change its colour.
4.5. Repeat the procedure of point 4.4 so as to have at
least three measurements.
4.6. At the end of the diffusion process, measure the
diameter of each droplet with a calliper.
4.7. Following the procedure from point 4.2 to 4.6,
prepare 150 mL of pH 3 and 150 mL of pH 2
FIGURE 105.14 Schematic representation of the diffusive processes taking place into the droplets in the various phases of the experiment: initial neutral
sodium alginate drop, jellification process after the dropping in a basic CaCl2 solution (pH ≈ 10.5), basic jellified droplet at pH ≈ 10.5, diffusion of H3O+ ions
after the immersion of the droplet in an acidic CaCl2 solution (pH < 3.5), acidic jellified droplet at pH < 3.5.
697
CaCl2 solutions and measure the turning times at
these pH values and the droplet diameters.
4.8. Repeat the same measurements for droplets with
B indicator.
In Figure 105.14, we show a schematic representation of the
diffusive processes taking place into the droplets in the various
phases of the experiment.
Results
We collect here the observations obtained for five drops of
alginate with phenolphthalein and five with bromothymol blue.
The data reported in Figures 105.15 and 105.16 show turning
times τ relative to the droplets with P indicator and to those with
B indicator. We observed that τ slightly changes depending on the
nature of the indicator. Indeed, the two indicators possess a different
turning pH range: 8.0–​10.0 for P and 6.0–​7.6 for B, resulting in
faster turning of phenolphthalein compared with bromothymol blue.
Figure 105.15 shows τ as a function of the droplet diameter.
Note that, on average, increasing the size of the spheres will
increase as well the time required for the pH inside the droplets
to change, in principle with a quadratic dependence on the radius.
In Figure 105.16 we show the turning time averaged over
droplet diameter (〈τ〉) versus CaCl2 solution pH. Note that 〈τ〉
increases with CaCl2 solution pH. This is due to the fact that
the pH difference between the droplets and the solution (ΔpH),
which is the diffusion driving force, decreases. Indeed, inside the
gel porosity, we can reasonably assume that there is molecular
diffusion, described by the second Fick’s law: this states that
the diffusion rate is proportional to the Laplacian of the concen-
tration, which is linked to the difference of H3O+ concentration
between the droplet core and the solution bulk. As expected, the
698 Lorenzo Soprani et al.

FIGURE 105.15 Turning time with respect to calcium alginate droplet diameter at pH = 3.5 (P = phenolphthalein; B = bromothymol blue). The uncertainty
is comparable to the dimension of the dots.

FIGURE 105.16 Turning time averaged over calcium alginate droplet diameter versus CaCl2 solution pH (P = phenolphthalein; B = bromothymol blue).
When the uncertainty bar is not visible, it is smaller than the dimension of the dot.

two curves have similar trends, and the one relative to B indicator
lies above that relative to P indicator.
Digital Spectrophotometric Analysis
Lastly, we simulated a digitalized spectrophotometric analysis
performed, during the diffusion process, on a sphere with P indi-
cator and on another one with B indicator. This analysis allows
us to count as a function of time, for each frame, the number
of coloured pixels, which is linked to the absorbance signal of
a real spectrophotometric measure. This signal is connected to
the number of pH indicator molecules that have still not turned,
NI(t), and hence, to the diffusion progress. However, we have
to take into account that we analyse circular 2D frames, while
a hypothetical absorbance signal is associated with a sample
volume. To tackle this problem, we start defining the fraction
xp(t) as the ratio between the number of magenta (for P indicator)
or bluish (for B indicator) pixels, Np(t), and the total number of
pixels in the cropped frame, Ntot:
N p (t )
x p (t ) ≡ .
N tot
As we explained earlier, Np(t) is proportional to the area of a
circle C (Figure 105.17), which is the 2D projection of the sphere
S containing the fraction of pH indicator that has not turned yet.
Ionic Diffusion: A Laboratory Experiment
C and S having the same radius, rS(t), we obtain
xp (t ) ∝ N p (t ) ∝ rS2 (t ),
and, considering that the volume of S, VS(t), is proportional to
rS3(t), we finally get
VS (t ) ∝ rS3 (t ) ∝ xp3 / 2 (t ).
The volume VS (t) is also proportional to NI (t), since
N I (t ) = VS (t )C I ,
where CI is the pH indicator concentration inside the droplet.
Therefore, xp3/​2(t) can be assumed to be proportional to NI(t),
and then to the absorbance signal.
C droplet; see Figure 105.16 at pH = 2.0).
rS
FIGURE 105.17 Representation of the circle C, 2D projection of the
sphere S, which contains the molecules of pH indicator not yet turned, and
its radius rS.
FIGURE 105.18 Simulated digitalized spectrophotometric analysis performed on a time-​lapse video of a calcium alginate sphere with P indicator immersed
into a pH 2.0 solution of CaCl2. The graph shows the fraction of magenta pixels (xp) raised to 3/​2 versus time.
699
In Figure 105.18, we show the fraction of magenta pixels (xp)
raised to 3/​2 versus time obtained from a time-​lapse video of a
droplet with P indicator immersed into a pH 2.0 solution of CaCl2.
Similarly, in Figure 105.19, we report the fractions of bluish and
yellowish pixels (xp) raised to 3/​2 versus time for a droplet with B
indicator put into the same CaCl2 solution.
The frames of the videos have been cropped around the sphere
itself, and the RGB values for each pixel have been collected.
These three colour values have been converted into luminance,
saturation and hue values (www.niwa.nu/​2013/​05/​math-​behind-​
colorspace-​conversions-​rgb-​hsl/​). All pixels without grey hue
values have been converted into colours using a “Hue Range Map”
(www.workwithcolor.com/​color-​names-​01.htm). In the case of P
indicator, we calculated the fraction of pixels containing orange
and brown, red or magenta, and for B indicator, those containing
cyan, green or blue (for the bluish) and yellow or orange and
brown (for the yellowish).
For both droplets, the turning times obtained by image pro-
cessing (Figures 105.18 and 105.19) are in good agreement with
those obtained previously (~6 min for P droplet, ~10 min for B
The image analysis is superior to the visual one, since it is
quantitative, providing the time evolution of the number of pixels
for different colours, which in principle could be used for testing
kinetic models of the diffusion process inside the sphere.
The trends in the fraction of magenta and bluish pixels,
correlated with the average H3O+ concentration inside the droplet,
reflect those predicted by Fick’s law; indeed, the diffusion
rate (i.e., the curve slope) decreases with time, resulting in an
exponential-​like decay.
700
FIGURE 105.19 Simulated digitalized spectrophotometric analysis performed on a time-​lapse video of a calcium alginate sphere with B indicator put into a
pH 2.0 solution of CaCl2. The two datasets represent the fractions of yellowish and bluish pixels (xp) raised to 3/​2 versus time.
Conclusions
We described a simple experiment that allows us to verify that
the hydrogel nature of calcium alginate confers physical prop-
erties that one normally associates either with a solid (well-​
defined shape) or with a liquid (fast diffusion of solvated ions).
Moreover, we provided a detailed procedure that can be followed
to organize an easy and inexpensive laboratory experience that,
even if shown to high school students, can be adjusted, with
proper modifications, also for other educational levels. Indeed,
alginate spheres can represent a useful tool to show how diffu-
sive processes occur into a gel: the H3O+ ion diffusion can be
qualitatively followed by eye, or quantitatively through image
processing, thanks to the presence of a pH indicator inside the
sphere.
Since the diffusion coefficient is about ten times greater for
H3O+ than for Ca2+, and the diffusion time is inversely propor-
tional to the diffusion coefficient, our results for H3O+ can be
used also to estimate the jellification time for sodium alginate
solutions, which is directly linked to the diffusion time of Ca2+
ions. This knowledge can be useful for performing partial jelli-
fication –​a typical practice in the field of molecular cooking –​
where the aim is to obtain a jellified external “skin” and a liquid
core, so that the flavour contained inside the sphere spreads once
it is bitten.
And finally … a bit of Molecular Gastronomy art!
Lorenzo Soprani et al.
Supplementary Material
Additional images, videos and information about our group are
available on the website https://​
spheresmoleculargastronomy.
netsons.org/​.
Acknowledgements
We are very grateful to Marco Fortini of the Studio Fotografico
Marco Fortini (www.marcofortini.com/​) for his friendliness, will-
ingness and professionalism: patiently accompanying us during
this experience, he was able to produce the marvellous pictures
we used in this work.
We would also like to express our gratitude to Sandra Stipa,
Alberto Mucchi and Stefano Cerini, the laboratory technicians
of the Department of Industrial Chemistry “Toso Montanari”,
University of Bologna, for their great help and support in setting
up the required equipment and preparing the reactants.
Roberto Berardi wants to thank Cristina Felloni, Marina
Seganti and Christa Spieckermann for their precious suggestions
and stimulating discussions that turned out to be part of the inspir-
ation for the birth and the development of this work.
Finally, we are very grateful to all the Editors of this handbook
and, in particular, to Hervé This and Róisín Burke for involving
us in this project and giving us the opportunity to make a contri-
bution to the fascinating field of Molecular Gastronomy.
Dedication
We dedicate this chapter to the memory of our beloved friend and
gifted scientist Roberto a.k.a. “Bebo” Berardi, who most sadly
left us in January 2020.
Ionic Diffusion: A Laboratory Experiment
REFERENCES
Barham P, Skibsted LH, Bredie WLP, Frøst MB, Møller P, Risbo
J, Snitkjaer P, Mortensen LM. 2010. Molecular gastronomy:
A new emerging scientific discipline, Chemical Reviews, 110,
2313–​2365.
Boontheekul T, Kong HJ, Mooney DJ. 2005. Controlling alginate gel
degradation utilizing partial oxidation and bimodal molecular
weight distribution, Biomaterials, 26, 2455–​2465.
Chuang JJ, Huang YY, Lo SH, Hsu TF, Huang WY, Huang SL, Lin
YS. 2017. Effects of pH on the shape of alginate particles and
its release behavior, International Journal of Polymer Science,
Article ID 3902704, 1–​9.
Cosgrove DJ. 2005. Growth of the plant cell wall, Nature Reviews
Molecular Cell Biology, 6, 8500–​8861.
Dettmar PW, Strugala V, Richardson JC. 2011. The key role alginates
play in health, Food Hydrocolloids, 263–​266.
Draget KI, Smidsrød O, Skjåk-​Bræk G. 2005. Alginates from Algae.
Vol. 1. In Steinbüchel A, Rhee SK (eds.), Polysaccharides and
Polyamides in the Food Industry. Properties, Production and
Patents,Wiley, Weinheim, Germany.
Dvir T. 2011. Nanowired three-​dimensional cardiac patches, Nature
Nanotechnology, 6, 720–​725.
Grant GT, Morris ER, Rees DA, Smith PJC, Thom D. 1973.
Biological interactions between polysaccharides and divalent
cations: The egg-​box model, FEBS Letters, 195–​198.
Gu M, Xiao XC, Liu G, Thevuthasan S, Bae D, Zhang JG, Liu J,
Browning ND, Wang CM. 2014. Mesoscale origin of the
enhanced cycling-​stability of the Si-​conductive polymer anode
for Li-​ion batteries, Nature Scientific Reports, 4, 3684.
Kovalenko I, Zdyrko B, Magasinski A, Hertzberg B, Milicev Z,
Burtovyy R, Luzinov I, Yushin G. 2011. A major constituent of
brown algae for use in high-​capacity Li-​ion batteries, Science,
4, 75–​79.
Lee KY, Mooney DJ. 2012. Alginate: Properties and biomedical
applications, Progress in Polymer Science, 37, 106–​126.
Li L, Fang Y, Vreeker R, Appelqvist I, Mendes E. 2007. Reexamining
the egg-​ box model in calcium alginate gels with x-​ ray
diffraction, Biomacromolecules, 8, 464–​468.
701
Liu M, Zeng X, Ma C, Yi H, Ali Z, Mou X, Li S, Deng Y, He N. 2017.
Injectable hydrogels for cartilage and bone tissue engineering,
Bone Research, 5, 17014.
Martinez CJ, Kim JW, Ye C, Ortiz I, Rowat AC, Marquez M, Weitz
D. 2012. A microfluidic approach to encapsulate living cells
in uniform alginate hydrogel microparticles, Macromolecular
Bioscience, 12, 946–​951.
McHugh DJ. 1987. Production, properties and uses of alginates. In
McHugh DJ (ed.), Production and Utilization of Products from
Commercial Seaweeds, 58–​115. FAO Fish.Tech.Pap, WHO,
Rome, Italy.
Niekraszewicz B, Niekraszewicz A. 2009. The structure of alginate,
chitin and chitosan fiber. In Eichhorn SJ, Hearle JWS, Jaffe M,
Kikutani T (eds.), Handbook of Textile Fibre Structure. Nature,
Regenerated, Inorganic and Specialist Fibres, Vol. 2, 266–​304.
Elsevier, Amsterdam, The Netherlands.
Shapiro L, Cohen S. 1997. Novel alginate sponges for cell culture
and transplantation, Biomaterials, 583–​590.
Sun JY, Zhao X, Illeperuma WRK, Chaudhuri O, Oh KH, Mooney
DJ, Vlassak JJ, Suo Z. 2012. Highly stretchable and tough
hydrogels, Nature, 489, 133–​136.
Tahtat D, Mahlous M, Benamer S, Khodja AN, Oussedik-​Oumehdi
H, Laraba-​ Djebari F. 2013. Oral delivery of insulin from
alginate/​chitosan crosslinked by glutaraldehyde, International
Journal of Biological Macromolecules, 58, 160–​168.
Tan WH, Takeuchi S. 2007. Monodisperse alginate hydrogel
microbeads for cell encapsulation, Advanced Materials, 19,
2696–​2701.
Woodward J. 1985. Immobilised Cells and Enzymes: A Practical
Approach. Vol. 13. In Practical approach series, IRL Press,
Oxford, UK.
Yoon J, Oh DX, Jo C, Lee J, Hwang DS. 2014. Improvement of
desolvation; resilience of alginate binders for Si-​based anodes
in a lithium ion battery by calcium-​mediated cross-​linking,
Physical Chemistry Chemical Physics, 16, 25628–​25635.
Simple Calculations Based on Cooking
Hervé This vo Kientza1,2
1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
For Some of Us, Calculation Is a Need That Can
Become Fun after Realizing That It Is Easy
Because modern science and technology are more and more
based on mathematics, calculation skills have to be developed
during education (Owenson, 2012; Feser et al., 2013). However,
some students attracted by a particular science (e.g., chemistry, or
physics, or biology) do not see the interest of the particular field
of mathematics, defined as the exploration of the mathematical
realm for its own sake, and would prefer to limit themselves to
learning the calculations needed for their topic of interest. This is
the case for food science and technology, in particular.
Nonetheless, training in calculation is needed for all, as said so
well by George Polya in the introduction to his wonderful How to
Solve It (Polya, 1945):
A great discovery solves a great problem, but there is a grain
of discovery in the solution of any problem. Your problem
may be modest, but if it challenges your curiosity and brings
into play your inventive faculties, and if you solve it by your
own means, you may experience the tension and enjoy the tri-
umph of discovery. Such experiences at a susceptible age may
create a taste for mental work and leave their imprint on mind
and character for a lifetime.
Thus, a teacher of mathematics has a great opportunity. If
he fills his allotted time with drilling his students in routine
operations, he kills their interests, hampers their intellectual
development, and misuses his opportunity. But if he challenges
the curiosity of his students by setting them problems pro-
portionate to their knowledge, and helps them to solve their
problems with stimulating questions, he may give them a taste
for, and some means of, independent thinking.
Also a student whose college curriculum includes some
mathematics has a singular opportunity. This opportunity is
lost, of course, if he regards mathematics as a subject in which
he as to earn so and so much credit, and which he should
forget after the final examination as quickly as possible.
Of course, the last sentence, including the word “mathematics”,
can be disputed, but anyway, the question of solving problems is
important. Questions about cooking are helpful, because culinary
activities attract many students. Such questions often involve
physical or chemical changes that can be investigated quantita-
tively using simple physical and chemical knowledge.
Simple Knowledge
In this chapter, we discuss the methodological reasons behind
many failures when students perform exercises, and we propose a
structured method for helping them to succeed. Specifically, we:
1. explain that a restricted number of laws can be used for
the analysis of many culinary questions;
2. show that having a good method (including strategy)
is of the utmost importance, for the resolution of the
exercises and questions: this will be given as a “back-
bone calculation”;
3. explain in one example how easy it is to solve exercises
using this method;
4. show that one can invent various solutions based on par-
ticular aspects of the question;
5. give a list of questions based on cooking to which the
answer can be found using scientific knowledge at
bachelor level. In particular, determinations of orders
of magnitude or calculations on lattices (this will be
explained later) can be used.
Let us begin with the knowledge needed. For all problems listed
later, only basic information is needed:
1. Density ρ:
m
ρ=
v
where m is the mass and v the volume.
Indeed, even if this notion is introduced in high school, the
difference between density and weight remains sometimes
confused, and this is sometimes associated with the confusion
between “light” (or heavy) and “less dense” (or more dense),
or because it is faster to use “light” and “heavy”. Also, when
asked why oil floats on water, some students will explain that
703
704 Hervé This vo Kientza

it is “because oil is less dense” than water, which is obviously Indeed, this is the law that many students have learned, but it
a tautology and not an explanation. All this shows that culinary would be preferable that students learn instead:
questions using density can be a good opportunity to construct
 dp
useful exercises; in particular, one can use the one about cocktails
(see list later). ∑ forces =
dt
Finally, let us observe that this definition is an opportunity to

observe that density is a useful intellectual tool (as the hammer is where p is the momentum:
for the blacksmith), creating new ideas and describing the world  
better than usual words. p = mv

2. Pressure P: 7. The law of ideal gases:

F PV = nRT
P=
A
where P is the pressure, V the volume, n the number of moles,
where F is the force applied on an area A. R the constant of ideal gases, and T the absolute temperature.
As could have been observed for the first item (about density, Students should also know the Avogadro–​Ampère law, stating
but it is also true here, and probably more visible), there are few that one mole of gas makes 22.4 L of vapor (at a temperature of
cases concerning real culinary questions in which this definition 0 °C and a pressure of 101,325 Pa, or 24 L at 20 °C), but this is
can be applied in a non-​differential form. However, it is a fact equivalent (students given detailed instructions would be able to
of experience that many students are not entirely easy with par- demonstrate it). Needless to say, this law is frequently used in
tial derivatives (for thermodynamics, for example) or even with food science and technology, because cooking usually involves
differential calculus. Indeed, this law would probably be better heating.
expressed as:
8. The heat (energy E) needed for heating a body of
d mass m and of heat capacity c p by ΔT:
P=
dA
(F )
E = mc p ∆T
For the exercises in the list given later, this expression is not
needed. About this law, we have to discuss the fact that the students with
weaker calculation skills find it difficult to use letters that they
4 3 did not learn initially. This could be the case with acceleration,
3. The area and the volume of a sphere, 4 πr 2 and πr ,
3 for which a or γ are often used, or molar mass, but here, there is
respectively, even more confusion, because there is a well-​known difference
between an energy and a difference in energy, on the one hand,
where r is the radius of the sphere. and also the fact that internal energy (often written U) is equal to
the sum of the work of forces W and the heat Q.
4. The power P is the energy E per unit of time t:
E 9. The surface tension:
P=
t ∂F
γ=
∂a
In the exercises listed later, one (about the cooling of coffee) calls
for the differential form. where F is the Helmholtz energy (free energy) and a the area.
Needless to say, this definition appears difficult to some, as it
5. The expression of kinetic energy Ek: includes both a differential form and the Helmholtz energy, the
1 2 use of which is sometimes confused with the use of the Gibbs
Ek = mv function. It provides the opportunity to make the reverse calcula-
2
tion, from differential to differences:
where m is the mass of the moving body and v its velocity.
∆F
γ≈
6. Newton’s law of dynamics: ∆a
 
∑
external forces = ma 10. The expression of the electric force:

where m is the mass of the body on which the external forces act, 1 qq'
 F=
and a the acceleration of the body. 4π ε0 r 2
Simple Calculations Based on Cooking
where ε0 is the permittivity of the vacuum, q and q′ the interacting
electric quantities (“charges”), and r the distance between them.
11. Less often, the Stefan–​Boltzmann law:
P = S σT 4
where P is the radiated power, S is the area of the emitting body,
σ is the Stefan–​Boltmann constant, and T the absolute tempera-
ture. This holds for a black body, but coffee is black, for example.
12. For those who are ready to use them, the Fourier
laws for heat in stationary conditions:
∆T ∂ λ ∂2
ϕ = −λ and T ( x, t ) = T ( x, t )
∆x ∂t ρc ∂x 2
where ϕ is the density of thermal flux, λ the thermal conductivity,
T the absolute temperature, x the distance, ϱ the density, and c the
heat capacity.
Much could be said about these laws, in particular the fact that
the calculations behind them are the same as for diffusion. About
these questions, the book by Cussler (1984) could be useful
reading.
13. The entropy:
S = kB ln (Ω );
where Ω is the number of accessible microscopic configurations
that a system can take.
While dealing with entropy in classical thermodynamics calls
for skilled manipulations of the various variables and functions,
it appears very simply in this expression; using discretization
of space can make the calculation easier in simple cases, as so
well demonstrated by Atkins (1984). This is important, because
entropic effects occur frequently in food science and technology,
e.g., the fact that oil does not dissolve in water, the effect of salt
for making ice creams, etc.
14. The Maxwell–​Boltzmann law, giving, at the thermal
equilibrium, the probability of a state of energy e:
 e 
p(e) = K exp  − . stand what one is doing and get more control of the calculation,
 kBT 
where K is a constant that one can get by normalization on all
possible states, kB the Boltzmann constant, and T the absolute
temperature.
Of course, this list is only a very limited one, but this charac-
teristic is important for helping the students to gain confidence in
themselves; if students learn to use successfully the 14 laws given
here, they will be able to come back to their courses and get more
“intellectual equipment” later.
Now, the question is how to apply these laws, but here the
answer is simple: in particular circumstances, the students are
705
invited to consider the laws one after the other and ask them-
selves if the law belongs to the same field as the question; for
example, if the question is about heat, one can probably skip the
expression of the electrical force. When the law is in the right
field, then one can try to apply it, and if it works, so much the
better, and if not, the next law is to be considered.
More generally, students seldom meet this situation of having
to choose the right physical or chemical concepts from their sci-
entific background in order to investigate a question, because
during their curriculum they generally had to apply the laws of
one particular course within the context of this course.
I propose that they should realize that in practical circumstances,
their knowledge is sufficient, but they have to learn about defining
the problem and making assumptions, which is sometimes diffi-
cult: for instance, students can seldom calculate the thickness of
the crust of a soufflé, a calculation that does not require complex
math. This is why a method is proposed later.
How to Calculate
Here, we propose to analyze first why so many students have dif-
ficulties with calculations; knowing the errors is a way to avoid
them in the future.
First, it is a fact that few students know that calculation and
natural language are almost the same, and that formalism was
introduced for making calculation easier, to simplify mental
b
processes. Writing “ ∫a
cos( x )dx = sin(b)-​sin(a)” is easier and
faster than writing “the area between the axis of the abscissas and
the curve representing the variations of the cosinus of the real
variable x, between the abscissas a and b, is equal to the diffe-
rence between the sine of b and the sine of a”; indeed, the same
idea was introduced in chemistry in 1782 by the French chemist
Antoine Laurent de Lavoisier (Lavoisier, 1782).
Many mathematicians contributed to the replacement of words
by letters and symbols, but the names of François Viète, René
Descartes, and Wilhelm von Leibniz have to be remembered.
Unfortunately, not all students appreciate how useful the intro-
duction of the mathematical and chemical formalism was, and
they consider that their life is made more complex by incompre-
hensible symbols. We have a duty to tell them that formal calcu-
lation is easy, simple, and powerful; moreover, simply churning
the equations often leads to the result, even if it is better to under-
because it is a logical sequence leading to a well-​defined goal.
Moreover, it helps students to know that, because formal cal-
culation is based on natural language, calculating means speaking
and thinking as usual, using natural language. One can think of
a formula as one would think in natural language, and the same
holds for algorithms and computer programs. A first consequence
of this observation is that in order to calculate well, one has to
think properly and to speak properly. Of course, speaking prop-
erly, but with logic, can be difficult for some who “think in all
directions”, being disturbed by the many perturbations of life:
having enough money for food or lodging, being in love, organ-
izing activities, answering their emails, and so forth. However,
706
calculation cannot be done chaotically; it is an activity that goes
rationally, logically, toward its goal.
One way to help students is to remind them that if a calculation
is like a story, one should divide it into:
• a starting point
• a path
• a goal.
This simple analysis has consequences:
1. If the student does not know where he or she is going
(the goal), the probability of reaching the goal is
nil (starting from Paris, the probability of reaching
Frankfurt is nil if one does not know that one wants to
go to Frankfurt).
2. If it is not known which way will be used to go from
departure to arrival, there are again few chances that the
goal will be reached. In other words, a “way” is needed.
Let us add that in Greek, the way is called methodon,
the word for “method”. Indeed, often, students can find
the goal, but they have difficulties in finding the way.
They can be helped by being invited to look first for a
“strategy” (the general idea) and only later for a “tactic”,
i.e., the practical way to move. For example, when it
has been decided to travel from Paris to Frankfurt, one
has to decide to go either by plane, or by train, or by
car. When the train (for instance) is chosen, one can
decide to walk from home to the tube, then to reach
the railway station, then to go into the train, then to go
down the train, then to take a cab to reach the precise
final destination.
All this being said, we are fortunate that for many questions that
are met in food science or technology, the solution is obvious when
the starting point and the goal are well defined. Often, it is enough
to simply translate each sentence of natural language into a formal
expression or a computer code, depending on the circumstances.
This method has the advantage that the final calculation will be
documented; one will be able to check it, and any reader (including
the professor checking it) will be able to follow it.
Finally, we propose to invite the students to always structure
their argument with:
1. the question
2. the way to reach the goal, including the choice between a
simple use of orders of magnitude, for simple questions,
or formal writing for more complex problems
3. the particular steps
4. the recognition of having reached the goal.
However, the practice of teaching shows that failures can occur
even when the general method is good. The reasons for failure
frequently include the following:
Hervé This vo Kientza
• mixing formal calculation with numerical calculation;
• using units that don’t belong to the International System;
• lack of structure of the documents;
• or marking poorly the formal symbols of figures, with
confusion between 9 and g, or 1 and 7, for instance.
But let us be positive, and let us now examine the various parts
of any calculation:
1. The question considered: in most exercises, the question
is simple to formulate with natural language, and it
is easy to later translate natural language into formal
language.
  In some problems (when they are not a succession
of exercises), one has to learn to ask questions of one-
self and to create a way from the initial data to the final
question. Here again, students have to learn the easy
step of introducing a formal symbol for any unknown
quantity. More generally, they have to get into the
habit of characterizing formally any “object” by one
or many symbols, whether the “objects” are known or
unknown.
2. The way to solve the problem: Often, using orders of
magnitude can be useful and fast (this is why the French
physicist Pierre Gilles de Gennes, Nobel prize winner
in 1991, trained every day in this regard).
  When such fast calculation is not possible, one has
to move to formal calculation. This is where software
for formal calculation such as Maple is useful, and the
possibility of dividing the document into sections and
subsections is particularly helpful, because it structures
the work.
3. The step-​by-​step way: Here, it is often necessary to put
ideas down on paper so that one can analyze the words.
This should be done in natural language, and it is only
when the job is finished that the formal symbols are
introduced. The structure should be:
• analysis of the question
• expressing the result
• numerical application.
In the analysis step, looking for a relationship between the already
introduced symbols is often enough to find the solution.
However, let us come back now to the “analysis of the
question”. This should be divided into:
• writing down the parameters (from the question)
• making a qualitative model; this is frequently a
synonym for making a picture and analyzing the
question physically or chemically. This is the place
for simplifications and assumptions;
• turning the qualitative model into a quantitative
model through the introduction of formal symbols
for describing the characteristics of the model. If a
picture is drawn, this is very easy, because one has
Simple Calculations Based on Cooking 707

simply to put letters for each aspect of the various
parts that one can see on the picture, such as shapes,
colors, surfaces, etc.
• looking for a strategy to find the solution; this is the
most difficult, and we shall come back to this par-
ticular question later;
• applying the strategy; this is often simple.
4. Validation:
  This word “validation” is probably not used enough
in education, because many students do not know it
even at Master level.
Which Tool?
Today, scientists and technologists seldom write using a pen;
rather, they use text software. In the same way, computers are
used today for calculations, not only digital calculation but also
formal calculation. Of course, habits have to change; for text
software, one has to learn typing; in the same way, for calcu-
lation, the choice of convenient software is important. Let us
differentiate between formal calculation and digital calculation.
For digital calculation (regressions, integration of signals, etc.),
calculators or software such as Matlab or Excel can be used,
but for formal calculations, only software such as Maple will
be suitable. The best feature of software for formal calculation
is that it allows calculation as one learned mathematics in high
school, without any change, in contrast to using rows (for soft-
ware such as Igor) or spreadsheets (Excel, Matlab). Later in
this chapter, the formulae that are executed with Maple give a
result in blue.
A First Summary, a “Backbone” for Calculation
Let us summarize our conclusions:
1. It is good to use software for formal calculation, such as
Maple, rather than Word or Excel;
2. calculating means speaking (writing), making sentences
in natural language;
3. calculations have to be done in a structured way;
4. successive steps have to be:
• express the problem in natural language;
• put down numerical data after translation into
SI units;
• translate the problem into formal language through
the introduction of names of variables;
• solve the problem, which means that the problem
in formal language is developed until the solution
is reached;
• introduce the numerical data into the formal
solution;
• validate, discuss.
Accordingly, in order to help the students, they are invited to use
the document in Figure 106.1 (copying and pasting the model,
then filling the slots).
The Quality of Calculation: Validation
Finally, imagine that the solution is found. How do we know that a
calculation is right? This question is important, because the devil
is hidden behind any calculation, even simple ones. One forgotten

FIGURE 106.1
The framework for calculation.
Title of the calculation
Author
Date
The question: in natural language, with the right words, in a concise and simple way (subject, verb, complement)
Analysis of the question (often questions are solved immediately when they are analyzed)
The data are introduced: one has to write down the available data (for example digital values given) and immediately, translate the data into units
of the International System.
Qualitative model: make a scheme (not a beautiful picture) representing the phenomenon studied, then write down the assumptions that were used,
including simplifications.
Quantitative model: describe formally the characteristics of the objects that you see on your picture, i.e., introduce formal symbols (letters) that will
be used for calculation.
Solving
Looking for a solving strategy: how are you going to reach the solution (often this step is useless because the next one is enough)?
Implementing the strategy; if there is no strategy, one can look for relationships (equations) between the symbols, and often the solution is reached.
Discussion, validation (a second, or even a third calculation should be different from the first: one should not follow the same way as initially, because a
mistake will be difficult to find)
Expressing the results
The formal result found is written down again
Finding digital data (if possible with references; here, the initial data are copied)
Introduction of data in the formal solution
Discussion, testing various parameters and assumptions in order to explore the solution space
Conclusions and perspectives
Conclusions
Perspectives
708
“−” sign, a letter z confused with a 3, a bad operation (even when
it is a simple one), interchanging two digits in a number, a bad
proportionality; all this can occur and, indeed, does.
It is funny to observe that remedies to such problems are
seldom proposed in spite of their high frequency. Of course,
in high school, one learns to check calculations by considering
orders of magnitude, but this is a very crude validation. Many
teachers tell students to perform their calculation twice, but
students who make mistakes can make them twice.
It is probably better to discuss this question openly and to pro-
pose other solutions. For example, considering the units is so
important that even “dimensional analysis” (Gibbings, 2011) can
be taught: when a formal result is obtained, the students would
be well advised to check that the units are the same on both sides
of the equation. Also, when the solution of an equation is found,
re-​injecting it into the equation should be common practice (this
is proposed by software such as Maple). For a formal equation,
particular cases can be looked for.
In summary:
• determine orders of magnitude
• repeat calculation completely
• apply the result of simple sets of data for which the
results are known
• use random data and explore the solution space
• test extreme cases
• validate by another calculation.
Here, the questions are given in order of increasing complexity
in terms of disperse system formalism (DSF; see the chapter on
this in this book).
A First Example: How Many Bubbles Form in a
Whipped Egg White?
The Question
Before discussing this particular question, let us consider the
way students answer when taking exams. I propose to distinguish
two cases:
• when the student knows the answer to a question that
he/​she is asked
• when he/​she does not know.
1. If he/​she knows the answer, it is easier, of course,
but then the issue is to deliver the answer as prop-
erly as possible, to leave no doubt about the fact
that the answer is known. Orally, one will give
it as directly as possible, trying to add interpret-
ation. In the written form, one should write very
precisely, with no spelling mistakes and careful
grammar.
2. If the candidate does not know the answer,
complete failure can be avoided, because there
remains the possibility of showing that a lot has
been learned anyway. For example, if a formal
relationship is not found, the possibility of calcu-
lating orders of magnitude remains. For example,
Hervé This vo Kientza
when asked “How many hairs on one human
head?”, there are three very bad answers:
• “I don’t know”: this is honest, but the
teacher cannot give any mark except the
lowest, because the student did not provide
anything;
• a random answer: this is silly, and the
teacher would give a negative mark if
possible;
• a wrong answer with much assurance: here
again, the examiner will be very strict,
because he/​she will have the feeling that
the student is fooling him/​her.
A much better way is to first repeat the question and
then rephrase it, as proposed in the “backbone cal-
culation” that was shown earlier. Often, discussing
the terms of the question leads to the solution.
Making a sketch –​again, as was proposed –​is another good step.
For the example about hairs, one could make a sketch with ran-
domly distributed hair. Then, the picture is simplified with a square
grid. And then, an assumption is made: for example, one hair per
millimeter, so that the order of magnitude of the number of hairs
could be found easily. Calculating orders of magnitude is what the
physicist Enrico Fermi asked students to do when they applied to
work with him, but the question was about how many piano repair
people there were in Chicago rather than how many hairs on a head.
In conclusion, a good method is helpful, and it often leads to a
result. This is what we propose to examine here.
Analysis of the Question
The Data Are Introduced
Here, no numerical data are given, but students can also realize
that no exact number is asked for. Accordingly, they should rather
look for an order of magnitude.
About the egg white, even if they do not know its mass, they can
easily find that the order of magnitude of the mass of an egg is not
10 g nor 1000 g; a guess of 100 g is not far from the right value
(about 60 g). The egg white can be assumed to be about half this
value. Now, the egg white is made of proteins and water. Can we
guess, as before, the concentration? Of course, complete ignorance
cannot lead to a good guess about anything, but students may know
that the maximum quantity of salt dissolved in water is about 350 g
per liter, and about 1 kg for sugar. The first is much less viscous than
an egg white, and the second is more viscous. Of course, viscosity
is not directly proportional to concentration only, and solutions of
proteins are different from solutions of simple molecules or salts,
but these two values can lead to a guess of the right order of mag-
nitude. Indeed, it can be useful for students in food science and
technology to know that egg whites contain 10% protein.
Qualitative Model
A whipped egg white is obtained by whipping an egg white, i.e.,
30 g of a 10% aqueous solution of proteins. Let us first make a
picture like Figure 106.2.
Here, we see that we can add air bubbles into a liquid until
they are closely packed, and this should lead us to consider that
Simple Calculations Based on Cooking 709

FIGURE 106.2 A model for a whipped egg white (a). Of course, it would be useful for the students to have observed a real microscopic picture of a whipped
egg white at the beginning of whipping (wet foam (b)) and at the end of whipping (dry foam (c)).

bubbles could fill up about 70% of the liquid if they were non-​ assume that the density of the egg white will not be very different
deformable and all of the same size. from the density of water, we can determine a volume equal to
On the other hand, if the bubbles had different sizes, the air 3 × 10–​5 m³.
fraction could reach a bigger limit. Now, for the radius of air bubbles, we can find this by making
the assumption that we can barely see the air bubble with unaided
Quantitative Model eyes, so that r is less than 0.2 mm. And there is no need to fix
Here, the students have simply to introduce symbols in order to a precise value; we can give an interval instead, for example
characterize the picture. For this, there is a liquid component (in between 1 mm and 10–​2 mm; let us use the values 10–​3 and 10–​6 m.
blue); let v be the volume of one egg white. And there are N air
bubbles, with a radius r, assuming that they are all alike. Let V be Introduction of Data into the Formal Solution
the volume of the whipped egg white. Using software is very convenient, because one copies and
pastes the final result, changing only some values in the expres-
Solving sion to be calculated. For example, with a radius of 10–​4 m:
  3 −V + v  
With N bubbles of air of radius r, the total volume of bubbles is: evalf  subs  V = 0.3e − 3, v = 30e − 6, r = 1e − 3. −
  4 πr 3  
4 
Vb = N ⋅  πr 3  6.455 × 10 4
3 
And with a radius of 10–​5 m:

The total volume of whipped egg white is the sum of the air   3 −V + v  
evalf  subs  V = 0.3e − 3, v = 30e − 6, r = 1e − 5, −
volume and the egg white volume v:   4 πr 3  
4  6.455 × 1010
V = Vb + ν = N ⋅  πr 3  + ν
3 
Discussion
Here, we look for N:
Of course, only the order of magnitude can be kept; the number is
3 V −v between about 100,000 and 10 billion.
N= ⋅
4 πr 3 A validation could be done in many ways.

1. For example, let us skip the assumption of a whipped

Expressing the Results egg white having a volume of 1/​3 of a liter, and let us
The formal result found is written down again: decide instead that air bubbles, all of the same size,
are packed into the liquid egg white.
3 V −v Now, the total volume of air in the liquid is given
N= ⋅
4 πr 3 using the close packing coefficient (p = 0.74%). This
means that if the volume of the egg white is v, the
Finding Digital Data
volume of air is:
Here, we need the values of V, v, and r.
For V, we know that the volume of a whipped egg white is less v1 ⋅ p
va =
than half a liter. Let us decide that it is one-​third of a liter (0.3 L, 1− p
i.e., 0.3 × 10–​3 m³).
To find v, we have to start from the initial data, i.e., the mass And if the volume of an air bubble is a, then the
of an egg white; we assumed that it is 30 g (30 × 10–​3 kg). If we number of bubbles is:
710
v1 ⋅ p
(1 − p)
a over which time no big difference was found between them. I had
This makes:
  vl ⋅ p  
  4
evalf  subs  p = 0.74, vl = 30e − 6, a = ⋅ π ⋅ (1e − 4 ) ,
3 (1 − p)  

  3 a 
   

2.04 × 107
This is the same order of magnitude as our previous
calculation.
2. Let us now consider that whipping is performed
using a whisk with 10 iron threads of 1 mm diam-
eter, and 10 cm dipping into the egg white at each
turn of the whisk. This would make 100 bubbles of
1 mm diameter per thread, i.e., 100 × 10 bubbles
for one turn. If we whipped for 3 minutes, with 1
rotation per second, this would make about 200
rotations, i.e., 2 × 105 bubbles.
With longer whipping (5 minutes) and bubbles divided by 2 at
each rotation, the number could increase much more.
Conclusions and Perspectives
This exercise was primarily a way to consider orders of mag-
nitude, as these were considered of primary importance not
only by Fermi but also by the French physicist Pierre Gilles de
Gennes, who wrote (De Gennes, 1998) that training in orders
of magnitude calculation was one way to become “smart” in
physics.
A Second Example: Why Do Oil Droplets Cream in
Mayonnaise Sauce?
Here, we propose to observe first that frequently, questions make
students “fall into them” without discussing the reality of the phe-
nomena considered. If it is true that milk, being a diluted emulsion,
creams, does creaming really occur in mayonnaise sauce?
In 1982, I conducted an experiment in order to check the
culinary precision that adding hot vinegar to mayonnaise sauce
would prevent phase separation. A mayonnaise sauce was
produced from one egg and one teaspoon of vinegar, with oil
emulsified with a fork until a “thick” consistency was obtained.
Then, the sauce was divided into five bowls:
(1) unmodified;
(2) with the addition of a spoonful of cold water;
(3) with the addition of a spoonful of hot water;
(4) with the addition of a spoonful of cold wine vinegar;
(5) with the addition of a spoonful of hot vinegar.
Hervé This vo Kientza
The bowls were kept in a fridge, and the sauces were observed with
the naked eye and using an optical microscope for up to 39 days,
to conclude that the culinary precision was wrong, and also that
the creaming, which is discussed here, does not occur as readily
as the question asked here would push the students to think. For
sure, as the French author Bernard Le Bovier de Fontenelle said:
“let’s make sure of the fact before we worry about the cause. It is
true that this method is very slow for most people who naturally
run to the cause, and pass over the truth of the fact, but at last we
will avoid the ridicule of having found the cause of something
that does not exist” (Fontenelle, 1687).
The Question: In Natural Language, with the Right
Words, in a Concise and Simple Way (Subject, Verb,
Complement)
In the experiment reported here, the mayonnaise sauce was thick,
as it can be when a large quantity of oil (up to 95%) is added, but
creaming can indeed occur in mayonnaise sauces with less oil.
And this is why the question deserves to be tackled anyway.
Let us begin by observing that many culinary systems are of
colloidal nature; i.e., they are made of particles of one phase (gas,
liquid, or solid) dispersed in another phase (gas, solid, or liquid)
(This, 2016; IUPAC, 2018). For example, “whipped egg whites”
are foams, made of air bubbles dispersed in the viscous envir-
onment of the 10% protein solution of egg white (This, 2009).
Many sauces (see the chapter in this book), like béarnaise and
hollandaise, are also dispersions of both “oil” droplets and pro-
tein aggregates in a liquid phase, the latter coming from egg
yolk (50%) or from vinegar (more than 90%) (see the chapter on
emulsions).
As the oil, air, or solid particles are generally different in
density from the liquid continuous phase, such systems are not
stable, and sedimentation or creaming can occur. In other words,
in one case, particles (solid particles) fall, but in the other case
(e.g., air bubbles or oil droplets), they move up to the upper part
of the sauce.
How fast? Always? The answer to such questions is important
for science and technology education, not only because they
allow us to describe common systems but also because they intro-
duce scientific developments that can be useful parts of scientific
culture.
Let us start with raw milk: how long should we wait in order
to recover cream? Or consider a mayonnaise sauce: how long can
we keep it until the emulsion is disrupted?
Often, a microscopic picture of the system is helpful. For
example, Figure 106.3 shows a mayonnaise when about 40% of
the volume is oil.
With more oil, the picture is different (Figure 106.4).
We want to calculate how objects can sediment or cream (the
same question) in a liquid.
Analysis of the Question
The Data Are Introduced
No data are given for this exercise.
Simple Calculations Based on Cooking 711

FIGURE 106.3 Oil was whipped using a whisk in a 50:50 mixture of egg
yolk and vinegar. The largest oil droplets have a diameter of 0.1 mm.

FIGURE 106.5 One particle in a liquid.

FIGURE 106.4 Mayonnaise whipped with a whisk when the oil content
is 80%.

Qualitative Model
In considering sedimentation and creaming, we build a model
FIGURE 106.6 Modeling the particle in the liquid. Assumptions are made.
with only one particle in a liquid (e.g., an oil droplet), assuming
first that there are no interactions between neighboring
particles. It can be observed that this assumption holds for In this picture, the particle was put in the center of the liquid,
diluted sauces, such as mayonnaise at the beginning of pro- on a vertical axis of symmetry, purely for aesthetic reasons.
duction, but not for packed emulsions like ready-​to-​use may-
onnaise. Then we can assume that the particle is much bigger
than the molecules of the liquid, so that the liquid seems to be
“continuous” (an assumption which we have to keep in mind for Quantitative Model
later) (Figure 106.5). Then, the problem has to be translated into calculation, introdu-
We now envision the successive steps of modelling. Of course, cing formal variables; to this end, it is sufficient to describe what
an oil droplet would not have this shape (on the contrary, it could is seen, i.e., a particle having a shape and a defined character
describe a protein aggregate), but it is a good practice to first be (shown in the picture by the color) and a liquid having a different
very general and then make explicit assumptions. character.
A good practice is to “simplify” the problem first, dropping For the shape of the particle, one variable is enough:
all secondary aspects (which is why it is important to be able to
distinguish different orders of magnitude). For example, here, the • radius r of the particle
number of dimensions was reduced from three to two. One can go
further, assuming a spherical particle (Figure 106.6). For its material:
712 Hervé This vo Kientza

• density of the particle ρp (it could be the mass M, For example, for the velocity:
instead)
d
v(t ) = z(t );
For the liquid: dt

• density of the liquid ρl So that the fundamental law of motion, assuming the particle is
• viscosity η. sedimenting, is:

For the liquid, the volume is not introduced because sedimenta- d

P − F − S = m⋅ v( t )
tion and creaming do not depend on the quantity of liquid. But dt
viscosity is introduced because if there is motion in a liquid, there
is some friction, and this is characterized by viscosity. It must Let’s also write down the relationship between the mass and the
also be noted that the flow of the liquid is not considered, which radius of the particle:
is an invalid assumption in some cases, such as when a solid is
sedimenting in a liquid inside a vessel whose radius becomes M = ρp V
comparable to the flow disturbance.
Finally, as there is motion, and because we assume this motion With:
to be in one direction, we have to introduce a vertical axis, toward
the bottom, for example (Oz); the point O can be the upper level 4
V= ⋅ π ⋅ r3
of the liquid. 3

Finally, the equation of the motion becomes:

Solving
4 4 d  4  d2 
Looking for a Solving Strategy ρ p πr 3 g − πr 3ρl g − 6 πη  z(t ) r = ρ p πr 3  2 z(t )
3 3  dt  3  dt 
Now that variables have been introduced, relationships between
them can be considered. Because we want to describe a sedimen-
tation or a creaming, i.e., a motion, we have to consider the laws This is a linear differential equation of the second order (the
of dynamics. highest-​order derivative is the second one), about the function z(t).
The first law, the “fundamental law of motion”, or “Newton’s Using software for formal calculation such as Maple, it is very
law”, indicates that the sum of external forces acting on a body is simple to get the solution of the equation:
equal to the product of the mass of this body by its acceleration.
This leads us very logically to observe that we did not introduce −
9 ηt
the mass m of the particle; only the radius and the density were
given. Also, the velocity v is missing, which is an opportunity to z (t ) = −
2
2 r ρ pe
2 r 2ρ p
_ C1
−
2
(
2 r g −ρ p + ρl t
+ _ C2
)
9 η 9 η
observe that one should be careful about pictures and kinetics.

The expression for z(t) includes two constants, “_​C1” and “_​C2”,
Implementing the Strategy that we now have to determine. We do this assuming first that the
Which forces act on the particle? initial velocity is nil, again using the software to calculate the
derivative of z(t):
1. First, the weight:
P = Mg  9 ηt
- 2 
( )
t
 2
d  2 r ρ pe
2 r ρp 2
_ C1 2 r g −ρ p + ρl 
2. Then, the buoyancy, equal to the weight of the − − + _ C 2
volume of liquid displaced by the body: dt  9 η 9 η 
 
F = V ρl g  

3. Finally, if the particle moves, there are frictions,

The result given is:
which can be described using the “Stokes drag”
within the frame of the hypotheses proposed earlier:
( )
9 ηt 2
−
S = 6 π η υ(t)∙r 2 r 2ρ p 2 r g −ρ p + ρl
e _ C1 −
9 η
Here some new variables appear, such as the velocity, v. One can
try to link them to variables already introduced.
Simple Calculations Based on Cooking 713

Now, we use the solve command to find the constant, using the ics:= z(0) = 0, D(z)(0) = 0
fact that the velocity at time t = 0 is equal to 0: We now apply the dsolve command:
dsolve({ics, eq1}).
 - 9 η⋅0

solve e
2 r 2ρ p
_ C1 −
2
2 r g −ρ p + ρl

= 0,_ C1
( ) And we get:
 9 η  9 ηt
  -

z (t ) = −
4
4 r ρ pe
2 r 2ρ p
(
g ρ p − ρl ) − 2 r g ( −ρ
2
p )
+ ρl t
And the result given is: 81 η 2 9 η
4
(
4 r g −ρ p + ρl ρ p )
(
2 gr 2 ρl − ρ p ) +
81 η2
9η

Here is now the first constant:

Expressing the Results

_C1 =
2
2 gr ρl − ρ p ( ) The Formal Result Found Is Written Down Again
9η Again, we give the solution that was found:

9 ηt
Now we use a second initial condition, expressing that the par- -
ticle is left at the origin O at time t = 0.
z (t ) = −
4
4 r ρ pe
2 r 2ρ p
(
g ρ p − ρl ) − 2 r g ( −ρ
2
p )
+ ρl t
81 η 2 9 η
 9 η⋅0
 2 gr 2 (ρl − ρn ) 
( )
− 2 4
4 r g −ρ p + ρl ρ p
 2 2 r ρp
r ρ pe  
 2 9η +
  81 η2
solve  −
 9 η

 Finding Digital Values

−
2
(
2 r g −ρ p + ρl ⋅ 0
+ _ C 2,_ C 2
) Observing the result, we see that we need:

9 η  • the radius
• the density of the liquid
• the density of the particle
And we find the value: • the gravitational constant, g
• the viscosity.
4r 4ρ p g ρl − ρ p ( )
81η 2 All these can be found online or assumed (with references).
For example, the radius can be obtained in This (2009), and the
density is to be assumed or determined using Hui and Sherkat
So that the final expression is
(2005), as well as the other data.
9 η⋅t
-
2 r 2ρ p  2 gr 2 (ρl − ρn ) 
2 Introduction of Data in the Formal Solution
r ρ pe  
2  9η 
z (t ) = − This can be done with the command subs(), and various results
9 η can be compared, for example when the radius is changed,

−
2
( +
4
)
2 r g −ρ p + ρl t 4r ρ p g ρl + ρ p ( ) which allows better illustration of sedimentation or creaming
behavior.
9 η 81η2
Discussion
Validation Numerical solutions are less interesting than the analysis of the
solution from a broader formal perspective. In the last equation,
Another possibility would be to introduce the initial conditions
one can see three terms:
(ics) directly in the dsolve command:
• a constant (the last term), which is quite uninteresting,
4 4 d  4  d2  because it tells only at which level the particle was
eq1 := ρ p πr 3 g − πr 3ρl g − 6πη  z(t ) r = ρ p πr 3  2 z (t) :
3 3  dt  3  dt  dropped at time t = 0 (with no velocity);
• a term including a negative exponential
714 Hervé This vo Kientza

−
9 ηt We now ask when this velocity is obtained:
2 r 2ρ p
e

p⋅ −
2
(
2 r g −ρ p + ρl ) 
which decreases very rapidly with time, so that it is  9 η 
important only at the onset of motion;
 9 ηt
- 2
• a term proportional to t
=

−
4
d  4 r ρ pe
2 r ρp
g −ρ p + ρl ( )
−
2
(
2 r g −ρ p + ρl t ) dt  81

η2
9 η 

which becomes very important at long times; after the −

2
( +
4
)
2 r g −ρ p + ρl t 4 r g −ρ p + ρl ρ p (


)
9 η 81 η2 
start, it becomes the main one.

If the motion is almost regular, it means that the velocity is The solution is:
almost at constant speed, and the acceleration is progressively
2 ln ( − p + 1) r ρ p
decreasing toward 0. The regime over long times is called “sta- 2
tionary” and can be reached when the acceleration is nil: t=−
9 η

4 4 d 
ρ p πr 3 g − πr 3ρl g − 6 πη  z(t ) r = 0 Then, we explore this solution for spherical particles with a
3 3  dt 
radius of 1 mm and density four times the density of water, with
viscosity η = 10−3 kg/​m/​s. We find 4.09. Which units? The answer
Here, we get only the last two terms, and the limit velocity is: is seconds, because we always use SI units.
With a lower density of particles (1200), one would find 1.23 s.

vlim
d
= −
 2
(
2 r g ρ p + ρl t ) 
+ _ C1 = −
2
(
2 r g −ρ p + ρl ) What about a smaller radius for the particles? For example, for
this second density but with a radius 10 times smaller, the time
dt  9 η  9 η would be 0.0123 s, which is faster, but there is something more;
having done three calculations, we also see that this exploration
Let us observe first that this expression is the same as previously; of the result is not systematic. Students can now understand that
this is part of validation. But one can also observe that: they might need to plot a “response surface”, for example.

• the greater the difference in densities, the faster the Why Does Stokes Drag Relate to r and Not r²?
sedimentation; As we have said, the use of software for formal calculation is
• with centrifugation, g is replaced by a bigger value in
transforming science education, because calculations are much
order to increase the phase separation;
easier. One can perform “experiments”, such as here, about
• the greater the viscosity, the slower the sedimentation;
Stokes drag.
• the smallest radiuses are associated with slow sedi-
mentation, but the variation is very significant, being This law is in r, which may seem strange, because we know well
related to r2! that if we extend a finger outside a moving car, we perceive less
force than if we extend the whole hand. In other words, we have
One can also observe that if one knows three of the four parameters the intuition that the Stokes force “should” be in r² rather than in r.
(r, ρp, ρl, η) and if one measures the limit velocity, then the fourth We shall not cover the whole discussion here, but only show
can be determined! how one can experiment.
Finally, it is worth considering the question: how important Let us consider the differential equation obtained before, but
is the limit velocity, and when does the particle reach this limit changing the friction expression:
velocity? Strictly speaking, this never happens, but one can ask
when it reaches 99% or 99.9% of it, for example.
For more generalized conditions, let’s use a velocity that is a
eq 2 :=
g⋅4
3
( ) d
⋅ π ⋅ r 3 ⋅ ρ p − ρl − 6 ⋅ π ⋅ η ⋅ z(t ) ⋅ r 2
dt
fraction of the limit velocity: ρp ⋅ 4 ⋅ π ⋅ r3 d  d 
= z(t )
V = p ⋅ vlim 3 dt  dt 

Again, we solve this equation using the Maple software:

Simple Calculations Based on Cooking 715

dsolve(eq2). box, which would dissolve the sugar contained in the solid net-
work of the meringues.
9 ηt Why it is hot above stoves? For this one, there should be a dis-
−

z (t ) = −
2ρ p re
2ρ pr
_ C1 2 gr ρl − ρ p t
+
(
+ _ C2
) cussion about “hands-​on” physics, because often, students make
9η 9η assumptions that are only guesses, and have no way to decide
between competing hypotheses. Calculating is a good way to
decide.
Here, it appears that the velocity of the particle is constant after
How can we demonstrate that even a long fork is not enough to
a first acceleration step, but this velocity does not depend on the
eat in the Devil’s company? This idea is based on a saying from
radius, and this is strange.
Alsace, but students find it funny to use it for calculations. Many
Now, using a force in r3?
different solutions can be invented: heat transfer in the metal,
radiation of heat by the black body of the Devil, etc.
eq3 :=
g⋅4
3
( ) d
⋅ π ⋅ r 3 ⋅ ρ p − ρl − 6 ⋅ π ⋅ η ⋅ z(t ) ⋅ r 3
dt
How much acidity is produced when the air in a bottle of wine
is turning ethanol into acetic acid? This question, again, is a trap,
ρ⋅ 4 ⋅ π ⋅ r3 d  d  because if the students simply calculate the pH of the acetic acid
= z(t )
3 dt  dt  created from the oxidation of ethanol, they will be wrong, as wine
is buffered.
Why it is very dangerous to pour water on flaming oil?
dsolve(eq3).
This exercise could indeed be given as an introduction to
laboratory work.
9 ηt

( )
−
2 ρ
9 ρp _ C1 2 g ρl − ρ p t
z (t ) = − + + _ C2
2 η 9 η
Now, the limit velocity would decrease when the radius increases,
which is even stranger! Finally, one can see that the use of soft-
ware for formal calculations is changing entirely the way science
and technology can be taught. In my experience, young students
can learn to use Maple in some hours and become more and
more skilled and fast, even playing with it. In the particular
case discussed here, such uses, which allow fast comparisons,
lead the students to new questions, such as how it is possible
that the Stokes drag is in r, and how Stokes obtained his result
(Stokes, 1856).
Conclusions and Perspectives
This exercise was discussed in more detail because we wanted
to show that simple knowledge is important for the initial dis-
cussion of common phenomena in the kitchen, but we also
wanted to show the limits of such simple treatments. It was
also the opportunity to see how the introduction of software
for formal calculation is changing the way students can learn
physical chemistry, in general, and molecular gastronomy, in
particular.
A List of Questions That Students Can Use For
Developing Their Problem-​Solving Skills
Here, we give a short list of exercises that can be readily studied
with the 14 initial definitions or laws.
About Gases
The softening of meringues, or why is it good to keep meringues
in closed boxes? One calculation could deal with humidity in the
About Liquids
How many molecules of water in a glass, and what is the distance
between them? This exercise is very simple when the method
is right, and it is a good demonstration of how the method can
be used.
Why has water no color, while whipped egg whites are
white? This one addresses a misconception about the “color” of
whipped egg white, and of many color changes during culinary
processes.
Is it possible to make dried tomatoes in an oven at 105 °C if the
door of the oven is closed? Here again, this is an opportunity to
change knowledge into skills, because the real difficulty is nil, but
students often don’t know how to treat the question.
Can the temperature reach 130 °C in a pan where water
is boiling? This exercise using the ideal gas law is helpful,
because it shows how sound reasoning can refute crazy culinary
precisions.
How much sulfur dioxide (SO2) added to wine increases its
acidity when oxidized? This is a culinary question for checking
knowledge about pH.
There is an experiment that involves pouring water over the
level of the upper edge of a glass. How much can we add? This
aims to train students about the use of surface tension.
Why does syrup accumulate at the bottom of a glass of water?
This one is helpful for demonstrating the interest of the Maxwell–​
Boltzmann distribution, but it also reveals pitfalls in some experi-
mental determinations.
When we look at a mayonnaise sauce under a microscope, can
we see Brownian motion? Microscopy is interesting in the kit-
chen, because most food items are colloidal.
How much does coffee cool by radiation? This exercise can
be solved in various ways, depending on the particular laws that
one applies.
When an egg (whole egg, in the intact shell) is dipped in vin-
egar, why does it become bigger after some days? This is a good
716
demonstration of osmosis, after some simple chemistry is used,
and leads to discussion of the shell’s dissolution.
About Foams
When sugar is added to a whipped egg white in order to make
a meringue, the size of the air bubbles decreases. How does the
density of whipped egg whites change with the size of the air
bubbles? This is a simple exercise, for which a good method is
needed.
How thick is the liquid layer between two oil droplets in a
mayonnaise or between two air bubbles in a whipped egg white?
This is the same kind of question as before.
What is the maximum volume of whipped egg white that one
can make from one egg? This one is important, because it helps
with thinking out of the box. Indeed, a simple, very large bubble
with a thin liquid limit is already a whipped egg white, and the
answer to the question in this particular case is interesting.
About Emulsions
Why does one have to whip oil to make a mayonnaise sauce? This
one can be solved in different ways, but of course, the surface
tension can be used.
How many surfactant molecules are there at the oil–​water
interface in a mayonnaise, and what is the mass of surfactant
needed to make the sauce? The maximum volume of mayonnaise
made from one egg has been shown to be 60 L, and this is the
way to predict it.
Why are Pastis, Ouzo, and Arak turbid only at the surface when
water is slowly added, but become turbid in the whole glass after
some time? This is a very good introduction to Brownian motion
and the use of the Maxwell–​Boltzmann distribution.
Is butter an emulsion? The answer is no, and simple mathem-
atics can be used to show it.
About Gels
How much gel can be made with a known mass of gelatin? The
question of the proportions of gelling agents frequently arises in
kitchens, and here is a way to get more insight into it.
About Suspensions
How “thick” can a white sauce be? Here, the Einstein law for
viscosity can be used.
More Complex Systems
How much oil in a French fry? Sometimes, we don’t perceive
well that we eat too much oil and sugar, and this simple calcula-
tion reveals how to avoid it.
How thick is the crust of a soufflé? Of course, the question
would be the same for bread, for example, and it is interesting
that this simple exercise, which can be solved in many different
Hervé This vo Kientza
ways, seems so difficult for students. The importance of the cal-
culation method appears again.
Why does fat leak out of foie gras when cooked? This is an
exercise to show that calculation is helpful when competing
hypotheses are discussed.
When cooking spaghetti in salted water, how much salt could
go into the cooked spaghetti at a maximum? Again, a question of
methodology.
Conclusions
The number of possible questions that arise from culinary
processes is probably infinite, and the key point is to select the
most useful ones, which can be solved using the simplest concepts
and laws. When these simple laws are not enough, this shows the
students why more studies are needed.
However, these exercises are intended primarily to demonstrate
that what was studied during the various courses is not useless –​
on the contrary –​and that it is a pity when something that was
learned is forgotten. Also, the difference between a knowledge
and the corresponding skill can be illustrated more clearly.
REFERENCES
Atkins PW. 1984. The Second Law, Scientific American Library,
New York, NY.
Cussler EL. 1984. Diffusion, Cambridge University Press,
Cambridge, UK.
Feser J, Vasaly H, Herrera J. 2013. On the edge of mathematics and
biology integration: Improving quantitative skills in under-
graduate biology education, CBE Life Sciences Education,
12(2), 124–​128.
Fontenelle BLB. 1687. Histoire des oracles, G. de Luyne, Paris,
France.
De Gennes PG. 1998. L’intelligence en physique, Pour la Science,
254, 11.
Gibbings JC. 2011. Dimensional Analysis, Springer Verlag,
London, UK.
Hui YH, Sherkat F (eds.). 2005. Handbook of Food Science,
Technology, and Engineering, CRC Press, Boca Raton,
Florida, 9–​12.
IUPAC. 2018. https://​goldbook.iupac.org/​html/​C/​C01172.html
Lavoisier AL. 1782. Considérations générales sur la dissolution
des métaux dans les acides. In Lavoisier AL, Mémoires de
l’Académie des sciences, Paris, France.
Owenson S. 2012. Maths and biology skills for life, https://​bbsrc.
ukri.org/​documents/​1204-​maths-​in-​school-​level-​biology-​pdf/​
Polya G. 1945. How to Solve It, Princeton University Press,
Princeton, NJ.
Stokes GG. 1856. On the effect of the internal friction of fluids on
the motion of pendulums, Transactions of the Cambridge
Philosophical Society, 9.
This H. 2009. Molecular Gastronomy, a chemical look to cooking.
Accounts of Chemical Research, 42(5), 575–​583.
This H. 2016. Statgels and dynagels. Notes Académiques de
l’Académie d’agriculture de France /​Academic Notes from the
French Academy of Agriculture, 12, 1–​12.
Teaching and Cooking with Culinary Teachers
Christophe Lavelle1,2
1 Genome Structure and Instability, CNRS UMR7196 /​INSERM U1154, National Museum of Natural History,
Sorbonne University, 75005, Paris, France
2 Institut National Supérieur du Professorat et de l’Education (INSPE), Toulouse University and Cergy-​
France
The world of cooking is full of pertinent statements on how to
do things and occasionally, why one should follow the advice
given. Such practices are questionable and provide rich research
material for various disciplines, including geography (distribu-
tion of food practices), history (evolution of food practices), and,
of course, physics and chemistry (what is the rationale behind
these practices?). Besides pure research, several outcomes are
expected to emerge, for instance regarding education and profes-
sional practices. Indeed, a better understanding of culinary phe-
nomena can not only be the goal of diverse pedagogical projects
in all schools (see other chapters in this book dedicated to this
issue), but it can more specifically also influence the way we cook
(and the way we teach cooking skills) in professional schools.
What should we teach to our students who are learning how to
cook? Let’s have a quick look at how teaching evolved in French
cooking schools in the last 60 years. In the 1950s, students mainly
learned recipes and were only supposed to reproduce these “trad-
itional” recipes. In the 1970s came “nouvelle cuisine”: new
dishes had a strong appeal for chefs and customers. Then in the
1980s, teaching content slowly shifted from recipes to techniques
(to invent recipes). In the 1990s came “molecular cuisine”: chefs
again had a strong interest in being innovative, and the innov-
ation was based on scientific investigations. In the 2000s, science
started to be injected into professional cookery books. Nowadays,
the tendency is growing, and culinary programs contain more and
more science for fewer and fewer recipes.
I provide here a concrete example from the French culinary
bachelor degree. In 1995, the official program, entitled “cooking
technology and methods”, asked the teachers to show the students
how to:
• peel, segment, slice, crush, chop, slash, prepare shell-
fish, dress and truss poultry, prepare fish, blanch.
• cook in a liquid: eggs, green vegetables, dried
vegetables, pasta and rice, fish, meat.
• steam: fish, vegetables.
• grill: white meat, red meat, fish, poultry, vegetables.
• roast: white meat, red meat, poultry.
• sear.
• sauté: vegetables, eggs, breaded items, fish.
Pontoise University,
• bind/​
thicken a liquid base with: a roux, a kneaded
butter, a fat, egg yolks, mashed vegetables.
• concoct basic sauces and stocks: tomato sauce, veal
stock, fish velouté, etc.
• produce cold emulsions.
In 2015, the official program was renamed “culinary science and
technology” and asked the teachers to:
• identify key physical/​chemical phenomena generated
by the culinary act (transformations of fat, protein,
carbohydrates, water; impact of temperature; properties
of foams and emulsions).
• develop a strategy adapted to the class level, avoiding
the strict reproduction of techniques.
• demonstrate the physical/​chemical phenomena as part
of the observed processes.
• further reveal these discoveries during experiments,
observations, demonstrations … and to make the link
with the science class.
Obviously, this is a big change! Teachers have to adapt; scholarly
books have to be re-​written. But this is probably the best way to
obtain not just “robots”, able only to reproduce the same dish
over and over, but instead true culinary artists who will maintain
cuisine as a living art. “The discovery of a new dish does more
for the happiness of mankind than the discovery of a star”, said
Brillat-​Savarin in his Physiologie du Goût (1825). Let’s all work
for this!
Note: I have been responsible for the scientific training of
culinary teachers at the (French) national level for 10 years now.
In order to fulfil their diploma and officially receive their nomin-
ation as a teacher in a culinary school anywhere in France, students
have to pass a master’s degree. All the dishes presented here are
part of tasting menus prepared by these “training-​teachers” for
real customers and served in the pedagogical restaurants of the
INSPE (French National Institute for Teaching and Education)
in Antony (under the supervision of Jacques Lamarque and
Bruno Cardinale). As a “game”, I usually ask them to use modern
ingredients/​tools/​techniques and always propose fresh ideas, not
717
718 Christophe Lavelle

FIGURE 107.1 Purple potato (vitelotte) mousse and chips, by Afonso FIGURE 107.4 Lobster, avocado purée, grapefruit gel, braised lettuce and
Herminio (2020). lettuce sorbet, by Marine Ricci (2018).

FIGURE 107.2 Oysters in a verbena jelly, carrot caviar, fresh dill, by FIGURE 107.5 Monkfish in bacon, radish jelly, radish chutney, beer foam
Thierry Rasoanaivo (2019). and broccoli/​almond soufflé, by Thierry Rasoanaivo (2018).

FIGURE 107.3 French meringue, beetroot sorbet, beetroot marshmallow, FIGURE 107.6 Trout tartar, tzatziki ravioli, olive oil hot mousse, algae ice
beetroot powder, Chioggia and yellow beetroots condiment, by Jamil cream, squid ink lace, by Sandra Chevallier (2017).
Fathallah (2020).
Teaching and Cooking with Culinary Teachers 719

FIGURE 107.7 Pigeon (chest and leg), confit carrots, polenta, by Jean-​ FIGURE 107.10 Chocolate sphere, litchi bubble, matcha frozen sponge
Baptiste Douce (2020). cake, ginger ice cream, red fruit coulis, by Davy Le Meur (2019).

only to surprise the customer but also to “challenge” their ability

to create and go beyond their “comfort zone”. The aim in this
contribution is not to try to reproduce the dishes (which would
be in exact contradiction with the creativity encouraged here!),
so the detailed recipes are not provided but only a short descrip-
tion, which helps to give a quick snapshot, through ten dishes, of
the work currently being done by these future teachers in French
culinary public schools.

FIGURE 107.8 Pineapple marshmallow, licorice jelly, yogurt sphere, milk

skin, rum air, meringue, by Stéphane Tulmets (2019).

FIGURE 107.9 Vanilla entremet with honey and chocolate ganache heart,
honey/​cinnamon biscuit, honey wafer, honey meringue in liquid nitrogen, by
Clarisse Joseph-​Louisa (2020).
The Monthly INRAE-​AgroParisTech Seminars on Molecular
Gastronomy
Hervé This vo Kientza1,2
1 INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
Is it true that the inner skin of lemons imparts a bitter flavour to
caipirinha? Is it true that massaging meat with butter can make
the meat more tender? Is it true that pouring the raka over ice
cubes gives a different result than adding the ice cubes into raka?
Is it true that roasted suckling pigs have a crisper skin when the
head is cut immediately after cooking?
Such traditional prescriptions, i.e., “culinary precisions”,
deserve testing, and this can be done in many different environ-
ments, from labs to schools.
Indeed, molecular gastronomy can be used as an important
tool for education at many levels, from primary school to
professionals, as this second part of the book demonstrates.
Depending on the particular context, various educational activ-
ities can be proposed. In Paris, since September 2000 (Figure
108.1), seminars of molecular gastronomy have been organized
every month, with various audiences: these meetings last for two
hours every third Monday of the month (except July and August)
(Figure 108.1). They have also been introduced or performed in
various cities of the world, including Montreal (Canada), Athens
(Greece), Nantes (France), La Havane (Cuba), London (UK),
New York (USA), Poligny (France), and Sao Paulo (Brazil).
Many Questions
The following list indicates some questions already explored
during the Paris seminars:
• Is there a difference in bitterness between artichokes of
which the tails were cut with a knife or torn apart?
• Is it true that cooked asparagus becomes firmer when
stored in hot water?
• Is it true that tiny brilliant spots can be observed on
oversalted sauces?
• Do almonds really thicken sauces?
• For a bearnaise (or hollandaise) sauce, is there a diffe-
rence if the yolk is cooked alone first, or if the sauce
containing egg yolk and emulsified butter is cooked
after emulsification of butter?
• Is it true that fried products are crisper when oil is added
to the batter into which the pieces are dipped before
frying?
• Can we distinguish sensorially between solutions of
citric acid and lemon juice?
• Is it possible to make hot and cold emulsions with
cooked egg yolk?
• Do apples taste different when peeled with a knife or
with a peeler?
• Does the presence of sugar influence the poaching
of eggs?
• Is it possible to make a mayonnaise sauce from eggs
that were once frozen?
• Does an overcooked hard-​boiled egg lose its sulfurous
odour after one night of storage?
• Is it possible to whip egg whites following freezing?
• Do strawberries lose their flavour when washed
with water?
• Is it useful to heat the genoise batter when preparing
the genoise?
• How can blue garlic be produced?
This is a very small sample, but one can see that almost any topic
of culinary technique can be studied. Also, in terms of techniques
discussed, one should indicate that the needed equipment is very
simple: ordinary kitchen equipment, a balance, pH paper and
thermocouples are generally enough. And having at hand a suffi-
cient number of participants can make sensory tests possible. In our
seminar in Paris, we very often use triangular tests for such studies.
Finally, these seminars help to, inter alia:
• clear up culinary practices
• promote scientific modelling of culinary practices
• explore culinary “precisions” (all the details that are
added to the “definitions” of dishes given through
recipes)
• modernize culinary teaching
• add research activity to culinary teaching
• present to the culinary world new ingredients, methods,
and hardware
721
722 Hervé This vo Kientza

FIGURE 108.1 In this seminar (October 2000), various equipment for whipping egg whites were compared.

FIGURE 108.2 Do corks prevent beans fracturing during cooking? Statistically, no.

In Practice A typical seminar includes:

Over the last 20 years, the Paris seminars have been organized in
different ways, but the latest ones took place on Monday (because
this day is often free for French professional cooks) between 4
and 6 PM (because this is between duties) in a culinary school
(Lycée Guillaume Tirel, Paris). Anyone is accepted after applying
by email, and in practice, the participants are interested people,
chefs, teachers, and scientists.
1. an introduction, discussing news about molecular gas-
tronomy activities in the world;
2. the experimental study of the month;
3. a discussion about note by note cooking, i.e., presenting
a particular product, demonstrating how to use it, and
discussing food safety matters;
The Monthly INRAE-AgroParisTech Seminars
4. choosing the topic of the next seminar: participants vote
for a particular culinary precision from a long list that
is given.
Shortly after the seminar, a report is sent by email using a large
distribution list.
Of course, the results that are obtained after such short
sessions are preliminary, scientifically speaking, but it should be
remembered that one goal of the seminars is only to get an idea
of real phenonema that can be studied correctly later. This being
723
said, it can be seen that the past results (one per month!) make
it worth the effort of organizing and performing these seminars
(Figure 108.2).
REFERENCE
Available on application to icmg@agroparistech.fr. The reports of
the seminars (in French) can be found at www2.agroparistech.
fr/​-​Les-​Seminaires-​de-​gastronomie-​.html
 Part III​

Culinary Applications
New Greek Cuisine

Georgianna Hiliadaki and Nikos Roussos

Funky Gourmet (Athens, Greece) www.funkygourmet.com

Greek Salad (Granita) 2. Put wet muslin on a strainer and empty the mixture into
the strainer. Let it drip for 1 minute. Throw away the
Ingredients
first water. Let it continue dripping in the fridge over a
• 500 g tomatoes, quartered strainer lined with muslin for 4 h. Freeze this mixture in
• 5 g capers a Paco container.
• 40 g calamata olives 3. Once frozen, pass it twice through a Pacojet. The result
• 120 g cucmbers will be a white granita.
• 8 g salt
• 30 g extra virgin olive oil
• 40 g onion
Cassoulet
• 100 g feta cheese
• 2 g dry oregano Ingredients for duck confit
• 50 g basic syrup
• 1 duck leg
Method • 10 g cooked duck skin
• 30 g demi glace
1. Mix all ingredients together with a handheld blender • 50 g onion, thinly sliced
• 30 g duck fat
until mixture becomes homogenized.

FIGURE 109.1 Greek salad (granita).

(Photo: Katerina Avgerinou)

727
728 Georgianna Hiliadaki, Nikos Roussos

3. After full pressure, lower the heat to number 1 and cook

for 15 minutes.
4. Remove from heat, let cool completely. Depress cooker
and open very carefully. Strain and drain beans over
kitchen paper.

Ingredients for duck confit balls in white bean paste

• 150 g white beans (from previous method)

• 20 g extra virgin olive oil
• 25 g lemon juice
• 5 g salt
• 50 g flour
• 3 g baking powder

Method

1. Mix all ingredients in the thermomix to make a

thick dough.
2. Take some bean paste, roll out on parchment paper to
FIGURE 109.2 Cassoulet.
4 mm thickness. Place atop the duck confit balls and
cover them completely with the dough, forming round
(Photo: Katerina Avgerinou)
croquettes.
3. Keep in the fridge until needed.
• 50 g port
• 2.5 g sherry vinegar Ingredients for white bean cream atop cassoulet
• 1 g fresh thyme, leaves only, finely chopped
• 100 g sunflower oil
Method • 200 g beans, soaked overnight in water
• 200 g bacon, chopped
1. Cook the duck leg in a vacuum bath with some sun- • 100 g duck sausage brunoise
flower oil at 82 °C for 14 h. • 10 g onion, chopped
2. Shred the meat off the bones and reserve 100 g of • 5 g garlic, chopped
• 50 g white wine
shredded meat.
• 70 g carrot, chopped
3. In a pan, caramelize onion in duck fat and deglaze with
• 40 g leeks, chopped
sherry vinegar and port until completely evaporated.
• 60 g onion, chopped
Add demi glace and thyme at the end. • 60 g fennel, chopped
4. Off the heat, add shredded meat. Mix well and cool in • 50 g celery, chopped
the fridge. • 10 g garlic, whole
5. Once hardened, cut into 25 g portions and roll into ball • 10 g coriander seeds, toasted and whole
shapes. • 2 l water
• 180 g butter
Ingredients for white bean paste • 300 g heavy cream
• 4 g white truffle oil
• 500 g white beans, soaked in water overnight • 7 g salt, white pepper
• 50 g carrot, coarsely sliced • 5 g lemon juice
• 50 g onion, finely sliced • 15 g xeres vinegar
• 2 g garlic, chopped
• 5 g coriander seeds Method
• 2 kg water
1. In a casserole, heat oil and saute bacon and sausage
Method until it renders some fat. Add onion, beans and garlic
and saute for a couple of minutes. Deglaze with the
1. Put all ingredients apart from beans in a muslin bag. white wine until evaporated.
2. Place everything in a pressure cooker, close the lid and 2. Empty the casserole with the bean mixture into a
bring to full pressure. pressure cooker and add the water. Put all the remaining
New Greek Cuisine 729

ingredients (from the 70 g of carrot to the 10 g of cori- Method

ander) in a muslin bag and place in the cooker.
3. Bring cooker to full pressure. Wait until it whistles and
lower heat to 1. Cook for 45 minutes. Strain and keep
the liquid.
4. In a pan, heat cream and butter until brought to boil.
5. Put the beans into the thermomix and add the cream
mixture. Heat until smooth. Correct taste by adding salt,
pepper, truffle oil, lemon juice and vinegar. Add some
cooking liquid if needed. Strain and keep warm in a
pastry bag.
1. Deep fry the duck confit balls at 170 °C until golden
in color.
2. On a plate, spoon some tomato jam. Sprinkle some pork
sausage brunoise and some julienned mange tout.
3. Place one fried duck confit ball and place some white
bean puree on top.
4. Sprinkle with some pork crackling.
5. Decorate with green cress and thyme flowers.

Ingredient for pork crackling

Dolmas
• 200 g pork belly For the crispy vine leaves

Method • 10 pieces of nice and whole vine leaves, preserved in brine

• 1 l simple syrup 1/​1
1. Cook the pork belly in a vacuum bag with some sun- • zest from 2 lemons
flower oil at 93 °C for 12 h.
1. Wash the vine leaves thoroughly until brine has been
2. Next day, shred the meat into fine strings and deep fry
removed.
at 190 °C until golden brown. Season with salt and keep
2. Put syrup in a big bowl and dip the vine leaves inside
over kitchen paper in a dry place.
until covered with the syrup.
3. Take them out one by one and dry them very well on
Ingredients for tomato jam
both sides on parchment paper. Sprinkle each leaf with
lemon zest on one side only.
• 500 g tomato flesh, no seeds or skin
4. Put the leaves on dehydrator trays. Each tray should be
• 2 g garlic brunoise
• 60 g onion brunoise wrapped with film.
• 100 g extra virgin olive oil 5. Dehydrate at 55 °C for at least 12 h. Use directly from
• 30 g sugar the dehydrator when needed.
• 4 g salt (or according to the taste of the tomatoes)
For the dolmas mixture
Method
• 75 g extra virgin olive oil
• 30 g ham fat, melted
1. In a pan, heat olive oil and saute onion and garlic • 200 g onion, brunoise
without coloring. • 5 g garlic, minced
2. Add the tomato flesh, salt and sugar and bring to the • 110 g scallions, minced
boil. Immediately reduce to low simmer and slowly • 50 g vine leaves (finely chopped without stalks)
cook for one to two hours until all liquids have • 280 g quinoa blanched for 10 minutes
evaporated. • 280 g chicken stock
3. Correct taste with salt and sugar. It needs to be quite • 10 g sea salt
sweet.
1. In a big pan, heat the olive oil and ham fat. Saute onions,
Ingredients for presentation scallions and garlic without coloring.
2. Add quinoa and saute.
• tomato jam 3. Gradually add the chicken stock and salt.
• rolled duck confit balls 4. Finish off by adding the vine leaves, stir and put mix-
• pork sausage, cooked and cut brunoise. ture in a bowl over an ice bath.
• white bean puree
• mange tout beans, blanched for 30 seconds and To finish dolmas mixture with taste
julienned
• pork crackling • 1.2 g white pepper
• thyme flowers, green cress • 4 g lemon zest
• 11 g mint
730 Georgianna Hiliadaki, Nikos Roussos

FIGURE 109.3 Dolmas.

(Photo: Katerina Avgerinou)

• 10 g parsley • 300 g egg and lemon sauce

• 10 g dill • 75 g yoghurt
• 15 g lemon juice, strained • 25 g syrup
• 150 g yoghurt lemon sauce
• 10 g egg and lemon sauce Mix all the ingredients in a bowl with a whisk.
To serve each plate
1. Weigh 600 g of mixture.
2. Add all the ingredients and mix well but softly. 1. On a plate put 50 g of quinoa mixture.
3. Keep in a bowl at room temperature covered with film 2. Put three dollops of yogurt on the quinoa.
until use. 3. Cover with a crispy vine leaf.

For the egg and lemon sauce

• 220 g whole egg Soft Boiled Egg

• 500 g chicken stock For the vanilla microwave cake
• 40 g lemon juice, strained
• 1.2 g white pepper, ground • 100 g flour
• 6.2 g salt • 100 g sugar
• 20 g dill, chopped • 100 g whole eggs
• 100 g butter, melted
1. Put eggs in a bowl and whisk well. • 7 g baking powder
2. Set another bowl over an ice bath and put in lemon, salt • 60 g milk
and pepper. • 40 g vanilla essence
3. In a pan, bring chicken stock to the boil and temper with
the eggs. Return to the pan and bring to 79 °C, whisking Method
constantly. Remove from heat immediately and strain
into the bowl with the lemon. Stir until dissolved. 1. Put eggs and sugar in the mixer and beat until smooth.
4. Strain again into a bowl, add dill and transfer into a 2. Mix flour and baking powder very well.
squeeze bottle. 3. Mix butter, milk and vanilla essence very well.
4. Add the liquids to the egg mixture and mix. Then add
For the yoghurt lemon sauce the flour mixture and mix again. Keep mixture in the
fridge.
New Greek Cuisine 731

FIGURE 109.4 Soft boiled egg.

(Photo: Katerina Avgerinou)

For the mango yolk 7. Once set, scoop out a small portion of the coconut
sorbet. Freeze again, and once set, adhere two pieces
• 300 g mango puree, strained into one to make a whole egg.
• 140 g passion fruit puree, strained
• 55 g Cointreau For the chocolate butter
• 7 g vanilla essence
• 500 g cocoa butter
Method • 500 g Ivoire chocolate
• 70 g Jivara chocolate
1. Mix all the ingredients together. • 0.2 g red color for chocolate
2. Empty mixture into 5 ml syringes.
Method
For the coconut sorbet
1. Melt cocoa butter in the microwave.
• 500 g coconut puree, strained 2. Add the chocolates to the melted hot cocoa butter.
• 165 g water 3. Let stand for 2 minutes and mix.
• 10 g glucose 4. Add the color, mix with a whisk and heat the mixture
• 30 g inverted sugar again in the microwave for one minute.
• 4 g super neutrose 5. Strain and set aside.
6. Hold the coconut “eggs” with a pin, dip into the choc-
Method olate butter twice and keep in the freezer.

1. Put water, glucose and inverted sugar into a pan. To finish (for one egg)
2. When water mix reaches 45 °C, add super neutrose and
mix until dissolved with a hand blender. • dehydrated coconut flakes for “salt”
3. Bring to boil whisking constantly for 3 minutes. • ground tonka beans for “pepper”
4. Strain and put the mixture in an ice bath. • 1 coconut egg
5. When it cools, mix with the coconut mixture and put it • one 5 ml syringe full of mango mixture
in a Pacojet overnight in the fridge. • vanilla cake
6. On the following day, freeze completely. Spin in the
Pacojet machine, set in plastic egg molds and freeze
again.
732 Georgianna Hiliadaki, Nikos Roussos

FIGURE 109.5 Orange explosion.

(Photo: Katerina Avgerinou)

Method 5. Put dry ice in a bowl and cover with orange leaves.
Place orange spheres on top. Add some boiling water
1. Empty one mango syringe into the coconut egg. infused with orange essence over the bowl and enjoy
2. Put 20 g of cake batter into a one-​use plastic coffee cup while it is still smoking.
and bake in the microwave for 30 s.
3. Serve with salt and pepper.
A Few Words by the Chef(s)
What Is Your Professional Background?
Orange Explosion
We both studied Culinary Arts in Manhattan, New York in 2002.
For the mixture
Ever since, we have worked in many different kitchens; the most
notable was a stage that Georgianna completed in the famous El
• 500 g orange juice, strained
Bulli restaurant in Spain. Nikos has also worked in Amsterdam
• 80 g syrup
• 5 g Cointreau in different places. Then in 2007, we started a private cheffing
• 10 g lemon juice business that evolved gradually into a restaurant. Funky Gourmet
• 15 drops of orange essence opened its doors as a restaurant in November 2009.

For the cocoa butter How Would You Define Your Cooking Style?
In Funky Gourmet, we create carefully designed degustation
• 200 g guanaja 70%
menus inspired by unique local products, captured at the peak
• 200 g cocoa butter
of their seasonality and freshness. We aspire to introduce our
guests to our own view of a refined Greek cuisine. It’s a combin-
For the golden powder
ation of Funky and Gourmet –​fine dining in an imaginative and
exciting way!
• 15 g malto texturas
• 7 g gold powder
The Recipe(s) You Are the Proudest of?
Method Every time is different, but for sure we love our “Greek salad”,
which is a signature dish that is always present in all our
1. Mix all ingredients from “For the mixture” and form degustation menus.
into hemispheric molds. Freeze.
2. Once frozen, combine two pieces to make a sphere.
Freeze again over dry ice. What Ingredients Inspire You More Particularly?
3. Drop into cocoa butter mixture and leave in the fridge. We get inspiration from many different ingredients! We love
4. After an hour, pass through the mixture of gold powder our local cheeses, the fresh buttermilk from Crete, which has a
and malto and leave in the fridge for 2 h to thaw. subtle flavor of sheep milk and grass! We love the lamb from
New Greek Cuisine
the northern part of Greece, and at the same time we love Asian
flavors like lime, ginger and lemongrass with their unique aromas
and acidity. They go very well in Greek cooking.
How to Be Creative and Keep “Fashion” in Cooking?
This is a big challenge in our occupation! I believe that as in
every form of art, you need to find yourself in a constant search
for inspiration from everything, i.e. traveling, reading, trying,
hard work, experimentation and documentation.
What Do You Do Now That You Did Not Do Ten
Years Ago?
Ten years ago, we were just starting off our adventure in the res-
taurant business, introducing Funky Gourmet to the public. As
is the case with every new endeavor, we were excited to show
our work and ideas while at the same time, being anxious about
the feedback we would receive. Over the years, and through
hard work, we managed to create a team of highly accomplished
professionals, which nowadays, after having successfully
launched two accredited restaurants, “Funky Gourmet” in Athens
and “Opso” in London, allows us to balance our time between
the chef and entrepreneur roles. We have now embarked on our
new endeavor, called “Pitta Bun”. The concept is a fast food
but slowly cooked walk-​in food joint where we aspire to give a
different meaning to the fast food concept.
What Would You Like to Be Able to Do in 10 Years That
You Are Not Able to Do Now?
A successful professional is someone who through his work is
able to inspire others and promote his field to new levels. With
this in mind, we hope that our hospitality group MGFG (www.
moderngreekfoodgroup.com), which is already active in Greece
733
and London, will be established as a worldwide hospitality com-
pany, offering a variety of services that span from fine dining
to everyday food and helping others with our ideas through
our consulting business “Horeca Consulting –​Restaurants and
beyond”.
Do You Collaborate with Scientists? Other People
(Artists, Designers, Engineers, …)?
In order to build a successful business and produce quality
results, you need to combine expertise from a number of fields,
each one acting as a building block of the end result. We actually
collaborate with a number of professionals, like architects for the
internal design of our restaurants, graphic artists for the design of
our marketing material, and artists and pottery makers for most of
our plates and serving tools. Each of our establishments needs to
welcome our customers and reflect the idea behind its existence.
At the same time, our dishes are unique and deserve the right
utensils to showcase them and bring out their uniqueness.
What Is Your Strategy to Innovate/​Create
New Dishes?
We believe that our dishes should create strong emotions in our
guests, as if they were in a theatrical play, so we are vigilant to
pay attention to every single detail. The starting point of every
new dish is inspiration. Inspiration can be found everywhere; in
art, nature, our childhood memories, during a walk through the
city. Once we grasp the idea, we will work hard on arriving at the
final flavor, texture, presentation and accompanying wine or drink
that will become part of our degustation menu. Then we search
for the ideal, and often custom-​made, plate to serve it in, while at
the same time we create a “story” to narrate to our guests, aiming
to offer a comprehensive experience. This procedure may take
from as long as one month to even a whole year to accomplish.
3D Printed Note by Note Recipe: Soya Lobster Prototype

Róisín M. Burke
School of Culinary Arts and Food Technology, Technological University Dublin, Ireland

The recipe in this chapter contains soy protein isolate 90%, which
is the only plant-​based complete protein source, consisting of all
the necessary amino acids. It has a high concentration of branched
chained amino acids (BCAAs) and can be used to create a Note
by Note food that is suitable for vegans who want to increase the
protein content in their diet. Also, it provides an alternative pro-
tein source for those who are allergic to dairy proteins. However,
it is not suitable for those who have an allergy to soya.

Recipe

• Cornflour –​66 g (high in amylopectin)

• Caster sugar –​2 g (sucrose)
• Soya protein isolate 90% –​ 28 g
• Salt –​1 g (sodium chloride)
• Vegetable oil –​14 g (olive oil, high in oleic acid)
• Water –​  120 g
FIGURE 110.1 The consistency of the Note by Note soya protein mix.

FIGURE 110.2 (a) Uncooked Note by Note soya lobster prototype and (b) cooked Note by Note soya lobster prototype.

735
736
Mix together, remove any lumps and put into the syringe in the
3D printer.
Print out and cook for 15 min at 180 °C.
A 3D Procusini 3.0 printing machine was used according to
the instructions outlined in the following.
1. The “Produce object” button was selected.
2. The food item was selected by clicking the relevant
button. In this case, “Pasta” was selected.
3. A prepared 60 mL cartridge was inserted containing
the soya mixture with the consistency shown in
Figure 110.1.
4. The device was calibrated.
Róisín M. Burke
5. The object for production was selected. In this case, the
lobster.
6. The production was started by clicking on the start
button.
After 6 min, the lobster was printed onto the silicone mat and was
then cooked for 15 min at 180 °C.
The recipe and shape used can be adapted and altered. The
protein content of the prototype lobster is approximately 10.9%.
Odorants and colours can be added according to what the con-
sumer wishes. Other shapes can be printed, either those available
from Procusini or customized shapes, e.g., 3D scanned images.
Cooking (with) Olive Oil
Christophe Lavelle1,2
1 Genome Structure and Instability, CNRS UMR7196 /​INSERM U1154, National Museum of Natural History,
Sorbonne University, 75005, Paris, France
2 Institut National Supérieur du Professorat et de l’Education (INSPE), Toulouse University and Cergy-​Pontoise University,
France
Known since ancient times for its medicinal virtues, olive oil
continues to bring into the kitchen its assets as part of a healthy
diet, inspired by the international reputation of the Mediterranean
cuisine. Extra virgin olive oil (EVOO) is extracted by a mechan-
ical process only (pressing, decanting, filtration, …) from fresh
olives, making it a pure fruit juice (see the chapter ”Extra Virgin
Olive Oil in Cooking” by Raffaele Sacchi).
Many international competitions are dedicated to the valoriza-
tion of this product. One of them, Olio Nuovo Days (www.olio-​
nuovo-​day.com; Figure 111.1) has a double particularity: (1) it
happens twice every year to compare only freshly pressed oils, in
January for the northern hemisphere and in June for the southern
FIGURE 111.1 Winners of the Olio Nuovo Days contest (northern
hemisphere, 2020).
(Photo ©Olio Nuovo Days)
hemisphere, and (2) it includes master classes and diners with
chefs to promote new ways to use EVOO in the kitchen.
Indeed, oil has been traditionally used mainly hot (for cooking)
or cold (for seasoning). For instance, in his Guide Culinaire
(1903), Escoffier mentioned olive oil only in four recipes: to
cook mushrooms in the Sauce Chasseur, to cook a hare in the
Civet de lièvre, to brown a fish in the Poisson à la Bonne Femme
and to season the Soupe à l’ail. The same is found in Le livre de
cuisine (1867) of Jules Gouffé, where ingredients are cooked in
olive oil (Cochon de lait, Brandade de morue, Filets de sole aux
anchois) or just sprinkled with oil for serving (Harengs saurs,
Salade allemande).
But EVOO could also be considered in itself as (one of) the
main ingredients of a recipe. This is what I demonstrate here with
a few examples I presented during some of the Olio Nuovo Days
master classes for chefs. Let’s go into the kitchen (Figure 111.2)!
FIGURE 111.2 Master class of EVOO cuisine at Ritz Escoffier cooking
school, Paris.
(Photo ©Olio Nuovo Days)
737
738 Christophe Lavelle

FIGURE 111.4 EVOO mousse served like a “Chantilly” on a butternut

soup.
FIGURE 111.3 Preparation of the jellified vinaigrette. (Photo ©Olio Nuovo Days)
(Photo ©Olio Nuovo Days)

Crunchy EVOO
Poor EVOO directly into liquid nitrogen, wait a few seconds,
then filter through a metal sieve and serve immediately.

EVOO Powder
Mix EVOO with half its weight of maltodextrin and a pinch
of salt; the preparation can be used immediately or kept in a
sealed box.

Jellified EVOO Vinaigrette

Boil together 100 g lemon juice + 30 g sugar + 100 g water + 1 g
salt + 6 g carrageenan, add 100 g EVOO, mix and pour on a flat
surface (or in a small glass for more volume); wait until set, cut
into the desired shape and serve (Figure 111.3).
FIGURE 111.5 Encapsulation of EVOO (served with the finished recipe
of Figure 111.4).
(Photo ©Olio Nuovo Days)
soap bubbles!), drop some EVOO on the film, and let the gravity
stretch the film and encapsulate the oil (Figure 111.5).

EVOO Cold Mousse

Put 200 ml liquid cream (30% fat) + 100 ml EVOO + a pinch of
salt/​pepper in a siphon. Keep cold until served (Figure 111.4).
EVOO Hot Mousse
Put 200 ml liquid cream (30% fat) + 100 ml EVOO + 200 ml egg
white + a pinch of salt/​pepper in a siphon. Keep for 1 h at 65 °C
in a water bath until served.
EVOO Encapsulation
Melt 200 g isomalt at 150 °C, then cool it to 120 °C. Bathe a
metal ring in order to form a film (as you would do to make
EVOO Ice Cream
Mix 300 ml liquid cream + 100 g sugar + 50 ml EVOO while
pouring liquid nitrogen into it until set. Serve immediately (liquid
nitrogen ice creams tend to melt quickly because they are made
from very tiny crystals; see the chapter on cryocooking by Peter
Barham).
EVOO Sponge Cake
Mix 150 g egg white + 25 g pistachio + 50 g EVOO + 75 g sugar
+ 25 g flour, pour in a siphon, let it rest in cold for 30 min, inject
gas (2 N2O cartridges) and pour 1 cm of the liquid into a paper
cup; cook for 45 s in a microwave oven at 900 W.
Cooking (with) Olive Oil
Frozen EVOO Soufflé
Prepare a sabayon with the juice and zest of 2 oranges + 5 egg
yolks + 100 g sugar + 2 tablespoons of rum, let it cool down
before incorporating 200 ml whipped cream + 200 ml EVOO,
add pieces of candied orange peel and meringues, pour into indi-
vidual cups and freeze. Serve with an orange coulis.
739
Of course, more generally, many recipes including a good
amount of any fat can be tested by replacing (part of) the fat by
EVOO. Have a try with pastries, chocolate mousse, soufflés, tarte
Tatin, custards, Hollandaise sauce –​Just play!
Cooking for the Elderly
Christophe Lavelle1,2
1 Genome Structure and Instability, CNRS UMR7196 /​INSERM U1154, National Museum of Natural History,
Sorbonne University, 75005, Paris, France
2 Institut National Supérieur du Professorat et de l’Education (INSPE), Toulouse University and Cergy-​Pontoise University,
France
Today, leading worldwide chefs are competing to be the most
innovative. Most of them look closely at scientific developments
and even sometimes work directly with scientists to help create
new textures, shapes and tastes. But this does not just concern
Haute Cuisine: science and modern techniques also provide
inspiration for innovative practices in collective catering, for
instance to master the texture of food served to the elderly in hos-
pital or nursing homes.
There is an increased prevalence of using modified texture
foods (MTF), due to the longer survival of those who suffer
from degenerative diseases that affect chewing and swallowing.
The modification of food texture and liquid thickness has
become a cornerstone of dysphagia management worldwide, as
it is important for the whole strategy to adequately nourish and
hydrate individuals.
The production of MTF involves the mechanical alteration of
the consistency of the original food so that it is easier to consume:
foods are chopped, minced, mashed or blended to compensate for
chewing difficulties or fatigue and to improve swallowing safety;
liquids are thickened to avoid aspiration of material into the
airway and improve transit to the oesophagus. While the control
of texture is not such a technical challenge in itself (numerous
mechanical tools and texturizing agents exist that help to obtain
the desired texture), a bigger issue is to preserve sufficient nutri-
tional and sensory quality to minimize the potential effect of
manipulating the food.
Food needs to be of an appropriate texture, particle size and
moistness for safe and comfortable swallowing. For moist food
(e.g. cooked vegetables and fruits), a thickener will often be
added during mechanical treatment to limit fluid loss; conversely,
a liquid (e.g. water, broth or milk) will be added to foods that
naturally contain less fluid (e.g. meats and cakes). Of course,
both the nutritional variability (due to overall dilution of the
nutrients and/​or changes in their bio-​accessibility) and the mois-
ture content (as food is a leading source of water for persons with
dysphagia) have to be carefully considered during this process.
To improve nutritional intake (regarding both energy and macro/​
micro-​ nutrients), products are usually enriched with various
constituents (e.g. milk powder, egg proteins, carbohydrate-​based
thickeners and enriched infant cereal).
In addition to a reduction of the texture (potentially removing
part of the pleasurable experience of chewing and manipulating
food in the mouth), visual appeal of MTF is often a concern, since
appearance also influences consumption and thus, nutrient intake.
Therefore, improving the nutritional quality of MTF obviously
goes in tandem with the improvement of its palatability (tex-
ture, taste and appearance): there is indeed no point in improving
something the patient won’t eat at the end!
I show here some examples of preparations made for eld-
erly people in nursing home and hospitals, using techniques/​
recipes mainly inspired by molecular cooking, in order to obtain
a desired taste and appearance and make the eating experience as
enjoyable and nutritionally beneficial as possible (Figures 112.1
and 112.2). These dishes were prepared thanks to a collaboration
with Julien Garnier (SENES SOLUTIONS, http://​senes.org) and
published monthly in a professional journal (www.edp-​nutrition.
fr/​parutions/​nutrition-​infos-​collectivites) in which recipes, along
with their scientific explanation and nutritional value calculation,
were provided to encourage their reproduction by chefs working
in catering for the elderly. Here, the idea is not to produce “fancy”
dishes, as one would expect in nice restaurants, but more basically,
to use modern texturizing agents to transform liquids made from
mixed ingredients into food whose shape either (1) reproduces
“real” food, therefore enhancing its appeal compared with a
puree (Figure 112.1), or (2) enables customers to grasp the food
with their hands (Figure 112.2).
There are still many solutions required in order to improve the
lives of elderly people; diet is a big part of this, which deserves
the best efforts of the best chefs!
741
742 Christophe Lavelle

FIGURE 112.1 Dishes prepared for elderly people using texturizing agents (alginate, agar, xanthan, egg white proteins, starch …) to reproduce “real” shapes,
as in a pot au feu (a), roast veal slices with vegetable puree (b), beetroot salad (c) or tomato salad (d).
(Pictures courtesy of Julien Garnier –​©SENES SOLUTIONS)

FIGURE 112.2 Various salted and sweet finger foods enabling the customers to eat with their hands, which is an efficient way to facilitate food intake by
some people.
(Pictures courtesy of Julien Garnier –​©SENES SOLUTIONS)
Culinary Constructivism and Note by Note Cooking
Pierre Gagnaire
Restaurants Pierre Gagnaire: Bordeaux, Chatelaillon, Courchevel, Danang, Dubai, Hong Kong, Las Vegas,
London, Nîmes, Paris, Shanghai, Seoul, Tokyo
At the end of the 1990s, because I had the feeling that molecular
gastronomy could be usefully applied for the modernization of
cooking, I called Hervé This … at a time when he had just been
invited by the president of the French Academy of Sciences (Guy
Ourisson) to deliver a lecture on molecular and physical gas-
tronomy during a working dinner. We decided to mix his lecture
with an “innovation dinner”, and we worked together in order
to put some of his applications of molecular and physical gas-
tronomy (inventions) “into culinary art”.
After this very successful “Science and Cooking Dinner” was
served in November 1999, we decided to collaborate on further
works, and this was my “Menu of the Year 2000”.
But, as the ideas were many, we soon decided to proceed with
one idea per month. Since then, the new food preparations invented
monthly by Hervé are described on my internet site (www.pierre-​
gagnaire.com/​pierre_​gagnaire/​pierre_​et_​herve) before he publishes
them elsewhere. Using his explanations, I turn these proposals
into real dishes that are served in my various restaurants around
the world. My recipes using these new ideas are published on my
internet site, demonstrating for 20 years now how culinary art and
molecular and physical gastronomy can usefully collaborate.
Of course, explaining all this material would make a whole
book, and it is not possible to discuss everything here, so I shall
focus on three examples only before giving a shortened list of our
published works, with brief explanations.
Debyes
In some cases, I ask Hervé about solutions to culinary issues, but
generally, he sends me a text explaining a new idea that he has
had. For example, for the preparations that he invented under the
name “debyes”, here is the text that he sent me first:
My dear Pierre
You know how much I love this egg yolk cooked at 67 °C:
it has an orange color that reminds me rather of the raw yolk
than the cooked egg yolk (tough, dry, sandy), but, especially,
it has more the softness of a yellow at 65 °C (that I introduced
years ago under the faulty name “perfect egg”), and it does
not yet have that hardness of a yolk cooked at 68 °C or so.
For me, who am neither a cook nor perfectly certain of
my real greed (I love probably too much science, equations,
calculations!), I see in this yolk at 67 °C the question of its
viscosity: why is it so? And I have promised myself for a long
time to test the hypothesis that I proposed more than fifteen
years ago: I think that, since only a minority of proteins have
coagulated at 67 °C, we have not formed a fully gelled system
throughout the mass, but only gelled aggregates; because of
this, this system is “pasty” (a “suspension”) because the con-
tinuous phase where the aggregates are dispersed is liquid.
The viscosity is increased by the presence of said aggregates.
Is it an “ointment”? No, because this type of system has
a fatty phase, and not aqueous one, where essential oils are
dissolved. So what is it? The word “batter” is too broad, since
it also designates pie dough as well as various suspensions.
Here, one could of course speak of a suspension of microgels
in aqueous phase, but it would be too long, and I proposed
to name it “debye”, after the Dutch physicist Peter Debye
(1884–​1966). When the dispersion is in an aqueous solution,
we have “hydrophilic” debye, but when the dispersion is in
the oil phase, then they are “hydrophobic” debyes.
Generalization
But this distinction that leads me to use words with more
than three syllables derails gluttony, and we will remain with
the debyes. How can we make them in practice? We have
seen that we need a liquid, which can be wine, orange juice,
spirits, alcohol, cider, beer, but also coffee, tea, broth … or
oil, whether from almonds, olives or pistachio, or even melted
butter, melted foie gras, melted cocoa butter.
In this liquid, we have to disperse gelled aggregates. Why
not start from a jelly that we have divided, using a mixer?
For example, let’s start with a velouté to which we will have
added agar-​agar; let’s take it, then mix the gel formed in oil.
We obtain a preparation that can be smooth as an ointment:
a debye.
Moreover, since we have previously macerated products
that have a lot of flavour in oils, why not use them. In sweet
as in salty. As for the gelling agents, there is plenty of choice,
between gelatin, agar-​agar, but also all the others, such in
flans, mousses, terrines, where the proteins meat, fish or
egg gel!
743
744 Pierre Gagnaire

Here, I dream: a ground citrus jelly in a beautiful olive 10 g chopped chives

oil? Or a chicken stock jelly with vodka in melted butter? Or a Mix egg whites, capers and chives. Cook the whole covered
lemon tomato gel and added vodka in a sunflower oil in which in a steam oven at 85 ° C for 15 minutes. Uncover and
lemon zest was marinated? let cool completely.
But I hasten to stop, because you know that my culinary art
is nil, and these proposals were above all smiling provocations, Final step:
attentive and friendly invitations. Mix very finely the caper/​chive gel while gradually adding the
anchovy oil. If necessary, pass the apparatus through a sieve to
Starting from this kind of text, we often have a meeting, including obtain a perfectly smooth debye; and add some pepper or a pinch
Michel Nave, who has been my sous-​chef now for more than of Espelette chili.
35 years, and after a small technical discussion, the issue is to
transform the proposal into a real piece of culinary art, a dish that
Debye chocolate/​orange
will be served in the restaurants. Of course, questions like the
particular season when this dish is served are important, as well For the orange oil:
as whether the proposals are practically possible in the specific
environment of restaurants. Sometimes, too much work needs to 100 g of sunflower oil
be performed correctly for all the guests, but sometimes, also, 5 g of grated orange peel
I don’t see the new possibilities that well. Culinary art is based on Mix the two elements and let macerate before filtering.
sensitivity, not entirely on reason!
In the following, I give the three recipes that were published For the chocolate gel:
for the debye proposal, but of course, there are many other
possibilities! 4 whole eggs
20 g unsweetened cocoa powder
¼ liter of milk
Debye Raspberry/​Basil 70 g of sugar
For the basil oil: Blanch the eggs with the sugar, add cocoa, mix well, finish
with milk. Put it through a fine sieve into bowls, cover
100 g of sunflower oil and steam cook at 85 ° C for 30 minutes. Uncover at the
10 g chopped basil leaves end of cooking and let cool completely.
Macerate the whole 48 hours before filtering by pressing
on the chopped leaves. Finishing:

For the raspberry gel: 200 g of chocolate gel

20 g of orange oil
250 g of raspberry Mix the chocolate gel by gradually adding the orange oil
100 g of spring water to get the debye.
30 g of sugar
Agar-​agar (24 g/​L)
Mix the raspberries, the sugar and the water. Filter, and add Diracs
agar-​agar. Let the gel harden in the cold.
Diracs is an old invention of Hervé as well. Again, the text by
Final step: Hervé is given before the recipes are explained.
For 250 g of raspberry gel and 40 g of basil oil: mix the gel
My dear Pierre,
by gradually adding the basil oil until you obtain a smooth and
I told you before that in vitro cultivated “meat” is being
creamy debye.
proposed for 1991, and regularly shown, in particular during
the summer, when political news is scarce. It sparked debates,
Debye Capers/​Anchovies some saying that, finally, we can stop slaughtering animals,
For the anchovy oil: other saying that it is not meat (and this was recognized at the
state level recently in some countries), others still highlighting
40 g of sunflower oil the exorbitant price of “steaks” produced …
20 g anchovies In short, once again, we were treated to a great hysterical
Grind the anchovies with the oil until you get a perfectly hustle and bustle … whereas “artificial meat” was invented
smooth paste. a long time ago: I published it on your internet site, but also
in my book Course of molecular gastronomy N° 1: science,
technology, culinary technique, which relations? (Editions
For the caper/​chive gel:
Quae /​Belin: a manual of technology and innovation, with
hundreds of my inventions described), the description of what
200 g of egg white
I had called “fibrés”, which are actually a reproduction of
30 g capers desalted with clear water
meat tissues.
Culinary Constructivism 745

Meat? It is made of “fibers” gathered in bundles, the Dirac Compact

latter being gathered in superbundles, and so on. Individual
Ingredients:
fibers are “pipes” made of a “skin”, the collagenous tissue,
with, in the center, water and proteins. And so hollow
100 g milk
noodles filled with a jelly gelatin, glued together by gelatin,
30 g egg white powder
make a physical system similar to meat: this is what I called
60 g sugar
“fibrés”.
Mane No. 1
But we can make other systems that “reproduce” the meat,
1 lime
including the “diracs” that we are discussing now.
Icing sugar
Let’s start from the following observation: meat is made
of 75 percent water and 20 percent proteins, plus fat, as a
first approximation. So if you make a batter with six spoonfuls For the crystalline lime
of water and four spoonfuls of proteins, and if you cook this
mixture, you get a system that is similar to meat, in particular Peel the limes, and freeze them before making very thin
from the nutritional point of view. slices using a ham machine.
Proteins? Not just any! Proteins that coagulate are needed: Add sugar to both sides of the lime slices and bake at
for example, egg white proteins (powdered egg white), milk 100 °C until they are crispy leaves.
proteins (but not just any), or vegetable proteins … since it has Keep away from moisture.
been known since 1801 that certain plants (peas, lentils, for
example) contain appreciable quantities of proteins, formerly For the compact dirac
called “vegetable albumen”.
In short, it is not difficult to take eight spoonfuls of water Mix sugar, milk, egg white powder, flavour with Mane N°
for every two spoonfuls of proteins, and to make a paste that is 1 and the lime peel.
heated in a pan, with a drop of oil so that the preparation does Mold into a greased baking sheet about 1 cm thick, steam
not stick to the pan. Oil: besides, can be added to the water cook at 60 °C for 15 minutes.
and protein mixture, so that it adds some tenderness. Unmold on leaving the oven.
Obviously, we must not forget the flavour: it is true that A consistency between marshmallow and lemon cream is
the cakes produced according to the process described above obtained.
will have no flavour if we use water, proteins, neutral oil. Cut using a 4 cm cookie cutter and dip each disc in the icing
However, for water, it is better to replace it with “water that sugar before serving between two lime crystallines.
tastes”: wine, broth, orange juice, coffee, tea … For the oil, we
can very well take a scented oil. For proteins, we discussed it Fibrous Dirac 1
earlier. And we can also add products to give color, taste (salt,
sugar …), smell (a beautiful essential oil …). Ingredients:
Then, when the object has been produced, it can be cut,
fried, grilled… In short, you can have fun cooking. 200 g tasty and seasoned beef broth
For example, if, instead of baking it en masse, you poured 150 g of egg white powder
the initial dough on a baking paper, and you scraped the layer 40 g of olive oil
formed with a comb, and then baking, microwave oven or beetroot powder
steam oven would give you fibers; if you rolled the fibred leaf Mix all the ingredients in order to obtain a homogeneous
on itself, you would get something similar to the bib … or the paste (avoid foaming). Let stand for approximately
surimis, since it is by this simple technique that we produce 1 hour.
the sticks which are often colored with paprika and scented Three different consistencies can be made:
with crab.
But here, I must stop, because the question is now one of First process:
culinary art! Spread on a greased silpat and cook in a 60 ° C steam oven for
10 minutes.
What to do with this? If the consistency question is easy to solve, Two possibilities are proposed:
the main one is the flavour. Indeed, this recipe was among “note
by note” proposals (see other chapters in this book), and we • after cooking, roll the sheet obtained on itself to form a
wanted to use the innovative odorant compounds in solution that roll about 3 cm in diameter. Then cut into slices 0.5 cm
one can now find on sale. At the very beginning of the note by thick. A soft, slightly fibrous gelatinous texture is
note adventure, we relied on odorant preparations prepared by obtained. It is to be used in the next dirac recipe.
the Mane company, each bottle of their box being made of pure
oil, with no flavour, and only one odorant compound. Here, the Second process:
product called “Mane N° 1” is a solution of benzaldehyde, with a In a mold 10 cm in diameter, put 1 cm of the first preparation, and
flavour between almond and cherry or pistachio; today, you could steam at 60 ° C for 15 minutes. When it is cooked, the consistency
use the Amerise product by the company Iqemusu. is the same as for a flan; stir in fresh butter until you get a tender
inner part inside a dry crust.
746
Third process:
By cooking directly with butter, one obtains a hard and dry mass.
Another Fibrous Dirac
Ingredients
100 g cooked black rice
20 g raw leek white cut into julienne
20 g roasted squash seeds
300 g of dashi
100 g beef (a little thick /​fat free)
100 g dirac roll cut into 32 slices
fresh butter
salt, pepper
Method
Season the beef, cook it with fresh butter and let it rest in
the heat before cutting it into thin slices.
Heat the dashi to 70 °C.
Place the hot black rice in the center of six hot soup plates,
place the minced beef, add the heated dirac rolls and the
leek julienne.
Pour the dashi, add the squash seeds and cover to keep very
hot until serving.
The “Chick Corea”
The first note by note dish ever served in a restaurant was
prepared in 2008 and shown in Hong Kong, at my restaurant of
the Mandarin Oriental, on 23 April 2008. It was called “Note by
Note N°1”.
It was not so easy to produce, but it clearly opened new doors.
Then, in 2010, it was decided to make another note by note dish,
with more culinary interest and more flavour complexity, and it
was this “Chick Corea” that was first shown at the Paris Book Fair.
It was again difficult to imagine because of the entirely new pos-
sibilities that the note by note technique was offering: not only in
terms of consistencies but also in terms of color, taste, odors, etc.
Here is the text by Hervé. Because the dish was the result of
many working sessions together, it was produced afterwards,
when the recipe was put online.
My dear Pierre,
I proposed note by note cooking in 1994, but I gave its
name only later. I showed it in lecture before the year 2000,
but I did not really start talking about it until after 2005, and
I chose that metaphorical name in 2006, when the world was
confusing molecular gastronomy and molecular cuisine.
Then, you have agreed to be the first to serve a note to
note dish in a restaurant in Hong Kong in 2009, and this
dish, named “Note to note No. 1”, became famous all over
the world, because it had been presented to the press, after
two dinners you had served at the Mandarin Oriental. The
recipe is on this site: http://​www.pierre-​gagnaire.com/​pierre_​
gagnaire/​travaux_​detail/​44
Pierre Gagnaire
Since note to note cooking has been explored by various
cooks, in many countries, because I have been constantly
promoting this new technique, which can be combined with a
truly new art. And that’s why, with restraint, however, I have
not stopped telling you the progress of note by note cooking,
anxious that you remain the first, not only from the artistic
point of view, as you are so almost unique, but also from a
technical point of view. Because it was also necessary to
learn the lessons of molecular cuisine: in French, the word
“cuisine” refers to both the piece, the practice, technique, and
style (we speak of “nouvelle cuisine”, for example); but, in
English, one distinguishes molecular cooking, the technique,
and molecular cuisine, this culinary style born with the new
technique. For it is true that new tools can have new uses, so
that new foods can result from a technical modernization, that
a new art can be born of new techniques.
And so, one day, you wanted another note to note dish.
This time, we have worked in a more “coordinated”, more
rational way, in that we have “drawn” (in the sense of design)
the dish, considering all its dimensions: consistencies, first,
their organization in space, colors, tastes. Of course, I did not
take part in the choice of the particularities of this dish, and
I simply answered technical requests.
This work lasted several months, and the result was this
dish which is today named “Chick Corea”, in honor of a
jazzman that we both like. I pass on the current constitution of
the dish, because you give it here, on the one hand, but, above
all, because I trust you: it changes often because I know that
your quest is tireless, and that you are never satisfied with a
dish that you have made.
On the other hand, let’s insist on some peculiarities, and,
especially, on the contrasts, very clear in this dish. Contrasts
of consistencies, contrasts of colors, contrasts of taste … but
it is not for me to speak about it, because I am not involved in
their choice.
On the other hand, this Chick Corea dish gives me the
opportunity to insist on the fact that after a few years of explor-
ation of note by note cooking, I see various ways, between
the realization of the dish entirely note to note, or simple
seasoning, or, as you did recently when the New York Times
came to discover cuisine note-​to-​note, building dishes “from”
compounds. For the Chick Corea, it is entirely note to note.
For convenience, Schweppes can be used instead of a solution
of quinine, and oil can be used instead of pure triglycerides,
vodka rather than pure ethanol … but this is only convenience.
For example, the oil is a mixture of triglycerides that does not
behave very differently from one of its fractions, and vodka is
truly a mixture of 40 percent ethanol and 60 percent water,
to the point even the odorous compounds which may be pre-
sent after distillation are removed, by passing the vodka over
active charcoal.
In short, the Chick Corea is perfectly a note to note dish …
and an extraordinary achievement, with very strange flavour
… as I love them: the maitre d’hôtel of your restaurant on
rue Balzac can testify that when I have the immense pleasure
of eating your food, I always ask for the most fun, the most
free jazz.
Free jazz, is not that what Chick Corea was playing?
Culinary Constructivism 747

Here is now the recipe for a sweet version of the “Chick Corea”: Method for the center:
2.1. Liquid lemon vodka
1. For the base: Bring the water to a boil with the isomalt.
1.1 The gel: Add the limonene or lemon peel off the heat.
100 g of glucose Infuse until completely cooled.
100 g of water Pass.
3 drops of limonene in solution in oil Add the vodka, color with beta carotene.
60 g of Schweppes Weigh 200 g of this infusion, add xanthan gum and cal-
5 g agar-​agar cium lactate.
1.2 The fiber: Let the mixture stand for 3–4 hours in the cold to hydrate
1 apple the xanthan.
200 g water Degas if necessary before making the beads.
70 g isomalt
8 g ascorbic acid 2.2. Alginate mixture
6 g agar-​agar Mix both, let stand overnight for optimal use.
3 drops of red dye
2. For the center: 2.3. Liquid pearls
2.1 liquid lemon vodka: Place some of the vodka/​lemon liquid in the alginate bath,
250 g of water let a film form, which will trap the liquid, and immerse
15 drops of limonene (which can be replaced by the pearl in clear water to stop the spherification.
lemon peel) Reserve the remaining pearls in the vodka/​lemon liquid.
60 g of isomalt
80 g of 40 percent ethanol solution in water (can be 2.4. Brittle green opaline
replaced with vodka) Heat the isomalt with water at 200 °C.
0.6 g xanthan gum Add the green dye and the menthol flavouring.
2 g calcium lactate Pour on a silicone sheet, spread with a sugar roller, let it
a pinch of beta carotene cool completely.
2.2 Alginate mixture: Break the green caramel using a grinder.
1 liter of water Lightly grease a silicone sheet, place on a sheet with a
5 g sodium alginate round imprint, sprinkle the pulverized sugar through a
broth, remove the imprinted foil, bake at 160 °C in a
3. Green brittle opaline: dry oven to form very fine caramel discs.
100 g water Let cool completely and keep in a dry place.
green dye
ascorbic acid Dressage
menthol
Cut the apple fiber into small cubes.
Mix 50 g of fiber cubes with 50 of gel.
Method for the base:
Arrange it in the bottom of a bowl, put a vodka/​lemon pearl
1.1. The gel in the center, cover with an opaline, and finish with a
Cook the glucose to form a caramel (indeed, a “peligot”). little bit of ascorbic acid.
Add water to this product. A soft sauce can be made by relaxing the gel with
Thoroughly melt the whole to obtain an amber syrup. Schweppes until creamy.
Weigh 140 g of syrup, add the limonene and Schweppes, Serve this sauce either in the bowl or in a gravy boat.
and stick together with the agar-​agar.
Harden.
Mix to obtain a thick gel.
The Whole Work
1.2. The “fiber” Enough with the presentation of a few. Let’s now just give the list of
Put the apple in a juice extractor, recover the pulp (cellu- all of them. There are three categories: “culinary constructivism”,
lose), finish squeezing it to obtain a dry material. “culinary precisions, knowledge and gourmandise”, and “note by
Add water, ascorbic acid and isomalt to the agar and bring note cooking”. Here is the list, which changes every month.
to a boil.
Add 50 g of apple cellulose to the hot liquid, followed by First for Culinary Constructivism
some red colorant, spread the preparation on a plate about 2018
1 cm thick, harden in the cold before detailing. Soft ticket: how to make an edible paper
748
Strawberry even more than strawberry!: a way to improve the
flavour of food ingredients in general
Debyes: see earlier
Perfumed oil: using knowledge about hydrophobicity and
chemical characteristics of odorant compounds
The cryptands: a generalization of chocolate sweets
2017
Firmed vegetables: how to get firmer vegetables using calcium
ions
Ultimate sweets
Many other puff pastries: the gerhardts
2016
Flaky butter: like a puff pastry, but entirely with butter
2015
Diracs: a reproduction of some characteristics of meat
2011
The beauties of the hard egg!
The kessels, the cousins of the scalded …
2010
Fried egg yolk: the title says everything
Acidity games: different acids give different tastes
Make the powder explode!: making powders with flavours
2009
Shitao: how to use capillarity in order to put some flavour
inside food ingredients
Leaves … spinach, parsley, mushrooms …: artificial leaves
The gold of the Great Kitchen!
2008
From a little evil, a great good … again
Ambroise Paré and emulsions
A glass of wine: the idea is to make a glassy material out
of wine
Crispy fish and meat leaves: the title explains
Priestleys: a generalisation of custard
Couhins Games: the Couhins is a wine from the Bordeaux
regions, and there are ideas for innovation out of wine in
general, and of Couhins wine in particular
Vauquelins: some foamy preparations
Double cooking: 144 ways of cooking differently
2007
Taste in the leaves: the name says it all
Bake in sugar crust: idem
Liebigs: there are physically gellified emulsions
Oil painting: painting with food
Geoffroy: an emulsion that you can generalize
The laquages: as in Chinese painting, but with dishes
Crystals in the oil: how to preserve the crackling sensation of
some salt crystals
Floating cubes: playing with density
2006
Tribute to Braconnot: strange gelifications
Hard under it: playing with consistency
The built is “beautiful”: an assumption about what is “good”
Chaptals: new dishes
The perfect scrambled egg
Pierre Gagnaire
The conglomerates
Interlaced layers: a generalisation of puff pastry
Sauce waters: how to produce clear sauces
An abstract dish: like abstract painting
2005
Welcome coffee: cocktails with 10 layers
A foam that holds: the würtz
Irreplaceable gelatin: to make pearls with a liquid core
Varnish and glaze in the kitchen
Cheese enfleurage: how to give some flavour to cheese
The kientzheim sauce: like a mayonnaise, but from brown
butter
2004
A clear tea jelly
Sweet salty
Two flavours from a single: using the fact that oil and water
don’t dissolve the same kinds of compounds
Roasted flour: and making balls with a chocolate flavour
Beginning of texture, epic cooking: how to delay food con-
sumption in order to make more beauty
Checkers with two and three dimensions: the Rubik cube out
of food
Constructions of space: again playing with the building
of food
Gradients and diffusion, controlled diffusion
Gay-​Lussac sauce, a sauce that is not classic: tradition never
produced foamy emulsions, but here is one
2003
The pastis effect: how to disperse oil in water
Pastel and contrasts: playing with art ideas; this was shown
after the publication of our book Cooking, a quintessential
art (The University Press of California)
The comfortable sauces: a base for making really “good”
sauces
Wind crystals: when you can make liters and liters of whipped
egg white from one egg white, you can make meringues
with a very light consistency, and Hervé called them “wind
crystals”
Taste in a gel
Maillard vegetables, half ice cream
Chantilly butter: like the famous “chocolate chantilly”
invented by Hervé in 1995, but with butter
The innate and the physiological: constant motors
Playing with tastes
2002
Ice salt: the name gives the description
The egg at 65 ° C: an invention by Hervé more than 10 years
ago, under the name “perfect egg”
Gradients in the kitchen: how to put some flavour into a dish
Juxtaposition: playing with contrast for improved pleasure
Inventing dishes by calculation: the example of Faraday
Culinary Precisions, Knowledge and Gourmandise
After 10 years of creative play that consisted of introducing each
month an “invention” (Hervé) that was staged (Pierre), a new
work (game?) was played, to better link knowledge and greed.
Culinary Constructivism
We start from the classical cultural background (culinary
precisions, proverbs, sayings, maxims, recipes, handshakes …)
that are the subject of Hervé’s daily analysis, and we go through
the sieve of knowledge in order to arrive at the kitchen. Both
gourmet and enlightened. Nothing beautiful is done without art
(Pierre).
All in all, this was the invitation to a new dialogue of culinary
art and scientific knowledge: long live enlightened gourmandise!
2017
The soufflés: the three rules for success
2012
Sayings, knowledge and greed: using cellulose in jams
Oh, roast chicken!: how to change its flavour
2010
The sole meunière is perfectible: improving a traditional dish
Kougelhopf and taste of yellow: how to increase the particular
brioche flavour
2008
Streamlined use for cream: now to make the best use of this
ingredient
2007
Salt and pepper in an omelette
Stories of hazelnut: innovations around brown butter
2005
Longer bouillons in the mouth
Note by Note Cooking
As we said before, the first note by note dish served in a res-
taurant was “Note to Note No. 1”, in Hong Kong, on 28 April
2006. Since then, there have been other note by note dishes.
One very important move took place when the New York Times
came for a note by note lunch in 2010, but we did not wait for this
to explore this new technique.
2015
Poiseuille: liquid channels into a gel
Chick corea: see earlier
2014
Berzelius: a note by note generalization of custard
Graham: a quintessence of creaminess
Capsaicin: the main pungent compound of chili
Cis-​3-​hexenol, the bouquet of the greenest olive oils
“Thio”: the spicy name of watercress, mustards (black),
rockets, wasabi
749
Piperine: this compound is the main pungent one in black
pepper; how to use it when pure
Complex gels: based on Hervé’s work on all the possible gels
Natural history and gustatory methional: methional has a
cooked potato flavour
2009
Note to note cooking: new tips
2008
Artificial oranges and orange blossoms: of course, one can
easily make alginate pearls, which some would name
spheres, but grouping them can make the equivalent of
oranges.
2007
Nollet salad: an artificial salad, including the dressing of
a “leaf”
2006
Note to note cooking: new hints
2005
Polyphenols and tannins: phenolics can be extracted from
grapes, and used in order to make sauces, such as the
“wöhler sauce”
Tartaric acid, elegant acidity, for “Pasteur”: using tartaric acid
as one would do with salt and pepper
Conclusions and Perspectives
Very early on, I realized that I did not know the classical
ingredients well. I observed them, I ate them, I touched them in
order to build my sensory universe and to write my pantry. Slowly,
the result became more harmonious. Slowly, I was drawing the
flavour toward a new universe. I need to put feelings and intelli-
gence in my dishes. Human beings need poetry, tenderness and
well-​made artifacts.
I am fascinated by paintings, and I love being captured by
them. The painter expresses his or her own intimacy with objects
that can be discussed. He or she offers to see, to share. And I like
this sharing, I need to put poetry in the plates.
The dressage teaches me harmony and makes me find peace.
I always need to organize space. My instinct is guiding me, but
also I learn by trial and error, which sometimes creates new
flavours. A dish should always be built, understandable, excep-
tional, and I need to find ways to give both emotion and pleasure.
One should evaluate culinary art in terms of tradition or innov-
ation; the key is the tenderness of the cook!
Decantation
Hervé This vo Kientza1,2
1 Ingénierie Procédés Aliments, AgroParisTech, INRAE, Université Paris-​Saclay, 91300, Massy, France
2 Groupe de gastronomie moléculaire, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
An Old Process for Cooking and Alchemy
Decantation is a simple and traditional process, but it can be
performed in innovative ways in the kitchen. A pinch of physics
and technology could help chefs to get easier results.
Decantation was initially a process for separating a liquid from
a solid, which was formerly used by alchimists (Delwiche, 2011;
Kahn, 2019). For example, Johannes de Rupescissa, in 1597,
discusses a “decantation at the sun” (De Rupescissa, 1597).
Later, when alchemy was transformed into “chymistry”, before
becoming chemistry, in the 18th century, decantation became
important in chemistry laboratories as well as in kitchens, as
shown by the encyclopedia published under the direction of
Denis Diderot and Jean Le Rond d’Alembert before the French
Revolution (Diderot, 1754, vol IV, p. 661 b):
[Decant, Decantation (chemistry & pharmacy) –​one uses
this term in order to describe the process of pouring slowly &
without creating turbidity a liquor that clarified spontaneously
from the deposit that formed at the bottom of the vessel in
which it lies; this is also what is called pouring by inclination.
Decantation is used either for separating a needed liquor from
feces that have to be discarded; or for recovering the solid
deposit from a useless liquid; or even because one wants to
separate the two materials that have to be treated separately
later. Decantation is used in all defecations, in the first case (see
Defecation): on the contrary, in pulverization, using water, it is
the fine powder deposited that one wants to keep & it is water
that has to be discarded. In falfisication of true precipitates,
the water clarified by rest & separated by decantation is usu-
ally useless. Washing of mines is a continuous decantation of
the second species. See Washing. When washing golden lime
treated by aqua forte & in decantation of the dissolution of
silver over this lime water, the liquor and the precipitates are
very precious, & the technician has to be careful about both.
[personal translation]
This encyclopedia made a bridge between alchemy (this word
appears in the first volumes) and chemistry (in the latest). It was
considered scandalous for many reasons, such as contesting
authority (of the king, of Catholic religion), but also because it
mixed, on the same level, technical activities and “philosophy”
(including natural philosophy, i.e. sciences of nature).
Later, the word “decantation” was also applied to the process
of separating two immiscible liquids (with different densities). For
example, when an aqueous solution and a liquid fat (“oil”) are put
in a beaker, two distinct layers form, with the oil layer floating on
top of the aqueous solution. The two liquids can be separated by
pouring the upper phase out of the container, leaving the denser
liquid behind. If done simply by tilting the vessel containing the two
liquids, or even a liquid and a sedimented solid, this old technique
produces an incomplete separation, as it is difficult to pour off all
of the top layer without pouring out some parts of the bottom layer.
Decantation has been frequently used in the kitchen, for
example for decanting drinkable water or for defatting meat
stocks, or for the clarification of wines (see later), but since the
beginning of the 20th century, the need to concentrate ores in large
facilities has triggered the design of systems for separating large
quantities of solids and liquids in continuous processes. In 1906,
for example, Dorr introduced a liquid-​solid continuous decanter
in a mine in South Dakota (Blazy et al., 1999): it was made of a
large circular vessel whose inclined bottom was scraped slowly,
using a rotating blade, while a clear liquid was flowing over the
walls of the vessel. Since the 1970s, a new lamellar decanter has
been introduced after a patent deposited in the USA in 1886.
Such ideas could be usefully transferred to kitchens.
The Old Dynamics
Decantation is based simply on the sedimentation of particles
in liquids. The classical description of the motion of particles
is given in the chapter on calculus at university in Part II of
this book. The calculation using a simple friction law (Stoke’s
Law) shows that, after an initial regime during which a particle
accelerates, the motion is at constant velocity:
vlim =
2
(
2 r g −ρ p + ρl )
9 η
where r is the radius of the particle, g the acceleration of gravity,
ρp the density of the particle, ρl the density of the liquid and η the
751
752
viscosity of the liquid. Of course, this applies only for isolated
particles, for specific radiuses, and in some particular regimes of
sedimentation.
More generally, the Reynolds number is useful for the classifi-
cation of the various descriptions. It is used for characterizing the
nature of flows (laminar, transient or turbulent), and represents
the ratio between inertial and viscous forces. For a spherical par-
2r vlim ρl
ticle of radius r, this number is defined as: Re =
η
The value Re = 10–​4 corresponds to the frequent separation
between particles and colloids. In practice, particles having a
radius less than 1 μm have sedimentation velocities less than
1 μm/​s, which is of the same order of magnitude as the thermal
molecular motion (Table 114.1). This is associated with a par-
ticular radius, called the colloidal limit:
(1/ 2)
 η2 k T 
rcol =  3 B2 
 ρg 
Using the values η = 0.001 P (for water), kBT = 4 × 10–​21 J,
g = 10 m.s−2 and ρ = 103 kg.m−3, one calculates a colloidal limit
of about 10–​6 m (Di Meglio, 2007).
The expression of the limit velocity changes when particles
have a surface electric charge, or when such electric charges are
screened by solutes in the liquid; using coagulants or floculants,
the small particles can aggregate into larger particles, for which
sedimentation is faster. For example, the cook of the French king
Louis XIV describes how one can rapidly get rid of particles
suspended in wine (“fining”, or “clarification”):
In order to clarify a wine that would be heavy or viscous, one
takes two ounces of beautiful fish glue, cut it in very small
pieces and melt it in a bottle of wine, without heating it,
simply by stirring; when melted, drop it in the barrel, and mix
the wine using a towel attached to a stick, then give a rest to
the wine, and it will be clear.
[personal translation]
For concentrated suspensions (volume concentration in particles
>0.5%), the inter-​particle interactions are no longer negligible,
and the decantation process is slowed. The limit velocity of
particles (for the laminar regime Re < 1 in a Newtonian fluid)
is then:
TABLE 114.1
Limit Velocity for Spherical Particles
Reynolds number Re Domain
<10 –​4 Stoke’s Law does not apply because of the
thermal motion of molecules
10–​4 to 1 Laminar regime
1 to 103 Transition regime
103 to 105 Turbulent regime
Hervé This vo Kientza
vlim =
( )
2.18 ρ p − ρl ( D − Dlim ) r 2
 1    1 
 D  + ρl   D +  ρ   η
    p  
where D and Dlim are the dilutions (volume of liquid per mass unit
of solids) of the suspension and the slurry, respectively.
Decanting in the Kitchen
All this being said, important ideas that one should keep in mind
in the kitchen are that small particles sediment sometimes very
slowly (days or even weeks) and that even liquids that appear clear
can be suspensions with particles too small to be seen, as has been
well studied with colloidal particles in wine (Zanchi et al., 2007).
In the kitchen, decanting is performed for various reasons, but
not only for liquids and solids; non-​miscible liquids can also be
decanted. A good idea is to take advice from chemists, who per-
form this process with test tubes: they know that tubes should be
placed at a 45° angle to allow sediments to settle at the bottom
of the apparatus. This allows the heavier particles to slide down
the side of the test tube while allowing the lighter liquid a path
to rise to the top. If the test tube were held vertically, the heavier
mixture component could block the container and not allow the
lighter liquid to pass as it rises.
For example, with wine, decanting the liquid from the potas-
sium bitartrate crystals avoids an unsavory taste. Also, cream
forms at the upper part of milk, so that decanting separates milk
from cream. Clarified butter is produced after a slow melting of
butter, the heavier aqueous phase sedimenting, before decanting
is performed. However, in the kitchen, one of the most frequently
practiced separations is with meat stocks; fat released from animal
tissues floats to the surface of stocks, and the cooks do their best
to separate it in order to get clear liquids. For such separations,
cooks would benefit from using separating funnels, as used in
all chemistry laboratories and in some areas of industry. A separ-
ation funnel has a valve at the bottom to allow the bottom layer to
be drained off, and can give a better separation between the two
liquids. Such systems are generally made of glass, which would
easily break in the kitchen, but some companies sell small Teflon
and quartz systems to which a large stainless steel vessel can be
adapted. In this way, decanting can be performed in seconds with
better results than with the traditional spoon or ladle.
Another important method for improving decantation is cen-
trifugation, which can greatly increase the rate of separation by
simulating a great increase in the force of gravity.
This can be easily seen in the expression of the limit velocity
(see expression earlier).
With some table-​ top equipment, acceleration more than
10,000 times higher than the natural acceleration of gravity can
be obtained, so that the velocity is increased by this factor: what
would take one day is reduced to less than 10 seconds.
Other Applications of This System
Having a separation funnel at hand, cooks can use it in various
innovative ways. In particular, it is useful to remember that in most
Decantation 753

food ingredients, there is a mix of taste, odorant or trigeminally

active compounds (among others), with various structures and,
accordingly, various solubilities. This is why I proposed as early
as 1984 to use separation funnels to separate such mixtures using
liquid–​liquid methods in order to “make twice more flavors than
before”.
In practice, this means simply putting one food ingredient
(divided solid or liquid) in a separation funnel with water and oil
(also ethanol could be used, for example). After the usual shaking
and possible heating of the whole, the liquids are separated,
dissolving different groups of compounds, which means different
flavors.
This is not the sole possibility for innovation. For example,
some cooks know that grinding plant tissues such as tomatoes
produces a red liquid, but clarification of this liquid removes col-
oured particles, leaving a transparent liquid with amber color.
Indeed, lycopene and other pigments are hydrophobic, so that
they do not color the aqueous solution. Centrifugation (after fil-
tration) can do the same as clarification, not only for tomatoes
but for many suspensions obtained by grinding plant or animal FIGURE 114.1 A curdled sauce can be clarified in order to serve a liquid
tissues. In particular, this is how I produced a clear sauce from that has the same flavor but a clear appearance.
a meat stew in 2005 from a sauce that curdled just before a ban-
quet. The clarified liquid was served near the plate, in a glass
Diderot D. 1754. L’encyclopédie, IV, 661 b, http://​enccre.academie-​
generally used for brandy (Figure 114.1). sciences.fr/​encyclopedie/​
Di Meglio JM. 2007. Colloides et Nanosciences, File j2130,
REFERENCES Techniques de l’Ingénieur, Saint-​Denis, France.
Blazy P, Jdid EA, Bersillon JL. 1999. Décantation, aspects Kahn D. 2019. Sources de l’alchimie, www.biusante.parisdescartes.
théoriques, Traité Génie des procédés, J3450 V1, Techniques fr/​histoire/​medica/​alchimie.php
de l’Ingénieur, Saint-​Denis, France. Zanchi D, Vernhet A, Poncet-​Legrand C, Cartalade D, Tribet C,
Delwiche J. 2011. Les alchimistes: la magie du feu, Bulletin de la Schweins R, Cabane B. 2007. Colloidal dispersions of tannins
Société Royale des Sciences de Liège, 80, 901–​906. in water-​ethanol solutions, Langmuir, 23, 9949–​9959.
De Rupescissa J. 1597. De consideratione Quintae essentiae rerum
omnium, Conrad Waldkirch, Basel, Switzerland.
Note by Note Recipes for a Press Conference and Tasting Organized
at ITHQ, 2012

Erik Ayala-​Bribiesca1 and Ismael Osorio2

1 Food Science and Technology Consultant and Professor at Cégep de St-​Hyacinthe, Quebec, Canada
2 Freelance chef and cook at the ITHQ, Montreal, Quebec, Canada

In October 2012, at the Institut de tourisme et d’hôtellerie du q.s. water to reach 300 mL
Québec (ITHQ) in Montreal, Canada, a press conference was 1.8 g citric acid
organized for a presentation of note by note cooking. With 2 g glycerol
students of ITHQ, we served the following bouchées during the 28 g sucrose
conference. The number of participants was about 150. Here, 0.5 g malic acid
there are slight changes to the initial recipes. 28 g sucrose
30 mL ethanol (94%)
3 g of a 10 ppm ethanol solution of sotolon
2 CO2 cartridges for the siphon
Bubbles q.s. apple flavouring
This is a sparkling cocktail made from water, sucrose, ethanol
and flavourings. It was carbonated using siphons and served dir- Method:
ectly in the glasses of the guests. The flavourings included apple
flavours and sotolon (3-​hydroxy-​4,5-​dimethylfuran-​2(5H)-​one) 1. Make a light brown caramel.
(Figure 115.1). 2. Deglaze with 100 mL water.
3. Add water in order to reach a total volume of 300 mL.
Ingredients: 4. Add citric acid and sucrose, mix.
5. Add the apple flavouring.
4 g table sugar (sucrose) for the caramel 6. Add the ethanol and the sotolon solution.
100 mL water for the caramel
7. Pour into a siphon, and charge with the CO2 cartridge.
8. Serve in wine glasses.

Cloud
These bouchées were made from fish protein coloured with
beetroot juice, sucrose, and earthy and woody odorant notes.
The bouchées were served between two clouds of minty cream
(Figure 115.2).

Ingredients for the “protein”:

500 g surimi (soaked, see Annex)

50 mL raw beetroot juice
25 g sucrose
q.s watermelon flavouring
1 g of a 10 ppm ethanol solution of the essential oil of gal-
FIGURE 115.1 Bubbles. banum (earthy and woody-​like notes)

755
756 Erik Ayala-Bribiesca, Ismael Osorio

FIGURE 115.2 Cloud. FIGURE 115.3 Tart.

Method for the “protein”:

Tart
1. Mix the beetroot juice with sugar and the essential oil This bouchée is a red seitan made using fructose and citric acid,
until the sugar is entirely dissolved. When used, the with a cherry and almond flavouring. It is served with a refreshing
galbanum odorant flavourings increase the earthy and isomalt tuile (Figure 115.3).
woody flavour of the dish.
2. Immerse the surimi in this solution, and soak for 1 h in Ingredients for the seitan:
the fridge.
3. Drain the surimi, and cut it in slices. 200 g gluten (powder)
4. Keep in the fridge. 200 g water
q.s red colouring
Ingredients for the green cream: 2 L sugar syrup 10%.

500 mL of cream 35% fat Method for the seitan:

8 g gelatine (sheets)
80 g sucrose 1. Mix the gluten, the water and the red colouring.
1 g of a 0.1% ethanol solution of L-​carvone (peppermint 2. Mix in a mixing machine at medium speed for 5 min
note) (until it becomes elastic).
1 g of a 0.1% ethanol solution of menthol (fresh mint note)
3. Using a pair of scissors, cut the mass into eight equal
q.s. solution of chlorophylls or bright green colouring.
parts.
4. Cook in boiling water for 30 min; discard the
Note: one can use an essential oil of peppermint instead of using
cooking water.
the solutions of carvone and menthol separately.
5. Again, cook in water for 30 min; discard the cooking
water.
Method:
6. Grind the seitan using a meat grinder.
7. Rinse the ground seitan.
1. Soak the gelatine in cold water.
8. Cook the seitan in the syrup for 45 min.
2. Heat the cream with the gelatine and the sugar at 37 °C.
9. Drain and keep in the fridge.
3. Let it cool in the fridge, then whip until it has a firm
consistency.
Ingredients for the flavouring syrup:
4. Pour the whipped cream in a piping bag, and keep in the
fridge.
100 g water
5. For serving, put a small quantity of cream on a plate, 150 g sucrose
and deposit the surimi on it. 6 g citric acid
6. Make disks of 3 cm radius with the cream on a plate 1 g of a 0.5% ethanol solution of benzaldehyde (instead,
cooled with liquid nitrogen. you can use an almond or cherry flavouring).
7. As soon as a thin layer of cream is frozen, put it on the
surimi.
Note by Note Recipes for a Press Conference 757

Method for the flavoured syrup: 200 g cold water

q.s. yellow colouring
1. In a pan, boil water, sucrose and citric acid. q.s. red colouring
2. Let it cool, and add the benzaldehyde solution. Cooking brine:
3. Pour the flavoured syrup on the ground seitan (250 g 1 L aqueous solution of salt 1%
syrup for 350 g seitan). Seasoning for the seitan:
4. Keep in the fridge overnight. 8 g whey powder
5. Drain the seitan before serving, season as needed. 2 g monosodium glutamate
q.s. salt (if needed).
Ingredients for the green tuile:
Here, the whey powder can be replaced with milk powder.
100 g isomalt
q.s. dark green colouring. Ingredients for the dusting:
1 g of a 1% ethanol solution of methylsalicylate.
15 g casein
Method for the green tuile: 5 g methylcelluose
80 g corn starch.
1. In a pan, melt the isomalt and remove from the heat.
Here, the casein can be replaced with milk powder
2. While still warm, add the colouring and the
methylsalicylate solution.
Method for the seitan:
3. Working with small quantities of isomalt, form into the
desired shape.
1. Mix all ingredients for the seitan.
4. Keep in a dry place.
2. Knead using a mixer at medium speed for 5 min
(until it becomes elastic).
Finally, on a small plate, arrange the seitan and the green tuile.
3. Using a knife, cut strips 1 cm thick.
4. In a pan, simmer water, and cook the seitan for
45 min.
Flama 5. Drain the seitan and change the water.
This one is a savoury seitan, cooked and fried with a dusting made 6. Again, simmer for 45 min.
of amylopectin (corn starch), casein, lactose and methylcellulose. 7. Drain the seitan.
It is served with smoky oil transformed into a powder using 8. Cut the seitan as needed.
maltodextrin and ground food charcoal, and sour and pungent 9. Season with the whey, glutamate and salt.
pearls (Figure 115.4). 10. In a mixing bowl, mix all ingredients for the dusting.
11. Dip the seitan in the dusting mixture.
Ingredients (yellow seitan, savoury): 12. Fry the seitan in neutral oil at 200 °C for about 30 s.
13. Place on absorbent paper, and keep hot.
200 g gluten
4 g salt
Ingredients for the smoky powder:

10 g liquid smoke
15 g neutral vegetable oil
100 g maltodextrin
1 g powder of food grade charcoal.

Method for the smoky powder:

1. Mix the oil and the liquid smoke, and heat at 80 °C.
2. Remove from heat and let stand until it cools down.
3. Recover the oily phase and add it to the maltodextrin.
4. Mix to adsorb the oil onto the maltodextrin.
5. Add the charcoal and mix well.
6. Keep in a dry place.

Ingredients for the pearls:

100 mL water
FIGURE 115.4 Flama. 1.5 g of a 1% ethanol solution of black pepper oleoresin
758 Erik Ayala-Bribiesca, Ismael Osorio

50 mL of a 5% aqueous solution of acetic acid Method for the blue surimi:

1.5 g agar-​agar
q.s. red colouring 1. Mix all the ingredients (except for the soaked surimi).
1 L cold neutral vegetable oil (stored overnight in a fridge). 2. Dry the soaked surimi in absorbent paper.
3. Using a brush, season the surimi (to taste), and store in
Method for the pearls: the fridge for 30 min.

1. Pour the water, the acetic acid solution and the agar-​
agar into a pan.
2. Mix well and bring to the boil.
4. Simmer for 4 min.
5. Add the red colouring and the black pepper oleoresin
solution. Mix well.
6. Using a pipette, add the solution, as droplets, to the very
cold oil. Allow the droplets (pearls) to gel.
7. Recover the pearls and keep them in the fridge.
Final assembly:
On a cold plate, put the fried seitan. Garnish with the smoky
powder and the peppery pearls.
Ingredients for the snow of pepino-​serum:
500 g cucumber (peeled, seeded)
50 mL cold water
Seasoning for 500 mL of pepino-​serum:
5 g salt
1.25 g citric acid
q.s. essential oil of grapefruit, diluted to 1% in ethanol.
Method for the snow of pepino-​serum:
1. Mix the cucumber with water at full speed for 1 min.
2. Sieve, so that you get a clear liquid.
3. Add the seasoning. Mix well.

Final assembly: put the cucumber water in a mixing bowl. Add

Ultra liquid nitrogen while whipping until you get the consistency of
Fish protein, capsaicin, salt, monosodium glutamate, blue/​ snow. Arrange the surimi and the pepino-​serum snow on Chinese
purple colouring. Garnish with a “pepino-​serum snow” made of spoons.
cucumber, salt, citric acid and grapefruit flavouring (Figure 115.5).
Ingredients for the blue surimi marinade:
Annex: Soaking the Surimi
225 g soaked surimi
The surimi that one can buy at grocers is already seasoned. This is
25 mL cold water
why we devised a method to soak it. We proposed to use sticks of
5 g monosodium glutamate
surimi, because they can be unfolded, in contrast to solid blocks
2.5 g salt
that would need to soak for too long.
q.s. blue colouring
q.s. purple colouring
1 g of a 100 ppm ethanol solution of capsaicin (when needed, Ingredients:
the capsaicin can be replaced by 3.2 g of an ethanol
solution of 0.1% Capsicum extract 500,000 SHU, or by 500 g surimi (fish protein)
an infusion of crushed habanero chili in ethanol). q.s. cold water.

Method:

1. Soak in very cold water in order to prevent microbial

proliferation.
2. Change the water three times during the first 30 min.
3. Soak for three more hours, changing the water every
hour.
4. Keep in the fridge during the whole soaking process.

Of course, surimi can also be made using 20% proteins (plant,

milk or egg) and 80% water: the dough that is obtained by simply
mixing them can be poured on a flat surface, scratched with a
spoon and cooked in the microwave oven. You can add a starch
batter to the protein one if you need more consistent surimi.

FIGURE 115.5 Ultra.

Using Liquid Nitrogen to Prepare Ice Creams in the Restaurant

Christophe Lavelle and Hervé This vo Kientza

with chefs André Daguin, Noël Gutrin and Philippe Labbé

Although liquid nitrogen has been used for more than a century Fruit Sorbet by Noël Gutrin
to make “instant” ice cream, relatively few of the general public
For more than 20 years, Noël Gutrin has been head chef of Le
know about it, and even fewer have had the opportunity to taste
Crystal, a restaurant located in the middle of the Futuroscope
the result in a restaurant. Because liquid nitrogen is very cold
(Poitiers, France), a leisure park dedicated to new technologies.
(−196 °C), it can be hazardous to work with, and one needs to
So, it makes sense to propose to the visitors/​customers some
be quite cautious when using it in the kitchen, and even more
“molecular cuisine” preparations, including sorbets made at the
cautious when using it in front of customers!
table. This is a very simple and straightforward application, but
To get all the details about the history, how it works and what
customers always enjoy it.
can be done, the reader should refer to the chapter “Cryogenics
Ingredients:
in the kitchen” by Peter Barham in this book. Here, we will illus-
trate three different preparations. 500 g of fruit puree
Before cooking, it must not be forgotten that one should respect 0.5 L liquid nitrogen
the safety instructions: store liquid nitrogen in a well-​ventilated
room, only use approved containers, and never try to keep liquid Preparation:
nitrogen in a closed bottle/​vessel (it will explode!); when using Mix fruit puree in a metallic vessel, pour on it 0.5 L of liquid
the liquid, it is advised to wear spectacles and also work in a well-​ nitrogen, slowly, while moving it with a whisk until done (you
ventilated room. More generally, when using a tool in the kitchen
(knife, stove, liquid nitrogen …), one has to learn how to use it in
order to avoid the risks associated with all tools (there is always
danger, but we have to be prepared and avoid the risks).

The Brandy Sorbet by André Daguin

In the 1970s, the late chef André Daguin had engineering contacts
at the centre for beef artificial insemination in Aubiet (France),
and this led him to use liquid nitrogen for making ice creams.
An example is this recipe, which is published in his book (André
Daguin, Le nouveau cuisinier gascon, Stock, 1982).
Ingredients:

½ L white Armagnac
½ L sugar syrup
1 L liquid nitrogen

Preparation:
Mix in a metallic vessel ½ L of white Armagnac (60°) with ½ L of
syrup (500 g of sugar and 0.6 L of water, which you boil together
and cook with a lid on for about 20 minutes). Pour on it 1 L of
liquid nitrogen, slowly, while moving it with a wooden spoon and
without being afraid of the dramatic vapour formation. Less than
one minute later, the sorbet is made. FIGURE 116.1 The equipment proposed in 1981 for using liquid nitrogen
in the dining rooms of restaurants.

759
760 Christophe Lavelle, Hervé This vo Kientza

FIGURE 116.2 Making an instant fruit sorbet with liquid nitrogen.

don’t see anything due to the vapour … but you feel the “resist-
ance” of the liquid becoming solid).

Crêpes Mademoiselle by Philippe Labbé

Philippe Labbé has been head chef of La Tour d’Argent, one
of the oldest restaurants in Paris, with a remarkable view on
Notre-​Dame cathedral. Among “traditional” dishes, the “Canard
au sang” in three services is undoubtedly the most famous and
has been numbered since the first one: the Prince of Wales
(who should have become Edward VII) had no. 328 in 1890,
King Alphonse XIII had no. 40,312 in 1914, Japanese Emperor
Hirohito had no. 53,211 in 1921 … and today, we have passed
the 1,200,000 mark! But beside keeping some historical dishes
on the menu, chef Philippe Labbé also injected some modernity, FIGURE 116.3 Preparing the pancakes.
including this Crêpes Mademoiselle recipe, which provides more (©Marco Strullu)
than a 10 minute show at the table, with “extra-​cold” (liquid
nitrogen) and “extra-​hot” (flaming) at the same time!
Ingredients:

30 g butter
30 g sugar
2 lemons
2 oranges
Some candied orange peel
3 pancakes (per person)
A mix of Grand-​ Marnier, Cointreau and Mandarine
Impériale
300 g fresh curdled milk

Preparation:
For the pancakes (Figures 116.3 and 116.5): heat a metal tray,
add the sugar and wait until caramelisation, then add the butter,
some orange and lemon zest, and the pancakes. Flambé with FIGURE 116.4 Preparing the ice cream.
alcohols.
(©Marco Strullu)
For the ice cream (Figures 116.4 and 116.5): put the curdled
milk in a special bucket designed with a transparent rim to
Using Liquid Nitrogen to Prepare Ice Cream 761

FIGURE 116.5 Ice cream and pancakes are prepared at the same time in front of the customers, here with Virginie Reynaud, sommelier and Stéphane Trapier,
restaurant manager.
(©Marco Strullu)

FIGURE 116.6 Grating the citrus zest as a final touch.

(©Marco Strullu)

prevent splattering and whisk energetically while pouring the Add some orange juice, candied orange peel and orange
liquid nitrogen until done. segments, arrange nicely on a plate, add the ice cream and grate
Serving (Figure 116.6): some lemon zest on the top.
A Note by Note Traditional Chinese Dinner Created and Served
in Singapore

Kelly Lee, Aaron Goh, Tony Choo, Nicolas Vergnole, Gn Ying Wei and Tais Berenstein
Chefs from the At-​Sunrice GlobalChef Academy

On 29 June 2018, a team of chefs from the At-​Sunrice GlobalChef
Academy created and served a complete seven-​course Note by
Note Chinese dinner to two different seatings at the At-​Sunrice
GlobalChef Academy. For the first time in Singapore, a Note by
Note menu was created and served.
It was quite a challenge to create a Note by Note menu incorp-
orating Chinese flavours, because the team basically had to iden-
tify the most important notes to each of the Chinese dishes we
chose to replicate.
When thinking of promoting Note by Note in Singapore, pro-
viding our diners with awareness and literally food for thought,
we are pleased with the outcome; if Note by Note is one of the
possible answers to reducing food waste and feeding the world
in the future, it is now time to start understanding Note by Note
cuisine and work on improving it.
textures, but at the same time complement each other as a com-
plete starter.
For “Mi”, we re-​created a sweet, light pink bao dough using
pure compounds. The final bao had a very thin crust that was
formed by steaming and frying. The result was a bao that was
pleasantly soft to the bite. For the inside, we created a moist,
juicy and sweet filling using vegetable fibre, protein and different
sugars. Aromatics chosen were fruity, vegetable notes. For many
guests, this one was deemed really wonderful.
Second Dish: Sea of Change
Inspired by Xiamen oyster pancakes, the Note by Note version
starts with pan frying pure starch to achieve the smokiness and
crispness of the original dish while combining it with a flavourful
gel and odorant solutions that capture the flavours of the sea
(Figure 117.2).

The Note by Note Menu, with Comments from

the Chefs
First Dish: Do, Re, Mi Come Together in a Mini Dim
Sum Bamboo Steamer
Do, Re, Mi is inspired by the traditions of Cantonese dimsum
(Figure 117.1). Do is a pot sticker/​gyoza type dumpling. Because
wheat flour is a mixture rather than a pure compound, we decided
to play with different starches in order to achieve similar texture
and mouthfeel to a conventional dumpling made with wheat flour.
For the filling, we mixed pure gluten (protein from wheat) and
cellulose that had been extracted from daikon. For the flavour
component, we used fruity and liquorous odorant compounds
(from the Iqemusu company; they are called “Evocations”) on
top, and we gave an ash colour to the dough using bamboo char-
coal. To complement and balance the overall flavour of “Do”, we
created a Note by Note dipping sauce that gave a chili kick and a
tangy sensation.
For “Re”, we developed a bright yellow multi-​starch dough,
which we moulded into round shapes to create circular “chips”
that were fried and seasoned with salted egg-​flavoured dust. We
seasoned with popcorn and grilled notes (Iqemusu Evocations).
We wanted Do, Re, Me to have different colours, flavours and FIGURE 117.1 Do, Re, Mi.

763
764 Kelly Lee et al.

FIGURE 117.3 Longevity.

FIGURE 117.2 Sea of Change.

Third Dish: Longevity

Gluten-​free noodles made out of a mixture of starch topped with
an extract of rice wine and chicken. Iqemusu products were used
to enhance the freshness of the noodle, and a piperine solution
was used to give a bit of a kick (Figure 117.3). This dish happens
to be gluten free. We used a combination of different starches
for the right noodle texture, egg albumin (protein) and vegetable
gum. We used green colorant to give a delicate mint tone to the
noodle. To complement all, we tossed it with an extract of rice
wine and chicken.

Fourth Dish: Lioness Head

This variation of the traditional Shanghainese Lion’s Head
meatballs is created with gluten, with savoury daikon cake as
white radish fibre (Figure 117.4). Pigments extracted from the
green part of leeks are used for colouring, along with water
and agar-​agar. The Iqemusu “Sfumo” was selected to enhance
the smoky taste of the meatball, and “Hertzon” to enhance the
taste of the “leek”. This variation of the traditional Lion’s Head
meatballs had three components:

• Lioness head: created with gluten, crumb from Note by
Note bread and cellulose from vegetable to create a soft
meatball-​like result. Sfumo was selected to enhance the
smoky taste of the meatball.
• “Daikon Cake”: we created a daikon patty using cellu-
lose extracted from daikon and a vegetable gum. Grilled
flavour came naturally when we used a frying pan to
prepare it.
• Note by Note leek: green colorant extracted from leeks
was applied to a semi-​ hard gel, giving it a natural
dégradé look. The gel was artistically cut into a leek
FIGURE 117.4 Lioness Head.
leaf shape, and Hertzon evocation (from Iqemusu) was
used to give the “leek” a fresh and grassy aroma. Leek
charcoal was sprinkled on top.
Fifth Dish: Save the Shark
This shark-​free version of traditional shark’s fin soup is made of
agar-​agar, with a dried seafood extraction added (Figure 117.5).
The soup base is an extraction of fowl and pork knuckle. The
A Note by Note Traditional Chinese Dinner 765

FIGURE 117.5 Save the Shark Soup.

FIGURE 117.7 The Last Kiss.

sweetened and salted coconut fibre combo. How to eat it? Knock
it hard to crack it open to see all its different elements. The
Iqemusu product chosen was “Copesca”, which gives coconut
and peach notes.

Seventh Dish: Last Kiss

Using the same amount of cocoa butter and sugar icing, blend
together to create a paste (Figure 117.7). For one type, the
“Piovo” Iqemusu evocation was added, which had a green bell
pepper flavour; the second was “Naha” evocation of pineapple
flavour. We then shaped it like a drop and coated it with coloured
cocoa butter to match the flavour. The final touch was to add one
drop of Naha and Piovo evocation on the top.

FIGURE 117.6 Hidden Heart. The Recipes

In the following recipes, the Note by Note idea is used exten-
sively, but not entirely: sometimes spices can be used (such as
odorant compound used had a mushroom odor (1-​octen-​3-​ol),
ginger powder). Of course, these mixtures can be replaced with
because it mixes well with shark’s fin. A mushroom helped with
pure compounds when they are available.
this dish, including a tiny quantity of monosodium glutamate
as well.
Do –​Pot Sticker, Note by Note
Sixth Dish: Hidden Heart Serves: 10
For our dessert, we wanted our guests to experience a number Dough:
of textures in the same preparation (Figure 117.6). We obtained 100 g wheat starch
this result by creating a hard meringue shell, inside which we 10 g gluten
included a foam, a gel and a sorbet. The components are crumble 8 g neutral oil (sesame)
(made with cellulose, salt and oil, and sugar); meringue shell 2 g salt
(made from protein powder); agar-​agar gel; sorbet; and foam. 1 g bamboo charcoal
The shell sat on Note by Note crumble made with a balanced To texture: boiling water
766 Kelly Lee et al.

Meat base: Dough:

40 g gluten powder 60 g (1/​2 cup) tapioca flour
100 g stock (artificial reproduction) 100 g (1/​2 cup) cornstarch
1/​2 tsp nutritional yeast (inactive) 30 g (3 tbsp) potato starch
1⁄4 tsp garlic powder 2 g (3/​4 tsp) salt
(1 tsp) neutral oil 6 g (1½ tsp) xanthan gum
12 g egg white powder
Daikon fibre: 107 g water (or to texture)
In a juicer, extract daikon fibre. Wash it to get rid of any flavour 23 g egg yolk powder
and starch (until the water comes out clean). 20 g (1½ tbsp) vegetable oil
To colour: a yellow food colorant
Marinade + meat base:
100 g “meat” base Salted egg seasoning:
15 g (3 tsp) soy sauce, Higashimaru Shoyu 30 g salted egg powder
7.5 g (1½ tsp) sesame oil or any vegetable oil 10 g maltodextrin
0.6 g ginger powder Salt to taste
20 g daikon fibre, wok fried Sugar (sucrose) to taste
Garnish with a solid dissolution of capsaicin on top and
Dumplings: popcorn Iqemusu evocation.
Dough
Meat base (see earlier) Method, dough:
Seasonings/​toppings: 1. Prepare the workplace.
2. Scale and collect all the ingredients and equipment.
Chinese vinegar or aqueous solution of acetic acid 6% 3. In a medium bowl, combine starches, salt and xan-
Sesame oil than gum, egg powders and oil.
4. Mix dry ingredients.
Method, dough:
5. Add water to texture.
1. Prepare the workplace. 6. This will feel much like pastry dough.
2. Scale and collect all the ingredients and equipment. 7. Let it rest for 10 min.
3. In a medium bowl combine dry ingredients: wheat, 8. Roll dough thinly; it will require lots of dusting with
gluten and bamboo charcoal. Mix. cornflour.
4. Add oil and water and mix until dough is well 9. Using a cutter, cut dough into small round shapes.
combined. 10. Deep fry in clarified butter or oil.
5. Rest the dough for 20 min and roll using a machine 11. Sprinkle with salted egg seasoning.
or rolling pin.

Method, “meat” base: Mi –​Bao Char Siu, Note by Note

6. Mix all well to develop the gluten. Serves: 10
7. Steam for 1 hour.
8. Crumble with fork till it looks like minced meat. Dough:
20 g wheat starch
Filling preparation: 22 g sugar
9. Mix “meat base” and marinate well; let it rest in the 1⁄4 tsp salt
chiller for at least 1⁄2 hour. 1⁄4 tsp baking powder
10. Stir fry filling and torch for more flavour. 12 g gluten powder
11. Chill before using. 4 g instant yeast
3 tsp neutral oil
Method, dumplings: To colour: pink colorant
12. Roll dough very thinly (better using Kitchen Aid). (55 g) Water (to texture)
13. Fill and fold dumpling.
14. Steam for 10 min. “Meat” base:
15. Shallow fry bottom of the dumpling on a frying 40 g gluten powder
pan. 100 g chicken stock reproduction or water
16. Serve with a drop of sesame oil and Chinese vinegar 1⁄4 tsp garlic powder
or Note by Note evocation and chili sauce. 4 g (1tsp) neutral oil

Daikon fibre:
Re-​Salted Egg Crisp, Note by Note In a juicer, extract daikon fibre. Wash it to get rid of any flavour
Serves: 10 and starch (until waters comes out clean).
A Note by Note Traditional Chinese Dinner 767

Char Siu sauce: 1 tbsp cooking oil

25 g (1½ tbsp) white sugar (sucrose) 1 tsp ginger juice
25 g (1 tbsp) honey, or a concentrated solution of glucose, To taste: white pepper or piperine
fructose or sucrose
23 g (1 tbsp) dark soy sauce Oyster seasoning:
15 g (2 tsp) soy sauce, Higashimaru Shoyu 100 g leek water
1 g (1 tsp) five spice powder To taste: white pepper or piperine
1 tbsp bacon oil, lard or vegetable oil To taste: Iqemusu Evocation “Onium”
To texture: corn starch, in water 10 g agar-​agar

Base meat + sauce: Chilli sauce note:

100 g base meat 50 g corn syrup
20 g daikon fibre 50 g glucose syrup
To taste: Char Siu sauce 3 g chili powder
To texture: corn starch, in water 5 g tomato powder
To taste: salt
Method, dough: To taste: tartaric acid
1. Prepare the workplace.
Method, batter:
2. Scale and collect all the ingredients and equipment.
3. Dissolve yeast in warm water. 1. Prepare the workplace.
4. Mix all dry ingredients. 2. Scale and collect all the ingredients and equipment.
5. Pour yeast, water and oil into dry ingredients and mix 3. Mix all ingredients and fry as a conventional pancake.
until incorporated.
Method, seasoning:
6. Prove for 1 hour or until it doubles it size.
1. Mix all ingredients.
Method, “meat” base:
1. Mix all well to develop the gluten. Method, sauce:
2. Steam for 1 hour. 1. Mix all ingredients and reduce in a pot.
3. Crumble with fork till it looks like minced meat.

Method, Char Siu sauce: Longevity –​Noodle, Note by Note

1. For Char Siu sauce, mix all ingredients but corn starch. Serves: 8

Method, filling: Dough:

1. Stir fry meat base, daikon fibre and Char Siu sauce. 60 g (1/​2 cup) tapioca flour
2. Thicken filling with starch and water till right 100 g (1/​2 cup) corn starch
texture. 30 g(3 tbsp) potato starch
3. Blow torch for Char Siu flavour. 4 g (3/​4 tsp) salt
6 g (1½ tsp) xanthan gum
Filling + dough: 12 g egg white powder
1. Portion dough. 107 g water
2. Fill it. 23 g egg yolk powder
3. Let it rest for 20 min. 20 g (1½ tbsp) vegetable oil
4. Steam for 10 min. To colour: green colorant (see earlier)

*add piperine powder (pepper) on top of bao. Chicken jelly:

Evocation on top: beetroot. 500 g chicken extract
gelatine, sheets
Sea of Change –​Oyster Pancake, Note by Note
Directions:
Serves: 8
1. Prepare the workplace.
Batter: 2. Scale and collect all the ingredients and equipment.
50 g potato starch 3. In a medium bowl, combine flours, salt and xanthan
10 g rice starch gum.
1⁄2 tsp. salt 4. Beat the egg white powder lightly and add the oil.
120 g cold water 5. Pour the egg-​oil liquid into the flour mixture and stir.
2 g baking soda (sodium bicarbonate) 6. This will feel much like pastry dough.
768 Kelly Lee et al.

7. On a pasta machine, set #1, fold a piece of dough into To taste: monosodium glutamate
thirds. Roll and fold nine times before starting to roll 8 g gellan gum
more thinly. Roll up to set # 6.
Note leek:
About this recipe: 200 g water
8 g elastin
• note that the dough is sticky at first and needs dusting. 2 g agar-​agar
In the roller, it feels crumbled, but once rolled a few To colour: green colour (see earlier)
times, texture is perfect for noodle making. To taste: Iqemusu Evocation Onium and Hertzon
(1-​cis-​hexen-​3-​ol)
To garnish: burned leek ash
Lioness Head, Carrot Cake, Note Leek,
Note by Note Method, “meat” base:
Serves: 6 1. Prepare the workplace.
2. Scale and collect all the ingredients and equipment.
“Meat” base: 3. Mix all well to develop the gluten.
40 g gluten powder 4. Steam for 1 hour.
100 g stock (chicken extract) or water 5. Crumble with fork till it looks like minced meat.
1⁄4 tsp garlic powder
(4 g) oil Method, crumb:
1. Dissolve yeast in warm water.
2. Mix all dry ingredients.
Daikon fibre:
In a juicer, extract daikon fibre. Wash it to get rid of any flavour Save the Shark, Note by Note
and starch (until water comes out clean).
Serves: 10
Note by Note crumb:
Soup:
20 g wheat starch
22 g sugar (sucrose) (This one was not Note by Note, as it used traditional ingredients,
1⁄4 tsp salt but one can change it using other ingredients.):
1⁄4 tsp sodium bicarbonate
12 g gluten powder 8 kg pork knuckle
4 g instant yeast 8 kg old chicken
3 tsp neutral oil 300 g dried scallops
55 g water (to texture) 500 g Yunnan ham
5 kg pork ribs
Meat balls: 100 g ginger
500 g minced meat base 200 g leek
100 g daikon fibre, wok stir fried To taste: salt
To taste: Chinese wine 80 L water
3 tbsp light soy sauce
Fin:
1 pinch ginger powder
1 pinch garlic powder 500 g water
100 g note by note crumb 15 g agar-​agar
2 tbsp neutral oil To taste salt
3 tbsp soy sauce
Broth and sauce: 1⁄2 tsp dark soy sauce
500 g water Gold leaves (optional)
To taste: monosodium glutamate
Method, soup:
To taste: rock sugar
To taste: gelatine sheet 1. Prepare the workplace.
To taste: salt 2. Scale and collect all the ingredients and equipment.
To taste: Iqemusu Evocation Sfumo (smoky odour) 3. Boil all ingredients in water until it reduces to half.
To colour: caramel colour
To texture: xanthan gum (for sauce) Method, fin:
1. Boil water and add in agar-​agar.
Daikon cake: 2. Bring to boil and add seasoning.
300 g daikon fibre 3. Blend in gold leaf.
500 g daikon, milk 4. Set in chiller.
To taste: salt 5. Use grater to grate fin.
A Note by Note Traditional Chinese Dinner 769

Hidden Heart, Note by Note Last Kiss –​Lollipop, Note by Note

Serves: 6 Serves: 10

Coconut agar gel: Filling:

Yields: enough for 10 meringue shells 200 g cocoa butter
200 g sugar (sucrose)
40 g sugar Flavours and colours:
250 mL coconut water or odorant solution of gamma
nanolactone To taste: odorant solution “Piovo” (Iqemusu)
3 g agar-​agar powder To taste: odorant solution “Naha” (Iqemusu)
5 drops odorant solution “Copesca” (Iqemusu) To colour: yellow
500 g water To colour: green
5 tbsp desiccated coconut
Method:
Method: 1. Prepare the workplace.
1. Prepare the workplace. 2. Scale and collect all the ingredients and equipment.
2. Scale and collect all the ingredients and equipment. 3. In a Robocoupe mixer, add cocoa butter and sugar and
3. In a pot, bring the 500 g water and 5 tbsp of desiccated mix until a paste.
coconut to a 5 min boil. Let steep overnight. 4. Separate into four batches: use two for filling and two
4. Strain the desiccated coconut. In a pot, bring 250 g of for colour coating.
coconut water (or water with gamma nanolactone solu-
tion added), sugar and agar powder to a boil for 3 min.
5. Pour the mixture into a plastic container and add 5
drops of the odorant solution “Copesca”. Whisk well,
let set in the chiller.
6. Cut the agar into small cubes and blend it into a gel.
Greek Diracs

Makis Kalossakas1 and Nicolas Nikolakopoulos2

1 Chef teacher at the Institut Le Monde, 45 Thessalonikis Str., Moschato, Athens 18346, Greece
2 Pastry chef at the Institut Le Monde, 45 Thessalonikis Str., Moschato, Athens 18346, Greece

Introduction The Menu

On 16 March 2018, Yannis Vlachos (professor of mixology), The menu served in Athens included:
Makis Kalosakas (professor of cooking), Nicolas Nikolakopoulos
(professor of pastry), Michel Ntenoutas (professor of cooking) Amuse-​Bouche Note by Note
and their students served a Note by Note Cocktail, followed by a Evocation Mediterranean seafood, crisp and tender
Note by Note Dinner, at a press event in the Ecole Le Monde in Evocation milk and cheese, foam
Athens, Greece. Evocation potato and tomato chips, crisp
This was the first time that Note by Note cooking was ever Evocation cucumber, French toast, garlic, foam
performed in Greece. A menu was devised, which contained no Evocation sausage, tempura, crisp and tender
meat, no fish, no eggs, no milk, no vegetables, no fruits … but only Evocation mustard and wasabi, foam
compounds: this was the culinary challenge for the chefs/​teachers
of the Ecole Le Monde and for the students who helped them. Entrée Note by Note
In order to make this Note by Note menu, the chefs/​teachers and
Ravioli evocation earth
their students used water, proteins, sugars and polysaccharides,
lipids, vitamins, odorant compounds, etc., i.e., the same
First Course
compounds that are found in ordinary food ingredients, but the
compounds used were obtained by fractionation of plant tissues, Cod /​beetroot and potato purée
which is a remarkable advantage from a sustainability point of Almond foam, black garlic jelly
view: instead of transporting vegetables and fruits, which can con-
tain as much as 99% water (in salads), only dry and light powders Sherbet Note by Note
are transported, so that nutrients are not damaged. Moreover, the
new food ingredients do not need to be stored in the cold, which Main Course
saves energy and avoids using gases that can damage the ozone Lamb under a crust of black olives, eggplant, red and
layer. Note by Note could be good for the climate! yellow pepper, thyme juice, and pearls of black olives,
Here, we give the menu, and the recipes for it. of green olives and of Gruyère cheese

FIGURE 118.1 Ravioli evocation earth. Notes: mushroom, 2,4-​

dithiapentane, goat, beetroot. Caramel solution, flavours mushroom and FIGURE 118.2 Note by Note sorbet. Evocation violet, geranium, pink rose,
beetroot. fresh meringues, evocation lemon butter.

771
772 Makis Kalossakas, Nicolas Nikolakopoulos

FIGURE 118.3 N-​Fruit, including imaginary fruit (evocation fruity and sour/​foamy), kernel (evocation caramel, honey, vanilla/​crisp and soft) and earth
(evocation coconut, milk, vanilla, smoke).

Dessert Note by Note Garlic jelly:

N-​Fruit 200 mL water
4 g gellan
titanium dioxide (white colorant, Manis Chemicals)
Mignardise Note by Note Iqemusu odorant product: Onium
Macaron freshness
Evocation French mint, lemon Dough for frying:
200 g water
milk serum proteins
140 g “Note by Note flour” (see the following)
The Recipes 20 mL neutral oil with Iqemusu odorant production:
Here we give the recipes for some dishes. Hertzon
Red colorant (Louis François)
To taste: salt
Evocation sausage-​tempura, evocation mustard-​wasabi
Iqemusu odorant product: Sfumo
This dish included a dirac sausage-​tempura and a garlic
jelly. Note by Note flour (portion):
760 g wheat starch
Dirac sausage-​tempura: 120 g gluten
1600 mL sausage water, made with water, Sosa sausage 8 g egg white powder
flavouring and Sosa smoke flavouring 5 g serum milk proteins
560 g egg white powder 27 g cellulose
640 g corn maltodextrin
16 g carob gum Note by Note bread:
20 g serum milk proteins 1 portion “Note by Note flour”
96 g psyllium cellulose 8 g lyophilized yeast
8 g monosodium glutamate (Manis Chemicals) 20 g salt
48 g methylcellulose 720 mL water
Red and green colorants (Louis François, Inc.) 3 g diglycerides
120 g garlic jelly cut in brunoise (see later) 10 g serum milk proteins
Iqemusu odorant products: Sfumo, Poppe, Onium 0.4 g monosodium glutamate (Manis Chemicals)
30 mL sunflower oil
Greek Diracs 773

Mustard foam: 40 mL olive oil (or sunflower with Iqemusu odorant

1400 mL water product Hertzon)
35 g milk powder 0% fat monosodium glutamate
Allyl isothiocyanate (solid solution, Mane S.A.) capsaicin (solid solution, Mane)
105 g white vinegar (solution of acetic acid in water) piperine (solid solution, Mane)
To taste: salt titanium dioxide (white colorant)
Colors yellow and brown jaune (Sosa) orange and red colorants (Sosa)
Iqemusu odorant products: Sfumo, Carez Iqemusu odorant product: Troot
Monosodium glutamate (Manis Chemicals)
Sablé dough (for the shrimp):
250 g Note by Note flour (see earlier)
Evocation Potato and Tomato Chips, Evocation 94 g oil
Cucumber, French Bread and Garlic, Evocation Milk 6 g mono-​and diglycerides of fatty acids
and Cheese 100 g water
salt
Evocation potato: sugar semolina
10 g egg white powder
180 g water
10 g milk powder 0%
40 g potato starch
carboxymethyl cellulose
monosodium glutamate (Manis Chemicals)
10 g baking powder
fleur de sel (after frying)
anisole
Iqemusu odorant product: Thian
active charcoal (Agugioro&Figna)
Evocation tomato:
Feta evocation:
175 g water
300 g feta water
40 g potato starch
100 g water
Red colorant (Sosa)
6 g kappa carrageenan
To taste: salt
8 g yoghurt powder
To taste: monosodium glutamate (Manis Chemicals)
Iqemusu odorant products: Berthome, Piovo Tomato sauce:
Evocation milk and cheese: 1 L water
80 g tomato powder
200 g water
monosodium glutamate
30 g egg white powder
82.5 g modified starch
20 g milk powder 0% fat
anisole
20 g dehydrated yogurt culture
sunflower oil
Fresh butter flavourings (Sosa)
40 g feta water (feta brine, i.e., salt with water in barrels
50 g corn maltodextrin
with the feta cheese)
6 g salt
Iqemusu odorant product: Onium
monosodium glutamate (Manis Chemicals)
Iqemusu odorant product: Chapre
Ravioli Evocation Earth
Emulsion cucumber:
Ravioli dough:
1800 g water
45 g water
360 g cooked “Note by Note bread”
5 g egg white powder
oil
200 g “Note by Note flour” (see earlier)
126 g modified starch
10 g olive oil (or sunflower oil with Iqemusu Hertzon)
salt
active charcoal dissolved in oil
360 g cooked Note by Note bread (the Note by Note bread
salt
was put in water, filtered, and the water recovered)
To taste: monosodium glutamate (Sosa)
Monosodium glutamate
Iqemusu odorant product: Chole
Colour yellow and green (Sosa)
Iqemusu odorant product: Conq Beetroot jelly:
200 mL water
Evocation Mediterranean Seafoods 15 g beetroot powder
Dirac shrimp bass: 4 gelatin sheets
salt
200 mL shrimp water
monosodium glutamate
150 g egg white powder
Iqemusu odorant products: Piovo, Troot, Onium
774 Makis Kalossakas, Nicolas Nikolakopoulos

Cheese foam: Meringue:

300 g water 60 g water
300 g feta water 7 g egg white powder
45 g milk powder 26% 40 g sugar
30 g yoghurt powder 80 g fructose
6 g iota carrageenan
6 g gellan Honey caramel caviar (14 g /​portion):
Iqemusu odorant product: Chapre 2.5 g agar-​agar
1 g carob gum
Garlic transparent gel: honey flavouring (Sosa)
1.5 L water 600 g caramel water
12 g agar-​agar Iqemusu orodant products: Poppe, Mappo
1 g carob gum Sunflower oil
citric acid cacao butter, colour brown
monosodium glutamate
salt Cigarette:
Iqemusu odorant products: Onium, Berthome 120 g “Note by Note flour”
100 g icing sugar
Foam caramel –​mushroom –​beetroot: 100 g sunflower oil
280 g beetroot water 6 g mono-​and diglycerides of fatty acids
160 g mushroom water 90 g water
60 g caramel water (sugar, caramel colours) 10 g egg white powder
38 g modified potato starch
monosodium glutamate N-​fruit skin:
salt 45 g gelatine
Iqemusu odorant product: Chole 500 g water
80 g glucose
Beetroot, mushroom and caramel juice: 1.5 g carob gum
6 L water 10 g apple cellulose
72 g beetroot powder Flavourings and colour (Sosa)
Iqemusu odorant compound: Chole
monosodium glutamate Vanilla earth:
600 g caramel water (200 g sugar, 400 g water) 50 g tapioca maltodextrin
salt 40 g cocoa butter
colorant: deep red (Sosa) 10 g oil
30 g icing sugar
vanillin
N-​Fruit, Imaginary Fruit, Fruity and Sour, Kernel
Iqemusu odorant products: Sfumo, Poppe, Monka, Frum
Evocation Caramel, Vanilla, Earth Evocation Coconut,
Milk, Vanilla and Smoke Sponge black:
N-​Fruit Mousse: 150 g water
40 g milk proteins
280 g water
110 g oil
50 g inulin
110 g water
50 g polydextrose
25 g egg white powder
30 g gelatine
1/​2 g soybean lecithin
4 mL malic acid dissolved in water 1/​1
190 g sugar
10 g xylitol
105 g corn starch
10 g trehalose
5 g salt
1 g ascorbic acid
vanillin
8 g cellulose from pears
colour, flavouring
5 g apple proteins
Iqemusu odorant product: Copesca
strawberry flavouring (Sosa)
apple flavouring (Sosa)
banana flavouring (Sosa)
colour yellow-​red (Sosa)
Iqemusu odorant products: Hertzon, Perqoi
An Eclipse Dish
Hervé This vo Kientza1,2
1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
In 2001, there was a total solar eclipse on the Earth, and this gave
us the opportunity to invent a “dish for the eclipse”. Here is the
recipe, based on the idea that compounds are differently released
by solutions, emulsions, gels … This idea is indeed an applica-
tion of the ideas that are developed in the chapters on disperse
system formalism (DSF), matrix effect and bioactivity, and it is
certainly based on this important picture of the ranking of inter-
action energy (Figure 119.1).
1. To begin, the appearance has to show the eclipse: a
white disc with a dark crescent. How to make it? It was
proposed to use squid ink on a clear pasta disc. For
example, use 200 g flour, 4 egg yolks, a spoon of olive
oil and 5 g salt, and knead. When the dough is smooth,
make two thin discs that you cook in boiling water for
about 3 min. Then paint the eclipse appearance.
2. But this is too simple, and we want also to make an
eclipse of flavour. In particular, an eclipse means that
something familiar (the sun) is losing part of itself.
Indeed, we shall use garlic, which has a strong flavour,
but boiling it five times will make it lose its strong fla-
vour. A purée made of it will be unusual.
3. But the eclipse is also associated with cold temperatures,
and we shall play with that. The two pasta discs will be
hot, and they will hide, in between, big cold shrimps
FIGURE 119.1 The eclipse dish is built using this ranking of interaction energy.
(langoustines): in a frying pan, cook the langoustines
in very hot oil, so that smoke appears. The shells have
to turn brown. Then let them cool, remove the shells
and keep the meat. In a pan covered by a lid (to keep in
the odorant compounds), cook the shells with onions,
carrots, tomato and oil; flambé with Pastis, add half
a litre of white wine, and cook with a lid on for one
hour. Then grind, filter, season with salt and pepper, and
emulsify a large quantity of butter (an ultrasonic probe
is very useful for this).
4. Now, we make a flavour that is eclipsed, i.e., that
disappears and appears again. We use the fact that
odorant molecules are often hydrophobic and volatile.
Such molecules are easily released when pure and hot,
but they are released less when they are “trapped”. For
example, let us use chervil, which we grind in order to
make a purée; part of it is spread on the black part of
the hot discs of pasta, and the other part is macerated
with oil for some hours. Then, this oil is used to make
an emulsion using the langoustine stock and gelatine
(this system is called a “liebig”). Put this emulsion in
the middle of the cold langoustines, and it will jellify.
You are now ready to taste an eclipse.
775
Modern Swiss Cooking

Denis Martin
Restaurant Denis Martin, Vevey, Switzerland

Sparkling Meringue /​Double Cream of Gruyère Do it again, and separate hemispheres of frozen double cream.
A sugarless dessert: The idea was to associate the memory of two
recipes that all Swiss people know, i.e., the Tiki sparkling bonbon Spicy Cigar
(acid and sparkling sweets with a citric flavour) and the cream 2 whole eggs
meringue. 1 cL of “five spices”
The double cream mixture with lemon is made without sugar 1 teaspoon of bitter cocoa
thanks to the vacuum evaporator: the odour of lemon is recovered 1 spoon of plant charcoal
without the acidity (Figure 120.1). 10 cL hot water
180 g flour
Double Cream Mixture
1. Mix and add the mixture into a siphon.
100 g of double cream 2. Charge with nitrous oxide and dispense into silicon
50 g egg white moulds.
30 g kirsch 3. Cook for 20 seconds in a microwave oven.
20 g lemon extract (made using a rotary evaporator)

1. Mix all ingredients. Sparkling Meringues

2. In a well-​aerated room, wearing glasses and protective 1 egg white
gloves, dip a spoon in liquid nitrogen, and after some 1 knife tip of sodium bicarbonate
seconds, put the convex side of the spoon in the double 1 knife tip of citric acid
cream.

FIGURE 120.1 Sparkling meringue /​double cream of Gruyère.

777
778 Denis Martin

FIGURE 120.2 Crisp fondue “Vacherin Fribourgeois”.

1. Whip the egg white with the mixture of bicarbonate and and it is reminiscent of the Swiss Alps. I have a passion for this
citric acid. It will foam. extract because during degustation the odours come at the end,
2. Immediately make thin meringues. and the olfactory sensation lasts much longer. It is a step-​by-​step
3. Dry for 4 hours at 58 °C in a dehydrator. dish, with a succession of flavours (Figure 120.2).

Chocolate Sorbet Ingredients

100 g dark chocolate 70% fat 3 g salt
20 g cocoa powder 12 g yeast
20 cL water 160 g rice flour (for tempura)
1 litre of liquid nitrogen 180 g icy water
10 g fir tree extract
1. Heat the water.
2 g powder of plant charcoal
2. Add the cocoa powder and the dark chocolate until
300 g Vacherin Fribourgeois
melted.
3. In a well aerated room, wearing glasses and pro-
tective gloves, add liquid nitrogen to the mixture while Recipe
whipping. For the black tempura dough:

Dressage 1. Mix the salt, the charcoal and the flour.

2. Dissolve the yeast in a small quantity of lukewarm water.
Make fragments with the hemispheres of double cream.
3. Add the flour to the yeast.
On one part, put the chocolate sorbet.
On the other part, put the sparkling meringues. 4. Slowly, add the icy water while whipping.
Put the spicy cigars in liquid nitrogen, and add them to the 5. Add the fir tree extract.
dish just before serving. 6. Let the preparation rest for at least 3 hours.

Finishing
Crisp Fondue “Vacherin Fribourgeois” 1. Cut the vacherin Fribourgeois into cubes 2.5 × 2.5 cm.
I am a real fan of the Vacherin fondue, which is to be served 2. Dip the cubes in the frying batter.
lukewarm. It is often complex to keep it at the right temperature 3. Fry at a temperature of 200 ° C for 3 minutes.
during the whole meal. With this crisp fondue, two consistencies 4. As soon as the cubes are out of the bath, put them onto
are obtained; the inside can be at the right temperature (the one absorbent paper.
that I like). The extract of fir tree comes late during degustation, 5. Eat hot with the fingers: the outside is hot and crisp, and
the inside is melted and lukewarm.
How Do Eggs Coagulate?
Hervé This vo Kientza1,2
1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
In this chapter, I shall discuss both the issue of egg coagulation
and the production of “eggs at 6X °C” that I proposed as a gen-
eralization of what I initially called “perfect hard boiled eggs”
(This, 1995a; 1995b; 2001a; 2001b; 2007).
Uncooking Eggs
Let us start from the common experiment of putting a chicken
(Gallus gallus) egg in boiling water (100 °C) for 10 minutes.
When the shell is removed, the egg white is a white soft solid, and
the yolk has turned into a brittle yellow solid. This transformation
does not occur for all species; some eggs, such as sea turtle eggs,
do not coagulate in boiling water (Vernes, 1875), as was checked
at my request by Jean Lescure in 1994 in Guyana.
Why does solidification of chicken eggs occur? Some modern
cooks and culinary teachers persist in saying that the hardening
of eggs is due to “albumin” coagulation (Maincent-​Morel, 2015),
but this is not correct, because “albumin” is a loosely defined gen-
eric name for a category of proteins, especially proteins of blood
plasma that transport various substances (IUPAC, 1997), rather
than one simple “principle” that would make up albumen, or “egg
white”. Indeed, egg white, from chickens for example, contains
about 90% water and many different proteins (Tables 121.1 and
121.2).
The coagulation of the egg white is indeed due to the aggre-
gation of proteins after they have been denatured (Gossett et al.,
1984); the transparent yellow-green liquid, which is called the
egg white, becomes opaque and solid when heated. This “soft
TABLE 121.1
The Components of Chicken Eggs
Fraction % of the total mass % of the dry matter
Shell 10.3 98.4
Egg white 56.9 12.1
Yolk 32.8 51.3
Source: Li-​Chan and Nakay, 1989.
solid” is a gel because it includes 90% water, coagulation
occurring even when water loss is prevented during the solidifi-
cation. However, what exactly happens?
Various hypotheses have been proposed to explain the
heat coagulation of the egg white. In the 1920s, for example,
hydrolysis and dehydration were viewed as likely reasons for
the heat-​induced coagulation of the egg proteins (Wu, 1925;
Belitz and Grosch, 1987), but mass measurement of eggs being
heated does not show any variation, on the one hand, and, on the
other hand, hydrolysis alone would not allow the formation of a
gel. Moreover, in the 1990s, the kind of gel being formed was
discussed, between so-​called “chemical gels”, based on covalent
bonds, and “physical gels”, based on weak bonds. Silicone breast
implants are an example of popular chemical gels, whereas a gel-
atin dessert is a tasty specimen of a physical gel. Which kind of
gel is obtained by heating an egg white –​physical or chemical?
In 1996, in order to decide which kind of gel the cooked
egg white is, I decided to explore the various forces that could
make the egg white gel (This, 1996). These forces are, ranked
by increasing order of energy: van der Waals forces, hydrogen
bonds, hydrophobic (pseudo) forces, disulfide bridges, cova-
lent bonds, and electrostatic bonds (This, 2009). An educated
guess regarding the thermal coagulation of the egg white is that
hydrophobic forces or hydrogen bonds are not energetic enough
to hold the gel; also, the chemical nature of the side groups of
the amino acid residues makes covalent bond formation very
unlikely. In this regard, the coagulated egg could be viewed as a
hybrid between “physical gels” (reversible with temperature) and
“chemical gels” (permanent). Indeed, disulfide bonds were better
Proteins (%) Lipids (%) Saccharides (%) Minerals (%)
3.3 0 0 95.1
10.6 0.03 0.9 0.6
16.6 32.6 1.0 1.1
779
780 Hervé This vo Kientza

TABLE 121.2
Egg White Proteins
% of the dry
matter of
Proteins egg white Molar mass Isoelectric point Comments
Ovalbumin 1,2 58 45,000 4.6 4 free thiol groups, 1 disulfide bridge, 3.2% saccharides,
phosphoric acid linked to serine residues, about 400 amino
acid residues, of which 355 are hydrophobic.
3 ovalbumins: A1 (two phosphate groups), A2 (1 phosphate
group), A2 (0 phosphate groups).3
Conalbumin 13 80,000 (another value 6.6 (6.0) Glycoprotein that can bind 2 Fe3+ atoms (or Cu2+, or Zn2+,
(ovotransferrin) was also found: or Al3+) per molecule. One chain of 686 amino acid
77,000) residues. One glycan on Asn 473: 3 mannoses and 6 N-​
acetylglycosamine residues.
2 parts held together by 15 disulfide bridges.
Ovomucoid 11 (10) 28,000 3.9 (4.1) No tryptophan residue.
186 residues, 3 domains with 3 disulfide bridges per domain.
Saccharides: D-​glucose, D-​mannose, 2-​amino-​2-​deoxy-​D-​
glucose, sialic acid.
Lysozyme 3.5 14,600 (14,300) 10.7 129 residues, 4 disulfide bridges
(ovoglobulin G1)
Ovoglobulin G2 4 30,000/​45,000 5.5 3.2% D-​glucose, 2.4% hexosamine
Foaming
Ovoglobulin G3 4 45,000 5.8 3.7% D-​glucose, 2.5% hexosamine
Ovomucin 1.5 (3) 210,000 (10,000,000) 4.5 (5) High sugar content, microfibrils of 2–​10 nm diameter,
0.2–​0.3 nm long, 3 subunits
Flavoprotein 0.8 35,000 (29,200) 4.1 (4.2) 219 residues, 15% saccharides.
Ovoglycoprotein 0.5 24,000 3.9
Ovomacroglobulin or 0.5 760,000/​900,000 4.5/​4.7 3.5% hexose, 5.2% hexosamine
ovostatin
Ovoinhibitor 0.1 49,000 (44,000) 5.2
Avidin 0.05 67,000 (63,800) 9.5 (10) 128 residues, 1 disulfide bridge, 10% carbohydrates
Cystatin 12,700 2 disulfide bridges

Source: Thapon and Bourgeois, 1994.

1 Fothergill and Fothergill (1970)

2 Thompson and Fisher (1978)

3 Li-​Chan and Nakay (1989)

candidates, as their bond energy is about 75% of that of covalent
bonds, and they can form and disintegrate with simple oxidation/​
reduction processes. Moreover, cysteine residues are present in
many egg proteins.
Were disulfide bonds responsible for the gelation of the egg
white? One simple way of deciding was to add a specific disul-
fide bridge cutting agent to a cooked egg white, so that it could
be observed whether this compound could reverse the coagu-
lation process. As disulfide bridges can be formed by an oxi-
dation process between two thiol groups, a reducing agent was
used. This experiment was both simple and demonstrative: first,
cook one egg white, then add a small quantity of water (1 mL
for one coagulated egg) and a few grams of a reducing agent,
such as sodium borohydride (NaBH4). In seconds, the solid
egg white disintegrates and a foam is formed; and after several
hours, a liquid uncooked egg is obtained. This uncooked egg is
not a strict “proof” that disulfide bridges are responsible for the
coagulation, because any bond obtained by oxidation could be
credited for it, but other chemical considerations based on the
chemistry of proteins lead us to assert that sulphur is the key
(This, 1996)!
Using This Knowledge to Make New Eggs
This first issue of deciding which kind of gel the egg forms is
concluded to be in the middle of physics and chemistry, because
disulfide bridges are “weak covalent bonds”. However, after
this work, a question remained: why do chicken eggs become
rubbery after more than 10 minutes in boiling water (Société des
cuisiniers de Paris, 1930)? In Figure 121.1, time is not a factor,
which shows that the description was insufficient.
Indeed, the formation of chemical bonds is not a question of
time but of energy, i.e., of temperature. And because the scientific
method is first to refute theories, it was considered how the new
coagulation theory based on the previous study was wrong.
Egg white contains different kinds of proteins, and each has
a particular denaturation temperature (Table 121.3), based in
How Do Eggs Coagulate? 781

particular on its particular composition in terms of amino acid

residues. When eggs are heated in boiling water, the temperature
increases from the outside toward the inside by conduction, so that
different proteins coagulate at different times in different parts of
the egg white and of the yolk. Coagulation should occur rapidly
in the egg white, and heat is mostly transferred by conduction.
How to check this theoretical description? In 1994, I predicted
that the coagulation of eggs heated at various temperatures would,
rather, follow a different scheme (Figure 121.2).
In order to test this idea, an egg was heated to 65 °C, and the
temperature in the yolk was followed using a K thermocouple
inserted using a syringe needle. The temperature profile is given
in Figure 121.3.
Because of the particular denaturation temperatures, it was
predicted that a first gel would be formed from ovotransferrin,
and another one would appear at 70 °C with ovomucoid, etc., and
so on with increasing temperatures (Table 121.3).
This table explains the important difference in consistency
between an egg heated until thermal equilibrium (about 1 h)
at 69 °C and at 70 °C (Figure 121.4): the second coagulation
generates a very different result, as a second network is created
inside the first (it is possible that the two protein networks are
independent, as the thiol groups from ovotransferrin should all be
engaged in disulfide bridges when ovomucoid denatures).
For the yolk, the description is slightly more complex because
of the high viscosity of the product, and probably also because
the quantity of livetin gamma is under the percolation threshold,
i.e., it is not enough to establish a continuous network in the
liquid phase, making a gel, as is well described by the theory
of percolation (Djabourov et al., 1988). The theoretical pre-
diction was confirmed; with increasing temperature, more and
more protein networks are formed, and the consistency becomes
harder and more opaque. This is the basis of the production of
what I proposed under the name of “perfect eggs” in 1995 (This,
1995a; This, 2002). Indeed, a preferred name was proposed
slightly later as “eggs at 6X °C”, because this could describe the
FIGURE 121.1 A simplistic theory of egg white coagulation: the denatured
(uncoiled) proteins (pink) link and make a network in which water molecules
whole category, X being equal to 1, 2 … n.
are trapped.

FIGURE 121.2 Considering that egg white (left) is a mixture of different proteins with various denaturation temperatures, one can understand that thermally
processing the egg white at a temperature between 61.8 °C and 70 °C would coagulate only one kind of proteins, making a soft gel (middle), in which the water
molecules and the other non-​denatured proteins would be trapped. Heating more would create a second network added to the first, making the mass firmer, and
so on, with more and more different proteins coagulated.
782 Hervé This vo Kientza

70
60
50

T°C
40
30
20
10
0 5 10 15 20 25 30 35 40
Time in minutes

FIGURE 121.3 Changes in temperature in the centre of the yolk for an egg (mass 58 g) put in a convection oven with temperature set at 65 °C.

TABLE 121.3
Denaturation Temperatures for the Protein of Gallus gallus Eggs
Protein Denaturation temperature (°C)
Egg white
Ovotransferrin 61
Ovomucoid 70
Lysozyme 75
Ovalbumin 84.5
Globulins 92.5
Yolk
LDL 70
HDL 72
Livetin alpha 70
Livetin beta 80
Livetin gamma 62
Phosvitin More than 140
Whole yolk 65–​70 (because of LDL)

Source: Li-​Chan and Nakay (1989) and Thapon and Bourgeois (1994).

HDL: high-​density lipoprotein; LDL: low-​density lipoprotein.

REFERENCES
Belitz HD, Grosch W. 1987. Food Chemistry. Springer Verlag,
Berlin, Heidelberg, 513.
Djabourov M, Leblond J, Papon P. 1988. Gelation of aqueous gel-
atine solutions. II. Rheology of the sol-​gel transition. Journal
de Physique, 49, 333–​343.
Fothergill JA, Fothergill JE. 1970. Thiol and disulphide contents of
hen ovalbumin: C-​terminal sequence and location of disul-
phide bond, Biochemical Journal, 116, 555–​561.
Gossett PW, Rizvi SSH, Baker RC. 1984. Quantitative analysis of
gelation in egg protein systems, Food Technology, 38, 67–​96.
FIGURE 121.4 (a) Egg cooked at 65 °C for many hours (here eight, but
IUPAC. 1997. Compendium of Chemical Terminology, 2nd ed (the
the result is no different with a different time, as long as water does not evap-
“Gold Book”). Blackwell, Oxford.
orate through the pores of the shell). Only one protein (ovotransferrin) has
Li-​Chan E, Nakay S. 1989. Biochemical basis for the properties of
coagulated in the white part. (b) Egg cooked at 71 °C. The difference in the
egg white, Critical Reviews in Poultry Biology, 2(1), 21–​58.
egg white is due to ovomucoid coagulation.
Maincent-​Morel M. 2015. La cuisine de référence. BPI, Paris.
Société des cuisiniers de Paris. 1930. La cuisine à l’usage des
familles (le livre de la profession). Eyrolles, Paris.
This H. 1995b. Le fin mot de l’oeuf dur, Pour la Science, 211, 20.
Thapon JL, Bourgeois M. 1994. L’oeuf et les ovoproduits. Tec et Doc
This H. 1996. Can a cooked egg white be uncooked? The Chemical
Lavoisier, Paris.
Intelligencer, 10, 51.
This H. 1995a. Public Lecture at the Cité des sciences et de l’industrie,
This H. 2001a. L’oeuf à 64 °C, Pour la Science, 290, 4.
Paris, 17 December.
This H. 2001b. L’oeuf dur parfait, Thuriès Magazine, 131, 85–​87.
How Do Eggs Coagulate?
This H. 2002. Molecular Gastronomy, a Scientific Look on Cooking.
Columbia University Press, New York.
This H. 2007. Let’s have an egg. In Davidson A (ed.) Eggs in
Cookery, Proceedings of the Oxford Food Symposium on Food
and Cookery 2006, Prospect Books, Totnes, 250–​258.
This H. 2009. Molecular Gastronomy, a chemical look to cooking,
Accounts of Chemical Research, 42(5), 575–​583.
783
Thompson EOF, Fisher WK. 1978. Amino acid sequence containing
half cystein residues in ovalbumin, Australian Journal of
Biological Science, 31, 433–​442.
Vernes J. 1875. L’île mystérieuse. Hetzel, Paris.
Wu H, Wu DY. 1925. Nature of heat denaturation of proteins, Journal
of Biological Chemistry, 64, 369–​378.
Vegetable Salad

Jean Chauvel
Restaurant Jean Chauvel, Paris, France

FIGURE 122.1 Vegetable salad.

(Photo ©Alban Couturier)

785
786 Jean Chauvel

Ingredients For the Vegetables

For the Broth
Syrup (1 L water + 300 g sugar)
Alcohols (a mix of peach cream, Grand Marnier,
Limoncello, Manzana, … to taste)
Curcuma
Select a collection of vegetables of your choice (samphire,
Brussels sprouts, nasturtium flower, spinach, broccolini,
asparagus, green pea, carrot, fennel, rhubarb, celery root, celery
stick, aloe vera, yacon, …) and wash/​cut to desired shape.

FIGURE 122.2 Mise en place.

(Photo ©Alban Couturier)

FIGURE 122.3 Plating.

(Photo ©Alban Couturier)
Vegetable Salad 787

For the Dressing

Kalamata black olives 50 g
Tomatoes 100 g
Olive oil
Xeres vinegar
Soy sauce
Salt
Pepper
Smoked paprika

Methods
For the Broth
Bring all the ingredients together to the boil and let the mixture
cool down.

For the Vegetables

Cut into small pieces and leave for 45 min at 45 °C in the broth
under vacuum in the Gastrovac®.

For the Dressing

Olive purée: mix olives with olive oil, salt and pepper.
Tomato gel: mix tomatoes, vinegar, soy sauce and salt;
squeeze through a thin cloth to extract a clear juice.
To serve: make a nice arrangement of the vegetables on a
plate; add small dots of olive sauce, squid ink and tomato
FIGURE 122.4 Gastrovac system (which enables impregnation of the
gel; sprinkle with some smoked paprika (Figures 122.1
vegetables).
to 122.4).
(Photo ©Alban Couturier)
Filtration

Hervé This vo Kientza1,2

1 Université Paris-​Saclay, INRAE, AgroParisTech, UMR 0782 SayFood, 75005, Paris, France
2 Group of Molecular Gastronomy, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,

F-​75005, Paris, France

After 1980, efforts were made to modernize culinary techniques

by transferring laboratory hardware into kitchens: this was called
“molecular cooking” in 1999. Among the techniques that could
benefit from modern equipment, filtration seemed important.
However, after almost four decades of promotion of modern fil-
tration tools, a failure is to be observed in this regard; most cooks
still filter and clarify as in the past. More effort is needed, and
perhaps the development of systems specifically designed for
FIGURE 123.1 Proline.
cooks is the key to success.
Filtration is a common technique of separation when the
goal is to separate a solid from a solution. In the kitchen, it was
used as early as the 14th century; in the Viandier (Tirel, 1319),
the process is discussed many times, using a fine cloth called
“estamine”. In particular, the practice of filtering meat stocks
in this way is already present, as it is in the Menagier de Paris,
published slightly later (Anonymous, 1393).
However, because such filtration techniques were not very pre-
cise, leaving particles in suspension, clarification techniques have
been performed before filtration, in particular for the production
of “consommés”, which are fortified and clarified meat stocks
(Carême, 1828). For such systems, whipped egg whites (or other
food ingredients such as ground meat or vegetables) are added to
the cold liquid, and the whole mass is filtered after coagulation
of egg proteins, trapping part of the suspended particles, as well
as other compounds, from the solution. This practice is similar
to the fining of wines, for which it has been shown that many
compounds, such as polyphenols (Ricardo-​da-​Siva et al., 1991)
FIGURE 123.2 Here, segments of leek (Allium ampeloprasum) stems are
and odorant compounds (Voilley et al., 1990), are removed because
lifted in water by visible steam bubbles.
of their adsorption on proteins. For polyphenols, the presence of
hydroxyl groups explains the adsorption: high affinity is observed
with amino acid residues such as proline (Figure 123.1), an imino density can be artificially reduced using gas bubbles that attach to
acid containing a carbonyl oxygen adjacent to a secondary amine the particles (Marchal et al., 2003).
nitrogen, which is a very good hydrogen bond acceptor. In the kitchen, this process is part of “clarification”, often
For odorant compounds, hydrophobic interaction with the because of steam bubbles produced at the bottom of the pan that go
hydrophobic part of proteins can occur, but more specific binding upward and aggregate on solids, driving them toward the surface
can also come into play, depending on the particular molecular (such steam bubbles are responsible for the flotation of gnocchis
structure of the odorants. However, when the liquid is heated and other culinary materials, as seen in Figure 123.2) (This, 2006).
with the egg white, flotation occurs as well; flotation is a solid–​ This traditional technique of “clarification” is still used, often
liquid separation process used when the density of the suspended using a sieve called a “chinois”, with a clean cloth folded in four;
particulates is lower than that of the liquid containing them. This the liquid that drains by gravity is clearer.

789
790 Hervé This vo Kientza

Another traditional technique is by pressurization; a suspen-

sion is put into a cloth, the ends of which are twisted in order to
compress the mass inside it and release the liquid. Of course, the
size of the particles that can flow with the liquid depends on the
particular nature of the material being used.

Modern Techniques
In laboratories, there is much more diversity for filtration systems,
adapted to the specific technical question to be tackled. Some
can be performed at atmospheric pressure, or under reduced
pressure, or in contrast, under high pressure. The liquids can be
hot or cold, and all sizes of pores can exist, from macroscopic to
molecular size.
In chemistry laboratories, filtration through special filtration
paper has been used, just as such papers are used for making
coffee, for example. Disks of special materials are folded in
order to fit the inside of a funnel. However, in laboratories, unlike
kitchens, the quality of the papers and the filtration grade (diam-
eter of pores) can be selected; special materials can even be
chosen for particular applications (hydrophilic or hydrophobic, FIGURE 123.3 A Büchner funnel. A disk of filter paper is to be put over the
for example) (Roth, 2018). grid before the liquid is sucked down using a pump.
In such systems, the liquid is simply poured into the funnel
in which the filter is placed. The liquid drains by gravity, and
the solid is kept in the paper. It should be noted that some tiny
particles can go through the filter; often, these are so small that
they sometimes remain unseen with the unaided eye, but they
sediment after some days anyway or are recovered by centrifu-
gation. A similar process can be implemented with cotton at the
entrance of the right part of the funnel or at the tip of a pipette.
This kind of filtration system is fast, but there is no control on
the size of the particles.
For faster filtration, sucking air from the vessel to which the
funnel is adapted can be an option.
A popular system in chemistry laboratories includes the use of
Büchner funnels (Figure 123.3), i.e., ceramic funnels with a flat
surface on which the filtration paper is placed. With pumps for
sucking air and reducing the pressure, fritted glass funnels can
alternatively be sucked (Figure 123.4).
These are more expensive, but the porosity is much better con-
trolled. The size of pores is given by a number between 0 and 5
for smaller and smaller pores (Table 123.1).
For the use of such systems in the kitchen, great care has to
be put into cleaning the fritted glass, and this was considered a
drawback when the technique was discussed with chefs. On the
other hand, this technique was compared for the clarification
FIGURE 123.4 A fritted glass funnel over the filtration system.
of cooked tomatoes and the preparation of a “tomato water”,
and a (visually) clearer result was obtained in a much shorter
time using a fritted glass (This and Westerman, 2004). When
there is a possibility of clotting, increased pressure systems Future Developments?
such as for making expressos can be used; in this case, clean These “traditional laboratory methods” are a tiny fraction of all
silica (boiled sand, for example) can be put in the filter of the the existing systems that have been designed in recent times in
machine. This has the advantage that the filtration process is science and technology. First, there are throwaway or reusable
accelerated, and also, the sand can be reused, after washing filters, filters for continuous or batch processes, and a range
and possibly burning (in order to avoid the accumulation of of pore sizes that can be chosen. For example, some systems
organic matter). include a steel frame, sometimes with large sizes, as needed in
Filtration
TABLE 123.1
The Various Grades of Filtration Systems
Porosity Size of pores (μm) Application
0 160–​250 Large precipitates
1 100–​160
2 40–​100 Crystalline precipitates
3 16–​40
4 10–​16 Very fine precipitates
5 1–​16 Bacterial filtration
the kitchen, where the filtration is performed by multiple layers
of a honeycomb-​type aluminum medium. Often, a cascade of fil-
tration systems is used instead of a unique filtration system. All
this is easy to see in any laboratory hardware catalog (Thermo
Fisher Scientific, 2018).
Recently, systems have been introduced in the food industry,
such as micro-​and nanofiltration, in order to produce microbio-
logically stable milk without heat treatment or to separate milk
components (Daufin et al., 2001). Such systems can be very
simple, including a pump and a filtering cartridge (This, 2012). In
particular, tangential flow filtration solves the issue of the clotting
of the filters. For example, ultrafiltration (UF) has progressed
from methods as diverse as chemical precipitation for sample
concentration or dialysis for buffer exchange toward pressure-​
driven cross-​flow purification systems based on UF membranes.
Such membranes are polymeric, with highly defined pore sizes to
separate molecules according to size. Simply put, UF procedures
rely on the use of fluid pressure to drive the migration of the
smaller molecules through a UF membrane with the simultan-
eous retention of larger molecules.
Separation with UF is based on mechanical rather than chem-
ical interactions, allowing concentration without the addition
of denaturing solvents or salts. Buffer exchange using dialysis
techniques uses large volumes of buffer, and since the only force
acting upon the solution is diffusion, the process can take several
days; in contrast, UF devices can rapidly perform either concen-
tration or buffer exchange procedures without the extensive hand-
ling required for many other techniques.
UF can be performed in one of two operational modes: direct
flow filtration (DFF) or tangential flow filtration. DFF works
well for small volumes (up to 30 mL) using centrifugal devices,
but membrane fouling can occur. To reduce the formation of a
gel layer, cross-​flow can be generated on the upstream side of
the membrane using a floating stir bar configuration (stirred
791
cell) or by creating a controlled laminar flow. While stirred cell
operations tend to improve the performance of UF, they are still
limited in achieving optimal performance since the velocity and
subsequent level of agitation are dependent on the sweep of the
bar, which varies along the radius of the sweep.
Today, chefs very seldom rely on such techniques, but much
progress can be made in order to move to more sustainable as
well as more precise techniques. It is a pity that the culinary
art did not benefit from the advances of membrane techniques,
both for health reasons (one should remember that in 2018 in the
United States, half of foodborne illnesses were due to insufficient
hygiene in restaurants; FDA, 2018) and for the improvement of
culinary preparations. In particular, membrane filtration could be
useful for sauces.
REFERENCES
Anonymous. 1393. Le ménagier de Paris, Chavane,Paris.
Carême MA. 1828. L’art de la cuisine française au XIXe siècle,
Bossange père, Paris.
Daufin G, Escudier JP, Carrère H, Bérot S, Fillaudeau L, Decloux
M. 2001. Recent and emerging applications of membrane
processes in the food and dairy industry, Trans. IchemE, 79,
C, 89–​102.
FDA. 2018. FDA report on the occurrence of foodborne illness risk
factors in fast foods and full service restaurants 2013–​2014.
Gésan-​Guiziou G, Van Audenhaege M, Omont S, Froelich D. 2011.
Séparations à membrane et fonctionnalités ciblées de fractions
protéines: vers une approche d’éco-​conception, Innovations
agronomiques, 13, 101–​115.
Marchal R, Lallement A, Jeandet P, Establet G. 2003. Clarification of
Muscat musts using wheat proteins and the flotation technique,
J. Agric. Food Chem., 51, 2040–​2048.
Ricardo-​da-​Siva JM, Cheyniere V, Souquet JM, Moutounet M,
Cabanis JC, Bourzeix M. 1991. Interaction of grape seed
procyanidins with various proteins in relation to wine fining,
J. Sci. Food Agric., 57, 111–​125.
Roth. 2018. Matériel de laboratoire, Roth Sochiel, www.carlroth.fr
Thermo Fisher Scientific. 2018. www.fishersci.fr/​fr/​fr/​home.html
This H, Westerman A. 2004. Public experiment at the Strasbourg
European Fair, Strasbourg, France.
This H. 2006. Molecular Gastronomy: Exploring the Science of
Flavor. Columbia University Press, New York.
This H. 2012. Cours de gastronomie moléculaire: la cuisine
note à note, www2.agroparistech.fr/​ podcast/​Gastronomie-
Moleculaire-2012-​partie-​3.html
Tirel G. 1319. Le Viandier, Techener, Paris, 1892, 248–​251.
Voilley A, Lamer C, Dubois P, Feuillat M. 1990. Influence of
macromolecules and treatments on the behavior of aroma
compounds in a model wine, Journal of Agricultural and Food
Chemistry, 38, 248–​251.
Waiter! There Is Garlic in My Meringue!

César Vega
Barry Callebaut Americas, 600 W Chicago Avenue, Chicago, IL 60654, United States

Meringues are the backbone of a range of both sweet and savory

dishes, such as soufflés, macarons, angel food cake and dried
shrimp patties, also known in Mexico as “tortas de camarón”, and
are responsible for much of the structure stability of these foods
(Vega & Sangvhi, 2012). Usually, meringues are made of egg
white (fresh and/​or re-​hydrated), sugar and in some cases, an acid
source such as lemon juice, vinegar or cream of tartar (McGee,
2004). These ingredients are then beaten to incorporate air while
concomitantly unfolding the protein structures in the egg white
to provide the required foam stabilization. For a thorough review
of the culinary, physical and chemical principles involved in mer-
ingue making and stabilization, see Vega & Sangvhi (2012).
Care must be taken during whipping, as it is very easy to
overdo it, which is manifested by the formation of evident tiny
“lumps” that forecast the imminent collapse of the foam (see
Figure 124.1); if this point is reached, there is no choice but to
start over with new egg whites. It is well known, however, that the
incorporation of an acid (best practice is about 1.5% w/​w) signifi-
cantly increases the stability of meringues against over-​whipping,
because, according to McGee (2004), the decrease in pH delays/​
inhibits the formation of disulfide bonds among ovalbumin
molecules (Figure 124.2). FIGURE 124.1 Close up of an overbeaten meringue showing lump forma-
tion, which predicts the collapse of the foam.
The addition of these acids decreases the pH of the whipping
mass from 9 to about 7 (Kamozawa & Talbot, 2010), which has
a few consequences: (a) it brings the pH closer to the isoelec- R R
tric point of the egg white proteins and reduces the electrostatic
SH oxidation S
repulsion between them; (b) proteins pack more closely at the air–​ + 2H+ + 2e–
water interface, promoting stronger films, which creates a more pH > 8.5 S
SH
stable foam (Rodriguez-​ Patino, Naranjo-​ Delgado & Linares-​
R
Fernandez, 1995; Oldham, McComber & Cox, 2000); and (c) the R
lower pH inhibits the oxidation of two sulfhydryl (S-​H) groups
and hence the formation of disulfide bonds (S-​S) between adja- FIGURE 124.2 Schematic showing that the oxidation of two sulfhydryl
cent ovalbumin molecules. groups into a disulfide bond is favored above pH 8.5, which is the case for
However, as already explored by Vega & Sanghvi (2012), the native egg white.
physical chemistry of egg white foams is rather complex, and it
is dangerous to settle for this explanation without systematically (NEM). Eight years later, it was up to me to bring closure to this
testing its validity. If, in fact, the key factor is the inhibition of question in a novel culinary manner.
S-​S formation, then this should be tested at the natural pH of egg Let’s first state the hypothesis:
whites (pH ≈ 9) to remove any parallel effect related to the pH
drop and the change in packing efficiency of the proteins. Vega Whipping egg whites at their natural pH of 9 in the presence
& Sanghvi (2012) proposed starting by whipping egg whites in of a thiol-​blocking compound should render foams as stable as
the presence of a thiol-​blocking group such as N-​ethylmaleimide those in the presence of acid at pH 7.

793
794
TABLE 124.1
Parameters for Preparation of Garlic Meringues
Control
2:1 water: egg white mix (i.e., 3.5% 200 g
egg white protein)
% active (w/​w) NA
Whipping conditions Speed 8 of Kitchen Aid for 5 min
pH 9.0
% overrun ~600
Drain in 30 min (g) 85.4
As % 43
Lo and behold, whipping a 2:1 water:egg white mixture at
pH 9 in the presence of 0.03% w/​w NEM resulted in a foam
of similar overrun (air inclusion) but significantly higher sta-
bility (measured as the amount of water drained from the foam)
compared with either water and egg white alone or the same with
0.5% added cream of tartar (Table 124.1). Not only that, but the
appearance of the NEM foam resembled that of shaving cream:
dense, silky and solid-​like. I couldn’t actually touch it or eat it,
as NEM is toxic – a little detail I forgot to mention! Due to a var-
iety of constraints, the experiment could only be run once, but
the results are considered unequivocal: S-​S inhibition is the key
to egg white foam stability!
Now, I needed to repeat the experiment, but this time, I had
to add something that was food grade. It took a while for the
culinary muses to pay me a visit.
One of my most vivid childhood memories is that of my mother
sauteéing onions. The aromas coming out of the pan are, still
today, tantalizing. Many years later, garlic, whether raw –​like
in aioli –​or roasted (spread onto bread), became one of my most
sought-​after flavors. It wouldn’t then surprise you that when the
book Garlic and Other Alliums: The Lore and the Science crossed
my path, I was equally tantalized. I got it and started reading it
in earnest. I was enjoying it, and then I read this (Block, 2010):
Cavallito goes on to speculate about the basis for the antibac-
terial properties of garlic: “The sulfhydryl group is postulated
to be a specific stimulator of cell multiplication. Since allicin
is considerably more bacteriostatic than bactericide in action,
it may operate by destroying –​SH groups essential to bacterial
proliferation, thus inhibiting growth”…allicin could react
readily with almost any –​SH group with which it comes in
contact.
I jumped off my sofa! The end to my quest was in sight! Garlic
contains an edible thiol-​blocking agent! I ran to the nearest super-
market and bought 2 kilos of garlic. Back home, I asked my wife,
Elizabeth, to help me peel garlic, as we were on a mission: put-
ting garlic to the test!
Garlic, onion and leek are all members of the genus Allium,
which is said to come from the Greek αλεω, “to avoid”,
because of its offensive smell (Boswell, 1883). Garlic (Allium
César Vega
Treatment
Cream of tartar NEM
200 g 200 g
0.5 0.03
7.0 9.0
~650 ~625
67.0 22.0
33.5 11
FIGURE 124.3 Chemical structure of allicin.
sativum) originated in the very tough climates of what is today
the Uzbekistan neighborhood and hence evolved some serious
chemical weapons to defend itself. Different alliums accumu-
late different sulfur compounds, which accounts for their varying
flavors. The stockpiles themselves are inert, but when the plant’s
tissues are damaged, enzymes therein convert them into reactive,
stinging molecules (Block, 2010). When garlic is crushed, alliin,
a non-​protein amino acid, and alliinase, an enzyme, interact to
produce allicin –​our main suspect (Figure 124.3).
I proceeded to make some garlic juice with the help of a centri-
fugal juicer (not the most efficient way, I must tell you) and went
on to whip, under similar conditions, the following:
200 g of egg white
200 g of egg white + 1.5% cream of tartar
200 g of egg white + 0.5% garlic juice
The results, shown in Figure 124.4, speak for themselves.
Meringues with added garlic juice (at pH 9) showed better sta-
bility against drainage than those with added cream of tartar (at
pH 7), which, as of yesterday, is no longer best practice. Just as in
the case of NEM, the foam made with garlic looks denser, silkier
and more solid-​like (as perceived by passing a spoon through it).
Confirming that allicin is indeed the chemical entity respon-
sible for the behavior observed deserves further investigation.
Knowing that allicin is only formed when garlic is cut, I cooked
peeled and intact garlic cloves for 20 min in boiling water to
inactivate allinase. I then tried to make garlic juice out of softened
cloves (very, very difficult indeed!) and repeated the experiment.
The foam drained rather early and copiously (not shown).
For those of you who, like many of the chefs I have shared this
with, wonder if the same takes place with onions, leeks or chives,
please take a deep breath and put your knives down. Acting on
that impulse is what I call the “cook and look” approach and must
be avoided at all costs. Recall that we arrived at using garlic to
Waiter! There Is Garlic in My Meringue! 795

FIGURE 124.4 Drainage (g of liquid expelled from a foam) as a function of time for three different foams: □ egg white; Δ egg white with 1.5% w/​w cream
of tartar added; ○ egg white with 0.5% raw garlic juice added. Foams were made with 200 g of egg white and whipped for 5 min at speed 8 using a Kitchen
Aid mixer.

stabilize meringues by virtue of a stated hypothesis and system-
atic enquiry. That’s a recipe for success.
Macarons, soufflés or even angel food cake can immediately
benefit from a garlicky twist. I can’t wait to get cooking! Can
you? After all, the proof is in the “pudding”.
REFERENCES
Block, E. 2010. Garlic and Other Alliums: The Lore and the Science.
Royal Society of Chemistry, Cambridge.
Boswell, J.J. ed. 1883. English Botany. George Bell & Sons,
London.
Kamozawa, A. and Talbot, A. 2010. Ideas in Food: Great Recipes
and Why They Work. Clarkson Potter, New York.
McGee, H. 2004. On Food and Cooking –​The Science and Lore of
the Kitchen. 2nd Edition. Scribner, New York.
Oldham, A.M., McComber, D.R., and Cox, D.F. 2000. Effect of
cream of tartar level and egg white temperature on angel food
cake quality. Fam. Consum. Sci. Res. J. 29:111–​124.
Rodriguez Patino, J.M., Naranjo Delgado, M.D., and Linares
Fernandez, J. 1995. Stability and mechanical strength of
aqueous foams containing food proteins. Colloids Surf.,
A. Physicochem. Eng. Asp. 99:65–​78.
Vega, C. and Sanghvi, A. 2012 Cooking literacy: Meringues as
culinary scaffoldings. Food Biophys. 7:103–​113.
Lobster and Juniper

David Toutain
Restaurant David Toutain, 23 rue Surcouf, Paris, France

This recipe is based on a noble product, which we are lucky to red, and dip it immediately in ice for 20 minutes before
be able to cook: no part of it should be omitted. When I designed breaking the shell and recovering the flesh.
this dish, I was fortunate to meet botanists with whom I could col- Divide the flesh into segments and poach for 5 minutes at a
laborate. For example, Fabrice Gabriel has often proposed wild temperature of 45 °C.
plants with remarkable flavour and changing with the seasons. Cook the lobster over hot coals and store. Just before
And a juniper vinegar gave me ideas for flavour associations. In serving, reheat in brown butter.
particular, the wild juniper that he brought us from Alsace had a
vegetal bitterness that seemed to me well paired with that of grape-
fruit; the acidity of this citrus also seemed convenient to equilibrate Juniper Gel
the dish (indeed, lobster, like langoustines, has some “sweetness”).
270 g sugar (sucrose)
In order to create a joyful dish, we decided to treat the various
30 g juniper berries
parts of the lobster differently, associating each with a different 5 g corn starch
juniper preparation. 1 g agar-​agar
Technically, the way we cook the lobster is key. Of course, the 240 g white vinegar
quality of the crustacean contributes to the success of the recipe,
but the particular consistency of the tail flesh is made possible Process:
thanks to the blowtorch (it could also be obtained using a deep
freeze, but with a different result). Then, the siphon is important Mix the berries with sugar, corn starch and agar-​agar, and
for the sabayon, not only because it makes the preparation much bring to the boil while stirring.
easier and faster, but also because the consistency that can be Keep the mixture in the refrigerator for 1 hour.
achieved is quite different from that of sabayon of the older When the gel is set, make a “debye” by mixing it with the
times. And finally, making “debyes”, i.e., grinding gels in liquids vinegar.
(aqueous solutions, like here, or “oils”), is important because of
the special soft consistency that can be achieved.
For the guests, there is fun with this dish, because all parts of
Grapefruit Caramel
the lobster are dissociated, and it provides more pleasure than
having a simple plain lobster. At the restaurant, the guests are 1 kg of grapefruit pulp
invited to use their fingers (and this is why part of the lobster is 250 g cream
200 g white vinegar
prepared on a juniper branch). The lobster claw comes second,
5 g agar-​agar
with more sourness and bitterness, after the sweetness of the
100 g lemon juice
body. And finally, the sabayon is comforting.
350 g sugar (sucrose)

Process:
Lobster
1 lobster (800 g) Make a caramel with lemon juice and sugar.
250 g brown butter Deglaze with the cream and the vinegar.
Mix this caramel with the grapefruit pulp and agar-​agar.
Heat to 80 °C, then store in the fridge for 1 hour.
Process:
When set, mix with the lemon juice to make another debye.
Kill the lobster by inserting a sharp knife through the head.
Then heat the shell with the blowtorch, so that it turns

797
798 David Toutain

FIGURE 125.1 The “Lobster and juniper” dish served in our restaurant in Paris.

Lobster Sabayon Juniper Oil

120 g lobster bisque made using the head 950 g neutral oil (sunflower)
75 g water (for the bisque) 500 g juniper berries
4 g salt
163 g egg yolks Process:
250 g butter
Mix all the ingredients together, and filter with a sieve.
Process:

Make a bisque from the lobster head and water.

Make a sabayon with the egg yolks, the water, salt and Juniper Cream
butter. 750 g cream
Add the lobster bisque to the sabayon. 30 g juniper berries
Put the mixture in a siphon. 20 g white vinegar
Adjust two N2O cartridges one after the other. 9 g salt
Store at 60 °C.
At the last minute, siphon it into cups.
Lobster and Juniper
Process:
Macerate the berries in cream for 30 minutes.
Mix all the other ingredients with this cream.
Filter.
For the final touch, serve the various parts in three different
bowls (Figure 125.1), the medallion on a juniper branch,
799
the claw along with juniper sauce and juniper oil, the
grapefruit caramel and the juniper debye, and finally,
the head in sabayon.
The guests are invited to use their fingers to taste the three
parts of the dish in this order.
Molecular Cooking

Róisín M. Burke and Pauline Danaher

School of Culinary Arts and Food Technology, City campus, Technological University Dublin, Ireland

In January 2003, the INICON project, which was funded by the decanting bulbs, rotary evaporators and eventually, ultrasonic
European Union (EU), commenced under the Innovation and probes. Table 126.1 lists examples of such tools, and detailed
SME 5th EU-​Research Framework. The focus of the project was descriptions are provided by authors such as Blumenthal (2009).
to introduce innovative technologies in cooking. As described Myhrvold et al. (2011b) and Potter (2015). Catalogues from com-
by TTZ-​Bremerhaven (2005), there was a technology transfer panies such as MSK (2019) and Sosa (2018), which specialize
from the Molecular Gastronomy Group INRA (France) and TTZ in providing equipment for molecular cooking, give details of a
Bremerhaven (Germany) toward four chefs from El Bulli (Spain), broad array of equipment and accessories.
the Fat Duck (UK), Grashoff (Germany) and Au Crocodile
(France), and the culinary school ESCF Ferrandi (France),
working together with food industry representatives from Alpha-​
Ingredients
Tec (plant manufacturer, Germany), Cosmos Aromática (flavour
manufacturer, Spain) and Iberagar (natural hydrocolloid manu- The number and type of speciality ingredients associated with
facturer, Portugal). molecular cooking have steadily increased since 2003, and
Three years after the project finished, molecular cooking was there have been numerous publications about their chemical and
defined by This (2008) as a culinary technique using ‘new’ tools, physical properties and their uses, such as those by Blumenthal
ingredients and methods (mainly from chemistry and physics (2009), Lersch (2014), Myhrvold et al., (2011c), Sanchez (2016)
laboratories), whereas molecular gastronomy is a scientific dis- and This (2014). Company catalogues by ingredient suppliers
cipline, which is part of food sciences. Subsequently, Cassi (2011) include those such as Iqemusu (2017), Louis François (2019),
proposed defining ‘molecular cuisine’ as a type of cuisine that MSK (2019), Texturas (2012) and Sosa (2019).
arises from collaborations between chefs and scientists. The term Common ingredient classes include hydrocolloids, which can
‘molecular cuisine’ has sometimes been criticized, but the reason be gelling agents or gums, acids, salts, chelating agents, colours,
for using it was that innovative cuisine had to be distinguished enzymes, emulsifiers and stabilizers, flavourings, foaming
from science, and in particular from Molecular Gastronomy
(This, 2013).
Early adopters of molecular cooking were the chefs Raymond
Blanc (Oxford, UK), Christian and Philippe Conticini (Paris), TABLE 126.1
Shirley Corriher (USA), Fritz Blanck (USA), Elizabeth Thomas Tools That Can Be Found in Modernist Restaurant Kitchens
(USA), Ferran Adria (Spain), Heston Blumenthal (UK), Pierre General Specific Measurement
Gagnaire (France), Andoni Luis Aduriz (Spain) and Denis Driers Aerator –​fish tank bubbler pH meter
Martin (Switzerland) (Myhrvold et al., 2011a; Myhrvold and Dehydrator Anti-​griddle Moisture analyser
This, 2018). Freeze-​drier Caviar box for spherification Refractometer
Spray-​drier Gastrovac Rheometer
Vacuum oven Kits for making macaroni and Mini weighing scales
Mixers spaghetti
Tools Hand-​blender Multi-​aromatizer
Pacojet Rotary evaporator
In many cases, new tools are designed for specific applications Thermomix Siphons –​nitrous oxide or
in the kitchen, e.g., caviar boxes for spherification. Other tools, Colloid mill carbon dioxide
such as a rotary evaporator, are normally found in a laboratory, Homogenizer Smoke gun
but it is now not uncommon to see them being used in a res- Separators Sous vide circulating water
Centrifuge bath, immersion circulator
taurant kitchen. Historically, the order of appearance in kitchens
Filtration unit
was first thermocirculators, followed by siphons, liquid nitrogen,

801
802
Colour
Taste
Flavour,
Odour
& Aroma
FIGURE 126.1 Examples of ingredients that are used in molecular cooking.
agents, odorants, sweeteners and ingredients to provide acoustic
sensations. Figure 126.1 provides examples of ingredients that
can be used to boost or enhance certain sensorial sensations,
while of course, combinations allow heightened multi-​sensory
experiences.
There are numerous recipes available online or in books
(Adria et al., 2008; Blumenthal, 2009; Myhrvold et al., 2011b;
MolecularR, 2019a) that incorporate these ingredients. Often,
these are accompanied by YouTube clips that guide the chef
through the methods step by step.
Famous recipes incorporating some of the ingredients in
Figure 126.1 include Spherical-​ I green olives (Adria et al.,
2008) and Hot and Iced Tea (Blumenthal, 2009). In the former,
Ferran Adria used sodium alginate, xanthan gum and calcium
salt to create spherical olives from a green olive juice (Adria
et al., 2008). In the latter, Heston Blumenthal used gellan gum,
sodium citrate, calcium chloride and malic acid to create a
tea that was hot on one side of the glass and cold on the other
(Blumenthal, 2009).
Róisín M. Burke, Pauline Danaher
Texture &
Consistency
Sound
Methods and Techniques
Centrifugation
Centrifuges separate mixtures based on differences in density; a
centrifuge spins tubes of equal weight around a central axis at
high rates of speed, causing the heavier elements in the liquid
to collect on the outer walls and bottom of the tubes, while the
lighter particles remain separated at the top (Sanchez, 2016).
When tomato juice was centrifuged, Blumenthal (2009) observed
that it resulted in three distinct components, as illustrated in
Figure 126.2.
Cryo-​Techniques Using Dry Ice and Liquid Nitrogen
Dry ice (frozen carbon dioxide) and liquid nitrogen have been
employed in kitchens in recent years, allowing rapid cooling at
low temperatures (−78.5 °C and −196 °C, respectively), which
prevents the formation of ice crystals. Blumenthal (2009)
Molecular Cooking
Froth composed of naturally occurring fatty compounds
in the tomato.
A transparent liquid with a tint of yellow, composed of
sugars, salts, acids and aroma compounds dissolved in
water.
A pellet composed of dense fruit solids such as cellulose,
pectin and heavy pigments.
FIGURE 126.2 Centrifuge tube containing three distinct components from
tomato juice.
FIGURE 126.3 Liquid nitrogen being added to an ice cream base.
FIGURE 126.4 Evaporation of a liquid using an evaporator.
highlights that dry ice is easier to buy and store than liquid
nitrogen, and unlike liquid nitrogen, it doesn’t boil away. Instead,
it sublimates, i.e., evaporates from a solid to a gas without ever
becoming liquid. In the case of liquid nitrogen, it freezes instantly
803
and can allow, e.g., fruit segments to shatter into individual pieces
(Edelstein, 2018). Conventional freezing allows the development
of large ice crystals, which damage frozen food, so more and
more chefs are now using, for example, liquid nitrogen to form
much smaller ice crystals in the making of ice cream and sorbet.
Chefs are also using liquid nitrogen to create dramatic theatre
when preparing dishes at the dining table; however, it is a dan-
gerous ingredient and must be handled with great care, adhering
to all safety precautions. The first chef to use liquid nitrogen was
André Daguin, who in the late 1970s, astonished the New York
food press when he demonstrated how to make instant ice cream
using liquid nitrogen (Vintage Insatiable, 1998).
Distillation
Evaporating a liquid from a mixture and then condensing it into
another space separates out solutions based on their physical prop-
erties. This can be done using a rotary evaporator, also known as
a rotavap, which allows distilling under vacuum. The pressure
inside the flask and condenser is lowered, which then lowers the
boiling point of the solvent and other volatiles in the solution.
Drying
Sanchez (2016) notes that chefs have been increasingly inspired
to utilize advanced equipment normally found in laboratories
and industrial kitchens, and techniques such as vacuum-​drying,
spray-​drying and freeze-​drying/​lyophilization. Dehydration has
fast become a popular drying technique for use by chefs. The
dehydrator shown in Figure 126.5a and b has a horizontal airflow
that allows fast, even drying.
Filtration
Filtration allows solids to be strained from liquids, such as sep-
arating pulp from pressed juices and turning cloudy liquids clear
(Potter, 2015). Techniques used by chefs include ice filtration
and vacuum filtration. The former technique is clearly explained
in a YouTube clip by Fotostudio Jan Bartelsman BV (2013),
which shows François Guerds, from FG restaurant in Rotterdam,
making a bouillon (thin stock solution). He blast chills and
cools a stock solution, which following sieving, is placed in a
muslin cloth. Then, he vacuum packs the cooled solution and
places it horizontally in a blast freezer. The pack is removed,
and the frozen sheet of stock is placed back in the muslin, which
is placed in a perforated tray over a non-​perforated tray. The
whole lot is covered in clingfilm and defrosted. Blumenthal
(2009) explains that as the stock melts, it trickles through a
fine network of gelatine strands and drips into the tray. Any
fat that is in the stock remains solid at fridge temperature and
becomes entangled in the gelatine. In addition, the pure ice is
also trapped, and the ice that contains salts, flavours and sugars
melts because of the lower freezing point. The result is a clarified
stock, which is more concentrated (because some ice is retained
in the filter) and with the volatile aromas preserved. In vacuum
filtration, a Büchner funnel is lined with very fine filter paper and
attached to a vacuum flask, where the filtered liquid is collected.
The pressure below the filter is reduced, thereby forcing liquid
804 Róisín M. Burke, Pauline Danaher

FIGURE 126.5 (a) A benchtop dehydrator; (b) shelves in the dehydrator; (c) a dehydrated carrot juice.

FIGURE 126.6 (a) Chocolate foam created using a fish tank bubbler; (b) a lime foam and aerated chocolate.

through the filter paper by the pressure of the atmosphere above
it (Blumenthal, 2009).
Foaming
As early as 1987, Hervé This developed a recipe for a gelified
foam, which he named a ‘würtz’ in honour of Charles Adolphe
Würtz, a famous chemist from Alsace (This, 2018). The method
is as follows:
1. in a large bowl, put 5 g gelatine
2. add 200 g aqueous solution
3. add 100 g sugar
4. heat until the gelatine is dissolved
5. whip extensively while cooling (put the pan in a larger
pan with cold water or ice) until a large volume of foam
is obtained
6. store in the fridge for gelification of gelatine.
Stable foams are commonly used as garnishes on many dishes
created in kitchens these days. The emulsifier lecithin may be
used to stabilize mixtures that would generally separate without
it. It is possible to create airs of sauces and soups, which can be
served warm or cold. Other emulsifiers are used to integrate a
watery medium into a fatty medium (Texturas, 2012). Another
ingredient that is useful for creating a very light-​textured stable
foam is a sugar ester, which is derived from sucrose; this is used
to prepare oil in water emulsions (Texturas, 2012) and is capable
of foaming alcohols. Apart from emulsifiers, another ingredient
in foam recipes includes xanthan gum, a thickener, which slows
the rate at which foams collapse (Figure 126.6).
Gelling
There are many gelling agents that chefs are utilizing in kitchens
today. These include low and high acyl gellan, agar-​agar, kappa
carrageenan, iota carrageenan and gelatine (Alicia Foundation,
2015; Blumenthal, 2009; Lersch, 2014; MSK, 2019). Gellan
Molecular Cooking
allows a chef to create, for instance, heat-​resistant jellies that
can be served on hot dishes, change the textures and viscosity
of liquids to create fluid gels, or turn fruit and vegetable juices
into smooth stable purees that can be made into tuiles. Gellan
also allows chefs to create flambé sorbets or ice creams. Agar-​
agar makes warm and cold jellies, and can be used in dishes
with temperatures up to 85 °C. It is used as a binding agent
in terrines and can also be used in siphons to make a stable
mousse that can be served on hot dishes. Agar-​agar also forms
thermo-​reversible gels.
Kappa carrageenan can be used as a vegetarian alternative
to gelatine, and is ideal to use for encasing liquid centres while
preventing separation. Iota carrageenan makes soft, elastic gels,
which are freeze/​thaw-​stable and can be used as a glaze for semi-​
freddo and parfaits; it is best used in milk-​based recipes, as it gels
in the presence of calcium ions (Potter, 2015).
Sous Vide Cooking
One of the first techniques of molecular cooking was the imple-
mentation of the theory of temperature control when cooking,
known as sous vide. This form of cooking allowed chefs to
achieve results that were not possible with traditional cooking
methods. Gisslen (2011) states that sous vide cooking in its sim-
plest form has two steps:
1. Vacuum packing the item with seasoning or liquids in
an appropriate plastic bag;
2. Cooking the item in the plastic bag at a constant low
temperature in a water bath.
Chefs are now cooking meats at lower temperatures for longer.
Garcia-​Segovia et al. (2007) reported that texture is the param-
eter most affected when cooking meat over a long period of time.
Extended cooking time causes collagen degradation and solubil-
ization (Dikeman and Divine, 2014), leading to gelatine forma-
tion and a decrease in meat toughness while keeping juiciness.
FIGURE 126.7 (a) Vegetables cooking in the waterbath; (b) sous vide cooked salmon; (c) plums and crème anglais cooked sous vide.
805
Using plastic bags in sous vide cooking allows the removal of air,
which improves heating uniformity and avoids some unwanted
flavour formation reactions that are oxygen-​dependent (Myhrvold
et al., 2011b).
Smoking
Smoking food involves cooking at lower temperatures for longer
periods of time, while woods and coals are used as sources
of smoke. There are two types of smoking: hot and cold. Hot
smoking cooks the product, and cold smoking does not. For
example, when fish is hot-​smoked, it is cooked to a minimum
of 62.8 °C for at least 30 minutes, while cold smoking requires
the highest internal temperature to be below 35 °C (Rahman,
2007). Culinary smoking is a method used for imparting flavour
rather than preserving, leading to companies creating innova-
tive equipment and tools necessary for this purpose. These tools
include upright electric smokers, water smokers (which incorp-
orate moisture), wood boxes, cocktail smokers and smoke guns.
The smoke gun has a small cage that holds wood shavings and
is attached to a motor-​operated fan, allowing the chef to smoke
single servings of food products (Dunnam, 2017).
Spherification
Chefs are using both direct and reserve spherification to create
jelly envelopes including liquids (see chapters on spherification).
Sodium alginate and calcium are the main ingredients used in
spherification, as the former is an extract of seaweed that reacts
with calcium to form a gel. In direct spherification, a liquid may
be coloured and/​or flavoured and dropped into a calcium chloride
or lactate bath using a syringe or caviar box. When the drop enters
the calcium bath, the process of gelification (due to calcium-​
induced cross-​linking of alginate molecules) begins, and small
beads with a liquid centre are formed, which ‘pop’ in the mouth
and release the trapped liquid. These are best served immediately,
as they will continue to harden even after rinsing.
806
FIGURE 126.8 (a) Spheres made using direct spherification; (b) a mango sphere on a larger yoghurt sphere, both made using reverse spherification.
MolecularR (2019b) described the reverse spherification
technique as dropping a liquid containing a mixture with cal-
cium lactate gluconate or calcium lactate into a bath of sodium
alginate. The gelation process is the same as direct spherification
except that the calcium in the product moves outward (Vega,
Ubbink and van der Linden, 2013). Reverse spherification is
more versatile than basic spherification, as it can make spheres
with almost any product. It is best for liquids with a high calcium
or alcohol content, which makes it great for cocktails and dairy
products like cheese, milk, cream and yoghurt. The resulting
spheres are long-​lasting and can be stored for later consumption.
In contrast to direct spherification, the process of gelification
can be stopped when the sphere is removed from the sodium
alginate bath and rinsed with water. This also allows the chef to
macerate the spheres overnight to add some extra flavour (e.g., in
aromatized olive oil or truffle water).
As mentioned by Caporaso and Formisano (2016), there are
challenges related to spherification, which include choosing the
correct acidity and calcium concentration and the most appro-
priate solution density and concentration of flavouring agent.
Conclusion
Even now, 16 years since the INICON project commenced,
molecular cooking is less and less mistakenly referred to as
molecular gastronomy. Despite this, a new style of cooking,
molecular cuisine, was developed, which caught the imagination
of the public across the globe. Initially, this type of cuisine was
confined mainly to Michelin starred restaurants, but it has now
become more mainstream. Elements of molecular cuisine such as
foams, gels and emulsions can be found as part of a dish on many
restaurant menus today. Restaurants are purchasing specialized
ingredients and equipment so that they can use molecular cooking
techniques to create dishes to tantalize the senses, allowing mem-
orable and enjoyable dining experiences.
Róisín M. Burke, Pauline Danaher
REFERENCES
Adrià F, Soler J, Adrià A. 2008. A Day at elBulli. London. Phaidon.
Alicia Foundation. 2015. A Chef’s Guide to Gelling, Thickening, and
Emulsifying Agents. Boca Raton. CRC Press.
Blumenthal H. 2009. The Fat Duck Cookbook. London. Bloomsbury.
Caporaso N, Formisano D. 2016. Developments, applications and
trends of molecular gastronomy among food scientists and
innovation chefs. Food Reviews International, 32(4), 417–​435.
Cassi D. 2011. Science and cooking: The era of molecular cuisine.
EMBO Reports, 12, 191–​196.
Dikeman M, Devine C. 2014. Encyclopedia of Meat Science. Volume
1. New York. Academic Press.
Dunnam R. 2017. Culinary trends and breakthroughs. Nutrition &
Foodservice Edge, www.anfponline.org/​docs/​default-​source/​
legacy-​docs/​docs/​cc012017.
Edelstein S. 2018. Food Science: An Ecological Approach.
Burlington, MA. Jones and Bartlett Learning.
Fotostudio Jan Bartelsman BV. 2013. Ice Filtration FG. www.
youtube.com/​watch?v=ymKnHWIXEHg.
Garcia-​Segovia P, Andres-​ Bello A, Martinez-​ Monzo J. 2007.
Effects of cooking method on mechanical properties, colour
and structure of beef muscle (M. pectoralis). Journal of Food
Engineering, 80, 813–​821.
Gisslen W. 2011. Professional Cooking. New York. John Wiley
and Sons.
Iqemusu. 2017. The 24 Notes. https://​iqemusu.com/​en/​the-​24-​notes-​
note-​by-​note-​cooking/​.
Lersch M. 2014. Texture A Hydrocolloid Collection. https://​
blog.khymos.org/​ w p-​ c ontent/​ 2 009/​ 0 2/​ h ydrocolloid-​
recipe-​collection-​v3.0.pdf.
Louis François. 2019. Louis François –​Food Ingredients Since 1908.
www.louisfrancois.com/​index_​en.html.
MolecularR. 2019a. MolecularRecipes.com. www.molecularrecipes.
com/​.
MolecularR. 2019b. Reverse Spherification. www.molecularrecipes.
com/​spherification-​class/​reverse-​spherification/​.
MSK. 2019. MSK catalogue. http://​ msk-​ ingredients.com/​
msk-​
catalogue-​2019/​?page=1.
Myhrvold N, This H. 2018. Molecular Gastronomy, Encyclopedia
Britannica. www.britannica.com/​topic/​molecular-​gastronomy.
Molecular Cooking 807

Myhrvold N, Young CA, Bilet M. 2011a. Modernist Cuisine. Volume This H. 2018. Atelier d’activités pratique note à note /​Note by Note
1. History and Fundamentals. Bellevue, WA. The Cooking activities. Available at: https://​hervethis.blogspot.com/​2018/​
Lab. 06/​atelier-​dactivites-​pratique-​note-​note.html.
Myhrvold N, Young CA, Bilet M. 2011b. Modernist Cuisine. Volume This H. 2014. Note by Note Cooking: The Future of Food. Translated
2. Techniques and Equipment. Bellevue, WA. The Cooking from French by Malcolm DeBevoise. New York. Columbia
Lab. University Press.
Myhrvold N, Young CA, Bilet M. 2011c. Modernist Cuisine. This H. 2013. Molecular gastronomy is a scientific discipline and
Volume 4. Ingredients and Preparations. Bellevue, WA. The note by note cuisine is the next culinary trend. Flavour, 2, 1.
Cooking Lab. Available at: https://​doi.org/​10.1186/​2044-​7248-​2-​1.
Potter J. 2015. Cooking for Geeks: Real Science, Great Hacks, and This H. 2008. Molecular gastronomy vs molecular cooking. Food
Good Food. Beijing. O’Reilly. Technol., 62(12), 108.
Rahman M. 2007. Handbook of Food Preservation. Boca Raton. TTZ-​ Bremerhaven. 2005. INICON: Introduction of Innovative
Taylor & Francis. Technologies in Modern Gastronomy for Modernisation
Sanchez J. 2016. Molecular Gastronomy: Scientific Cuisine of Cooking. https://​ttz-​bremerhaven.de/​en/​research/​food/​
Demystified. Hoboken, NJ. Wiley & Sons. research-​projects/​893-​inicon.html.
Sosa. 2019. Sosa Catalogue. www.sosa.cat. Vega C, Ubbink J, van der Linden E. 2013. The Kitchen as a
Sosa. 2018. Sosa Catalogue: Catering Tools Catering Equipment. Laboratory: Reflections on the Science of Food and Cooking.
http:// ​ p ublic.sosa.cat/​ c atalogues/​ c atering-​ t ools-​ c ooking-​ New York. Columbia University Press.
equipment/​cat-​sosa-​tools-​2018-​1.0-​low-​ca-​es-​en.pdf . Vintage Insatiable. 1998. The Pride of Gascony. [online]. Available at
Texturas. 2012. Texturas Albert y Ferran Adria. http://​ www.insatiable-​critic.com/​Article.aspx?ID=1293&keyword
albertyferranadria.com/​eng/​texturas.html =The%20Pride%20of%20Gascony:% 20 The%20Daguins.
Note by Note Cooking and Note by Note Cuisine
Hervé This vo Kientza1,2 and Róisín M. Burke3
1 UMR 0782 SayFood, AgroParisTech (INRAE), Université Paris-Saclay, F- 91300, Massy, France
2 Groupe de gastronomie moléculaire, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
3 School of Culinary Arts and Food Technology, Technological University Dublin, City Campus, Ireland
What Is Note by Note Cooking?
Cooks traditionally make dishes using food ingredients that are
primarily derived from plant and animal tissues, but very early in
history, fractionation techniques for such materials appeared. For
example, meat stocks were produced (“pit cooking”), because our
ancestors implictly realized that keeping part of animal tissues in
water was a good way to improve the use of food ingredients,
avoiding losses in the fire over which the food ingredients were
heated (about 30% of meat can be lost in this way, including
water and various nutrients; Wandsnider, 1997).
Progressively over time, new techniques were introduced in
order to fractionate more precisely some plant tissues (but also
animal tissues, producing, e.g., cream) and prepare extracts
such as oil, salt and sugar, which are either pure compounds or
mixtures of related compounds (This, 2018). For example, oil is
mainly a mixture of triglycerides, and egg white powder is a mix-
ture of the proteins from egg white, but table sugar is almost pure
sucrose, and salt is usually almost pure sodium chloride. More
and more pure fractions have been produced, in particular by the
milk and cereal industries, such as whey from cheese-​making
being fractionated into a range of protein-​rich fractions (Linden
and Lorient, 1995).
But even before the developments of plant tissue fraction-
ation, Note by Note (NbN) cooking was proposed by H. This
as a culinary technique based on the use of pure compounds in
order to make dishes (This and Kurti, 1994; This, 2014). Using
this method, any aspects of dishes can be designed: shape, con-
sistency, nutritional properties, colour, odour, taste, trigeminal
sensations, temperature and the recent modalities of flavours.
For example, in the last decades, the perceptions of calcium ions
and of long-​chain unsaturated fatty acids have been discovered
(Laugerette et al., 2006).
The development of this culinary technique already has a
history, moving from a time when even the food industry was
strongly opposed to the idea to the situation today, with innova-
tive companies adopting the technique and, for example, coup-
ling it with 3D printing devices (see the chapter by Ross, Burke
and Kelly in this book).
A Short Prehistory and a Birth
NbN was an answer given by one of us (This) to an incorrect
idea proposed by the late Nicholas Kurti in 1969, during a lecture
at the Royal Institution in which he discussed the roots of what
would be called “molecular and physical gastronomy” (by him-
self and by This) in 1988. In this speech (Kurti, 1969), he said
(and wrote):
Let me first explain why I should speak about the physicist in
the kitchen. My main reason for limiting my subject to physics
is that contributions of the chemists to culinary art and to many
other activities designed to increase our sensual pleasures are
well known. The chemists analyse and explain fragrances and
flavours and often invent new ones: a vast chemical industry is
concerned with producing dyes to please our eyes.
In later discussions between Nicholas Kurti and This, Kurti
considered that developments were possible for physics but not
for chemistry (contrary to This’s opinion). The chemistry of fla-
vour was already well developed at that time, with flavourings
being used more and more, but one should also recall that, as
late as 1990, chefs were often reluctant to use products such as
purified gelatine, although they themselves had produced it by
cooking calf feet and clarifying the extracts for pastry making.
The idea of using “chemicals” in food was considered to be
insane, even by chemists.
Then, in the 1980s, the idea of NbN cooking appeared slowly,
in particular after it was reported that storing wood with ethanol
could dissociate lignin, producing various odorant compounds
such as vanillin and other similar compounds (Deibner et al.,
1976; Puech, 1984). One of us (H. This) proposed to add van-
illin directly to brandies, and he gave public lectures in which he
showed how the addition of five drops of cheap vanilla extract to
one litre of cheap whiskies transformed the “harsh” flavour of the
cheap products into “rounded” flavour.
Slightly later, when it was published that paraethylphenol is
partly responsible for the flavour of leather in old Burgundy wine
(Etievant, 1989), tests were also organized to add paraethylphenol
to wines, but the most interesting results were obtained when this
809
810
compound was added to whisky, because when the right amounts
of the compound were used (about 4 ppb), results analogous to
the character of Lagavulin or Laphroig whiskies were obtained.
In 1993, i.e., shortly after the first international workshop
on molecular and physical gastronomy, This was invited by the
editor in chief of Scientific American to write an article about the
freshly created discipline, and as always, he invited Kurti to join
in (This and Kurti, 1994). Because a conclusion was needed for
this article, and because This was generally doing the first draft of
the common texts, he proposed a conclusion in which the use of
pure compounds in dishes would be the future of cooking.
As stated earlier, at that time, This was already making such
additions publicly, but it was considered heretical or provocative;
however, Kurti accepted the conclusion of the article, and the text
generated many negative reactions, in particular by wine makers,
who could not admit that their products could be “adulterated”.
The response to them was that there was a difference between
adding a compound to food ingredients for personal consump-
tion, as we do with salt and sugar, and adulterating wines, which
is strictly forbidden by law.
Why Is NbN Different from Molecular Cooking?
This practice of not only adding compounds to dishes but making
dishes entirely from compounds was named “Note by Note
cooking” in 1996 by This. It is very different from what This called
“molecular cooking” (in 1999, i.e., many years after its develop-
ment, and only because there was confusion with the scientific
discipline named “molecular gastronomy”). Indeed, molecular
cooking was introduced as a new culinary technique using
hardware from (mostly chemistry) laboratories and using more
rational methods than in the past. For example, it was proposed
to use thermocirculators in order to get precise temperatures
during cooking, or to use ultrasonic probes to make emulsions.
Molecular cooking still used traditional food ingredients such as
whole plant tissues (vegetables, fruits) or animal tissues (meat,
eggs and fish).
With NbN cooking, the tools are not very important, and
indeed, both traditional and modern tools can be used, but
ingredients are the key! For NbN cooking, it is proposed to use
only pure compounds, and, of course, these compounds are gen-
erally components of traditional ingredients: water, saccharides,
amino acids and proteins, particular lipids, polyphenols, etc.
One should be aware of the slight difference between “pure
Note by Note cooking” and “practical Note by Note cooking”.
For pure NbN cooking, only pure compounds are used, one by
one. However, for practical NbN cooking, simple mixtures can
be used as well. For example, oils are mixtures of many different
compounds, but most are of the family of triglycerides. Similarly,
corn starch is not pure amylose but rather, 85% amylose and 15%
amylopectin, the behaviour of which is not different from that of
pure amylose. This difference is the same as that between pro-
ducing music using computers and assembling music wave by
wave, and making music with simple synthesizers, still synthetic
but with fewer possibilities.
Hervé This vo Kientza, Róisín M. Burke
Why Is It Important?
The lawyer Jean-​Anthelme Brillat-​Savarin wrote that the dis-
covery of a new dish contributes more to the happiness of human-
kind than the discovery of a new star (Brillat-​Savarin, 1825), and
certainly NbN cooking can be at the root of many new dishes,
because any food is now within reach: any shape, any consist-
ency, any colour, any odour, any taste, etc. In addition, a new style
and a new art (“Note by Note cuisine”, but also “Note by Note
mixology”), can also emerge.
However, the interest in NbN cooking goes much further than
this. Describing all interests would be too long, and deserve a
whole book in itself, so that we are only giving a summary list
of ideas:
• if an energy crisis happens, it is better to transport dried
material than fresh products (a tomato, for example,
contains 95% water);
• if the fractionation of plant tissues is performed imme-
diately on production sites (fields or farms), then the
water can be kept for irrigation, while added value is
obtained through at least two mechanisms: (1) the
prices are more regular (because the new products
can be stored), which is important for the producers;
(2) new products can be designed through fractionation,
which contributes to innovation;
• for nutrition and toxicology, the exact composition of
food can be decided, which can contribute to the fight
against obesity and malnutrition;
• allergies can be avoided by not using the particular
compounds responsible for such reactions;
• when compounds are considered useful for human con-
sumption, the origin of such compounds has no tech-
nical importance (from natural source or synthetic),
and perhaps some specific plants can be selected for
synthesizing particular compounds (for example, per-
haps alfalfa is more efficient than spinach for extracting
chlorophylls, and genetically modified plants could
improve the production even more);
• if proteins are extracted from pulses (this is already
done today), the non-​protein part that remains after
recovery of starch can be given as food to insects, so
that more proteins are produced and extracted;
• the ever purer fractions that the food industry has
been increasingly using lead to greater availability of
ingredients to facilitate NbN cooking;
• avoiding the transport of fresh products can reduce
spoilage, thereby increasing the global output of agri-
culture, which will be needed to feed 9–​ 10 billion
people in 2050;
• food ingredients with no water can be stored without
using non-​sustainable cold technologies;
• foods created using NbN cooking have chemical and
microbiological traceability, as all the constituent
compounds are known;
• appetizing foods can be designed for those with eating
difficulties, e.g., dysphagia;
• personalized foods and diets can be created for those
with special dietary requirements;
Note by Note Cooking and Cuisine
• NbN can be used in conjunction with new and emerging
technologies such as 3D and 4D printed foods and vir-
tual reality meal experiences;
• NbN opens up the possibilities of creating innovative
and entirely novel foods and drinks.
Overall, we are discovering new advantages of NbN cooking day
after day.
A Short History
A Comment on the Name
It was said that the root of NbN cooking is in experimental
additions of particular pure compounds to food and drinks, but
the idea of giving a specific name to the generalization of this
practice came later, probably in 1996. Indeed, it was observed
that there is a close parallel between NbN cooking and synthetic
music, and this was the basis for the choice of a name.
To begin with music, it is a fact that in the old times, musicians
used traditional instruments that make predefined sounds, with an
acoustic spectrum that depends on the particular structure of the
instruments, i.e., flute, violins, trumpets, etc. After the seminal
work by the French mathematician Joseph Fourier (1768–​1830)
and after the advent of recording systems, it became possible to
analyse such sounds and to distinguish between “fundamental”
and “harmonic” vibrations, the intensity and time-​course evolu-
tion of which determined the “timbre” of the particular musical
instruments. More recently, the advent of electronic techniques
made it possible to synthesize music, assembling pure acoustic
waves into new timbres and assembling particular notes or, more
generally, acoustic signals. This led to the development of elec-
tronic, or “synthetic”, music, which is now found worldwide.
For food, the same occurred, but at a later time. The analysis
of traditional food and food ingredients began at almost the same
time as acoustic analysis, shortly after the beginning of modern
chemistry. Today, a lot is known about the chemical composition
and physical structure of food. This makes it possible (but per-
haps pointless) to reproduce traditional food synthetically, but it
also opens new, more interesting possibilities, of dishes designed
entirely de novo from both a chemical and a physical point of
view. Of course, such synthetic food can have the exact properties
that one wants to achieve.
So, what name should be given to such a new technical activity?
In the 1990s, there was much confusion in culinary circles
about the expressions “molecular cooking” and “molecular gas-
tronomy”, and there was also much criticism from traditional
chefs and journalists about the presence of “science in the kit-
chen”. Because cooking is not a scientific activity (as far as nat-
ural sciences are concerned), and because it was considered that
it would be better to get closer to art than to science, the correct
expression “synthetic cooking” was avoided and a comparison
with another art, music, was decided. Comparing NbN cooking
to electronic music, strictly speaking, the right name should have
been “wave by wave cooking”, but this was difficult to say, and
instead, “Note by Note cooking” was chosen. In the future, as
811
NbN cooking becomes more and more familiar, perhaps we can
finally use the name “synthetic cooking”.
First Lectures
As said, as early as the end of the 1990s, the idea of NbN cooking
was shown in public lectures, along with experiments of adding
compounds to beverages or making specific flavours by heating
specific mixtures of saccharides and amino acids, with a view
to obtaining specific glycation products. These experiments
developed more and more into making synthetic food.
However, at the end of 1998, it was observed that the number
of invitations to lectures and of interviews by journalists was
dropping for one of us (This), and it was guessed that this
perceived “fear of chemistry” had something to do with a fear of
the computer bug of the year 2000. Lectures were changed into
simply analysing scientific advances from molecular gastronomy,
which in retrospect was probably a bad decision.
Hong Kong
After the year 2000, when it was felt that this decision to “step
back” had been a mistake, new lectures were given that showed
NbN cooking. However, it was realized that a public event was
needed to present the idea more broadly. In 2004, one of us (This)
proposed to the French chef Pierre Gagnaire that he should be
the first to serve an NbN dish in a restaurant, and together they
decided that this would be done in Hong Kong, at Gagnaire’s
Mandarin Oriental restaurant, with two dinners and a press con-
ference. Many international media covered the events, during
which the dish “Note by Note number 1” was served. This
included crisp disks (named “peligots”) made from thermally
processed glucose, alginate spheres containing synthetic aqueous
solutions, and a granité made from water, glucose, citric acid,
glucose, colorants and odorant compounds (Figure 127.1).
Pioneers
What next? As Pierre Gagnaire is not a man fixed on one par-
ticular cooking trend, he had no reason to go forward entirely in
the direction of NbN. On the other hand, shortly afterwards, in
2010, This used a lecture at the meeting of the Japanese Society
for the Promotion of Science in Strasbourg to invite two Alsatian
cooks, Hubert Maetz and Aline Kuntz, to make two NbN dishes.
They prepared these two dishes in front of the audience, assisted
by This.
Then, still in 2010, because This is the president of the
Educational Committee of the Institute for Advanced Studies
in Gastronomy, he organized an NbN educational dinner at the
Ecole du Corbon Bleu in Paris. The chefs were Patrick Terrien,
Patrick Caals, Frédéric Lesourd, Bruno Stril, Philippe Clergue,
Marc Thivet, Franck Poupard, Patrick Lebouc, Jean-​François
Deguignet, Jean-​ Jacques Tranchant, Nicolas Bernardé and
Xavier Cotte.
The menu was the following (Figure 127.2):
Royale de sous bois, blanc-​
manger truffé et bouillon
légèrement mousseux
812 Hervé This vo Kientza, Róisín M. Burke

(Savoury wild mushroom custard, truffled blancmange and (Napoleon of fresh goat cheese mousse from a siphon)
a light foamy bouillon) Guimauve en deux textures
Profondeur iodée de poulpe et Saint-​Pierre, écume et trans- (Marshmallow in two textures, from raspberry powder)
parence de spaghettis aux cèpes Ardoise “This”
(Deep-​sea octopus and John Dory, cep foam and trans- (“This” dessert platter, chocolate-​free chocolate mousse
parent spaghetti) with cocoa nibs and egg white substitute, egg white
Pigeonneau en deux cuissons, sa compotée de cuisses, and sucrose-​ free meringue, egg yolk and oil-​ free
potimarron fondant, gelée aux polyphénols, asperges mayonnaise)
virtuelles Sucrette glacée au parfum de Menton
(Squab in two ways, slow-​cooked sqab leg, “fondant” red (Menton lemon-​flavoured lollipop, sucrose and lemon-​free
kuri pumpkin, polyphenol jelly, virtual asparagus) lemon sherbet)
Mille-​feuille de chèvre frais au siphon
It can be observed that these dishes still contained animal and
plant tissues, but one has to remember that, at that time, cooks
did not have easy access to appropriate dilutions of odorant
compounds. However, even this practical NbN cooking was
already considered as a big innovative step.
We can now jump over the large number of lectures, shows,
talks and interviews that followed, and consider only the most
important steps. The next one was in 2011: for the International
Year of Chemistry, the official partner was the Dow Chemicals
Company, which agreed to fund an NbN banquet the day before
the official opening at UNESCO in Paris on 26 January. A team
from the catering company Potel & Chabot, under the direction
of the chef Jean-​Pierre Biffi, served this NbN menu:

Sur une idée d’huitres: huitres de tapioca, bavarois

FIGURE 127.1 The first NbN ever served in a restaurant: a preparation
served by Pierre Gagnaire at the Mandarin Oriental of Hong Kong, after a
design by This and Gagnaire.
d’amylopectine, tapioca de citron vert, eau de mer
gelée, crème d’huitres, cristal de vent
(The idea of an oyster, foamy bavarois of amylopectin,
lime, seawater jelly, oyster cream and wind crystal)
Soufflé au homard, sauce wöhler et gelée de framboises
(Lobster soufflé, wöhler sauce and agar-​ agar jelly of
raspberries)
Fibres de bœuf, capellini, cylindres orange
(Beef fibres with orange cylinders)
Boule de cassis
(Blackcurrant ball)

FIGURE 127.2 Some partial NbN dishes served at the Cordon Bleu in 2010.
Note by Note Cooking and Cuisine
Here, the names of the dishes should be discussed. For example,
there was no lobster in the “lobster” dish, but an artificial
flavouring of lobster. Lime was not present either. For raspberries
and blackcurrants, it was the powder of these fruits, which is not
entirely NbN. Also, there was a real mistake at the end, because the
culinary team put real fruits on the plate around the main element.
However, it was so successful that Potel & Chabot served it again
in April 2011 to 450 guests (mostly chefs) gathered for the pres-
entation of French Michelin stars at the Espace Cardin in Paris.
That same year, in October 2011, another NbN dinner was
served by the chefs-​teachers of the Cordon Bleu School in Paris.
The chefs were Patrick Terrien, Patrick Caals, Philippe Clergue,
Frédéric Lesourd, Patrick Lebouc, Franck Poupard, Bruno
Stril and Marc Thivet, Jean-​François Deguignet, Xavier Cotte,
Nicolas Jordan and Jean-​Jacques Tranchant, and the menu was
(Figure 127.3):
Réminisence d’une meurette
(Note by yolk, reminiscent of meurette)
Mille feuilles terre et mer trois couleurs, souligné des deux
sauces Kientzheim et crustacés
(Tricolour surf and turf Napoleon with Kientzheim sauce
and shellfish sauce duo)
Recherche note à note en pot-​au-​feu
(Note by Note exploration of pot-​au-​feu)
Reconstitution d’une mozarella, huile d’olive et mâche
(Reconstitution of a mozzarella, olive oil and lamb’s
lettuce)
Le dessert Cordon bleu
(The Cordon Bleu dessert)
One month later, the Association of Chefs Toques Blanches
Internationales held their first innovation workshop on NbN
cooking: Jean-​Pierre Lepeltier (Hôtel Renaissance La Défense,
Paris), David Desplanques (Hôtel Crowne Plazza République,
Paris), Michael Foubert (Hôtel Renaissance Arc de Triomphe),
Marie Soyez (Hôtel Renaissance La Défense, Paris), David Crenn
(Hôtel Renaissance La Défense, Paris), Vincent Vitasse (Hôtel
Concorde Lafayette, Paris), Julien Mercier (Hôtel Pullmann
Bercy, Paris) experimented after some products were shown.
This workshop led, in December of the same year, to culinary
courses given by the chefs of the same association during the
charity event for Duchesne myopathy; Jean-​ Pierre Lepeltier
FIGURE 127.3 Some NbN dishes served in 2011: they were closer to the pure NbN theory.
813
(Hôtel Renaissance La Défense, Paris), David Desplanques
(Hôtel Crowne Plazza République, Paris), Michael Foubert (Hôtel
Renaissance Arc de Triomphe), Marie Soyez (Hôtel Renaissance
La Défense, Paris), David Crenn (Hôtel Renaissance La Défense,
Paris), Vincent Vitasse (Hôtel Concorde Lafayette, Paris) and
Julien Mercier (Hôtel Pullmann Bercy, Paris) educated the public
who had paid for the courses and tasting of the dishes.
Then, in Montreal in April 2012, a series of lectures and
press conferences were organized at the Institut du Tourisme
et d’Hôtellerie du Québec (ITHQ). For the first press confer-
ence, Ismael Osorio and Erik Ayala Bribiesca, along with chefs
and students of ITHQ, served four NbN bouchées to about 150
journalists. A strange phenomenon appeared: the new dishes
divided the tasters into groups of “like” and “don’t like”, but
these groups were not the same for all bouchées. The next day, an
NbN meal that was less “art moderne” was served to international
journalists, but because the flavours were more familiar, no rejec-
tion was observed by the journalists.
In July 2012, in Dublin, Ireland, This’s lecture was followed
by the production of NbN food samples by the chef David
Desplanques, and in August of that year, NbN cooking was
shown to the students of the Eramus Mundus Master Program
Food Innovation and Product Design (FIPDes), at AgroParisTech,
Paris. The chef Jean-​ Pierre Lepeltier (Hôtel Renaissance La
Défense, Paris) came to show new NbN dishes.
Acceleration: Book and Courses
In the same year (2012), free public courses were given on
molecular gastronomy at which NbN cooking was discussed.
During three podcasted days of lectures, chefs participated:
Philippe Clergue, from le Cordon Bleu, and Jean Pierre Lepeltier,
the president of the Toques Blanches Internationales Association.
In August 2012, a press conference was held with
demonstrations when the book La cuisine note à note en 12
questions souriantes (later translated into English as Note by
Note cooking) was shown to the press; dishes were prepared
by chefs Jean-​Pierre Lepeltier, chef Hôtel Renaissance Paris La
Défense, Laurent Renouf, sous chef Hôtel Renaissance Paris La
Défense, Julien Lasry, chef de partie Hôtel Renaissance Paris
La Défense, Marie Soyer, chef de partie Hôtel Renaissance Paris
La Défense, Mickael Foubert, chef Hôtel Renaissance Arc de
Triomphe, Lucille Bouche, sous chef Hôtel Renaissance Le parc
814
Trocadéro and Yannick Jaouen (sous-​chef Hôtel Mariott Rive
Gauche Paris).
International Contests
In 2012, in order to foster the development of NbN cooking, it
was decided to organize international contests for Note by Note
cooking. The first was organized in 2013 in Paris, without any
particular topic. Pierre Gagnaire came and showed the dish
named “Chick Corea”, which had been shown for the first time
some weeks earlier at the Book Fair in Paris.
Now, in 2021, the International Centre for Molecular
Gastronomy is organizing the ninth event, following the previous
events:
• 2014: using methional
• 2015: playing with proteins, 1-​octen-​3-​ol
• 2016: using cellulose (and/​or derivatives of cellulose)
and trigeminal compounds
• 2017: fibrous consistencies and acidities
• 2018: crackling and crispiness
• 2019: diracs
• 2020: playing with pectins
For all of these, there are three categories (students, lay people
and professionals), and the final event is on the first Friday of
June in AgroParisTech, Paris.
Products
With NbN, one issue for chefs was initially to find the specific
ingredients that can be used. Indeed, for consistency, the issue
was almost solved, because the development of molecular
cooking had led to the creation of many companies selling
such compounds, e.g., gelatine, agar-​agar and carrageenans, for
example. A large choice of colorants is also available..
However, the question of odorant compounds remained. In
July 2013, the Mane company produced boxes of about 20 pure
compounds that were offered to some French chefs, allowing
training of chefs in the restaurant Akrame and then at the Plaza
Athénée. Later, other kits were given to chefs of more countries
each time This was lecturing there. These kits were also useful for
the development of courses on NbN to master’s students in France
and Ireland (Burke and Danaher, 2016; 2018). After lessons on
structuring food at any scale, from the molecular level up to the
macroscopic one, students were trained to create NbN dishes. In
the same year, a wealthy Georgian introduced an NbN dish in a
fast food chain that he owns, while the chef David Desplanques
was designing NbN dishes with Elham Tehrani, student of the
Erasmus Mundus master programme “Food Innovation and
Product Design” (FIPDes).
In Canada, a company was designing kits for children, and
the Dublin Institute of Technology (now called Technological
University Dublin) was inviting new students to work on an NbN
educational programme.
Hervé This vo Kientza, Róisín M. Burke
However, the main step occurred in April 2018, when the
French entrepreneur Michael Pontif officially created the com-
pany Iqemusu specifically for selling pure odorant compounds at
the right concentrations (Pontif, 2017).
Creating an NbN Dish
So, how to start creating a Note by Note dish? There are many
ways, but it is probably more rational to have first an idea, a
theme or a concept. The schematic in Figure 127.4 shows the
stages that are followed by students at Technological University
Dublin who develop NbN dishes each year as a project-​based
learning exercise.
One of us (R. Burke) has noted that many students at
Technological University Dublin (some of whom are qualified
and working as chefs, and some have science qualifications) start
by drawing and creating pictures, and some use a mood board of
what they envision their dish will look like, e.g., an underwater
scene or a forest floor.
Following this, they research the ingredients and consider
the equipment that will be required. It is very important to have
access to information about the ingredient suppliers and their
products, as ingredients can come in different forms. To take one
example, calcium can come as calcium chloride, anhydrous or
dihydrate. As outlined in Table 127.1, to obtain the same concen-
tration of calcium ions in solution, different amounts of calcium
salt have to be used, depending on the molecular weight (MW).
Another ingredient that can be used in NbN cooking is
the emulsifying agent wrongly called “lecithin” by produ-
cers. Scientifically speaking, lecithins are a particular class
of phospholipids, and, more precisely, choline esters of
phosphatidic acids (PAC, 1995). However there are many
different products with this name, each having different
emulsifying properties. These properties are related to their
“hydrophilic-​ lipophilic balance” (HLB; see the chapter on
emulsions and surfactants), the parameter that describes the
relative composition of the water-​loving (polar) and fat-​loving
(non-​polar) elements of an amphiphilic emulsifier molecule
(American Lecithin Company, 2003).
FIGURE 127.4 The stages in the development of a Note by Note dish.
(Source: Burke and Danaher, 2020)
Note by Note Cooking and Cuisine 815

temperature. In TU Dublin, it is usual to optimize the structure

TABLE 127.1 of a NbN dish first. Functional properties such as emulsification,
Levels of Calcium Salts Needed to Achieve Similar Calcium Ion gelling and foaming are examined. Following this, the odours,
Levels in Solutions colours, flavours, trigeminal sensation, consistency and texture
Salt used Amount Calcium concentration are worked on before and after cooking and other temperature
treatments. Production and recipe optimization should include
calcium chloride, anhydrous 5 g/​L (= 0.5%) 0.045 M
(MW = 110.98) 5 g/​L (0.5%) 0.034 M sensory analysis testing.
calcium chloride, dihydrate Finally, the dishes are presented for sensorial valuation by the
(MW = 147.01) lecturing staff. Students take photographs to include in their TU
calcium gluconate 5 g/​L (0.5%) 0.012 M Dublin project report and also for the International Note by Note
(MW = 430.37) 25 g/​L (2.5%) 0.058 M competition. Recent examples of winning dishes include:
50 g/​L (5.0%) 0.116 M
19.4 g/​L (1.94%) 0.045 M (same as 0.5%
2018 Crackling and crispiness (Figure 127.5)
calcium chloride) 2019 Diracs
calcium lactate 5 g/​L (0.5%) 0.022 M The dishes created by the two students met the judges’ cri-
(MW = 218.22) 25 g/​L (2.5%) 0.114 M teria. David Hurley created a cocktail that appeared to be an
50 g/​L (5.0%) 0.229 M eggnog but tasted of bacon, and what appeared to be a bacon
9.8 g/​L (0.98%) 0.045 M (same as 0.5%
crisp had a flavour of eggnog. His main dish included a Note by
calcium chloride)
Note beetroot-​inspired protein cake, horseradish-​inspired jelly
Source: Lersch, 2014. and beetroot-​inspired cremeaux. It was presented in the form of a
meat muscle and put under a smoked filled lid.
For her part, Eugenia Xynada created a Note by Note version
of a breakfast dish with what appeared to be eggs, layered pork
sausage and jellied beans, and bacon flakes, and the tomato
element was in the form of a Note by Note “Bloody Mary”
cocktail. The dish was sweet tasting rather than savoury and
included flavour compounds such as citronellyl acetate, octyl
acetate, pentyl pentanoate, ethyl lactate, ethyl acetate and
cis-​3-​hexenalacetaldehyde.

Recent Developments
Since 2012, the development of this field has rapidly accelerated,
with about 200 lectures per year, all over the world, showing NbN
cooking. Only some of these are described.

• In Denmark, in 2014, the staff of the chemistry depart-

FIGURE 127.5 “ Crackling and crispiness” . Ruth Kelly won joint first place
for her dish, “ A Dish Reminiscent of Black Forest Gateaux (Schwarzwälder
Kirschtorte) incorporating an isomalt dome with a crisp and crunchy tex-
ture and honeycomb providing a crunchy and crackling texture” . The diner
shatters the dome with a spoon and flavoured smoke is released, which gives
the impression of a misty forest atmosphere.
During the development of recipes, it is vital to check the max-
imum permitted levels and safety of the pure compounds and/​or
fractions that will be incorporated. Potential allergenic and toxic
effects of ingredients should be investigated. This information is
available from national and international regulatory authorities.
Following recipe development, the next step is to carry out kit-
chen trials. As outlined earlier, there are many aspects to consider
in the design of the dish, including shape, consistency, nutri-
tional properties, colour, odour, taste, trigeminal sensations and
ment of the University of Aarhus and chefs produced a
Note by Note meal served to the king’s family.
• In Copenhagen, Denmark, in Estoril, Portugal, in Boston
and New York, in some French culinary schools, and in
Dublin, Ireland (since 2012), Note by Note cooking is
being taught.
• In Japan, 2015, a collaboration of the Corbon Bleu and
Ritsumekan University led to NbN sushis, created by
the chef Guillaume Siegler, being shown to the Japanese
press (see the recipe in Part III of this book).
• Then, in 2015, when a journalist from the New York
Times came to Paris in order to prepare an article on
Note by Note cooking, the chef Pierre Gagnaire agreed
to make a whole menu in which all dishes were based
on one single odorant compound.
Now followed important steps. In 2017, the chef Andrea Camastra,
of the restaurant Senses in Warsaw, Poland, transformed his res-
taurant entirely from molecular cuisine to the NbN domain.
He is being followed. For example, on 21 February 2018, the
Alsatian chef Julien Binz delivered the first 100% NbN meal in
816 Hervé This vo Kientza, Róisín M. Burke

FIGURE 127.6 Winning dishes in the NbN student category in 2019. Joint first prize in the student category went to David Hurley and Eugenia Xynada.

his restaurant in Ammerschwihr (France). On 16 March 2018, a
big NbN event was organized in Athens, Greece, by the Ecole Le
Monde, with NbN cocktails, an NbN dinner and a huge training
lecture; the chefs who worked on this were Yannis Vlachos (cock-
tail), Makis Kalosakas, Nicolas Nikolakopoulos and Michel
Ntenoutas (see chapter “New Greek Cuisine”). And in July
2018, chefs of the At-​Sunrice GlobalChef Academy in Singapore
prepared a dinner and began organizing classes on NbN cooking
(see chapter “A Note by Note Traditional Chinese Dinner Created
and Served in Singapore”).
What will be the next steps? Of course, there is no certainty,
but the fact that NbN cooking was measured (by Julien Binz) as
being twice as fast and half as expensive to produce (compared
with ordinary restaurant meals) makes it a good argument in
favour of the development of this new art for fine dining, whereas
the increasing world population will more and more call for
new developments in this direction by the food industry and by
domestic cooks. Projects for “sustainable food without waste”
should become more and more important in the future.
REFERENCES
American Lecithin Company. 2003. American Lecithin Company
[online]. Available at: www.americanlecithin.com/​ leci_​ Moss GP, Smith PAS, Tavernier D. 1995. Glossary of class names
appfood. html.
Brillat-​Savarin JA. 1825. La physiologie du goût. Chez l’auteur, Paris.
Burke R and Danaher P. 2020 Assessment and evaluation of stu- Pontif M. 2017. https://​iqemusu.com/​en/​note-​by-​note-​cooking
dent learning through a project-​based assignment on note by Puech JL. 1984. Characteristics of oak wood and biochemical
note cooking. International Journal of Food Studies (ISEKI).
Accepted for publication.
Burke R and Danaher P. 2018. Project-​based learning and Note by
Note cooking: Two ingredients to enhance student learning
and creativity. In: Proceedings of the 5th International
ISEKI_​Food Conference, University of Hohenheim, July 3–​
5, Stuttgart, Germany.” The Food System Approach: New
Challenges for Education, Research and Industry p. 54. ISEKI
Food Association, Stuttgart. Available at: https://​arrow.dit.ie/​
tfschcafcon/​54.
Burke R and Danaher P. 2016. Note by Note: A New Revolution in
Cooking. Available at: https://​arrow.dit.ie/​dgs/​2016/​May31/​1/​
Deibner L, Jouret C, Puech JL. 1976. Substances phénoliques des
eaux de vie d’Armagnac. I. La lignine d’extraction et les
produits de sa dégradation. Industries alimentaires et agricoles,
93, 401–​414.
Etievant P. 1989. L’odeur de cuir: responsabilité de la fermentation
malo-​lactique. Revue des oenologues, 53, 39.
Kurti N. 1969. The physicist in the kitchen. Proceedings of the Royal
Institution, 42, 99, 451–​467.
Laugerette F, Gaillard D, Passilly-​Degrace P, Niot P, Besnard P.
2006. Do we taste fat? Biochimie, 89, 2, 265–​269.
Lersch M. 2014. Texture –​a hydrocolloid recipe collection (v.3.0.,
February 2014). [online] Available at: https://​ blog.khymos.
org/​recipe-​collection
Linden G, Lorient D. 1995. Biochimie agro-​industrielle. Masson,
Paris.
of organic compounds and reactivity intermediates based on
structure, IUPAC Recommendations 1995, 67, 1307–​1375.
aspects of Armagnac aging. American Journal of Enology and
Viticulture, 35, 77–​81.
Note by Note Cooking and Cuisine
This H. 2014. Note by Note Cooking. Columbia University Press,
New York.
This H. 2018. Who discovered the gluten and who discovered its
production by lixiviation? Notes Académiques de l’Académie
d’Agriculture de France/​ Academic Notes from the French
Academy of Agriculture, 3, 1–​11.
817
This H, Kurti N. 1994. Chemistry and physics in the kitchen,
Scientific American, April, 66–​71.
Wandsnider LA. 1997. The roasted and the boiled: food compos-
ition and heat treatment with special emphasis on pit-​hearth
cooking. Journal of Anthropological Archaeology, 16, 1–​48.
Spherification
Sasa Hasic
R&D Chef, Croatia
Spherification, i.e., the formation of liquid spheres inside a thin
gelled membrane, is described theoretically in the chapter of this
handbook by Luck, and also by Berardi et al. Here, I give infor-
mation about the main ingredients, as well as technical advice,
before giving recipes using the technique for either the direct or
the reverse method.
Some Technical Information
Sodium Alginate
Sodium alginate is a polysaccharide extracted from the ocean’s
brown algae, first mentioned in 1881 by the British chemist
E.C.C. Stanford. It has the properties of a hydrocolloid (see the
chapter by Edwards-​Stuart and Barbar) and is used in the food
industry as a stabilizer and thickening agent due to its ability to
retain water. When the alginate moiety reacts with calcium ions,
it makes a non-​thermoreversible gel, which is why it is used in the
process of “spherification” (Adria, 2010; Caballero et al., 2003).
Instructions for Use
The minimum quantity to obtain the gelling of solidum alginate
around liquid spheres is 10 g/​kg (more than is used for making
jams, jellies, and preserves). In basic spherification, small
proportions of alginate (0.4% to 0.7%) are used in the product to
include in the spheres, and a calcium salt (e.g., chloride) (0.5%
to 1% concentration) bath is used. In reverse spherification, the
calcium salt is added to the liquid phase to include in spheres, and
the bath contains the alginate.
Stirring is used for dissolution, but heating is not necessary,
and the alginate solution is left in a refrigerator overnight to
hydrate. If it is strongly stirred, air bubbles are formed, which are
lost when the solution is left to rest, while leaving it to hydrate
slowly avoids air absorption. Heating in the presence of sugars
aids the hydration process, and it is possible to remove air from
the solution with a vacuum packing machine.
In practice, alginic acid can precipitate if pH < 3.65, so
trisodium citrate can be used to adjust the pH to reach pH 5
(Table 128.1).
The Role of Calcium
Calcium ions are naturally present in a number of food
ingredients, such as milk and milk-​derived products, as well as
plant and animal tissues (Table 128.2).
For spherification, various calcium salts can be used: chloride,
gluconate, glucose lactate, and lactate. They are often extracted
from dairy and mineral products. Calcium chloride is bitter,
and so spheres should be washed before use. Bear in mind that
calcium ions can be “trapped” by sequestrants with sodium
gluconate, sodium tripolyphosphate and calcium disodium
ethylene diamine tetra-​acetate, and other phosphates. Calcium
chloride and calcium sulfate tend to absorb water from the air,
but the quantities given here are for the water-​free (anhydrous)
forms. Different calcium salts contain different amounts of cal-
cium ions per mass unit, and this is the basis on which the cal-
cium bath is prepared. The most common calcium salts used in
spherification and the concentrations to create the calcium bath
are given in Table 128.3.
To have the same concentration of calcium ions in the solution,
different calcium salts must be used in different quantities, based
on the molecular weight (MW). Table 128.4 shows the amounts
of various salts to add, as well as concentrations obtained when
adding the salts at 0.5%, 2.5%, and 5% levels; e.g., to obtain the
same calcium concentration that you get with a 0.5% solution of
calcium chloride, you must prepare either a 1.94% solution of
calcium gluconate or a 0.98% solution of calcium lactate.
TABLE 128.1
Required Levels of Trisodium Citrate to Reach pH 5 from Solutions
at Various initial pH
Initial pH Trisodium citrate to add (g/​L)
2 2.7
2.5 0.85
3 0.27
3.5 0.082
Source: Lersch, 2014.
819
820 Sasa Hasic

TABLE 128.2
Content of Calcium in Some Food Ingredients (Lersch, 2014).
Food product Calcium (mg/​100 g) Calcium conc. (mg/​100 g) Food product Calcium (mg/​100 g) Calcium conc. (mg/​100 g)
Sesame seeds 980 0.244 Garlic, raw 181 0.045
Mozzarella 720 0.179 Instant coffee 141 0.035
Brie 510 0.127 Rhubarb 140 0.035
Tahini 426 0.106 Yoghurt 128 0.031
Spinach, frozen 280 0.069 Ice cream (dairy) 120 0.029
Dried figs 250 0.062 Hazelnuts 114 0.028
Peppermint, fresh 243 0.060 Pistachios, raw 105 0.026
Almonds 240 0.059 Milk 100 0.025
Soya flour 210 0.052 Olives 88 0.022
Molasses 205 0.051 Blackcurrant 65 0.016

Source: Lersch, 2014.

liquid (juice, puree, etc.) with a hand immersion

TABLE 128.3
Sources of Calcium to Add to Make a 0.04% Solution
Calcium source
Calcium chloride (anhydrous)
Calcium sulfate (anhydrous)
Calcium lactate
Calcium gluconate
Source: Myhrvold et al., 2011.
TABLE 128.4
Required Levels of Different Calcium Salts
Salt used
calcium chloride, anhydrous
(MW = 110.98)
calcium chloride, dihydrate
(MW = 147.01)
calcium gluconate
(MW = 430.37)
calcium lactate
(MW = 218.22)
Source: Lersch, 2014.
General Principles
In order to make successful spheres, the following directions
should be observed:
1. Prepare liquid for basic spherification like most
hydrocolloids. Disperse the sodium alginate in the
blender. Sodium alginate is very “sticky”; thus, for
thicker liquids like purees, it may be necessary to first
add some water to the liquid in which to dissolve the
Calcium content
in the salt Quantity to add sodium alginate.
2. The liquid needs to be cold, and if it is too acidic,
36.1% 0.11%
sodium alginate will not hydrate and will make a thick
29.4% 0.14%
18.4% 0.22%
solution.
9.3% 0.43% 3. Above all, check the pH of the liquid and adjust it to
4.5. For pH correction, you can use trisodium citrate,
in quantities that you control using a pH meter or pH
papers.
4. The liquid should not contain calcium ions; otherwise,
the gelation will occur in the bath before you make
spheres. Also, sodium alginate will not hydrate prop-
erly in brandies, and it is very important that you first
Quantity Calcium concentration disperse alginate in water, juice, purée, etc. in order
5 g/​L (= 0.5%) 0.045 M to allow alginate to hydrate, before adding to alcohol;
5 g/​L (0.5%) 0.034 M otherwise, liquid will leak out and break in the process
of cooking, and you will not have the right effect of
spherification.
5 g/​L (0.5%) 0.012 M
25 g/​L (2.5%) 0.058 M
50 g/​L (5.0%) 0.116 M For the solution to be included in spheres:
19.4 g/​L (1.94%) 0.045 M (same as
0.5% calcium a. Add sodium alginate into 30% of the prepared liquid for
chloride)
making spheres.
5 g/​L (0.5%) 0.023 M
25 g/​L (2.5%) 0.115 M
b. Disperse with hand immersion blender; after several
50 g/​L (5.0%) 0.23 M minutes for incorporating alginate into the liquid, a very
0.045 M (same as sticky solution will be produced.
0.5% calcium c. Then add the rest of the prepared liquid for spherification;
chloride) disperse with the hand immersion blender and hydrate
in the fridge for 2 h.
d. This rest is important for avoiding air bubbles, so that
the spheres will be able to sink and “cook” properly. As
mentioned earlier, one can also use a vacuum packing
machine for faster hydration in some recipes.
e. Prepare the calcium bath with distilled water to achieve
the right thickness. Tap water containing too much cal-
cium cannot be used.
Spherification
For the calcium bath:
a. Prepare the calcium bath by adding calcium salt to the
distilled water; e.g., use 0.5% to 1% calcium chloride
to water.
b. Dissolve in water with stirring. It will dissolve very
quickly, or a magnetic stirrer may be used.
c. Adding a small amount of sugar to your calcium salt
bath will stop spheres from sticking to the bottom of
the container in the process of spherification. The cal-
cium solution can be stored in the refrigerator for sev-
eral days before using.
Process of spherification in calcium bath:
a. Take your liquid with the content of sodium alginate
from the fridge.
b. Also take calcium chloride bath, syringe, spherification
spoon, paper towels, and timer.
c. Prepare one more bowl with water for washing and
rising spheres. This step will remove all calcium con-
tent from the surface of the spheres.
d. Using a measuring spoon for best results, take the liquid
juice or puree with sodium alginate added.
e. Gently pour liquid from spoon into the sodium chloride
bath. Transfer all liquid from spoon into bath, and start
to move the spheres around with spoon in a circular
motion without touching the spheres.
f. This will allow to cook your spheres and to start the for-
mation of a gel from the top to the bottom.
g. Stir gently for 2 min in a calcium chloride bath.
h. Collect the sphere with a spherification spoon from a
calcium bath and transfer to a bowl with water.
i. Start stirring the water around the spheres with a spoon
to wash the calcium from the spheres. Be careful when
TABLE 128.5
Quantities for the Various Spherification Methods
Components of the setting bath Added to the sphere base
Direct
Calcium chloride 0.5–​1% Sodium alginate 1%
or calcium gluconate 2.5%
Reverse
Sodium alginate 0.5% Calcium lactate 3%
Tartaric acid 0.2%
Carrageenans (direct)
Potassium phosphate 5% Iota carrageenan 2%
Low methylated (LM) pectin (reverse)
Pectin 2% Calcium lactate 5%
Gellan (direct)
Calcium gluconate 6% Low acyl gellan 0.2%
Note: For the various cases, xanthan gum can be added, as it thickens liquids, and trisodium citrate is helpful when the liquid is too acidic or high in calcium
(Myhrvold et al., 2011).
821
moving the spheres with a spoon from a calcium bath
or the membrane will break easily, because it is thick
outside and liquid inside.
j. Dry with paper towel under a spherification spoon to
remove any remains of water or gel around the spheres.
In the process of basic spherification, you must serve your basic
spheres immediately, because the process of gelling is still going on.
Adding xanthan gum to the liquid can increase liquid vis-
cosity; when the liquid is thick, it separates in the bath at the
gelling point (Table 128.5).
Recipes
Basic Spherification Recipe
Ingredients for a calcium bath:
100% deionized water
0.5% calcium chloride
Set a glass beaker on the magnetic stirrer, add water to the glass
beaker, and start stirring on medium speed, slowly adding cal-
cium chloride. Stir for 3 min and transfer solution into the con-
tainer to the fridge before usage.
Ingredients for a sphere (for a base of diced tomatoes):
0.5% sodium alginate
0.20% xanthan gum
malic acid to taste
fructose to taste
1.4% salt
0.20% “Onium” odorant solution (Iqemusu)
0.10% red colorant, hydrosoluble
Possibilities of changes
Possibility of adding xanthan gum 0.2–​0.5% (xanthan gum thickens liquids)
Use trisodium citrate when the liquid is too acidic or high in calcium
Possibility of adding xanthan gum 0.2–​0.5%
Hydrate carrageenan for at least 5 h; can also make reverse spherification.
Can also make direct spherification.
Sodium hexametaphosphate 0.1%
Best made with low-​acid moderate-​calcium liquids; make frozen spheres and set
in calcium bath; can also make reverse.
822
Method:
1. Cut tomatoes, and transfer them into the food pro-
cessor; blend for 3 min.
2. Take four centrifuge bottles; put one bottle on scale
and transfer the tomato puree in bottle to weight of
150 g; do the same with the three other bottles.
3. Transfer the bottles into the centrifuge, and set at
4000 rpm for 12 min.
4. Prepare a clean measuring glass cup, spoon, and filtra-
tion paper.
5. Store bottles from the centrifuge so that the tomato
juice and the pulp separate.
6. Prepare the glass beaker on a working surface and set
filtration paper over the filter, then open bottles and
with the spoon hold the pulp from the tomatoes in the
bottle, at the same time carefully passing the rest of the
liquid through the filtration paper into the glass beaker.
7. When you have a clear tomato juice, you are ready
to make red tomato juice. Prepare a pH indicator (pH
paper or pH meter), magnetic stirrer, refractometer,
pipette, spoon, and paper towels.
8. In the glass beaker with magnetic stirrer on, add salt
to taste and adjust the sweetness of the juice by adding
fructose while stirring constantly; take some juice
with the pipette between additions of fructose, and
check the content of fructose in your juice (in Brix)
with the refractometer. Repeat this procedure a few
times until you get 20 °Brix.
9. Adjust the acidity of the juice by adding some Citras
(Texturas Inc.) while stirring constantly. You have to
reach a pH of 4.6.
10. Disperse the red water-​ soluble colorant and the
odorant solution in the juice with the hand immersion
blender.
11. Disperse sodium alginate in 30% of tomato juice with
the hand immersion blender, then add this solution to
the rest of the juice and disperse for 2 min more.
12. Pass through a fine mesh sieve.
13. Remove air from solution with vacuum packing
machine or allow the solution to hydrate for 1 h.
14. Set up the calcium bath and use the measuring spoon
to pour some juice solution into the calcium bath; wait
for about 3 min, moving the spheres around in the bath
solution with the spherification spoon.
15. Take the spheres out of the bath, gently rinse in water
and serve.
Adding xanthan gum to the liquid can increase liquid viscosity.
When the liquid is thick, it separates in the bath at the gelling
point.
Reverse Frozen Spherification
Ingredients for the sodium alginate bath:
100% distilled water
0.5% sodium alginate
Sasa Hasic
Method for the bath:
1. Set cold deionized water in a jug and add sodium
alginate; disperse it with a hand immersion blender
for 2 min.
2. Pass the liquid through a fine mesh sieve to prevent
clumps and create a smooth-​textured liquid.
3. Seal the mixture into a vacuum packing bag and
refrigerate to hydrate for 2 h.
4. After this rest, transfer the liquid to a food storage con-
tainer and use it like a bath.
Ingredients for the spheres:
70% Greek yoghurt
20% kefir yoghurt
10% Mediterranean yoghurt powder (Sosa)
0.20% calcium salt
salt to taste
0.22% xanthan gum
solution of 0.1% lactic acid
200 mL olive oil
Method:
In this recipe, a lower proportion of calcium is used, because
dairy products like cheese, milk, and yoghurt are already rich in
calcium.
1. Put the Greek and kefir yoghurt in a glass beaker.
2. Add all substances to the yoghurt except the lactic
acid and the odorant solutions.
3. Disperse with a hand immersion blender for 2 min.
4. Pass the solution through a fine mesh sieve to get a
smooth solution.
5. Adjust yoghurt solution with odorant solutions, and
adjust the acidity with lactic acid, adding the solution
of lactic acid drop by drop. Each time you add a drop,
stir the solution with a spoon.
6. Pass the yoghurt solution one more time through a fine
mesh sieve into a squeeze bottle and hydrate for 1 h in
the refrigerator.
7. Prepare a bowl with clean water, paper towels,
spherification spoon, and double fruit baller spoon,
and prepare olive oil in a small container.
8. Transfer the container with the sodium alginate bath
from the refrigerator to the working space. Prepare
squeeze bottle with the yoghurt solution.
9. Take the baller spoon in one hand and set the bottom of
the spoon into the alginate bath, and with the other hand,
start pushing yoghurt solution from the squeeze bottle to
the top of the baller spoon, while at the same time moving
the spoon towards the bottom of the alginate bath. When
you reach the bottom of the bath, turn the spoon around
and move from the alginate bath, making slow, circular
movements around the spheres for about 3 min.
10. Remove the spheres from the bath with the spherification
spoon and rinse them in water for 10 seconds.
Spherification
11. Transfer the spheres to the container with olive oil to
preserve spheres for up to several days in the refriger-
ator before usage.
In a modification of the process of reverse spherification, frozen
reverse spherification is possible. Chefs use this technique when
alcohol is in the spherification recipes. In this case, the solu-
tion with sodium alginate is added into hemispheric moulds and
frozen in a deep freezer before usage.
Cryogenic Cooking –​Cryo-​Diffusion for Making
Frozen Spheres Which Are Then Gelled in a Sodium
Alginate Bath
Cryogenic cooking, also called cryo-​cooking, has become more A Recipe with Cranberries
popular with the molecular cuisine movement (see the chapter
by Barham in this book). Chefs have been working with frozen Preparation of the alginate bath:
carbon dioxide or liquid nitrogen for a number of years in high-​
end restaurants.
One can control frozen textures, foams, pearls, etc. without
any water crystals or air inside the solution. Liquid nitrogen is a
very good way to preserve the flavour, consistency, colour, and
nutritional value of foods, especially for delicate products. The
most interesting part is that, in the process, you can remove the
air from the solution used.
When using liquid nitrogen, safety rules are extremely
important, because liquid nitrogen has an extremely low tempera-
ture, and if it comes into contact with the skin, hands, or tissue Ingredients for the cranberry purée:
for more than a couple of seconds, it can cause serious injury.
Individuals should always wear eye protection and use cryo-​
gloves when handling this material. Store the dewar containing
liquid nitrogen at floor level, and do not leave the tank in the way.
The general process:
1. Combine gelling substances in appropriate
proportions. Ensure that the pH is not too low, and
add trisodium citrate until the pH is above 4.5.
2. Prepare the alginate bath; disperse water and sodium
alginate with a hand immersion blender.
3. Remove air from the solution with a vacuum packing
machine or leave to hydrate for 1 h in the refrigerator.
4. Disperse the various ingredients of the liquid you will
cryo-​cook (puree, juice, etc) using the hand dispersion
blender.
5. If needed, pass this liquid through a fine mesh sieve.
6. Transfer the liquid containing calcium (juice, puree,
etc.) into a squeeze bottle or syringe.
7. Take out the alginate bath from the refrigerator and
bring it to your working surface; prepare an additional
metal bowl.
8. Put on gloves and safety glasses before you start to
transfer liquid nitrogen from the dewar to the cryo-​
bowl. Handle liquid nitrogen with extreme care.
9. Start to drop small droplets of the prepared liquid
solution from the squeeze bottle or syringe into the
cryo-​bath, stirring the cryo-​bath with a whisk, and at
823
the same time, with the other hand, start to drop liquid
in the cryo-​bath and wait for 2 min.
10. Collect the shaped frozen spheres with a large
spherification spoon from the cryo-​bath.
11. Transfer the frozen spheres to a metal bowl and set to
−40 °C in a fast cold freezer.
12. Take some frozen spheres from the metal bowl with
a big spherification spoon, drop into an alginate bath,
and allow to react for 2 min.
12. Collect the cryo-​spheres with a spherification spoon
from the alginate bath after 2 min, rinse in water,
and serve.
1. Prepare a sodium alginate bath with 100% deionized
water and 0.5% sodium alginate; disperse using a hand
immersion blender for 2 min.
2. Pass this liquid through a fine mesh sieve to create a
smooth texture.
3. Vacuum and seal this mixture to avoid air bubbles, or
hydrate for 2 h in a refrigerator.
4. Transfer the solution to a food storage container.
100% fresh cranberries
malic acid to taste
fructose to taste
0.5% calcium chloride
0.30% low acyl gellan
0.25% “Frum” odorant solution (Iqemusu)
0.25% “Balqin” odorant solution (Iqemusu)
0.20% xanthan gum
Method:
1. Wash the cranberries and make a puree in a jug with a
hand immersion blender (mixing for 3 min).
2. Pass the puree twice through a fine mesh sieve until
the consistency of the puree is almost silky.
3. Disperse calcium chloride, xanthan gum, and low acyl
gellan into the puree.
4. Pass through a fine mesh sieve.
5. Adjust the pH of the cranberry puree with malic acid,
adding small amounts at a time and measuring the pH
with a pH indicator until the pH is equal to 3.9.
6. Add fructose to the puree while stirring; after each
addition, transfer a drop of the puree over the surface
glass of the refractometer and measure the sugar con-
tent until you get 25 °Brix.
7. Add the odorant solutions to the puree.
8. Remove air from solution with a vacuum sealer.
9. Transfer the cranberry puree into a squeeze bottle and
set in the refrigerator before usage.
824
REFERENCES
Adria F. 2010. Modern gastronomy A to Z: a scientific and gastro-
nomic lexicon, CRC Press, Boca Raton.
Caballero B, Finglas P, Toldra F (eds.). 2003. The encyclopedia
of food sciences and nutrition (2nd ed.), Academic Press,
Cambridge, MA.
Sasa Hasic
Lersch M. 2014. A hydrocolloid recipe collection, https://khymos.
org.
MolecularRecipes. 2020. www.molecularrecipes.com/​spherification-​
class/​basic-​spherification
Myhrvold N, Young C, Bilet M. 2011. Modernist cuisine: the art
and science of cooking, The Cooking Lab; Slp Spi Ha edition,
Köln, Germany, 4.
The Raspberry Pear Viennoiserie

James A. Griffin
School of Culinary Arts and Food Technology, College of Arts and Tourism, Technological University Dublin,
City Campus, Dublin 1, Ireland

This pastry is an exquisite blend of the finest butter laminated crimson red/​purple coloured pear. The pears should be rinsed and
Viennoiserie filled with crème patissiere, raspberry jam and patted dry using a clean cloth before placing them on the proofed
poached raspberry pears infused with vanilla. fermented dough at a later stage in the process.

The Four Stages of Materials Preparation The Crème Patissiere

Stage 1. The raspberry pear syrup is prepared and cooled, Stage # 1:
and tinned pears are placed in the syrup, which is placed
in a stainless-​steel bowl and covered for a minimum of 50 g Caster sugar
two days in a fridge at 3 °C. 50 g Cornflour
Stage 2. The crème patissiere is prepared, cooled and piped 4 Egg yolks (100 g approx.)
into fingers of 4–​5 cm length on silicone paper using a 150 g Milk
1-​cm round piping tube and frozen at −18 °C. Mix together to form a paste
Stage 3. The laminated pastry element is prepared first by
mixing a stiff, sweet, yeasted dough. Stage # 2:
Stage 4. Special lamination croissant butter containing a 450 g Milk
minimum of 84% fat is made into a rectangular block 50 g Sugar
by passing it through a dough sheeter or rolled by hand Vanilla Pod or Essence
between sheets of plastic. The butter block should be
chilled overnight. This process can involve the one-​ Total weight 754 g.
minute croissant butter block technique (Griffin, 2015).
Method:

Raspberry Pear Preparation
Method: Juice from the tinned pears is strained into a saucepan.
The sugar and raspberry purée are added to the pear juice, and
the mixture is brought to the boil and allowed to simmer for 5
minutes. The juice is strained to remove raspberry seeds, and
the vanilla is added and stirred into the mix. The syrup should
be set aside to cool, at which point the raspberry alcohol is
added, and finally the pears. The pink syrup mixture with the
pears should be placed in a sealed container, covered and put
in the fridge at 3 °C for two days to ensure correct colouration
of the pears.
The Science: The raspberry pear juice is a super-​saturated sugar
solution, having a higher sugar content than that of the pears.
Osmosis takes place over the two days of soaking, whereby the
pears, immersed in the super-​saturated solution, draw the super-​
saturated pink pear solution towards their centre, resulting in a
• Boil the milk, sugar and vanilla pod, then whisk in the
paste to form a smooth batter.
• Bring back to the boil to thicken.
• Transfer to a large stainless-​steel bowl.
• Sprinkle with icing sugar to prevent skinning.
• Store in a refrigerator when cool, pipe into fingers as
previously described and freeze.
Preparation and Cutting of the Pastry
For the raspberry pear viennoiserie, the pastry is sheeted to a
thickness of 9 mm and chilled again for 20 minutes to stiffen
the butter and the pastry dough. This enables cutting, without
shrinking the pastry, with a large pear-​shaped or teardrop-​shaped
cutter. The pear-​shaped pastry pieces are placed on a tray with
silicone paper, egg-​washed and proofed at a temperature of 27 °C

825
826
FIGURE 129.1 Poached raspberry pear viennoiserie.
for 60–​90 minutes at 75–​80% relative humidity. The temperature
is critical, as if it exceeds 28 °C, the butter will liquefy and run
out of the dough, destroying the lamination layers.
When proofed, the frozen custard fingers are placed in the
centre of the proofed pastry and given a gentle shove downwards
to embed them into the pastry. The function of the frozen custard
fingers is to prevent the custard from leaking out over the edges
James A. Griffin
as the pastry rises. Raspberry jam is then piped onto the frozen
custard, and finally, the rinsed, dried raspberry pears are sliced
five times from right to left but leaving the upper part of the pear
attached. The sliced raspberry pear is placed on the custard and
pressed down to prevent it falling over in the oven while baking.
Using a ventilated or fan oven, pre-​set the temperature at
230 °C, load the pastry and close the oven door. Reset the tem-
perature to 175 °C and bake for 22–​24 minutes. The baked pastry
should be allowed to cool, then glazed with nappage or apricot
jam. Next, use a bench scraper, with the blade placed 3 mm from
the edge of each pastry at a 45 ° angle. The pastries are then
dusted using a small sieve with icing sugar. The same is done at
the slender tip side of each pastry using a small sieve and adding
a dusting of raspberry powder. Finally, a fresh raspberry is placed
at the top of the baked pear on the pastry. If required, red choc-
olate hearts may also be sprinkled on the baked raspberry pear
pastry to finalize the garnish. To see the raspberry pear pastries
spring to life in the oven using time-​lapse photography, refer to
the link in the bibliography (Griffins Bakery, 2016).
REFERENCES
Griffin JA. 2015. One minute croissant butter block technique,
Galway, James Griffin.
Griffins Bakery. 2016. Poached Raspberry Pear. Galway, Griffins
Bakery. www.facebook.com/​Griffinsbakery/​videos/​1015542
8360388984/​?v=10155428360388984
Molecular Mixology: Welcome Coffee, a Cocktail with Ten Layers
Hervé This vo Kientza1,2 and Pierre Gagnaire3
1 INRAE-​AgroParisTech, UMR SayFood, Université Paris-​Saclay, 91300, Massy, France
2 Groupe de gastronomie moléculaire, INRAE-​AgroParisTech International Centre for Molecular Gastronomy,
F-​75005, Paris, France
3 Restaurants Pierre Gagnaire
Cocktails were once simple mixtures of aqueous solutions
(mainly fruit juices) and spirits. Progressively, other ingredients,
such as cream, sugar, or herbs, were added, but with the advent
of molecular cooking, new tools have been used, and various
additions such as alginate pearls have been proposed. Here, we
describe a cocktail with ten layers, i.e., many more than the three
of the famous Irish coffee. The technological idea is by H. This,
and we give a recipe using this idea by P. Gagnaire.
Cocktails are defined as alcoholic drinks consisting of a spirit
or spirits mixed with other ingredients, such as fruit juice or
cream (Oxford Dictionaries, 2018). Many of them are said quite
arbitrarily to contain at least three flavours, one of which is
ethanol, or to contain alcohol, a sugar, and a bitter/​citrus com-
ponent (Robuchon, 2007). When a mixed drink contains only a
distilled spirit and a mixer, such as soda or fruit juice, it is said
to be a highball. When a mixed drink contains only a distilled
spirit and a liqueur, it is a duo, and when a mixer is added,
it is a trio. Additional ingredients may include sugar, honey,
milk, cream, and various herbs. Mixed drinks without alcohol
that resemble cocktails are known as “mocktails” or “virgin
cocktails”.
The origin of the word a“cocktail” is disputed, but the first
recorded use of cocktail not referring to a horse with cut tail is
found in 1798 (The Morning Post and Gazetteer in London, 1798;
Brown and Miller, 2009), and the Oxford English Dictionary
cites the word as originating in the United States (Oxford
Dictionaries, 2018). The first recorded use of “cocktail” as a
beverage (possibly non-​alcoholic) in the United States appears
in 1803 (Wondrich, 2007). A definition of a “cocktail” known to
be made in reference to an alcoholic beverage appeared in 1806
(Croswell, 1806).
After 1980, new equipment (often from chemistry labora-
tories), techniques, and know-​how were proposed to be applied
to food and drinks under the names “molecular cooking” and
“molecular mixology”, respectively, and they have been quickly
adopted by chefs and bartenders to create interesting cocktails.
In particular, alginate systems, siphons, and liquid nitrogen
have been extensively used all over the world, being now
complemented by “note by note mixology”.
Here, we describe how cocktails of many layers can be
produced using the simple scientific concept of density. After the
creation of the drink called “Welcome coffee” shown by one of us
(H. This) at the bar of the Ritz Hotel in Paris (2005) at a training
seminar, a savoury recipe using the same technical idea was used
by P. Gagnaire.
More Layers than Usually Done: The Welcome
Coffee
The Welcome coffee was designed as an educational tool in
1997, but it was created only in October 2005 by the bartender
of the Paris Ritz hotel, because a seminar for bartenders had been
organized in this hotel. In the Welcome coffee, very simple ideas
were used, and in particular the ranking of layers by order of
densities: water, oil, ethanol, and air. This first group of possi-
bilities is increased when layers at different temperatures are
used and also when sugar is dissolved. Moreover, colloidal layers
can include more than one fluid and have different densities. For
example, a foam is generally less dense than an emulsion.
For the initial “Welcome coffee”, the bottom layer was
proposed to be gelled coffee, prepared using gelatine and a
coffee infusion. Because it was easy to make many layers of
gels, another one (of chocolate) was put over the coffee gel. On
top of this, a cold chocolate drink layer was poured, over which
the same liquid, but this time heated, was added. Because of the
variations of density with temperature, these two layers did not
mix immediately, and the bottom gelled layer was protected from
melting. But of course, in order to avoid the melting of a gelatine
gel, agar-​agar can be used as a gelling agent instead of gelatine.
Over this hot chocolate layer, a liquid with density lower than
water (present in chocolate) had to be used; this was an emulsion
obtained from coffee beverage, gelatine (as surfactant), and an oil
in which coffee had been macerated for one week. This emulsion
has a density almost equal to the density of oil, so that it could
float over hot chocolate without mixing.
For the next layer of lower density, vodka was used; poured
slowly on the coffee emulsion, it could remain there without
827
828 Hervé This vo Kientza, Pierre Gagnaire

mixing, and finally, a foam was made: gelatine was dissolved Recipe 3 –​Custard with Pistachio
in coffee and foamed using a siphon with a nitrous oxide (N2O) • Whip two yolks and sugar.
cartridge. • Boil the milk.
But here again, many foams can be added without mixing, and • Add the boiling milk to the mixture of yolks and sugar.
even other liquids could be used. In the first “Welcome coffee” • Cook the custard.
originally served at the Ritz, only a “wind crystal” (a very light • Dissolve the pistachio paste in the hot custard.
meringue obtained by adding coffee and sugar to whipped egg • Cool, and distribute in the glasses.
white before cooking) was added. Finally, some grilled almonds
Recipe 4 –​Olive Oil with Citrus
with a mixture (2/​1, w/​w) of tartaric acid and sodium hydrogen
carbonate were sprinkled just before serving for a sparkling • Prepare 10 g of orange peel and 10 g of lemon peel.
effect. • Heat the olive oil with the 20 g of peels at a temperature
of 80 °C.
• Let this cool.
• Filter.
A Cocktail for Restaurants • Soak 4 g of gelatine in cold water.
After this cocktail was proposed in one of our monthly discussions, • Press the orange and the lemon, filter the juice.
• Heat the juice, and dissolve the soaked gelatine.
another recipe was designed with layers for daily serving in a res-
• Make a gel on ice.
taurant. This was called “Strawberry, foie gras, pistachio”.
• Using a whisk, emulsify the olive oil into it.
• Pour this emulsified emulsion into the glasses (about
Ingredients for six people two spoons per glass).
500 g of strawberries • Keep in the fridge.
200 g of raw foie gras • When the gelled emulsion is set, add a teaspoon of
60 g of chicken broth kirsch.
30 g of red Port wine
Recipe 5 –​Strawberry Mousse
200 g of full fat milk
2 egg yolks • Soak 4 g of gelatine in cold water.
15 g sugar • Bring to the boil water and sugar, and dissolve the
20 g pistachio paste soaked gelatine.
1 untreated orange • Whip on ice.
1 untreated lemon • Put the foam on the kirsch.
200 g of olive oil
12 g gelatine Finishing
3 teaspoons of kirsch spirit
Into each glass, put a strawberry wind crystal (obtained by
20 g of roasted pistachios
whipping one egg white with a strawberry juice and sugar, and
tartaric acid
sodium hydrogen carbonate cooking like a traditional meringue), pistachios, and a small
6 wind crystals quantity (for example, a quarter of a teaspoon) of the mixture of
salt, pepper tartaric acid and sodium hydrogen carbonate (2/​3; 1/​3).
6 tumbler glasses Of course, producing such a system of layers takes time, and
even if the elementary steps are simple, the production should be
careful in order to get a really fine cocktail.
Preparation
Recipe 1 –​Gelled Strawberry Juice REFERENCES
• Mix the strawberries and filter. Brown J, Miller A. 2009. Spirituous Journey: A History of
• Soak 4 g of gelatine in cold water. Drink, Book Two, Mixellany Limited, New York. ISBN
• Heat 240 g of strawberry juice and dissolve the soaked 978-​0-​9760937-​9-​4.
gelatine. Croswell H. 1806. The Balance and Columbian Repository. Archived
• Cool, and pour into six glasses. 13 July 2014 at the Wayback Machine, May 13, 19 (V) 146.
• Keep in the fridge until the gel is made. www.barnonedrinks.com/​tips/​articles/​shawn-​soole/​molecular-​
mixology.html
Recipe 2 –​Foie Gras Cream Oxford Dictionaries. 2018. Available at: https://​en.oxforddictionaries.
• Using a sieve, transform the foie gras into a paste. com/​definition/​cocktail.
• Boil the chicken stock and the Port wine. Robuchon J. (ed.). 2007. Larousse gastronomique, Larousse, Paris.
The Morning Post and Gazetteer in London. 1798. March 20, 1798, 3.
• Add the hot liquid to the foie gras and mix, so that a
Wondrich D. 2007. Imbibe!: From Absinthe Cocktail to Whiskey
smooth cream is obtained.
Smash, a Salute in Stories and Drinks to “Professor” Jerry
• Add salt and pepper to taste. Thomas, Pioneer of the American Bar. Perigee Trade, New
• Pour this cold cream on the strawberry jelly. York. ISBN 978-​0-​399-​53287-​0.
• Store in the fridge.
Cube of “Chicken-​Carrot” with Chips of “Basil-​Lemon”
Pasquale Altomonte and Dao Nguyen
The Kitchen Lab, 23 chemin de la Tour, Meyrin-Village, Geneva, Switzerland
This dish was created to showcase the different textures achiev-
able when using Note by Note cooking. Instead of having a soft
centre as found in most dishes, we wanted the centre to be chewy
and fibrous (meaty texture). From an optical point of view, we
were looking for vibrant and vivid colours that are commonly
found in conventional foods (for example, green like vegetables
and fruits, yellow like egg yolks). The “egg yolk” look, as well
as the “kiwi” layer, are there to comfort people with something
that is familiar to them. Closing their eyes and eating the cube
will make people travel back to their childhood, remembering a
FIGURE 131.1 Cube of “chicken-​carrot” with chips of “basil-​lemon”.
family dish of chicken, pork stock and carrots with a hint of basil
and lemon.
Visually, this dish seems to be jellified, but in fact, it is not: it
contains a sphere of “chicken-​carrot” in the middle, with a fibrous
consistency. There are two gelatinous consistencies, one above
the other. The one around the sphere is made of pork gelatine, so
that it melts when a hot liquid is poured on it. The upper layer,
with its “kiwi” seeds, has a more elastic consistency in order to
be reminiscent of pork jelly. The chips over the cubic part have a
flavour of basil and lemon.
Sphere of “Chicken-​Carrot”
80 g carrot juice
80 g fibres (for example, carrot or apple)
40 g proteins (for example, egg white powder)
20 g flavoured oil (for example, roasted chicken)
20 g starch (for example, maize starch)
Mix the ingredients, make spheres in moulds and cook in the
steam oven; then put the spheres in moulds containing gelatine
(see “gelatine cube” in the following). In order to get a smooth
aspect to the sphere, add kappa carrageenan around the sphere.
Mix the kappa carrageenan in cold water (12 g/​l), bring up to
boiling point and then remove from the heat. Dip the sphere
(use a tooth pick) three to four times into the kappa carra-
geenan mix.
Gelatine Cube “Pork Stock”
10 g pork stock (cube)
6 g gelatine (powder)
500 g cold water
Dissolve the gelatine in cold water first, then add the cube of pork
stock, bring up to boiling point and pour into moulds. When it is
almost jellified, add the previous sphere.
829
830 Pasquale Altomonte, Dao Nguyen

Mix the ingredients, pour small quantities into a hot frying pan,
“Rind with Kiwi Seeds” Layer cook for two or three minutes, and store on tissue paper.
100 g tapioca starch
100 g water
10 g flavoured oil (for example, pork rind) “Tomato Granité” Juice
10 g ground pepper
2 g colorant (for example, chlorophylls) 80 g tomato juice
salt 3 g agar-​agar
50 g water
Mix the ingredients, bring up to boiling point, add to a cube 2 g gelatine (powder)
Some drops of Tabasco
mould and let it cool before cutting thin layers.
Mix the tomato juice, tabasco and agar-​agar. Bring to the boil, let
it cool until it starts to gel and then mix (keep on the side). Mix
“Basil-​Lemon” Chips the gelatine and water for a couple of minutes in order to activate
30 g wheat flour the gelatine, then heat until the water is clear, then add to the
80 g neutral oil (for example, grape seed) tomato–​agar mixture and put into moulds. Let cool, then cut into
240 g water brunoise (small cubes). It looks like a granité. In contact with a
10 g flavoured oil (for example, basil) hot liquid, the “granité” will melt.
2 g colorant (for example, chlorophylls)
2 g tartaric acid
Some of the Easiest Note by Note Recipes Served at Senses
Restaurant

Andrea Camastra
Restaurant Senses, Bielanska 12, Warsaw 00-​085, Poland

In 2017, the restaurant Senses in Warsaw, Poland, became the first

entirely note by note (NbN) restaurant in the world. It took about
six months to make the transformation, after lectures in which
Hervé This demonstrated NbN cooking in Poland. At first, some
traditional ingredients remained among the pure compounds. But
then, the decision was taken to focus on one traditional ingredient
… but this would only be present provided that the main part of
the dish was NbN.
Since then, more and more NbN dishes have been introduced,
and they are served daily. Here, we show only the simplest of the
NbN dishes served at Senses.
Technical notes for all recipes:

1. The proportions of odorant solutions are according to

the weight of the water or to the total grams if described.
At Senses, we make all recipes with 250 g of water,
because this is the minimum weight for a good result;
doubling it up is even better.
2. These odorant solutions are pure odorant compounds
dissolved in a solvent (it can be oil); they are described
either using the name given by the producer, or by ref-
erence to a traditional ingredient of which they are a
reminisence, or by their chemical name (remember that
the solutions have to be very dilute).

Recipe 1. NbN “Cucumber” and Buttermilk Spheres FIGURE 132.1 NbN cucumber.
(Figure 132.1)

250 g water Process:

8.5 g skimmed milk proteins
12 g lactose 1. Put all the ingredients (except lactic acid and odorant
9.75 g neutral oil solution) in a Thermomix.
0.9 g xanthan gum 2. Set the temperature at 70 °C and blend together.
0.5 g lactic acid 3. Once the mixture is homogeneous, add the odorant
white colorant solutions and the lactic acid while the Thermomix is
to taste: odorant solution “Onium” (Iqemusu) running.
to taste: odorant solution “Piovo” (Iqemusu) 4. Add salt to taste.

831
832 Andrea Camastra

5. Take the whole mixture and freeze in a spherical silicon

mould with a dispenser.
6. While the mixture freezes, prepare the cocoa butter.

For the Coating

cocoa butter
green food colorant

Process:

1. In a bain marie, melt the cocoa butter with the colorant;

disperse well.
2. When the spheres are frozen, extract them from
the mould.
3. Attach them to a toothpick and dip them into the cocoa
butter when the cocoa butter is around 60 °C.
4. Remove from the toothpick.
5. Let them defrost in the fridge slowly.

Recipe 2. NbN “Nuts” Air Bubbles (Figure 132.2)

250 g water
5.5 g egg white proteins (albumin)
0.3 g xanthan gum
2 g salt
5 g fructose
to taste: dilute solution of sotolon
Process:
1. Combine all the ingredients except the egg white
proteins.
2. Blend with a hand blender.
3. Heat at 80 °C in a Thermomix or in the pan (in this
case, control with a thermometer or a thermocouple).
4. Once it is hot, cool down to around 20 °C.
5. Then, add the egg white proteins and blend.
6. Let it rest in the fridge for 2 hours.
7. Put the liquid in a bowl big enough that the liquid fills
half of the bowl.
8. Take an aquarium pump.
9. Put the tip of the aquarium pump tube in the cup and
switch the pump on; big bubbles will form.
10. Let these bubbles fill up the cup and burst them, as the
first bubbles are not clear in colour.
11. From this point on, make bubbles when needed, and
take them gently from the cup with a spoon and serve.
Recipe 3. NbN “Goat Cheese” Frozen Pearls
250 g water
12 g lactose
8.5 g skimmed milk proteins
FIGURE 132.2 NbN nutty bubbles.
2.5 g xanthan gum
3.5 g salt
to taste: odorant solution “Chapre” (Iquemusu)
to taste: odorant solution “Earthy” (Iqemusu)
75 g glucose syrup
40 g neutral oil
2 g pork gelatine (powder)
Process:
1. Blend all the ingredients except the neutral oil, the
odorant solution and gelatine.
2. After blending well, put all of the mixture in a
Thermomix set to 60 °C.
3. In a pan, put together the neutral oil and the odorant
solutions, and warm up to around 75 °C.
4. When the Thermomix is hot, put it on a medium-​low
speed and emulsify the oil mixture.
5. Then, add the gelatine.
6. Give a last blend on high speed for around 20 seconds.
7. Put the whole mixture in a squeeze bottle and let it
cool down in the fridge for around 3 hours.
8. When chilled, give bottle a good shake.
9. Prepare a liquid nitrogen bath, put safety glasses on
and make sure the room is well ventilated.
10. Drop some “goat cheese” mixture in the liquid
nitrogen to form small pearls.
11. Remove the pearls from the nitrogen and store them in
the freezer on a pre-​frozen cup; use when needed.
Easy Note by Note Recipes from Senses 833

Recipe 4. NbN “Beetroot” and NbN 9.75 g neutral oil

“Gorgonzola” Rocks 2 g gelatine leaf (180 bloom)
0.7 g xanthan gum
Part 1 7.5 salt
250 g water 0.25% odorant NbN “Blue cheese” (heptanone)
5 g fructose
2 g salt Process:
35 g glucose syrup
0.4 g xanthan gum 1. Blend together water, xanthan gum and odorant solu-
red colorant tion “Blue cheese”.
3 g egg white powder (albumin) 2. Put in a pot and bring it to around 65 °C.
1.5 g gelatine leaf (180 bloom) 3. When this is hot, reduce it to around 50 °C.
to taste: odorant solution “Beetroot” (Iqemusu) 4. Except for the addition of the glucose, use the pro-
cess steps from the “Beetroot” recipe in exactly the
Process: same way.
5. When both siphons are chilled, prepare a tray by
1. Blend together water, xanthan gum and odorant solu- brushing with a little neutral oil and cover the tray with
tion “Beetroot”. baking paper.
2. Bring to the boil. 6. Flow the two siphons alternately into the tray.
3. Once boiled, add glucose syrup and mix well. 7. When this is done, place the tray immediately in the
4. Cool down to around 50 °C, and put in the Thermomix, blast chiller to freeze solid.
which has been pre-​set to 50 °C. 8. Cut or break with your hands, as you wish.
5. While the Thermomix is running, add all the other 9. Keep them in the freezer until they are ready to serve.
ingredients and blend well.
6. transfer the mixture into a siphon
7. Close and charge with two charges of nitrous oxide
(N2O). Recipe 5. NbN Bread Grissini (Figure 132.3)
8. Let rest in the fridge in a ice water bath for at least 3 250 g water
hours. 41 g egg white proteins (albumin)
3 g xanthan gum
9 g isomalt
Part 2 1.5 g salt
250 g water 0.6% of the total grams white vinegar (acetic acid solution
0.5 g lactic acid in water of 4.5% of the volume)
8.5 g skimmed milk proteins 0.3% of the total grams odorant solution “Bread crust”
12 g lactose

FIGURE 132.3 NbN bread.

834 Andrea Camastra

Process:

1. Blend together water, milk proteins, lactose, xanthan

gum, odorant solution “Potatoes” and kuzu.
2. Bring the mixture to the boil to thicken the kuzu.
3. Once thick, put the mixture in a Thermomix.
4. With the motor running, add the maltodextrin and
the oil.
5. Take out the mixture, and with a spatula spread it thin
and let dry at around 55–​60 oC until fully dry
6. Set a fryer at around 170 °C and fry the tuile until
crispy; this should take around 5 seconds.

Recipe 7. Smoked Bell Pepper NbN Ketchup

(Figure 132.5)
250 g of water
1.5 g xanthan gum
12.5 g kuzu (Sosa)
3 g salt
5 g fructose
1 g white vinegar or acetic acid solution 8% of the volume
of water
0.1% odorant solution “Sfumo” (Iqemusu)
0.35% odorant solution “Piovo” (Iqemusu)
0.1% odorant solution “Onium” (Iqemusu)
red colorant
FIGURE 132.4 NbN potato.

Process:
Process:
1. Take water, xanthan, kuzu, salt and fructose.
1. Blend all the ingredients except the odorant solution 2. Blitz and boil.
“Bread crust”. 3. Put in a Thermomix and blend in all the other ingredients
2. When all is blended well with a hand blender, put the at 60 °C.
mixture in a Kitchenaid bowl with the whisk attachment 4. Cool and serve, or store in the fridge. If bubbles form,
and whip the mixture until very light and fluffy; it pass it through a vacuum machine in a tray and remove
should be like a meringue. the air, making sure the solution is cold before starting.
3. At this point, add the odorant solution drop by drop on
the medium-​low speed.
4. Put the mixture in a piping bag and form grissini shapes
Recipe 8. Orange Blossom Sorbet
in a dehydrator tray covered with baking paper.
5. Dehydrate at 60 oC for 12 hours or overnight. 250 g water
6. Before drying, sprinkle with any kind of seeds if you 1.25 g xanthan gum
wish, but this is optional. 25 g fructose
5 g malic acid
50 g glucose syrup
0.3% of total grams NbN orange blossom water or solution
Recipe 6. NbN “Potato” Tuiles (Figure 132.4) of linalool
250 g water red or orange colorant
0.85 g skimmed milk proteins
1.2 g lactose Process:
1 g neutral oil, or reproduction of truffle oil
1.25 g xanthan gum 1. Blend all the ingredients together.
10 g kuzu (thickening agent, Sosa brand) 2. Put it in a Pacojet canister.
2 g maltodextrin 3. Freeze until solid.
0.2% of total grams odorant solution NbN “Potatoes” 4. Run in the Pacojet, and sorbet is ready.
Easy Note by Note Recipes from Senses 835

FIGURE 132.5 NbN ketchup.

FIGURE 132.6 NbN orange blossom sorbet.

The Forest Floor

Sophie Dalton
School of Culinary Arts and Food Technology, College of Arts and Tourism, Technological University Dublin,
City Campus, Ireland

The “Forest Floor” dish was created under mentoring by Pauline of Note by Note cooking. Each recipe is presented separately in
Danaher and Róisín M. Burke for the 4th International Note by the following section.
Note Competition, which was held in Paris in 2016. The com-
petition brief required the use of cellulose and/​or derivatives
of cellulose and also trigeminal compounds. In the following
Mushroom Meringue Recipe
recipes, methylcellulose and the trigeminal compound allyl iso-
thiocyanate were used. Allyl isothiocyanate is a constituent of Ingredients (yield: 22 mushrooms)
mustard, horseradish and wasabi, and certain other vegetables. Ingredient Breakdown of ingredients Quantity Supplier
Note by note cooking requires a different approach than trad- Egg white Powdered hen’s egg proteins 60 g Louis
itional cooking. Instead of recreating conventional foods, it was powder Stabiliser: E415 Francois –​
decided to create a dish that captured and represented a place and Acidifier: E330 Blanc Gallia
time in nature. In order to achieve this, the odours, textures and Expansion Agent: E1505
feeling of the dish had to evoke a sense of the chosen concept: a Sucrose Caster sugar 74 g N/​A
forest floor in autumn. Water Tap water at room temperature 120 g N/​A
The creator used familiar flavours and unfamiliar ingredients “Umami” Tomami: 100% tomato 4.5 mL Tomami:
flavour concentrate tomato
to make a dish that was both recognisable and challenging. By concentrate
creating elements that represent or reflect what is often found in Orange Water 4.2 mL Wonder –​
our environment, mushrooms, soil, leaves and moss, they reduced colour Colour –​  E110 Orange
some of the alien concepts of cooking with “chemicals” while Acidity regulator: citric acid SY Liquid
also showcasing the benefits and endless possibilities for the uses Preservative: potassium Colour
sorbate
Mushroom 1-​octen-​3-​ol 1 mL Mane
odour

Equipment list:
Equipment Quantity Model
Steel bowl (medium) 2 N/​A
Kenwood mixer with whisk 1 Major
attachments
Disposable piping bags 2 Plastic
Dessert spoon 2 Steel
Balloon whisk (medium) 1 Steel
Plastic pipette 2 Disposable
Rubber spatula 1 N/​A
Weighing scales 1 Calibrated scale (any)
Weighing scales: mini 1 Mini scales: 0.1 g
increments
Silicone parchment paper 1 20 × 40 cm
Steel measuring jug 1 1 L Jug
Steel bowls (small) 3 N/​A
FIGURE 133.1 The Forest Floor dish is made up of four elements: Deck oven 1 Tom Chandley Deck-​Oven
“Mushroom meringues”, “bacon soil”, “horseradish, basil and pea sponge”, Convotherm oven 1 OEB 6.10
and “basil and matcha tea leaves”.

837
838 Sophie Dalton

Equipment Quantity Model

11. The white meringue is to act as a mushroom stem and
the orange as the bulbous top.
Flat oven tray 1 Large
12. The piping bags have no nozzle but are cut using a
Office knife 1 Wustoff 3″
sharp knife. The white meringue piping bag is cut so
Method: that the diameter of the hole is 1 cm. The orange mer-
ingue piping bag is cut to a diameter of 1.5 cm.
1. Carefully weigh the ingredients. 13. The meringues are piped directly onto parchment
2. Add the egg white powder and water into a medium paper. The orange mushroom caps were 4 cm in diam-
steel bowl. eter and 3 cm in height. The white stems of the mush-
3. Use the balloon whisk to smooth the mixture and room were 2 cm in diameter and 3 cm in height. They
ensure there are no lumps. were piped so that the base was thicker and came to a
4. Place into the Kenwood mixer and whisk on high point at the top.
speed for 20 seconds. 14. Using a clean plastic pipette, gently dab the orange
5. Add the sugar and whisk on high speed for one minute meringues with white meringue to create a polka dot
(the mixture should look glossy). effect.
6. Add the mushroom flavour and whisk for 10 seconds 15. Pipe the meringue onto a small piece of silicone-​based
on high speed to incorporate it. parchment and place on a microwave-​safe plastic tray.
7. Split the mixture 2:1 into two bowls. 16. Microwave the meringues until they have expanded
8. The larger portion should then be coloured a deep (15 seconds in an industrial microwave at full power).
orange colour and flavoured using Tomami flavour. If they are cooked for too long, they will shrink and
9. The orange meringue can then be placed into a become wrinkly.
piping bag. 17. They then need to be left on the tray to cool. They
10. The white meringue should be placed into a separate must not be touched initially, as touching damages the
piping bag. surface finish of the product.

Bacon Soil Recipe

Ingredients:
Ingredient Breakdown of ingredients Quantity Supplier
Cocoa butter Fat: cocoa butter 97 g DGF Service
Olive oil Olive oil: extra virgin 55 g Don Carlos
Orange colour Water 4.2 mL Wonder, Orange SY Liquid Colour
Colour: E110
Acidity regulator: citric acid
Preservative: potassium sorbate
Red colour Water 2 mL Mallard Ferrierre –​Rouge Framboise
Carmoisine (E122) 1.7%
Green colour Water 2.5 mL Mallard Ferrierre –​Vert Pistache
Colour: tartrazine (E102) 0.71%
Colour: patent blue (E131) V 0.49%
Sodium benzoate (E211) 0.2%
Smoked bacon aroma Invert sugar 0.5 g Sosa Bacó Fumat
Aroma
Glycerine (E422)
Bacon flavour 2,4,6-​tris (2-​methylpropyl)-​1,3,5 dithiazinane 1g Mane
Salt NaCl 2g N/​A
Soy lecithin Soy lecithin 12 DE 2g Sosa Lecithin
Maltodextrin Maltodextrin 12 DE 40 g
The Forest Floor 839

Equipment list: Method

Equipment
Steel bowl (medium)
Dessert spoon
Balloon whisk (medium)
Plastic pipette
Weighing scales
Steel measuring jug
Small pot
Thermo-​mix
Vacuum pack bags (medium)
Steel bowls (small)
Quantity Model
1. Weigh out all ingredients, keeping maltodextrin
2 N/​A
separate.
2 N/​A
2. Place the fats into a medium metal bowl and melt over
1 N/​A
a bain marie.
3 Disposable
1
3. Once melted, add the colours and flavours to the fats,
1 1 L Jug
and also add 2 g of soy lecithin.
1 Stainless steel, heavy bottom 4. Blend in a Thermomix at speed 7 for 2 minutes to emul-
1 TM5 sify the colours and fats.
2 Heat safe 5. Allow to cool for 10 minutes at room temperature in a
3 N/​A medium steel bowl.
6. Mix with the maltodextrin and break into small pieces
that resemble soil; break into small crumbles for plating.

Horseradish, Basil and Pea Sponge Recipe

Ingredients:
Ingredient Breakdown of ingredients Quantity Supplier
Water N/​A 120 g N/​A
Olive oil Fat: Olive oil 12 g Don Carlos
Basil flavour Glycerine (E422) 4 drops Sosa
Natural basil extract
Recommended dose: 0.2 g/​kg
Chlorophyllin Glycerine (E422) 6.2 g Sosa
Natural colour: obtained from vegetables Cu-​chlorophyllin
(E141)
Green colour Water 2.5 mL Mallard Ferrierre –​Vert Pistach
Colour: tartrazine (E102) 0.71%
Colour: patent blue (E131) V 0.49%
Sodium benzoate (E211) 0.2%
Egg white powder Powdered hen’s egg albumin 15 g Louis Francois –​Blanc Gallia
Stabiliser: E415
Acidifier: E330
Expansion agent: E1505
Pea aroma Aroma 6 drops Sosa
Inverted sugar
Glycerine (E422)
Methylcellulose Methylcellulose 5g Sosa Methylcellulose F50
Pea protein Pea protein isolate 10 g Pulsin
Allyl isothiocyanate Allyl isothiocyanate solid solution on starch 0.1 g Mane
Bacon flavour 2,4,6-​tris (2-​methylpropyl)-​1,3,5 dithiazine 0.1 g Mane

Equipment list:
Equipment Quantity Model
Steel bowl (medium) 2 N/​A
Dessert spoon 2 Steel
Balloon whisk (medium) 1 Steel
Plastic pipette 3 Disposable
Weighing scales 1 Mini scales 0.1 g increments
Steel measuring jug 1 1 L jug
Plastic tub 1 Microwaveable
Small siphon 1 iSi 0.5 L siphon
N2O charges 2 N/​A
Microwave oven 1 Industrial microwave oven: Buffalo programmable commercial microwave oven (1500 W)
Small plastic tray 1 N/​A
840 Sophie Dalton

Method: Equipment list:

Equipment Quantity Model
1. Weigh out all of the ingredients and place into a medium
Steel bowl (medium) 2 N/​A
steel bowl.
Plastic pipette 2 Disposable
2. Whisk the ingredients together until a smooth paste is
Offset palette knife 1 Small size, 1 cm wide
formed. The consistency of the sponge batter should
Deck oven 1 Tom Chandley deck oven
be runny. Weighing scales 1 Calibrated scales
3. Pour the batter up to the line inside the siphon jug. Silicone parchment paper 1 20 × 40 cm
4. Close the lid securely and add one N2O charge to the Steel measuring jug 1 1 L jug
siphon. Card board 1 1 mm cake board
5. Shake vigorously for 30 seconds. Steel bowls (small) 3 N/​A
6. Test by piping some of the mixture into the plastic con- Flat oven tray (large) 1 Heavy steel
tainer. The mixture should be highly aerated, like a Office knife 1 Wustoff 3″
mousse.
7. The second charge may be required to achieve this tex- Method:
ture. Once added, the siphon will require 30 seconds of
further shaking. 1. Carefully weigh all of the ingredients and place all
8. The sponge is then microwaved in an industrial micro- together in the medium steel bowl.
wave at full power for 1 minute. 2. Mix, and allow to sit in the fridge for an hour or two or
until set into a thick paste.
Once cooked, the sponge is tipped out of its plastic container onto 3. Cut leaf shapes into the card to use as a template.
a clean plastic tray and allowed to cool. 4. Prepare the oven tray, and prepare a sheet of silicon
parchment paper.
5. Using the small offset palette knife, smooth the paste
Basil and Matcha Tea Leaves Recipes into the template, leaving a leaf shape on the paper.
6. Place in an oven at 200 °C for 7 minutes.
Ingredients:
7. Allow to cool.
Breakdown of
Ingredient ingredient Quantity Brand Dish assembly:
Water N/​A 70 g N/​A
Olive oil N/​A 12 g Don Carlos Build the dish by creating a small bed of soil and placing small
Methylcellulose Methylcellulose 4g Sosa pieces of the green sponge moss on top. Place the mushroom in
Matcha green tea Organic Japanese 2g Organic the soil and carefully arrange the autumn leaves to represent the
powder Koyu Matcha 100% Japanese
matcha green tea
forest floor.
Egg white powder Powdered hen’s egg 7g Louis
albumin Francois –​
Stabiliser: E415 Blanc Gallia
Acidifier: E330
Expansion agent:
E1505
Whey powder Whey protein isolate 5g Pulsin
Basil flavour Glycerine (E422) 4 drops Sosa
Natural basil extract
Recommended dose:
0.2 g/​kg
A Note by Note Macaron
Julien Binz
Restaurant Julien Binz, Ammerschwihr, Alsace, France
Ammerschwihr is a small village in the middle of an Alsace vine-
yard, located between France and Germany. Here, a restaurant
was created in December 2015: the culinary style is modern, but
based on traditional and seasonal products. For example, the egg
+ green asparagus, with truffle foam, the lobster with Gruyère
cheese broth, the sea bass with a sauce bouillabaisse and the
modernized chestnut torch are now signature dishes.
In February 2018, a special event was organized: a 100% note by
note dinner for paying customers, because I wanted to understand
exactly what it was and to take my cooking in a new direction. Note
by note cooking has been theorized for about 14 years but seldom
practiced. I decided to make tests starting from scratch in order to
make brand new dishes, and I wanted to invite my guests to do the
same, as they were ready to eat entirely new dishes. I did not want
to demonstrate that note by note dishes are “better”, but I wanted
to show that one can make a dish in an entirely differently way.
I wanted to show “dishes of the future” to my guests.
About two months of daily experimentation were needed
to create this new menu (on top of the regular work at the res-
taurant); I now feel that with a research laboratory and a special
dedicated team, the possibilities are huge. As there is no book
entirely focused on note by note recipes today, I made systematic
tests until I got satisfactory results. The menu that we served can
still be improved, but we are proud to have made it, and this is why
I accepted the idea of serving it to the guests of the restaurant. We
are pleased to have contributed to this exciting enterprise.
For this menu, I used ingredients such as plant proteins from
peas, milk proteins, egg proteins, polysaccharides such as agar-​
agar, maize starch (85% pure amylopectin), carrageenans, water,
sucrose and glucose, but also various emulsifiers, choosing, of
course, perfectly safe food ingredients.
For the odours, I used “evocations” from the Iqemusu com-
pany: these products are odorant compounds dissolved in organic
oil. This is brand new, because the cook can now create his/​her
own flavour, based on the mixture of many odorant compounds,
as perfumers create perfumes. Instead of using ready-​ made
flavourings, one can make an infinite number of new flavours!
And the word “evocation” deserves its name: for example,
only one odorant compound is in the bottle called “Amerise”,
corresponding to the evocation of cherry and almond. Some of
the Iqemusu “evocations” are wonderful, such as Frum (liquor,
fruity, rum), Mapo (cacao, plant, fruit) or Chapre (goat cheese,
leather, fat). Today, more than 30 “evocations” are available,
and the number of possible odorant compounds is in the tens of
thousands; this is exciting. For my menu, I used about 15 different
odorant compounds. And I want to thank the Groupement
d’études des protéines végétales, and the Ingredia Company, for
providing me with proteins.
Note by Note Cooking Can Teach a Lot
I love fresh food ingredients, and our restaurant will not change
entirely. Of course, the new techniques that were used for this
exceptional dinner will be further used in some other recipes …
because they are so good. And we shall use the new techniques
for making dishes when guests have allergies, for example. Last
week, a lactose-​intolerant guest was able to taste the “revisited
egg”, and she was so happy, so surprised by the new textures and
perfumes. This is another advantage of note by note cooking. Let
us now consider the menu.
Evocation No. 1: Craquante (Figure 134.1)
This amylopectin crackling tuile, Kroepoeck and gel (evocation
grilled rice/​popcorn) sponge cake (evocation nut) was served
with a Crémant d’Alsace from Q.V. Paul, Bott Frères.
Evocation No. 2: Voluptueuse
This sorbet (evocation beetroot) cloud (evocation almond/​cherry)
was served with a note by note cocktail.
Evocation No. 3: Cremeuse
In a shell, custard (evocation cheese/​ egg), siphon (evocation
truffles/​forest). In the plate, “evocation fried egg”: egg proteins,
evocation grilled rice, gel evocation fenugreek/​curry, yellow evo-
cation hay/​celery, grilled emulsion, served with a Riesling lieu-​
dit Scherwiller, Achillée, 2016.
Evocation No. 4: Fondante
Plant proteins, evocation smoke, carrot cellulose, veil of gel
(evocation roasted chicken), emulsion (evocation mushroom/​
841
842 Julien Binz

FIGURE 134.1 The first dish of the 100% NbN menu.

forest), emulsion grilled bread, proteins croutons. Served with a

Faugères, Cuvée Amour, Prés Lasses, 2015.

Evocation No. 5: Aerienne

This tuile of sucrose/​white (evocation foamed, evocation caramel/​ FIGURE 134.2 On this board for mignardises, the green macaron is accom-
panied by a “bonbon ultime”.
strawberry jam), granite (evocation coconut/​milk) was served
with a Muscat, Grand Grand Cru Altenberg de Bergbieten, (Picture S. Kauffer)
Mochel, 2015.
Gelified slice in the middle:
Evocation No. 6: Moelleuse
It included a guimauve (evocation cola-​lemon), a “Bonbon ultime” To the basic gel, add the Pine evocation (Iqemusu).
and a macaron (evocation cucumber/​pine tree). Pour on a table to give a thickness of 3 mm; wait until
you can cut disks.
Here, we give only one recipe, because the technique was
entirely original: the final green macaron (Figure 134.2).

NbN macaron:

The Recipe for the Macaron 294 g water

For the basic gel, in the middle: 40 g egg white proteins
80 g sugar (sucrose)
500 g water 1 g xanthan gum (one should be careful to test before use,
3 sheets of gelatine because depending on the various brands, different
20 g glucose powder results can be obtained)
20 g sugar (sucrose) Green colorant (Louis François)
xanthan gum (Louis François)
5 g agar-​agar (Louis François) Whip all the ingredients until the consistency of a mer-
green colorant (“pistache”, Louis François) ingue is reached, put on Teflon and store in the dehydration
Evocation Cong (Iqemusu) machine for 4 h at 60 °C.
Evocation Hertzon (Iqemusu) Of course, the flavour is difficult to explain … and the
quantities of the various evocations are important, but dif-
Boil all the ingredients together for some minutes, add the ficult to describe in a recipe.
colorant and the evocations. Let it cool until jellification.
Mix in order to make a soft gel.
Note by Note Cooking
Michael Pontif
Iqemusu Inc., Paris, France
Note-​by-​note cooking (NbNC) is a way of cooking that uses
pure compounds to create dishes (This, 2014). As with molecular
cooking in the 1990s, it is a big breakthrough, but of a different
kind: where molecular cooking was implementing new tools and
techniques imported from the chemistry and physics laboratory
(mainly), NbNC uses new food ingredients. The cook needs to
discover the various compounds, their sensory properties and
their interactions with other compounds. It involves a sense of
formulation, which is not intuitive and only comes from experi-
ence. To be able to create note by note flavours, cooks need a
guide to know how to proceed.
Introduction
Where did the name of note by note cooking come from? Hervé
This vo Kientza chose this name in 1997 to evoke the way
musicians compose music using pure waves in electroacoustic
or synthetic music (This, 2014). Here, with NbNC, the question
is to invite chefs to compose their dishes in the same way, but
with pure compounds instead of defined acoustic frequencies.
The chosen name is interesting, because it brings cooking closer
to another artistic discipline: perfumery. Indeed, perfumers call
their working material “notes”, and they compose perfumes with
notes and chords.
If the parallel with the notes found in perfumery seems appro-
priate, we cannot use the same definition for each category of
notes in cooking. In perfumery, the base notes last for several
days or even weeks, the middle notes last for one day and the top
notes last for only a few hours (Poucher, 1993). Classifying notes
according to these categories does not make sense in cooking,
since a dish is consumed in minutes.
Another definition should be used for different categories of
flavour compounds. The base notes will group all the compounds
that have an action on the tongue (Macdonald, 2007). Among the
best known are sucrose and sodium chloride, sour compounds like
citric acid or acetic acid, and compounds having an effect on the
trigeminal nerve, like piperine, menthol and many others. These
compounds are found in all foods and compose the basis of taste.
We shall call middle notes the compounds that create the
odorant profile of the food; they are fundamental for overall
flavour, because they enable us to differentiate pork from beef,
banana from apple, or salmon from shrimp. There are thousands
of them, and their combined presence creates all the variations of
flavour that can be found in nature.
The last category, the top notes, will only include very vola-
tile compounds. These top notes are of great importance for the
product’s fragrance. Some are much more subtle once in the
mouth, while others will be used for their olfactory contribution
but not tasted, as we will see later.
To be able to create note by note flavours, it is necessary to have
in mind these three structural parts of flavour. This will help to
proceed methodically and to be able to correct any errors in order
to arrive at the desired result (Burdock 2009; InformedHealth.org
2006; Maarse 1991; Rowe 2000).
The Base Chord
The base chord contains all the compounds that create what chefs
call “the balance” of a dish. These compounds are in direct inter-
action with the taste buds of the tongue and provoke sensations
that everyone is able to recognize without any particular training:
sweetness, saltiness, acidity, bitterness, spiciness, freshness,
umami, fattiness, etc. (Poucher, 1993).
The evaluation of the base chord can be done by tasting (in
particular when the nostrils are pinched). However, some com-
plementary tools allow the results to be confirmed in order to
obtain regularity in the making of a dish. A refractometer will
give the precise percentage of sugar in a solution, a conductivity
meter the salinity, and a pH meter the acidity or alkalinity. With
the help of these tools, the search for balance is more efficient and
the reproducibility of dishes assured.
Some of the most common sugars are D-​glucose, D-​fructose
and sucrose, this last being our common table sugar.
For acidity, we can use malic acid (2-​hydroxybutanedioic acid,
the main acid of apples), citric acid (2-​hydroxypropane-​1,2,3-​
tricarboxylic acid, lemon), acetic acid (vinegar), ascorbic acid
(vitamin C), lactic acid (2-​hydroxypropanoic acid, dairy products)
or phosphoric acid, which is very interesting for its acidity but
also to provide a “dry” taste. Each acid can be used alone for its
taste, but combinations with other acids will create new acidities.
In foods, we would never find a single acid but a combination of
several of them. Of course, the right dilutions have to be used.
Salt is an essential flavour modifier, and a concentration of
about 0.5% is common in the dishes we eat.
843
844
Bitternesses are generated mainly using quinine, while umami
is a name that was given to the taste of sodium monoglutamate.
Other compounds can have other taste effects, such as ethanol
and sodium bicarbonate.
In traditional occidental cooking, black pepper is used daily
in combination with salt: it brings not only taste and odour but
also a pleasant sensation of heat that spices the dish, increases
the flavours and gives a feeling of well-​being due to a release of
endorphins (hormones of pleasure) (Lee, 2012) related to inges-
tion. In fact, one compound in black pepper is mainly respon-
sible for this sensation of heat: piperine (McNamara, 2005). Its
effects are remarkable and can be used in almost any context,
since it provides the “heat-​sensation” without any flavour. One
of its particularly interesting uses involves its integration into a
fresh liquid or an ice cream. Well balanced, this provides first the
freshness of the product and then some time later, the heat from
the piperine. It makes the brain want to take another spoonful
or sip to refresh this heat. This creates an endless cycle of satis-
faction, where the pleasant heat is momentarily softened by the
coolness and then comes back.
In contrast to piperine, menthol (5-​ methyl-​2-​
(propan-​
2-​yl)cyclohexan-​1-​ol) and eucalyptol (1,3,3-​trimethyl-​2-​
oxabicyclo[2.2.2]octane) create a cool sensation. They find their
use in the creation of cocktails and drinks. They can also be used
in perfumes on the edges of a glass, as their decongestant prop-
erties will increase the olfactory capacity of the taster and leave
them with a pleasant freshness on the lips.
The balance of these flavours is fairly well documented in trad-
itional cuisine, but the use of pure compounds allows total con-
trol and regularity. Sea water is never used to add salt to a dish,
because pure salt is more convenient to handle and more effective
in the deployment of its properties. Acids, sugars and other pure
compounds can also now be used in their pure versions for the
creation of delicious base chords.
The Middle Chord
The creation of a middle chord is slightly more complex, because
apart from some specialists, we have not been trained to develop
our sense of smell, and this is the main sense at stake here. The
middle chord will consist of two types of notes: the main notes
that determine the “skeleton” of the flavour and the secondary
notes that give nuances (cooked, raw, creamy, fruity, ripe, green,
fresh, etc.). The tasting exercise of a wine goes in search of these
secondary notes, the main notes constituting the skeleton of the
wine being detected rather easily.
Take the example of the almond: benzaldehyde will mainly
be used because in this case, it represents quite well the skel-
eton of the aromatic profile of almonds. However, benzaldehyde
is also present in cherries, and alone, it leaves the brain of the
taster the choice to evoke cherry or almond. To get closer to
the aromatic profile of the almond, we can combine benzalde-
hyde with other compounds, the secondary notes, like pyrazines
such as 2,3-​dimethylpyrazine, which will give it a nutty tone,
dihydrocoumarin, which can bring mellowness, and pentan-​2,3-​
dione, which will give a creamy and buttery side. Adding these
Michael Pontif
three notes in the right proportions can be a good start for the
creation of an almond middle chord.
Some skeletons come from the combination of more unex-
pected compounds. Isoamyl acetate has a sweet smell and banana
flavour, and adding to it a little bit of eugenol (a major component
of clove oil) gives a perfume with much more depth and authen-
ticity. However, it should not be overdosed, and therefore small
consecutive additions of eugenol should be tested (Wright, 2004).
Complementing with secondary notes such as ethyl propionate
should give a very ripe fruit tone, while small doses of furaneol
(4-​hydroxy-​2,5-​dimethylfuran-​3(2H)-​one) and maltol can also be
used for sweet tones.
For vegetables, among the most recognizable notes are trans-​2-​
cis-​6-​nonadienal, with the characteristic smell of fresh cucumber,
and 2-​ isobutyl-​3-​
methoxy-​ pyrazine, also known as galbazine,
which has the odour of green bell-​peppers and nutty tones if
concentrated. Once it is diluted, the pea and green bean tones
of the galbazine will take over. This skeleton of flavour can be
completed with cis-​3-​hexenol and cis-​3-​hexenyl acetate, which
both provide green tones of leaf and cut grass to improve the
overall profile.
To create a chicken fragrance, sulphur compounds can be used
for the middle chord, for example a combination of trithioacetone
and 2-​ methyl-​3-​furanthiol. This skeleton will be completed
with different side notes to get the desired flavour profile.
Acetylthiazole and furfuryl-​mercaptan will work on the roasted
appearance, and 2,3 dimethylpyrazine will give a slight hazelnut
smell suggestive of the Maillard reactions; finally a hint of fat
will be given by trans-​2,trans-​4-​decadienal.
For dairy products and cheeses, the middle chord will be
creamy. A good skeleton of Roquefort cheese will be obtained
with 2-​heptanone, with a small amount of methylthiobutyrate for
the cheesy aspect. The secondary notes will consist of pentan-​
2,3-​dione (butter) and lactones such as gamma-​nonalactone and
gamma-​decalactone, which will bring creamy and fruity tones.
More generally, there are hundreds of other notes that
can be used to form middle chords. Among these, guaiacol
(2-​methoxyphenol) has a woody fragrance, evoking this smoky note
found in products that have been smoked. Different terpenes, such
as limonene (1-​methyl-​4-​prop-​1-​en-​2-​yl-​cyclohexene), myrcene
(7-​methyl-​3-​methylideneocta-​1,6-​diene) or gamma-​terpinene
(4-​methyl-​1-​(1-​methylethyl)-​1,4-​cyclohexadiene), make it pos-
sible to obtain citrus or hop notes. Of course, all this does not
imply that one has to use traditional ingredients, and the mixture
of pure compounds can otherwise lead to a wealth of new odorant
structures.
The Top Chord
Often, one tastes first with the eyes, then with the nose. The cooks
can hardly change the smell of the dishes that they make, and
these dishes smell of what they contain. Yet, in the same way as
with visual perception, the smell of the dish conditions the brain
of the person who tastes. By working on the top chord, we can
highlight certain characteristics of the dish by accentuating the
odours of the food it contains (in the same way as we would seek
Note by Note Cooking
to strengthen the green of vegetables or the red of a raspberry
coulis). It is also possible to create a smell that is not present in
the dish and to spray it on the spoon or on the side of the plate so
that the brain mixes it into the flavour during the tasting.
The work on the top chord uses only odorous compounds. It’s
about doing the work of a perfumer, creating a scent that will
bring a new olfactory component to the dish. This perfume will
be sprayed at the last moment on the dish, or on the container. We
will then find notes that have a strong olfactory impact, because
it is sprayed in small quantities on the surface of the dish without
altering the initial flavour composition.
For some notes, the use in a top chord is interesting to enable
them to express their potential to the fullest. If they have a slight
scent and very little power in the mouth, spraying them on the
dish will make the best use of their olfactory characteristics.
Fresh notes such as cis-​3-​hexenol (freshly cut grass), L-​carvone
(spearmint) or eucalyptol (eucalyptus) are among these.
Other notes have a sweet scent but too much power in the
mouth, which is undesirable in the final dish. This is the case for
the rose, which has a very fresh and sweet fragrance, strongly
characterized by the presence of phenethyl alcohol. Its smell is
extremely interesting, but the presence in the mouth gives an
unpleasant bitterness and makes it difficult to properly dose the
phenethyl alcohol. The use of phenethyl alcohol as a perfume
only, on the edge of the plate or on the handle of the fork, allows
the taster to smell the scent of a fresh rose while eating without
having the inconvenient taste of this compound.
How to Work to Create Note by Note Flavours?
The creation of a dish or of a 100% note by note flavour requires
trials and failures. Speed and accuracy of creation can only be
obtained with experience and good knowledge of the notes’
properties.
A first approach to experience note by note creation in a fun
way is to start from an existing liquid product. This can be a fruit,
a vegetable juice or an infusion, for example. The cook then
faces a definite assemblage of compounds, as the base chord and
the skeleton of the middle chord are already present. They can
now work to improve them, thanks to note by note techniques.
845
Consistency, volume and texture of this liquid will be later added
to create a fluid foam, crunchy mousse, sorbet, gel or emulsion.
The second step is to train on recreating known flavours. By
breaking down a flavour and reproducing it with pure compounds,
a cook will gain an understanding of the power of these
compounds and how they interact with each other. However, the
true purpose of note by note cooking is revealed once the basics
are mastered. From the moment one is able to work on existing
flavours, why not try to create new ones? Nature has chosen a par-
ticular blend of compounds to create apple or salmon. Now that
these compounds are at our disposal, we can consider inventing
new tastes, new fruits or new meats. With more than 2,500 food-
grade compounds authorized by the European Union, possibil-
ities are endless (EU No 872/2012).
REFERENCES
Burdock GA. 2010. Fenaroli’s Handbook of Flavor Ingredients, 6th
edition, CRC Press.
InformedHealth.org. 2011. How does our sense of taste work?
Cologne, Germany: Institute for Quality and Efficiency in
Health Care (IQWiG), Cologne, Germany. Available from:
https://www.ncbi.nlm.nih.gov/books/NBK279408/
Lee JS, Kim SG, Kim HK, Baek SY, Kim CM. 2012. Acute effects of
capsaicin on proopioimelanocortin mRNA levels in the arcuate
nucleus of Sprague-​ Dawley rats. Psychiatry Investigation,
9(2), 187–​190.
Macdonald S. 2007. Composing savory flavors. Perfumer and
Flavorist, Available from: www.perfumerflavorist.com/​flavor/​
application/​savory/​12117421.html
McNamara FN, Randall A, Gunthorpe MJ. 2005. Effects of piperine,
the pungent component of black pepper, at the human vanilloid
receptor (TRPV1). British Journal of Pharmacology, 144(6),
781–​790.
Maarse H. 1991. Volatile Compounds in Food and Beverages, CRC
Press.
Poucher WA. 1993. Poucher’s Perfumes, Cosmetics and Soaps
Volume 2: The Production, Manufacture and Application of
Perfumes. 9th ed., Chapman & Hall.
Rowe DJ. 2004, Chemistry and Technology of Flavors and
Fragrances, CRC Press.
This H. 2014, Note-​by-​Note Cooking: The Future of Food, Columbia
University Press.
Wright J. 2004. Flavor creation. In Flavor Development: Composition
to Innovation, Allured Books.
Note by Note Sushis

Guillaume Siegler
Chef-​teacher at Le Cordon Bleu

This recipe was designed at the School Le Cordon Bleu Tokyo. Method: Mix all the ingredients, bring to a boil and add it to a
Some months later, we proposed it at the 4th International siphon; spray in a trail.
Contest for Note by Note Cooking in Paris in 2016, and we had
the pleasure of winning the Contest in the “Chefs” category.
Obviously, as for any sushi, the recipe has two parts:
“Note by Note” Covering Sheet
• a reproduction of the rice 200 g water
• a reproduction of the fish a small quantity of dithiazine (“Bacon” Note by Note No. 10,
Mane S.A.
8 g agar-​agar
0.5 g monosodium glutamate
Consistency of Rice “Note by Note” 0.5 g tartaric acid
200 g water to taste: salt
12 g agar-​agar 2 g glucose
to taste: tartaric acid 1 g xanthan gum
3 g glucose 1 g allyl isothiocyanate (dispersed in starch, Mane S.A.)
2 g monosodium glutamate Colorants: red, black, yellow
2 g xanthan gum
5 g corn starch Method: Mix all ingredients, bring to a boil, and ladle into a tray
some drops of a “basmati” odorant solution (Note by Note at the appropriate thickness.
No. 8 by Mane S.A.) These sushis were tested during a public lecture in front of the
some drops of acetic acid in solution 10% Japanese press in Kyoto, Japan, in a meeting at the Ritsumeko
University, a partner of the School Le Cordon Bleu, on 29
October 2015.

FIGURE 136.1 Note by Note sushi served at an event in Kyoto, Japan.

847
Slowly Cooked Lamb Neck with Fermented Flour Pancakes,
Sunchoke Puree and Beer Glaze

Alex Tsionitis
CTC (Athens, Greece) –​http://​ctc-​restaurant.com

For the Lamb Apply cling film, place the lamb on top, add salt, pepper and
transglutaminase. Roll tightly with the film and leave at least 12
2 lamb necks (deboned) hours in the refrigerator for the enzyme to work.
salt, pepper Remove the film, place in vacuum bags together with duck fat
duck fat and cook at 82 °C for 12 hours.
Activa® (transglutaminase)

Debone the lamb necks; thoroughly remove the nerves and fibers For the Sunchoke Puree
and cut into strips; weigh in portions of 75 g.
1 kg of sunchokes
½ bunch of lemon thyme
1 L milk
500 ml chicken broth
Beurre noisette
Salt

Peel the sunchokes, cut them with the help of a mandolin into thin
slices (be careful to do this quickly to avoid oxidation).
Place the milk in the pot, along with the sunchokes, the chicken
broth and the lemon thyme.
Simmer until well cooked.
Remove the lemon thyme, strain (keep the broth of cooking)
and blend to get a very smooth puree. Dilute if necessary with the
broth and adjust taste with a little beurre noisette and salt.

For the Fermented Flour Pancake

450 g flour 00
50 g Greek style pasta made from fermented milk (trahanas)
20 g yeast (moist)
10 g salt
11 g sugar
20 g olive oil
250 g water
6 g thyme

Blend the trahanas in a Thermomix® in order to turn it into flour.

Place both types of flour in the mixer bin along with water, sugar,
oil and yeast and knead at low speed for 5–​6 min. Add the salt,
FIGURE 137.1 Slowly cooked lamb neck with fermented flour pancakes, speed up and knead for 3 more min, and finally, incorporate
sunchoke puree and beer glaze. the thyme.

849
850 Alex Tsionitis

Let the dough rest for 25 min, then roll it like pizza in a thin initially in Lyon, where I studied at Institut Paul Bocuse, then
sheet of 3 mm thickness. Cut 10 cm diameter circles. in Paris, at Ferrandi and in various restaurants including Alain
Allow to rise for 10 min and fry in a pan with olive oil (you Passard, Eric Frechon and Pascal Barbot. After that, with a few
can also grill). additional stops in Denmark, Belgium and China, I decided to
return to my home country, where I took over the reins of the
6* hotel Elounda Peninsula. Today I have my own restaurant in
Athens, CTC Urban Gastronomy.
For the Confied Sunchokes
500 g small sunchokes
Beurre noisette How Would You Define Your Cooking Style?
Salt I like to say that a cook’s kitchen is a mirror of his course. So
I think I’m cooking Greek but inspired by France and the
Wash the sunchokes well and put them in vacuum bags with Mediterranean.
butter and salt. Close and cook at 90 °C for about an hour.
The Recipe(s) You Are the Proudest of?
The one I have not discovered yet!
For the Frosting of Black Beer
250 g lamb broth What Ingredients Inspire You More Particularly?
125 g black beer (Crazy Donkey from Santorini)
200 g butter I love citrus fruits because they have many faces at the same time;
0.2 g xanthan they can be sweet, sour, bitter …
100 g sunchoke puree
How to Be Creative and Keep “Fashion” in Cooking?
Boil the lamb broth and add the beer. Incorporate butter, xanthan
You need to know your style and evolve. You do not have to be
and sunchoke puree with a hand mixer.
creative just to impress but to look for the essence.
To serve, sprinkle with a little bit of lime zest and finely
chopped dill.

A Few Words by the Chef

What is Your Professional Background?
I started my career in Greece, where I took my first cooking steps,
and then I travelled to France, where I lived for almost four years,
Index

Absinthe, 233 Alcoholic beverages see also beer; distilled Angularity, plating preferences, 459
Absolute waters, 271 drinks; ethanol; fermentation; liquors; 2,5-Anhydro-D-mannose, 565
Absorption, 265–266 wine Animal husbandry, 415, 416
Acacia gum, 385 see also gum arabic background noise and consumption levels, Animal tissues
Acacia spp., 399, 400 569 chemistry of cooking, 160
Acetaldehyde, 14 bubbles (recipe), 755 as gels, 344
Acetate esters, 330 production methods, 19–21, 22, 23 structure, 158
Acetate pathway, 275 trends and innovations, 21–23 Anise-flavoured liquors, 232–233
Acetic acid yeast flavours, 328–330 Anisidine index, 307
coffee, 143, 478 Alcoholic strength By Volume (ABV in % v/ v), Anogeissus latifolia, 400
kimchi, 321 214, 215 Anthocyanidins, 13, 14, 151, 165
sourdough bread, 60 Aldehydes, 632 Anthocyanins, 13–17, 42, 151, 338, 437, 565
Acetobacter aceti, 683 Aldohexoses, 564–566 Anthocyanin–tannin adducts, 15
2-Acetyl-1-pyrroline, 487 Ales, 19, 21 Antimicrobials, ash, 25
Acetylformoin, 565 Algae see also seaweeds Antioxidants
Acetylthiazole, 844 agar-agar, 387 coffee, 146
Achatz, Grant, 570 alginates, 391, 689 educational activities, 684
Acid-base behaviour, educational activities, carotenes and carotenoids, 152 fats and oils, 308–309
683–684 carrageenans, 389 frying, 316
Acids chlorophylls, 153 meat, 425
chlorophylls effects, 154 structure, 689 nitric oxide (NO), 425
firmness perception, 7 Algebra, 210–211 nitrite salts, 425
maceration, 157 Alginates, 385, 391–395, 538, 589 see also pasteurization, 455
note by note cooking, 843 calcium alginate; sodium alginate tea infusions, 436
pectin hydrolysis, 159 Alginic acid, 385, 392, 393, 689–690 vitamin E, 425
usage with chemical leaveners, 41–42 Aligot, 117–119 Apigeninidin, 13
Acrylamide Alkalinity Apolipoproteins, 299
barbecue, 64 chlorophylls effects, 154 Appert, Nicolas, 87
coffee, 145, 146 Maillard reactions, 42 Apple juice
formation, 67 Alkaloids, physiological function in plants, 276 purification, 337
frying, 315, 365, 366 Alkylpyridines, 68 spray freeze-drying, 174
Maillard reactions, 68, 81, 84–85 Allicin, 794 Apples
structure, 67, 69, 482 Almonds, 844 air content, 241
Actin, 160, 415, 416 Altrose, 563 browning, 684
Actual compound release, 73–74 Amadori, Mario, 633 sauce, 505–514
Acylpyridines, 68 Amadori products, 64, 67, 82, 193, 632 storage treatment, 438
Adenine, 481, 482 Amherst College, 683–686 Aquavit, 233
Adenosine triphosphate (ATP), meat maturation, Amines, 632 Arabinogalactans, 141, 476
30 Amino acids see also proteins Arabinose, 142, 475
Adenylate, 543, 622 cheese browning, 114 Arak, 233, 667
Adria, Ferran, 172, 255, 431, 608, 802 classification, 463 Arginine (Arg), 463, 464, 480
Aesthetics, seaweeds, 517–518 essential, 465 Argumentation, 643–649
Agar-agar, 385, 387–389, 396, 589, 651–653, 805 language usage, 631 Armagnac, 217
Hidden Heart (recipe), 765, 769 microwave cooking, 431 Aroma see odour
Save the Shark (recipe), 764–765 residues, 112, 188, 229, 302, 437, 453, Art, in cooking, 2
Agaropectin, 387 463–465, 468, 582, 631 Artificial intelligence (AI), brewing, 22
Agarose, 387 sourness, 7 Arzark, Juan Mari, 177
Ageing stocks, 76 Ascorbic acid
alcoholic beverages, 328 structure, 463, 464 meat curing, 424, 425
cheese, 411 sweetness, 543 pasteurization effects, 456
meat, 415, 416, 621 umami, 7, 543 prevention of fruit discolouration, 164–165,
AGEs (Advanced Glycation End products), 82, Ammonium chloride, Salmiakki, 128 201, 202
632 Ammonium salts, browning effects, 437 Ash, 25–33
Aioli, 496 Amylase, 58 Asparagine, 67, 84, 366
Air drying, 204 Amylopectin, 53, 159, 160, 402, 471 Aspartame, 463
Air humidity, 203–204 Amylose, 53, 159, 160, 402 Aspartic acid (Asp), 7, 463, 464
Albisu, Victor, 527 Anabolism, 266 Astaxanthin, 152
Albumins Analytical methods see also imaging methods Astringency
eggs, spray freeze-drying, 174 matrix effects, 345 chocolate, 134–136
milk, 182 particle size, 524 coffee, 147
Alcohol acetyltransferases (AATases), 330 in situ quantitative nuclear magnetic grape seeds, 587
Alcohol distillation, 213–219 resonance (is q NMR), 74–76 taste effects, 587
Alcohol fermentation, 59 Anchovies, debye capers/anchovies (recipe), 744 tea infusions, 436

851
852 Index

Astronauts’ food, 195, 205, 352 Slowly Cooked Lamb Neck with Fermented Borax, 437
Attention Flour Pancakes, Sunchoke Puree and Beer Botulism, 88, 426, 435
“sonic seasoning,” 571 Glaze (recipe), 849–850 Bouquet garni, essential oils, 276–278
Temporal Dominance of Sensations (TDS), trends and innovations, 21 Brain, taste sensation, 586
575–578 Beet distilleries, 339 Braising, 157, 533
Aubergines see eggplants Benzaldehyde, 329, 844 Brandy, 23, 221, 223
Autoxidation, 305–307, 454 Benzo[α]pyrene, 64, 529 brandy sorbet (recipe), 759
Avocados, 684 Berries, freeze-drying, 355–356 Bread see also fermentation; baking
Avogadro– Ampere law, 704 Betacyanins, 152 chemical leaveners, 41–42
Avogadro number, 265 Betalains, 151–152 crust colour, 437
Betanins, 152 heat transfer, 383
Bacillus botulinus, 88 Betaxanthines, 152 injera, 43–46
Bacillus cereus, 533, 534 Beverages matrix effects, 343
Back-slopping, 59–60 alcoholic, 19–23, 328–330 see also beer; melanoidins, 82
Bacon and egg ice cream, 178 distilled drinks; ethanol; wine microwave cooking, 91
Bacteria see also fermentation; specific species carbonation, 586 note by note (recipe), 772
chlorophylls, 153 cryo-carbonation, 177 odorants, 486
kefir, 331–333 emulsifiers, 400 sourdough, 57–60
lactic acid, 58–60, 114 emulsions, 233 toasting, 487
meat curing, 424–425 fruitiness perception, 10 Brettanomyces, 21
nitrates/nitrites, 425–427 hibiscus, 16 Brewing, beer, 19–21
pathogens, 426 ice cubes, 172 Brillat-Savarin, Jean-Anthelme, 3, 92, 569, 635,
spoilage, 426 kefir, 331–333 666, 671, 717, 810
thermobacteriology, 88–89 liquid creamers, 301 Brochothrix thermosphacta, 426
Baking see also bread non-alcoholic, 330 Bromelain, 469, 470
butter, 193–194 pasteurization, 452 Broth, definition, 284
definition, 157 BHA, 308 Brownian motion, 716
matrix effects, 345 Bicarbonate of soda see baking soda Browning see also glycation; Maillard reactions
proteases, 469 Bifidobacterium bifidum, 184 anthocyanins reactions, 14
starch gelatinization and texture, 53–55, Bioaccessibility, 71–73, 345, 456 butter, 83–84
343 Bioactivity, 71–79 see compound release cheese gratin, 113–115
3D printing, 610 Bioavailability, 71–73, 345, 435 frying, 365
Viennoiserie pastry lift, 47–48 Bioconversion, 345 language usage, 631
Baking powder, 41–42 Bioefficacy, 345 meat, 421
Baking soda, 41–42 Biot number, 688 melanoidins, 64
Bancroft’s rule, 227 Biphasic systems microwave cooking, 431
Barbecue, 63–69, 420 evaporation, 284–285 oxidation, 684
Barley partition coefficient, 259–260 Bubbles
malting, 339 Biscuits, chemical leaveners, 42 calculation of quantity in whipped egg white,
plant-based drinks, 230 Bisque, 160 705–710
Barley malt, 19 Bitterness Champagne, 97–103
Bay leaves see laurel leaves chocolate, 133–136 cheese melting, 114, 115
beaumé, 222–224 coffee, 147 educational activities, 639, 640
Beaumé, Antoine, 221, 223 irritation effects, 586 foams, 233–235, 243, 244, 246, 357–358,
Beans mineral ions effects, 434 360, 363, 716
baking soda, 42 note by note cooking, 844 ice cream, 194
cassoulet (recipe), 727–729 olive oil and milk interactions, 317 injera eyes, 43–46
chlorophylls, 153, 165–166, 660–661 perception, 7, 9–10 “Nuts” Air Bubbles (recipes), 832
colour, 437, 660–661 sound effects, 571 size and production method, 361
odour, 844 tea infusions, 436 soufflés, 292–293
Béarnaise sauce, 241, 242, 499 umami effects, 622 Bubbles (recipe), 755
Béchamel sauce, 241, 301 Blackcurrant juice, proteases, 469 Buckwheat, beers, 21
Beef Black pepper, 844 2,3-Butanedione, 487
ageing, 416, 621 Blair process, 154 Butanoic acid, 193, 486
diracs (recipes), 745–746 Blanching Butter
inosinate (IMP), 621 colour effects, 154 beurre noir, 193
nitrite salts, 425 ultrafiltration (UF) of juices, 339 beurre noisette, 193
odorant compounds, 284 Blueberry-oat ice cream (recipe), 237 browning, 83–84, 309
pasteurization, 452, 454 Blue colours Chantilly, 106
roasting, recipe robustness, 167 anthocyanins, 14 clarification, 752
salting, 443, 443 pyranoanthocyanins, 15–16 consumption, 193, 195
sous vide cooking, 452, 533–534 Blumenthal, Heston, 172, 178, 398, 431, 802 Dispersed System Formalism (DSF), 195–196
Beer Boiling fermentation, 193
carbonation, 235–236 definition, 157 hollandaise sauce, 499–504
category blurring, 23 ebullition, 382 imaging methods, 410
foams, 235–236 heat transfer, 383 odour, 844
low-alcohol, 338 latent heat, 172 production, 191–192
microfiltration, 337 milk, 185–186 structure, 192, 195–196, 207, 230, 630
production methods, 19–21 vapour pressure, 265 texture, 192–194
proteases, 469 Boltzmann constant, 124, 705 vitamins, 230
Index 853

Butter milk, 192 Câreme, Antonin, 499, 633 saltiness perception, 7

Butyric acid, 286 Carob, roasting, 484 spreadability, 115–117
Carotenes, 152–153, 192, 456 stretching, 117–119
Cabbage Carotenoids, 152–153, 313–314, 456 taste-taste interactions, 9
kimchi, 321–324 Carrageen, 518 Cheese analogues, food matrix composition
red, 13, 14, 16 Carrageenans, 194, 229, 385, 389–391, 537, 589, effect on perception, 7
Cafestol, 144, 146, 482, 484 805 Cheese and wine tasting, 577–578
Caffeic acid, 314, 477, 480 Carrots Cherries, 844
Caffeine bioactivity analysis, 75–77, 78 Cherry extracts, 424
chocolate, 133, 134 calcium treatment, 438 Chevreul, Michel-Eugène, 163
coffee, 143, 145, 146, 147 carotenes and carotenoids, 152 Chewing gum, 266
structure, 481 cassoulet (recipe), 727–729 Chia, plant-based drinks, 230
tea infusions, 436 chemistry of cooking, 159 “Chick Corea” (recipe), 746–747
Caffeinic acid, 477, 480 Dispersed System Formalism (DSF), 377 Chicken
Cake filtration, 335 juice pasteurization, 456 inosinate (IMP), 621
Cakes stock, 75–78, 91 nitrite salts, 425
EVOO sponge cake (recipe), 738 L-Carvone, 845 odour, 844
expansion, 293 Caryophyllales, 151 pasteurization, 454
‘lighter than air,’ 431 Caseins sous vide cooking, 533
sucrose, 549–555 cheese, 107, 112–113, 117 Chickpeas, hummus, 231, 666–667
Calcium flama (recipe), 757–758 Chicory root, 566
alginate interactions, 392, 393–394 micelles, 182–183, 186, 188, 191, 411 Chinese dinner, note by note cooking, 763–769
browning effects, 437 milk gels, 200 Chipotle, 530
cheese content, 107, 110, 411 milk structure, 229 Chitin, 385
cheese fondue, 112, 113 Cassoulet (recipe), 727–729 Chitosan, 229
coffee, 145 Catabolism, 266 Chitose, 565
colostrum, 188 Catalysts see enzymes Chloride ions, water, 436
colour effects, 437 Catechins, 14, 133, 134, 436 Chlorogenic acids
drinking water, 436 Catechol, 477, 480 coffee, 143, 145, 146, 147, 476, 477
food content examples, 820 Cathepsins, 30, 32 fruit discolouration, 165
gels with MTGase, 302, 303 Celery, nitrates, 423–424, 427 olive oil, 316
spherification “baths,” 392–393, 537, 694, Cells, matrix, 344 Chlorophyllases, 154
819–820, 821 Cellulose, 63, 64, 599–604 Chlorophyllides, 154
taste receptors, 193 Forest Floor (recipe), 837–840 Chlorophylls, 153–155, 660–661
tea infusions, 436 Cellulose derivatives, 403 Chocolate
Calcium alginate, 689–700 Centrifugation, 752, 802 capillarity, 91
Calcium chloride, 434, 539, 819, 820 Cephalopods, 412, 541 chantilly (recipe), 105–106, 211, 496, 681
Calcium lactate, 438, 537, 539, 819, 820 Ceviche, 683 cocoa percentage, 127
Calcium lactate gluconate, 537, 539, 819, 820 Chamberlain, William, 171 cryo-shaping, 176–177
Calcium sulfate, 819, 820 Champagne, 22 debye chocolate/orange (recipe), 744
Calculations, education, 703–716 scientific tasting, 97–103 differential scanning calorimetry, 124,
Calpains, 30, 32 taste-taste interactions, 9 125–127
Campersterol, 144 Chantillys, 105–106, 211, 496, 681 educational activities, 675
Candida spp., 58, 60, 331 Charcoal, 63 history, 121, 122
Candy see confectionary flama (recipe), 757–758 inclusions, 355
Cannelés, 293 see also expansion Charqui, 204 lactose free, 471
Canning/cans, 22, 87–89, 154, 312–313 Cheese microwave melting, 431
Capillarity, 91–94, 492 ageing, 411 oral processing, 123, 124, 125–126, 131–138
Caramel alternative proteases to rennet, 200 Salmiakki, 128
microwave cooking, 431 ash usage, 25, 27–29, 31 science, 121–124
note by note (recipes), 774 bitterness perception, 7 Sparkling Meringue / Double Cream of
Caramel confectionary, 545–547 calcium content, 107, 110, 411 Gruyère (recipe), 777–778
Caramelization, 81, 84, 632 Chantillys, 106 structure, 131
Carbohydrates, use of term, 631 see also sugars; chymosin, 469 sugar replacers, 127–128
polysaccharides colostrum, 186 synaesthesia, 572
Carbonation composition, 114 tasting and music, 577
beer, 235–236 Crisp Fondue “Vacherin Fribourgeois” thermodynamics, 124–125
cryo-carbonation, 177 (recipe), 778 3D printing, 605–607, 614
Carbon dioxide fondue, 109–113 Cholesterol, milk, 182
Champagne, 97–98 gratins, 113–115 Choline chloride, browning effects, 437
cryo-carbonation, 177 Greek salad (granita) (recipe), 727 Choux, 293–294
dry ice, 172, 174, 177, 178, 179 imaging methods, 410, 411 Chromatography, 339
Carboxylates, 437 microstructure, 110 Chymosin, 181, 200, 411, 469
Carboxymethylcellulose (CMC), 599–601 MMV process, 337, 338 Cider, purification, 337
Carboxypyranoanthocyanins, 15 mould development, 27–29 Citraconic acid, structure, 476
Carcinogens see also acrylamide note by note (recipe), 773 Citrates, 425
barbecue, 64–68 odour, 844 Citric acid
cured meats, 423, 435 production process, 107–109, 191, 229 coffee, 143
frying, 366 ripening effect on sourness, 7 fruitiness perception, 10
smoking, 529 saltiness, 433 saltiness perception, 436
854 Index

sources, 7 caramelisation, 81, 84 Sparkling Meringue / Double Cream of

sourness, 7 chlorophyll preservation, 153–154, 165–166, Gruyère (recipe), 777–778
structure, 476 660–661 structure, 192
tomatoes, 436 coffee, 147, 475 whipping, 234, 496
Citrus fruit educational activities, 660–661 Creamers, 301
essential oils, 268, 271 glycation reactions, 81, 82, 85, 684–685 Creaming
limonene, 217–219 Maillard reactions, 64, 81, 82 calculation of causes, 710–715
Clarification, 335, 337, 469, 752, 789 mineral ions, 436–437 emulsions, 228, 404, 500
Clostridium botulinum, 88–89, 154, 426, 435, misconceptions, 715 mayonnaise, culinary precision, 710
453, 533 nitrates/nitrites, 424 milk, 181, 229
Clostridium perfringens, 427, 533 pasta, 448, 449 oil and water, 257–258
Cloud (recipe), 755–756 pasteurization effects, 455 Cream of tartar, 41, 42
Clove, 427, 844 pastry, 48 Crêpes mademoiselle (recipe), 760–761
CML (Nε-carboxymethyllysine), 82, 83–84, 85 spherification, 538–539 p-Cresol, 528
Coacervation, 404–405 Colourants see also pigments Crisp Fondue “Vacherin Fribourgeois” (recipe),
Coagulation anthocyanins, 13–17, 151 778
cheese production, 107, 108, 191, 411 betalains, 151–152 Crispness
cysteine, 463 carotenes and carotenoids, 152–153 auditory sensation, 205, 518, 569
eggs, 779–782 chlorophylls, 153–155 dried food, 205
egg sauces, 502 Comminution, sausages, 232 water content, 203–204
protein cooking, 160 Communication, language choice, 627–633 Crisps
Coalescence “Complex disperse systems” (CDS) formalism, frying, 315
butter, 192 207 see also Dispersed System Formalism humidity, 581
emulsions, 228, 244, 500, 502 Compound release, 73–74, 378 Critical thinking, 647, 682
Cocktails Compounds, use of term, 631 Croissants, 35, 37, 47
educational activities, 660 Concentration polarization, 336 Crossmodal correspondences, 570–571
liquid nitrogen, 178–179 Conching, 132, 133 Crust formation, 287, 367, 716
Strawberry, Foie Gras, Pistachio (recipe), 828 Condensation, 14, 15, 350 Cryogenics, 171–180, 759–761, 802–803
Welcome Coffee, 827–828 Conduction, 157, 381, 429 Cryospherification, 538, 823
Cocoa, 121, 122, 484 Confectionary Crystallization
Cocoa fibres, Pickering emulsifiers, 229 food matrices, 344–345 butter, 193, 230–231
Coconut, plant-based drinks, 230 hydrocolloids, 385 chocolate, 122–123, 124, 132–133
Coctions, 157 ice blossoms, 560, 561 cryogenics, 173
Codex Alimentarius, 241, 331 invertase, 685 erythritol–sucrose mixtures, 559–561
Coffee sugar crystallization, 545–547 essential oils, 272
acrylamide, 81 3D printing, 607–608 fractionation, 296–297
brewing, 91, 145–147 Confit, squid, 543, 544 from glass state, 557–559
composition, 141–145 Consistency, 384, 433, 454, 472, 496, 507, 509, ice cream, 236
extraction, 140–141 512, 513, 581, 629, 741, 745, 781, 797 Ostwald ripening, 174
flavour, 147 Consommés, 789 polymorphism, 295–296
freeze-drying, 349, 354, 355 Consumer testing, Temporal Dominance of sugars, 545–547
grinding, 139–140, 523–524 Sensations (TDS), 575–578 Cuajada, 199
liquid creamers, 301 Convection, 381 Cucumber
melanoidins, 82, 145, 146, 481 Cooking cryo-slicing, 176
roasting, 473–484 art of, 2 Greek salad (granita) (recipe), 727
volatiles, 481 chemistry, 158–160 kimchi, 321–324
Welcome Coffee, 827–828 culinary precisions, 163–168, 292–294, 442– note by note (recipe), 773
Cognac, 219 443, 491, 645, 666–672, 710, 721 odour, 844
Collaboration, 733 definition, 157 ultra (recipe), 758
Collagen relationship to science, 2 Culinary constructivism, 743–749
cephalopods, 412, 541, 542 thermal processing methods, 157–158 Culinary definitions, 163–168
chemistry of cooking, 160 Co-pigments, 13–14, 16 Culinary heritage, 665–666
contraction during cooking, 91, 417, 419, Copper, chlorophyll metallo-complexes, 154–155 Culinary myths
453–454 Cork popping, 97 ginger milk curd, 199
conversion to gelatin, 419, 454 Corn mayonnaise, 163, 165, 246
glycine, 463 nixtamalization, 25 onions, 163
meat, 92, 93, 416, 417, 419 pasteurization, 455 osmosis, 442, 443
physiological function, 344 Coupe du Monde de la boulangerie (CDM), 48 pears, 164, 165
structure, 467 Courgettes see zucchini Culinary precisions see precisions
Collagenase, 454 Cows, diet, 192 Culinary schools, 2, 655–656, 670
Colloidal systems see also hydrocolloids Craft beer, 21 Culinary science
definition, 385 Cranberries, spherification, 823 public understanding of science, 655–658
Dispersed System Formalism (DSF), 207–211 Cranberry powder, 427 use of term, 633
language usage, 629–631 Cream see also butter Culture
surfactants, 260–261 colostrum, 186–187 food pairing, 348
Colorants, natural, cured meat, 424 composition, 630 texture, 588
Colostrum, 181, 186–189 confectionary, 546 Curculin see neoculin
Colour Dispersed System Formalism (DSF), 105 Curcumin, 152–153
bread crust, 437 formation, 191 Curing agents, 423–427, 435
calcium chloride treatment, 437 imaging methods, 410–411 Custard, 241
Index 855

Cutlery, 372 scientific usage, 210–211 undergraduate courses, 545–547, 655–656,

Cutting, 523 Disproportionation, 235, 241–246 669, 673–678, 683–686
Cuttlefish, 412 Distillation, 213–219, 266–268, 269, 270–271, EDUF process, 339
Cyanidin 3-(sinapoyl)diglucoside 5-glucoside, 14 803 Effervescence, 572
Cyanidins, 13 Distilled drinks, 23 see also specific drinks Egg albumin, spray freeze-drying, 174
Cysteine (Cys), 463, 464, 467, 481 Diterpenes, coffee, 144, 146 Eggplants (aubergines), 316, 667
Do, Re, Mi (recipe), 763, 765–767 Eggs see also mayonnaise; meringues; soufflés
Daguin, André, 171, 759 Dolmas (recipe), 729–730 baumé, 221, 223
Dairy products see also butter; cheese; cream; ice Dominance, definition, 578 boiling, 222, 224
cream; milk; yogurt Dondurma, 685–686 coagulation, 779–782
milk gels, 181–189 Dorure, 48 composition, 779–780
mineral ions taste effects, 433–435 Dough cryo-peeling, 177
odour, 844 baking, 47–48 cysteine, 463
wastewater recovery, 339 folding methods, 35, 36 denaturation, 468
Dalton’s law, 213 formation mechanisms, 37 Dispersed System Formalism (DSF),
Danish pastries, 35, 37 microwave cooking, 431 207–208
Dashi, 518, 520, 619 proportions calculations, 636 educational activities, 636–642
De Bruyn, Lobry, 632 tree of doughs, 158 “eggs at 6X °C,” 222, 781–782
Debyes (recipes), 743–744 Drinks see beverages emulsions, 245, 410, 502
Decantation, 751–753 Drum drying, 205 geoffroy, 222, 224, 262
Deficiency diseases, pellagra, 25 Dry-hopping, 21 gibbs, 223, 225, 431–432
Defrosting, microwaves, 431 Dry ice, 172, 174, 177, 178, 179 hollandaise sauce, 231, 499–504
Dehydration, 203–205, 352, 411 Drying, 204–205, 430, 447–448, 449, 803 hydrogen sulfide formation, 160
Dehydrocafestol, 482, 483 see also freeze-drying ice cream, 194
Dehydrokahweol, 482, 483 Dual Attribute Time Intensity (DATI), 575 matrix effects, 345
Delphinidin, 13 Duck microwave cooking, 431
Density, 381, 703–704 ash-ripened Danish duck (recipe), 30–32, 33 “minus one century eggs,” 221–222, 224
3-Deoxyanthocyanidins, 13 cassoulet (recipe), 727–729 one century eggs, 221–222
Desalination, 336, 444 pasteurization, 454 osmosis, 442
4-Desmethylsterols, 144 roast duck Pravez-Cointreau, 431 pastry egg washing, 48
Desserts Dulse, 518, 519, 520 position of yolk, 637
Hidden Heart (recipe), 765, 769 Durham popovers, 293 Ramsden emulsions, 263, 357–363
orange explosion (recipe), 732 Durum wheat, pasta, 447–450 thenard, 222, 224
soft boiled egg (recipe), 730–732 Dutrochet, Henri, 441 transformation innovations, 221–225
Dewar, James, 171 Dynagels, 76, 378–379 ultrafiltration (UF), 338
Dewars, 179 Dynamats, 379 uncooking, 779–780
Dicalcium phosphate, 42 Dynamers, 378–379 vauquelin, 223, 225, 431
Dielectric heating, 429–430 Dynamics, Newton’s laws, 704 yolk layers, 208
Differential scanning calorimetry (DSC) Dysphagia, 610, 741 Ehrlich pathway, 328, 329
chocolate, 124, 125–127 Einstein, Albert, 637
colostrum, 186 Earl Grey ice cream (recipe), 237 Elasticity, 685–686
Pickering edible oil foams, 359 Ebullition, 382 Elderly people
Diffusion Eclipse dish, 775 cooking for, 741–742
compared to capillarity, 91–92 Education food design, oleosomes, 302
educational experiments, 692, 695–700 argumentation and inquiry, 643–649 3D printed foods, 610
Fick’s laws, 243, 491–492, 697 communication and language choice, 627–633 Electric force, 704–705
Digestion, matrix effects, 345 critical thinking, 647, 682 Electrodialysis, 336, 339
Dihydrochalcone derivatives, 316 culinary schools, 2, 655–656, 670 Ellagitannins, 16
Dihydrocoumarin, 844 culinary teachers, 717–719 Emotion transference, 571
2,6-Dimethoxyphenol see syringol Flavour Experimental Workshops, 635–642 Emulsifiers
2,3-Dimethylpyrazine, 844 gelling agents, 651–653 efficiency factors, 227
Dim sum, Do, Re, Mi (recipe), 763, 765–767 heat transfer, 687–688 foams, 234
Dipeptides, sourness, 7 “Kitchen Stories” Project, 645–649 hydrocolloids, 386, 399–404
‘Dippin dots’™, 175 Lebanon, culinary precisions and science, stability, 500–502
Diracs (recipes), 744–746, 771–774 665–672 types, 228–229
Discolouration master degrees, 669–670 Emulsions see also ice cream; mayonnaise;
anthocyanins reactions, 14, 16 mathematics, 703–716 pseudo-emulsions
fruit browning, 164–165, 201, 684 note by note cooking, 814–815, 843–845 Chantillys, 105–106
Dispersed System Formalism (DSF) public understanding of science, 655–658 cheese fondue, 109–113
butter, 195–196 reducing oil absorption, 663–664 colloidal systems, 207, 399
comparative molecular gastronomy, 168 schools in Denmark, 655, 656 definition, 257, 524, 629
evaporation, 281 schools in France, 2, 168, 635, 659, 668 destabilization, 242, 253, 438
food matrices, 343–345 secondary schools, 659–661 evaporation, 285–286
gels, 375–379 seminars, 2, 721–723 filtration membranes, 337
ice cream, 195–196 science and cooking workshops 1, 2, 163, 168, food matrices, 344
language usage, 630 171, 347, 550, 635–642, 651–653, 660–661, geoffroy, 222, 224, 262, 497
milk and cream, 105 666, 670–671, 679–682, 810, 813 imaging methods, 410–411
principles overview, 207–210 Singapore, molecular gastronomy, 679–682 language usage, 629–630
reference size, 208 spherification, 689–700 lecithin, 228–229, 252–256, 410
sauces, 495–497 sustainability, 657 liebig, 496–497
856 Index

LSW theory, 244 soufflé expansion, 292 perception, 7

microemulsions, 232–233, 244, 262–263, 630 thermodynamics, 382, 383 yogurt, 183
milk, 229–230, 630 EVOO see olive oil Fischer, Emil, 81, 632
mineral ions effects, 438 Expansion, 291–294 Fish
nanoemulsions, 244, 263, 399, 630 Extraction acid maceration, 157
odour, 285–286 coffee, 140–141 canning, 312–313
olis, 496 water mineral content, 436 ceviche, 683
Ostwald ripening, 228, 241–246 Extrusion, pasta, 447, 449 freeze-drying, 349, 356
principles overview, 227–228 Exudate gums, 399 frying, 316
Ramsden, 263, 357–363 marination, 314–315
ratios, 500 Fat bloom, chocolate, 123, 125, 126 nitrite salts, 425
salt, 245 Fat content olive oil cooking, 312–313
sausages, 581–582 cheeses, 110, 114, 116 polyunsaturated fatty acids (PUFAs),
structure, 252–253 sourness perception, 7 312–313
surfactants, 245, 257–263 Fats see also oils smoking, 527, 529, 530
taste perception effects, 7–8 chemistry of cooking, 160 sous vide cooking, 532–535
Encapsulation, 385, 402, 537, 538–539, 738 ice cream, 236 umami, 621
Endomysium, 416, 417 low-density lipoproteins (LDLs), 299 Fizziness, cryo-carbonation, 177
Endorphins, 844 meat marbling, 416, 417 Flama (recipe), 757–758
Energy, 704, 705 oleosomes, 299–303 Flavanols, wine, 14
Enterobacteriaceae, 426 oxidation, 305–309 Flavenes, 14–15
Entropy, 227, 705 pastry lamination, 37–38 Flavones, 316
Enzymes physiochemical properties, 295–297 Flavonoids, membrane filtration, 338
educational activities, 685 storage, 308 Flavonols, 316
Michaelis constant, 685 Fatty acids Flavour see also taste
proteases, 200, 463, 464, 469–470 butter, 192 barbecue smoking, 63, 64, 68
Epicatechin, 133, 165 chocolate, 122, 132 coffee, 147
Epimysium, 416 coffee, 144, 145 fermentation, 327, 328–329
Equipment formulas, 306 language usage, 627–629
cryogenics, 179 language usage, 631 matrix effects, 345
size reduction, 524 long-chain (LCFA), language choice for taste, nitrates/nitrites, 425
Eragrostis tef, 43 628, 629 physiological process of tasting, 193
Erythritol oils, 227 smoking, 527
chocolate, 127, 128 oleogustation, 629 texture effects, 286, 585–590
pastry, 549 residues, 7–8, 83, 123, 132, 192, 193, tsukemono, 597
sucrose mixtures, 557–562 295–297, 252, 301, 631 Flavour release
Erythrose, 478 sourness perception, 7–8 gellan gum gels, 398
Escherichia coli., 453 trans fatty acids, 296 matrix effects, 281, 286, 345
Escoffier, Auguste, 2, 499, 633, 737 Favre, Joseph, 633 oral processing, 345
Essential oils Fermentation Flavylium cation, 13, 14–15
antioxidants, 308 acetaldehyde formation, 14 Flax, plant-based drinks, 230
applications, 276–278 alcohol fermentation, 59 Flour
distillation, 217–219 alcoholic beverages, 19 note by note (recipe), 772
evaporation, 281–282 butter, 193 pastry, 36–37
history, 266–268 carotenoids, 152 sourdough bread, 57
manufacturing processes, 269–274 chocolate, 134 starch, 58
partition coefficients, K, 269 educational activities, 674 teff, 43
safety, 275–279 Ehrlich pathway, 328, 329 Flow, rheology principles, 184–185
smell (perception), 265 gellan gum, 395 Flowers, essential oils, 267
yields, 274 injera, 43 Fluorescent labelling, 409
Esters, 329–330 kefir, 331–333 Foams
Ethanol kimchi, 321–324 Adria recipes, 255
beer foam, 235 lactic acid fermentation, 59 beer, 235–236
distillation, 213–219 milk, 229 cream conversion to butter, 192
essential oils, 271 sourdough bread, 57–60 definition, 631
fermentation process reactions, 328 starters, 43, 60 disproportionation, 235, 241–246
oxidation, 14 Viennoiserie pastry, 47 educational activities, 638–642
thenard eggs, 222, 224 yeasts, 327–330 emulsifiers, 228–229
Ethephon, 437 yogurt, 183, 185, 189 espresso coffee, 147
Ethyl acetate, 329 Fetu, Emile, 2 food matrices, 344
Ethyl-carbamate, 365 Fibre Hidden Heart (recipe), 765, 769
Ethylcellulose (EC), 604 cellulose, 599 language usage, 631
Ethyl hexanoate, 329 hydrocolloids, 385 note by note (recipes), 774
3-Ethylphenol, 529 Ficin, 469 particle stabilizers, 357–358, 359
Ethyl propionate, 844 Fick’s laws, 91–92, 243, 491–492, 697 Pickering edible oil, 357–363
Eucalyptol, 844, 845 Filtration, 789–791, 803–804 siphons, 255
Eugenol, 528, 844 Filtration membranes, 335–340 size and production method, 716
Eutectic point, 173 Firmness structure, 233–235, 363
Evaporation calcium treatment, 438 whipping, 234, 361, 638–642
principles overview, 281–287 chemistry of cooking, 159 würtz, 804
Index 857

Foie gras Fruitiness, perception, 10 properties and uses, 589, 804–805

Chantilly, 106 Fruit juices Gels see also spherification
fat leakage, 716 food matrix, 344 colostrum, 187–189
sous vide cooking, 452 osmotic evaporation (OE), 338 compound release, 378
Strawberry, Foie Gras, Pistachio (cocktail pasteurization, 455–456 debyes (recipes), 743–744
recipe), 828 proteases, 469 definition, 375, 387, 630–631
Fond, 166 purification, 337 Dispersed System Formalism (DSF), 375–379
Food matrices sodium alginate, 395 dynagels, 76, 378–379
acid release, 7 Frying educational activities, 675
bioactivity, 71–79 definition, 157, 365 flavour perception, 286
cooking effects, 345 fish, 316 food matrices, 344
definition, 343 oil absorption, 365–366, 367, 663–664 fractal particle gels, 583–584
Dispersed System Formalism (DSF), 343–345 olive oil, 315–316 ginger milk curd, 199–202
effects in the kitchen, 343–345 oxidation prevention, 307–308 graham, 377
Food pairing, 347–348, 520, 577–578 physiochemical changes, 366–367 hysteresis, 389
Food poisoning potatoes, 287, 315–316, 365, 366–367, language usage, 630–631
botulism, 88–89, 426 663–664 marker diffusion, 492
Listeria, 426–427 vacuum frying, 366 matrix effect, 378
Salmonella, 426 water vaporization, 287 milk, 181–189
Food science Fucoidans, 385 mouthfeel, 583
compared to molecular gastronomy, 1 Fudge, 545 oleogels, 359, 360, 363, 432
history, 163–164 Fumaric acid, 478 oleosomes, 302–303
Forest Floor (recipe), 837–840 Fumigation, 266 plating effects, 372
Formalism, 705–706 see also Dispersed System Funori, 518 sponge phases, 377
Formalism Furaneol, 844 statgels, 378–379
Formic acid, 143, 478 Furanones, 68 volume calculations, 716
Formulas, Dispersed System Formalism (DSF), Furanose, 564–565 Geoffroy, Claude Joseph, 163, 222–223
207–211 Furans, 68, 193, 365, 366, 528, 566 Geoffroy family, 222–223
Fourier laws, 705 Furcellarans, 385 Geoffroys, 222, 224, 262, 497
Fourier’s law, 381 Furfural, 14, 528–529 Ghee, 192
Fractal particle gels, 583–584 Furfuryl alcohol, 147 Gibbs, Josiah Willard, 223, 441–442
Fractionation Furfurylic acid, 482 Gibbs system, 431
fats, 296–297 Furfurylic alcohol, 481 Gibbs emulsion, 223, 225, 377, 431–432, 497
filtration membranes, 335–340 Furfuryl-mercaptan, 844 Gibbs–Marangoni effect, 245
Fragrance see odour; smell Fusel acids, 329 Gilbert, Phileas, 2
Free energy, 124, 209–210, 242 Gin, 23
Freeze-drying, 174, 177, 195, 205, 349–356 Gagnaire, Pierre, 94, 106, 166, 347, 743–749, Ginger milk curd (recipe), 199–202
Freezing see also cryogenics 811 Ginger proteases, 200, 470
educational investigations, 688 Galactomannans, 141, 229, 476 Glasses (state)
eutetic point, 173 Galactose food matrices, 344–345
freezing point depression, 351–352 coffee, 114, 142, 475 freeze-drying, 351–352
industrial methods, 174 Maillard reactions, 471 frozen meat, 421
meat, 421 structure, 563 sucrose, 557–559
Fried foods, acrylamide, 81 Galbazine, 844 Glasses (vessels)
Frozen Florida, 431 Gallic acid, 134 cleanliness, 235
Fructans, 582 Gamma-terpinene, 844 gastrophysics, 372
Fructose Garlic shapes for Champagne, 99–100
lactic acid bacteria fermentation, 58 aioli, 496 Glass transitions, 352, 557–559
structure, 476, 563 crushing, 524 Glaze (stock), 284
yeast fermentation, 58 meringues, 794–795 Gliadins, 36–37, 447
Fructoselysine (FL), 82 Garlic chives, kimchi, 321–324 Globulins, 410
Fruits Gases, ideal gas law, 704 Glucans, 235
anthocyanins, 16 Gastrophysics, 371–372, 459, 541–544, 656–657 Glucono-delta lactone, 42
beers, 21 Gay-Lussac, Louis-Joseph, 87 Glucose
in chocolate, 126–127 Gelatin Maillard reactions, 471
cryo-carbonation, 177 collagen conversion, 419, 454 structure, 476, 563
discolouration, 164–165, 201, 684 dynagels, 378–379 yeast fermentation, 58
Dispersed System Formalism (DSF), 201 educational activities, 651–653 Glucose syrups, 337
drying, 204 as emulsifier, 229 Glühwein ice cream (recipe), 237
essential oils, 268 as gelling agent, 385, 387 Glutamate, 520, 543, 619
freeze-drying, 355–356 hydrogels, 375–376 Glutamic acid (Glu), 7, 30, 463, 464
fruit sorbet (recipe), 759–760 properties and uses, 589 Glutaric acid, 476
note by note (recipe), 774 Gelatinization, starch, 53–55, 343, 448, 449 Gluten
nutrients, 454–455 Gellan gum, 229, 385, 395–399, 537, 589, 805 dough formation mechanisms, 36–37
osmosis in syrups, 442–443 Gelling agents durum wheat pasta, 447, 449
pasteurization, 454–456 educational activities, 651–653 flama (recipe), 757–758
purées, 505–514 hydrocolloids, 387–399 Lioness Head (recipe), 764, 768
ripening, 510–511 hydroxypropylmethylcellulose (HPMC), tart (recipe), 756–757
storage firmness, 438 602–603 Gluten-free cereals, teff, 43
3D printing, 610, 611 methylcellulose (MC), 602–603 Glutenins, 36–37, 447
858 Index

Glycation, 81, 82, 85, 631, 632, 633, 684–685 ice cream, 383–384 Ice
Glycation reactions see Maillard reactions jelly refrigeration, 383 crystal formation, 173, 178, 236
Glycemic index, pasta, 449 meat, 419–421 formation, 351–352
Glycerin, foams, 357 potato cooking, 383 melting, 172, 351–352
Glycerol residues, 122, 181–182, 259–260 principles overview, 381–383 Ostwald ripening, 174
Glycine (Gly), 463, 464 Hedonic evaluation, 577–578 sublimation, 350, 353–354
Glycogen, meat, 416 Heme see haem iron Ice blossoms, 560, 561
Glycolic acid, 478 Hemicellulose, 63, 64, 528, 689 Ice cream
Glycolipids Hemoglobin, 467 bacon and egg ice cream, 178
lecithin, 251 Hemp, plant-based drinks, 230 blueberry-oat ice cream (recipe), 237
physiological function, 251 Henry’s law, 97, 243 brandy sorbet (recipe), 759
Glycolysis, 59 2-Heptanone, 844 consumption, 196
Glycoproteins, 135 Herbs crêpes mademoiselle (recipe), 760–761
Glycosidases, 14 cryo-grinding, 175 cryogenics, 171–172, 175, 178, 195, 759–761
Glycotoxins, 85 freeze-drying, 356 Dispersed System Formalism (DSF), 195–196
Gobley, Theodore Nicolas, 249 Heterocyclic amines (HCA), 64–68, 69, 365 Earl Grey ice cream (recipe), 237
Goussault, Bruno, 452 Cis-3-Hexenol, 844, 845 emulsifiers, 229, 236
Goût, 627–628 Cis-3-Hexenyl acetate, 844 freeze-dried, 352
Graham, Thomas, 441 Hexoses, 563–566 fruit sorbet (recipe), 759–760
Graham gel, 377 Heyns, K, 633 glühwein ice cream (recipe), 237
Grains, drying, 204 Heyns products, 82 heat transfer, 383–384
Grapes see also wine Hibiscus beverages, 16 history, 194
anthocyanins, 15 Hidden Heart (recipe), 765, 769 hydrocolloids, 399
calcium chloride, 437 High-temperature short-time (HTST) processing, inclusions, 194
cyanidin, 14 154 mango-mint olive oil ice cream (recipe), 237
ethephon, 437 Histamine, anthocyanin reactions, 151 olive oil, 317, 738
juice pasteurization, 456 Histidine (His), 463, 464, 465, 467 Pacojet method, 237
raisins, 204 HMF (hydroxymethylfurfural), 14, 81, 83–84, production methods, 236–237
waste thermal processing, 455 160, 365–366, 482, 565–566 savoury, 195
Grass, cows’ diet, 192 Hodge, John E., 82 science, 194–195
Grating, cryo-grating, 176 Hollandaise sauce, 231, 241, 499–504 structure, 195–196, 207, 236, 630
Gratins, 113–115 Home brewing, 19–21 texture, 236
Greek cuisine, new, 727–733, 771–774 Homogenization, milk, 191, 229 Ideal gas law, 704
Grilling Honey IGF-1 (insulin-like growth factor 1), colostrum,
definition, 157 cryo-shattering, 176 181
salt, 92, 491–494 hexose dehydration, 565, 566 Imaging methods, 409–412, 524, 542
Grinding Hops, 19, 23, 235 Imines, 64, 67, 69, 82, 632
coffee, 139–140 Horchata, 230 Immunoglobulins
cryogenics, 174, 175 HTST (high-temperature short-time) processing, colostrum, 181, 186
size reduction, 523 154 milk, 182
spices, 174 Humectants, 285 Inclusions, 194, 355, 395
Guaiacols, 63–64, 66, 527, 528, 844 Humidity, 581, 715 Indole, butter, 193
Guanine, 481, 482 Hummus, 231, 666–667 Infusions, water mineral content, 436
Guanylate (GMP), 619 Humulone, 235 Injera, 43–46
Guar gum, 194, 229, 589 Hydrocolloids Innovation, transformation codes, 221–225
Guinness, 235–236 classification, 385, 386 Inoculum, 59
Gum arabic, 385, 386, 399–400, 589 as emulsifiers, 399–404 Inosinate (IMP), 30, 619
Gum ghatti, 385, 400–402 as gelling agents, 387–399 Inquiry, 643–649
Gum karaya, 385 overview, 385–386 Interesterification, 297
Gums, exudate, 399 polysaccharide interaction, 404–406 Intersolubility, 296
Gum tragacanth, 385, 399 spherification, 392–395, 537–539 Inulin, 582–583
Gustation, 628 Hydrogels, 54, 375, 517, 691 see also gels Invertase, 685
Gutrin, Noël, 759 Hydrogenation, oils, 296 Iron
Hydrogen sulfide, eggs, 160 coffee, 145
Haem iron, 435, 454, 463 Hydrolases, 14 colour effects, 437
Ham Hydrophobic effect, 227 meat, 425, 435
juice fractionation, 339 Hydroquinone, 477, 480 taste effects, 437
nitrites, 425 4-Hydroxy-2,5-dimethyl-3(2H)-furanone, Irritation
plant nitrates and starters, 424 486 sourness, 7
Harrison, William, 171 Hydroxycinnamic acids, 14, 15 taste effects, 586
Harvard Science and Cooking Program, 655–656 Hydroxycinnamic aldehydes, 14 temperature effects, 587
Health programmes, 168 Hydroxycinnamoylquinic acids, 316 Irritation effects, odour effects, 587
Heat, physiological process of tasting, 193 5-Hydroxymethyl-2-furanaldehyde, 147 Isoamyl acetate, 329, 844
Heat transfer Hydroxypropylcellulose (HPC), 603–604 2-Isobutyl-3-methoxy-pyrazine, 844
bread, 383 Hydroxypropylmethylcellulose (HPMC), Isoeugenols, 528
calculations, 704 601–603 Isoflavones, coffee, 146
cryogenics, 172 Hydroxytyrosol, 316 Isoleucine (Ile), 464
dehydration, 204 Hydroxytyrosol-elenolic acid, 316 Isomalt, 557
educational activities, 674, 687–688 Hygroscopicity, 204 tart (recipe), 756–757
freeze-drying, 353–354 Hypoxanthine, 481, 482 Isomaltol, 566
Index 859

Iso-α-acids, 235 Maillard reactions, 471 polymorphism, 295–296

Itaconic acid, 476 membrane filtration, 338 taste receptors, 193
IUPAC (International Union of Pure and Applied milk content, 191 Liqueurs, 23
Chemistry), 627 Lagers, 19 Liquid creamers, 301
Lamb, Slowly Cooked Lamb Neck with Liquorice, Salmiakki, 128
Jam, odour, 566 Fermented Flour Pancakes, Sunchoke Liquors, anise-flavoured, 232–233
Jamming, emulsions, 228 Puree and Beer Glaze (recipe), Listeria monocytogenes, 426–427, 453, 533
Japan, 163 849–850 Liver, nitrite inhibition, 425
Jellies see also gels Lamination of pastry, 35–38, 47–51, 194 Liver sausage, 581–584
kiwi, 645–646 Landgraf, Alberto, 459 Lobsters
pineapple, 469 Language colour, 152
refrigeration, 383 formalism, 705–706 lobster and juniper (recipe), 797–799
Jerusalem artichokes, 566 use of terms, 627–633 Locust bean gum, 194, 388, 589
Juices Laplace equation, 93, 244 Logics, 210–211
food matrix, 344 Last Kiss (recipe), 765 Longevity (recipe), 764, 767–768
osmotic evaporation (OE), 338 Latent heat, 172, 204, 382 Low-density lipoproteins (LDLs), 299
pasteurization, 455–456 Laurel leaves, 276–277, 277 LSW theory, 244
proteases, 469 La Varenne, François Pierre, 47 Lucas–Washburn equation, 92
purification, 337 Lavoisier, Antoine-Laurent de, 163, 211, 627, Lupin, plant-based drinks, 230
suspensions, 233 635–636 Lupulic acid, 235
Juniper: lobster and juniper (recipe), 797–799 Leaveners Luteolinidin, 13
Junket, 199 chemical, 41–42 Lycopene, 313, 314, 425, 456
water during baking, 37 Lyophilization see freeze-drying
Kahweol, 144, 146, 482, 484 Lebanon, culinary precisions and science Lysine (Lys), 463, 464, 465, 481
Katsuobushi, 520, 619 education, 665–672
Kefir, 330, 331–333 Lecithin Maceration
Reverse Frozen Spherification (recipe), chocolate, 131, 133, 137 fish, 157
822–823 composition, 249–252 wine, 174
Kelvin equation, 243 definition, 249 Magnesium
Keratin, 467 emulsions, 228–229, 252–256, 410 browning effects, 437
Kerner, Justinus, 88 foams, 234 chlorophylls, 153
Ketoenols, 193 labelling, 255–256 coffee, 145
“Kientzheim” sauce, 262 margarine, 231 drinking water, 436
Kimchi, 321–324 nutrients, 255 tea infusions, 436
Kimjang, 323–325 physiological function, 249, 251 Magnesium chloride, 434
Kinetic energy, 704 sources, 228 Maillard, Louis Camille, 64, 81, 631, 632–633
Kipferl, 47 Leeks Maillard reaction products (MRPs), 81–82, 85
“Kitchen Stories” Project, 645–649 ash, 25–27, 28–30 Maillard reactions see also Glycation reactions
Kiwi jellies, 645–646 capillarity, 93 acrylamide, 84–85
Kiwi proteases, cheese, 200 cassoulet (recipe), 727–729 alkalinity, 42
Kluyveromyces marxianus, 327 diracs (recipes), 745–746 cheese gratin, 113–114
Kluyveromyces spp., 331 Lemon juice, ascorbic acid prevention of educational activities, 684–685
Kombucha, 330 discolouration, 164–165 frying, 365
Konbu, 518, 519, 520, 619 Lemons history of scientific investigations, 632–633
Konjac, 229 drinks taste perception, 10 language usage, 631–633
Kurti, Nicholas, 291, 429, 431, 809 lemon tart (recipe), 552–555 meat, 421
Leucine (Leu), 464, 465, 486 mechanisms, 64, 67, 81–83, 632, 684–685
Labbé, Philippe, 760–761 Leuconostoc mesenteroides, 321 melanoidins, 314
Lactic acid Leuconostoc spp., 331 phosphate effects, 437
coffee, 143, 478 Levogalactosan, 565 product yields, 83
kimchi, 321 Levoglucosan, 565 rice pudding, 471
sources, 7 Levoglucosenone, 565 Maize see corn
Lactic acid bacteria, 58–60, 114, 424–425, 427 Liebig emulsions, 496–497, 775 Maleic acid, 478
Lactic acid fermentation, 59 Lignans, 311 Malic acid
Lactobacillus, 58–60 Lignins, 63–64, 65–66, 527, 529 coffee, 143
Lactobacillus acidophilus, 184 Limes, Nitro Poached Green Tea and Lime saltiness perception, 436
Lactobacillus casei, 184 Mousse, 172, 178 sources, 638–634
Lactobacillus delbruekii subsp. bulgaricus, 183 Limonene, 133, 217–219, 844 structure, 476
Lactobacillus fermentum, 424–425 Linearity, plating preferences, 460 tomatoes, 436
Lactobacillus plantarum, 321, 424–425 Ling, AR, 632 Malondialdehyde (MDA), 454
Lactobacillus sakei, 425 Linoleic acid, 482, 483, 484 Malt industry, 339
Lactobacillus spp., 331 Linolenic acid, 484, 485 Maltitol, chocolate, 127
Lactococcus spp., 331 Lioness Head (recipe), 764, 768 Maltol, 528, 529, 565, 844
Lactones Lipids Maltose, lactic acid bacteria fermentation, 58
butter, 193 coffee, 144, 146, 482 Malvidin 3-glucoside, 14
coffee, 477, 478–480 hydrogenation, 296 Manganese, coffee, 145
note by note cooking, 844 milk, 181 Mango-mint olive oil ice cream (recipe), 237
Lactose oleogustation, 629 Mannose, 142, 475, 563
cheese, 114 oxidation, 305–309, 425, 453, 454 Maple syrup, sucre à la crème, 545–547
intolerance, 471 pasteurization, 454 Margarine, 231, 296
860 Index

Margerine, structure, 37 smoking, 527, 529 Microparticles, 245

Marigolds, carotenes and carotenoids, 152 sous vide cooking, 532–535 Microscopy, 25, 26–27, 409–412
Marination spoilage bacteria, 426 Microwave cooking, 91, 157, 382, 429–432, 674,
meat, 92, 314–315, 470 tenderness, 415–417, 454, 470, 533 684
sous vide cooking, 532 thawing, 421 Microwave volumetric heating (MVH), 456
Marshall, A. B., 171 3D printing, 616 Milk see also butter; cheese; ice cream; yogurt
Maslinic acid, 316 umami, 621–622 boiling, 185–186
Mass filtration, 335 Mediterranean molecular cuisine, 312, 313 cheese production, 107
Mass transfer, 204, 354, 419 Mediterranean sofrito, 316 composition, 181–183, 191, 630
Mastic, 233, 266, 685 Mège-Mouriès, Hippolyte, 231 Dispersed System Formalism (DSF), 105
Mastication see also oral processing Melanoidins emulsions, 229–230, 630
sound, 569, 570 beer, 235 gels, 181–189
Mathematics, education, 703–716 coffee, 82, 145, 146, 481 ginger milk curd (recipe), 199–202
Matrix effects discovery, 633 history, 181
definition, 343 Maillard reactions, 64, 81, 82, 314 kefir, 331–333
flavour release, 281, 286, 345 Melon proteases, cheese, 200 lactose-free, 692
food analysis, 345 Melons, storage treatment, 438 microfiltration, 337, 791
gels, 375 Melting microwave cooking, 431
nutrients, 345 butter in the mouth, 193 note by note (recipe), 773
texture, 345 chocolate, 123, 124–127, 131, 136 olive oil interactions, 316–317
Maxwell–Boltzmann law, 705, 716 ice, 172, 351–352 pasteurization, 452
Mayonnaise latent heat, 172 protein concentration, 338
creaming, calculation of causes, 710–715 Membranes rice pudding (recipe), 471–472
culinary myths, 163, 165, 246 filtration, 335–340 safety, 191
definition, 164 milk, 182 structure, 192, 229–230
definition variation, 241 osmosis, 441–444 treatment methods, 191
Dispersed System Formalism (DSF), 208 Menstruation, culinary myths, 163, 165 UHT compared to pasteurised, 82, 191
fat content and odour, 286 Menthol, 844 Milk fat globule membrane (MFGM), 191, 229,
imaging methods, 410 Menthol breath strips, 395 299
oleosomes, 300–301 Meringues Milkshakes, 177, 233
recipe robustness, 167 garlic, 794–795 Millet, beers, 21
stability, 243, 245 Hidden Heart (recipe), 765, 769 Mineralization, cheese, 107, 113, 116, 117
surfactants, 262 note by note (recipe), 774 Minerals
surfactants calculations, 716 Sparkling Meringue / Double Cream of coffee, 145, 146
volume from one yolk, 500 Gruyère (recipe), 777–778 cooking effects, 433–438
Meals, 3D printing, 608–609, 611, 612–613 stability, 793–794 macrominerals, 433
Meat storage, 715 physiological function, 433
ageing, 415, 416, 621 sugar/egg ratio, 549 trace elements, 433
barbecue, 64–69, 420 Mesaconic acid, 476 Mint, mango-mint olive oil ice cream (recipe),
browning, 421 Mesoporous ultrafiltration membranes (UF), 335 237
capillarity, 92, 93 Metabolism Miraculin, 8, 674
charqui, 204 plant odours, 266 Miscibility, 265
chemistry of cooking, 160 plant pathways, 275 Modernism, 675
collagen, 416, 417, 419 Methional, 486 Modified texture foods (MTF), 741
composition, 63, 453 Methionine (Met), 464, 465, 481 Moisture, 203–204, 285
connective tissue, 416–417 Methlycellulose, 195 Mole, definition, 265
contraction, 417, 419, 453–454 Methoxyphenol see guaiacols Molecular cooking
cooking methods, 416 2-Methyl-3-furanthiol, 844 compared to note by note cooking, 810
curing, 423–427, 435 3-Methylbutanal, 486, 487 definition, 1, 3
ethanol extraction, 222 Methylbutanoic acid, 486 ingredients, 801–802
freeze-drying, 356 3-Methyl-butanol, 486 methods, 802–806
freezing, heat transfer, 421 3-Methylbutyric acid, 487 tools, 801
freezing methods, industrial, 174 Methylcellulose (MC), 403, 589, 601–603 Molecular cuisine
glycogen, 416 flama (recipe), 757–758 definition, 1
grading schemes, 417 Methylerythritol phosphate pathway, 275 history, 253–255
heat transfer, 419–421 Methylguaiacol, 527, 528, 529 Molecular gastronomy
marbling, 416, 417 N-Methylpyridinium, 481, 482 comparative, 168
marination, 92, 314–315, 470 Methylthiobutyrate, 844 creation of discipline, 164
maturation, 30, 32 Mevalonate phosphate pathway, 275 definition, 1, 371
microwave cooking, 431 Meyer, Karl Friedrich, 88 local culture, 682
mincing, 175 Mezcal, 23 objectives, 164
muscle structure, 29, 32, 415–416 Micelles 3D printing, 614
nitrate/nitrate reduction, 423–427 caseins, 182–183, 186, 188, 191, 229, 411 Molecules
osmosis myth, 442 critical micellar concentration, 261–262 educational activities, 637, 639
oxidation, 453, 454 enzymes, 200 use of term, 631
pasteurization, 452, 453–454 phospholipids, 251 Mollusc shells, 437
pathogens, 426 surfactants, 245, 260–262 Momentum, 704
protein content, 453 Michaelis constant, 685 Monocalcium phosphate, 42
roasting, recipe robustness, 167 Microemulsions, 232–233, 244, 262–263, 630 Monolignols, 63
salt, 435, 442, 491–494 Microfiltration (MF) membranes, 335 Monosodium glutamate (MSG), 193
Index 861

Morbier cheese, 25 Note by note cooking measurement in the mouth, 286

Mousses see also Chantillys base notes, middle and top chords, guidance note by note cooking, 843, 844
cryo-poaching, 178 for formulation, 843–845 perception, 265, 283
cryo-shaping, 176 “Beetroot” and “Gorgonzola” Rocks (recipe), plant metabolism, 266
EVOO mousses (recipes), 738 833 seaweeds, 518–519
Nitro Poached Green Tea and Lime Mousse bread, 772 smoking, 527–529
(recipe), 172, 178 Bread Grissini (recipe), 833–834 sourness perception, 10
Mouthfeel bubbles (recipe), 755 texture effects, 586–587
chocolate, 121, 123, 126, 131–138 challenges, 590 toasting, 487
coffee, 147 “Chick Corea” (recipe), 746–747 yogurt, 184
food culture, 588 cloud (recipe), 755–756 Ogonori, 518
Japanese terms, 519 compared to molecular cooking, 810 Oil palm fruit, carotenes and carotenoids, 152
pineapple, 469 Cube of “Chicken-Carrot” with Chips of Oils see also fats
seaweeds, 519 “Basil-Lemon” (recipe), 829–830 debyes (recipes), 743–744
texture, 581–584, 585–590 “Cucumber” and Buttermilk Spheres (recipe), drying behaviour, 307
MTGase (microbial transglutaminase), 302, 303, 831–832 evaporation, 281–282
849 current developments, 815–816 low-density lipoproteins (LDLs), 299
Mucins, 135 definition, 3, 809 oleosomes, 299–303
Muscle structure, meat, 29, 32, 415–416 dish development, 814–815 oxidation, 305–309
Mushrooms Dispersed System Formalism (DSF), 207, 211 physiochemical properties, 295–297
note by note (recipes), 774 education, 814–815 Pickering edible oil foams, 357–363
umami, 619, 622 flama (recipe), 757–758 storage, 308
Music, 569, 570, 571–572, 577, 811 flour, 772 wastewater recovery, 338, 339
Mustard, spread, 301 Forest Floor (recipe), 837–840 Okra, 155
Mycobacterium tuberculosis, 453 Gagnaire/This collaborations, 743–749, 811 Oleanolic acid, 316
Myofibrils, 29, 30, 415, 417, 453–454 gels, 375 Oleic acid, 192, 311, 315, 484, 485
Myoglobin, 419, 425, 466, 467, 532 “Goat Cheese” Frozen Pearls (recipe), 832 Oleocanthal, 311
Myosin, 160, 415–416 Greek diracs, 771–774 Oleogels, 359, 360, 363, 432 see also geoffroys;
Myrcene, 844 history, 809–810, 811–814, 843 diracs (recipes)
Myricetin, 134 ingredients, 814 Oleogustation, 629
Myristic acid, 359, 360, 362 macaron (recipe), 841–842 Oleosins, 299, 300
Myths menus, 811–813, 841–842 Oleosomes, 230, 299–303
ginger milk curd, 199 “Nuts” Air Bubbles (recipes), 832 Oleuropein, 316
mayonnaise, 163, 165, 246 orange Blossom Sorbet (recipe), 834 Olfaction see smell
onions, 163 “Potato” Tuiles (recipe), 834 Olive oil
osmosis, 442, 443 potential benefits, 810–811 antioxidants, 308
pears, 164, 165 Smoked Bell Pepper NbN Ketchup (recipe), Chantilly, 106
834 composition, 311
Nanoemulsions, 244, 263, 399, 630 sushis (recipes), 847 cooking, 315–316
Nanofiltration (NF) membranes, 336 tart (recipe), 756–757 cryogenics, 175
Nanoparticles, 245, 299 3D printing, 735–736 extra virgin (EVOO), 311–317, 737–739
Naringenin, 313 traditional Chinese dinner, 763–769 foams, 357
Neoculin, 8 ultra (recipe), 758 health benefits, 311–312
Nervous system, trigeminal lingual NRTL model (Non-Random Two Liquid), 216 mango-mint olive oil ice cream (recipe), 237
system, 7 Nuclear magnetic resonance (NMR) imaging, meat marination, 314–315
Nettles, chlorophylls, 153 74–76, 409 milk interactions, 316–317
Neurogastronomy, 371, 541 Nutmeg, 427 novel cooking recipes, 737–739
Newton’s law, convection, 381 Nutrients Olio Nuovo Days, 737
Newton’s laws, 704 bioactivity, 71–79 Pickering foam recipe, 363
Niacin, release of bound niacin by ash, 25 compound release, 73–74 wastewater recovery, 338
Nicotinic acid, 143, 146, 481, 482 matrix effects, 345 Olives
Nitrate salts, 423–427, 436 microwave cooking, 431 freeze-drying, 177
Nitric oxide (NO), 424, 425 pasta, 448 Greek salad (granita) (recipe), 727
Nitric oxide synthase (NOS), 425 Nuts, plant-based drinks, 230 spherical-I green olives, 802
Nitrite salts, 423–427, 435 Onions
Nitrogen Oats capillarity, 93
beer carbonation, 235–236 blueberry-oat ice cream (recipe), 237 cassoulet (recipe), 727–729
cryogenics, 171–172, 195, 759–761 plant-based drinks, 230 compound extraction, 78
Nitro Poached Green Tea and Lime Mousse Obesity, 168 culinary myths, 163
(recipe), 172, 178 1-Octen-3-one, 486 dolmas (recipe), 729–730
Nitrosamines, 365, 423, 435 Octopus, 412 as gels, 378
Nixtamalization, 25 Odour Greek salad (granita) (recipe), 727
Nollet, Jean-Antoine, 441 Champagne bubbles, 100, 103 kimchi, 321–324
Nomenclature, 627 chocolate, 123, 136 red, cyanidin, 13
(E)-2-Nonenal, 487 coffee, 147, 482 structure, 77–78
“Non periodical organization of space” (NPOS) evaporation, 281–287 thermal treatment, 511, 512
formalism, 207 fermentation, 327, 328–329 Oral processing see also mouthfeel
Noodles, Longevity (recipe), 764, fogging with cryogens, 178 chocolate, 123, 124, 125–126, 131–138
767–768 food pairing, 347–348 flavour release, 345, 398
Nori, 518, 519, 520, 622 language usage, 628, 629 sound, 569, 570
862 Index

Orange Pastry anthocyanins, 151

crêpes mademoiselle (recipe), 760–761 baking, 47–48 beer foam, 235
debye chocolate/orange (recipe), 744 colour, 48 cheese fondue, 112
drinks taste perception, 10 egg washing, 48 coffee, 146, 147
essential oils, 272 expansion, 293–294 educational activities, 683–684
orange explosion (recipe), 732 flour, 36–37 gellan gum, 395, 398
Order of magnitude in Dispersed System lamination, 35–38, 47–51, 194 hydrocolloid-polysaccharide interactions,
Formalism (DSF), 208, 210 proportions calculations, 636 404, 405
Oregano, 425 Raspberry Pear Viennoiserie (recipe), meat maturation, 30
Organoleptic properties see also flavour; 824–825 milk gels, 181
mouthfeel; odour; taste sucrose, 549–555 nitrite effects, 425
perception, 283 Viennoiserie, 47–51, 824–825 salivary flow rate, 8
“Osmazone,” 222 Pathogens, 426, 453, 533 sourness perception, 7, 8
Osmosis, 91, 441–444, 716 Pearl ash, 41 spherification, 689, 692–700
Osmotic evaporation (OE), 336, 338 Pears tea infusions, 436
Ostwald ripening, 174, 228, 241–246 culinary myths, 164, 165 yogurt, 183
Ouzo, 232–233 kimchi, 321–324 Phenethyl alcohol, 845
Ovalbumin, 246 lemon juice, 164–165 Phenolases, 469, 684, 685
Ovens, meat cooking, 421 Raspberry Pear Viennoiserie (recipe), Phenolic acids
Ovotransferrin, 410 824–825 coffee, 146
Oxazoles, 68 reddening in acid cooking, 17, 165 fruit, 165
Oxidation storage treatment, 438 Phenols
colour changes, 14, 16 Peas butter, 193
educational activities, 684 canning, 154 chocolate, 133, 134–136
lipids, 305–309, 425, 453, 454 cryo-grinding, 175 pasteurization effects, 455–456
wine, 715 nutrients, 76 Phenylacetaldehyde, 486, 487
odour, 844 Phenylalanine (Phe), 464, 465, 486, 487
Packaging plant-based drinks, 230 2-Phenylethanol, 329, 486
freeze-dried foods, 354–355 Peat, 529 Phenylpyranoanthocyanins, 15
sous vide cooking, 531–532 Pectin methyl esterase (PME), 76, 438 Pheophytinisation, 153
Pacojet machines, 175, 195, 237, 524 Pectin Pheophytins, 153, 155, 660–661
Palmitic acid, 192, 482, 483, 484 calcium interaction, 437–438 Phosphates, taste effects, 433–434, 436
Palm oil, fractionation, 297 chemistry of cooking, 159 Phosphatidylcholine (PC), 229, 300
Papain, 469, 685 emulsifiers, 229, 404 Phosphatidylethanolamine, 229
Papaya, 469 fruit gelling, 404, 689 Phospholipids
Paprika, 530 fruit ripening, 511 chocolate, 133
Paraethylphenol, 809–810 hydrocolloids, 385, 404 cured meats, 435
Paré, Ambroise, 163, 207, 257 oleosomes gels, 302–303 emulsifiers, 228–229
Parsley, 277–278 properties and uses, 589 lecithin, 249–251
Particle morphology, 509 spherification, 537 milk, 182, 229
Particle size Pediococcus acidilactici, 424–425 oleosomes, 299–300
cocoa solids in chocolate, 127 Pediococcus pentosaceus, 425 Photosynthesis, 266, 275
coffee grinding, 139–140 Peeling, cryo-peeling, 177 Phycogastronomy, 517–521
cooking for the elderly, 741 Pelargonidin, 13 Pickering edible oil foams, 357–363
distribution in food powders, 523–525 Pellagra, 25 Pickering emulsifiers, 229, 402–403 see also
emulsions using membranes, 337 Pentan-2,3-dione, 844 Ramsden emulsions
Particulate emulsifiers, 229 Pentosans, 235 Pickles, tsukemono, 411, 520, 593–597
Partition coefficients, 259–260, 265, 269, 282 Pentosidine, 82 Pigments see also colourants
Pasta Pepper (black), 844 anthocyanins, 13–17, 151
durum wheat, 447–450 Peppers (green), 316, 844 betalains, 151–152
educational investigations, 687–688 Peppers (red), 456 carotenes and carotenoids, 152–153
note by note (recipe), 773 Peptide bonds, 464 chlorophylls, 153–155
salt requirements for cooking, 716 Peptides lipids oxidation, 307
3D printing, 607 pork loin, 8 olive oil, 311
Pasteur, Louis, 41, 87, 451 separation, 337, 339 pasteurization effects, 455
Pasteurization sourness, 7 pyranoanthocyanins, 15
beverages, 452 Perception Pineapple
discovery, 87, 451 firmness, 7 jellies, 469
fruit, 454–456 flavour/goût language usage, 627–628 mouthfeel, 469
meat, 452, 453–454 saltiness, 7 Pinoresinol, 311
milk, 191, 193, 452 sound, 569–573 Piperine, 844
novel methods, 456 sourness, 7–11 Pistachio, Strawberry, Foie Gras, Pistachio
pathogens and safety, 453 Perfumery, 843, 845 (cocktail recipe), 828
physiochemical modifications, 453–456 Perimysium, 416, 417 Pizzas, 107, 113–117
purpose, 451 Permeate flux, 336 Plant-based drinks, 230, 299
sous vide cooking, 452–453, 531 Peroxidases, 14 Plants
types, 451–453 Pervaporation, 336 cell walls, 505, 506
vegetables, 454–456 Pfeffer, Wilhelm, 441 chemistry of cooking, 158–159
wine, 451 pH essential oils, 266, 267–268
Pastis, 232–233 agar-agar gels, 388 exudate gums, 399
Index 863

oleosomes, 230, 299–303 sous vide cooking, 532–535 Radiation (heat transfer), 157, 381–382
photosynthesis, 266, 275 umami, 621 Radish
plant tissues as gels, 344 Powders, 345 imaging methods, 412
saccharide exchange, 78 Power, 704 kimchi, 321–324
Plasticity, 581 Pralus, Georges, 452 Raisins, 204
Plates, musical, 570 Precisions Raki, 233
Plating science, 372, 459–461 culinary, 163–168, 292–294, 442–443, 491, Ramsden emulsions, 244, 245, 262, 263, 285,
Polarity, water, 227 645, 666–672, 710, 721 344, 357, 629–630 see also Pickering
Polarized light microscopy, 25, 26 Escoffier’s use of term, 633 emulsifiers
Polya, George, 703 Presentation, plating science, 459–461 Raoult’s Law, 351
Polyaromatic hydrocarbons (PAH), 64, 68, Pressure, 157, 704 Raspberries
529–530 Primary education, France, 2, 168, 635 debye raspberry/basil (recipe), 744
Polymers, dynamers, 378–379 Proanthocyanidins, 14 Raspberry Pear Viennoiserie (recipe),
Polymorphism, 295–296 Probiotics, yeasts, 330 824–825
Polyols, 557 Proline (Pro) Ready meals, 356
Polyphenol oxidase (PPO), 201, 202 anthocyanin reactions, 151 Rebaudioside A, 549
Polyphenoloxidases, 14 protein structure, 465 Recipes, robustness, 163–168
Polyphenols structure, 463, 464 Redox chemistry, 684
antioxidant role, 308, 425, 684 Proline-rich proteins (PRP), 136 Reducing agents, 684
beer, 235 Proteases, 200, 463, 464, 469–470 Refrigeration, 688
chocolate, 134 Proteins see also amino acids Regreening, 155
coffee, 147 anthocyanin reactions, 151 Regulations
olive oil, 312 cheese content, 110 acrylamide, 85
Polysaccharides chemistry of cooking, 160 colourants, 155
coffee, 141–142, 475 coffee, 145, 147, 481 lecithin, 255–256
emulsifiers, 229 colostrum, 186 nitrates/nitrites, 423
hydrocolloid interactions, 404–406 concentration, 338 toxicology, 276
oleosomes combination, 302 denaturation, 419, 453, 468, 502, 683 Rehydration, 204
Polysorbate 80, 229 dietary requirement, 465 Remoulade, 245
Polyunsaturated fatty acids (PUFAs), 312–313, durum wheat pasta, 447–450 Rennet
315 educational activities, 674 cheese, 191, 411, 469
Pomaine, Edouard Pojersky de, 2, 163 emulsifiers, 229 milk gels, 181, 199–200
Pomegranate, anthocyanins, 16 freezing, 174 Rennin see chymosin; rennet
Pomegranate juice, 338 gluten-forming, 36–37 Retention rates, membrane filtration, 336
Popcorn, 431 ice cream, 236 Retrogradation, 53
Pork meat, 419 Reverse osmosis (RO), 335–336, 338, 444
ageing, 416 milk, 182 Reynolds number, 752
cassoulet (recipe), 727–729 physiological function, 463 Rhamnose, 142
crisp skin, 165 in saliva, 135, 136 Rheology
inosinate (IMP), 621 sourness/sweetness perception modification, 8 hydrocolloids, 385
nitrite salts, 425 structure, 63, 66, 463–467 principles overview, 184–185
pathogens, 426 Protein/starch gels, 54–55 salt content effects, 435
peptides, 8 Protons, sourness perception, 7, 8 yogurt, 185–186
spoilage bacteria, 426 Pseudo-emulsions, butter, 195 Ribose, 475
Porosity, 344 Psychophysics, 372, 459 Rice
Port, cryo-shattering, 176 Psyllium, 385 diracs (recipes), 745–746
Portisins, 15 Puddings plant-based drinks, 230
Potash see potassium carbonate colostrum, 186 rice pudding (recipe), 471–472
Potassium rice, 471–472 Rice vinegar, taste-taste interactions, 9–10
browning effects, 437 Pumpkin, 316, 456 Ripening
coffee, 145 Pumpkin seeds, plant-based drinks, 230 cheese, 107
tea infusions, 436 Purees, fruit, 505–514 fruit, 510–511
tomatoes, 436 Purification, filtration membranes, 337–338 Risk assessments, 179–180
Potassium bitartrate, 112 Pyranoanthocyanins, 15–16 Roasting
Potassium carbonate, 30 Pyranone structures, 565 carob, 484
Potassium chloride, 435 Pyrazines, 68, 147, 193, 484, 844 cocoa, 484
Potatoes Pyridine, 481, 482 coffee, 473–484
aligot, 117–119 Pyrogallol, 477, 480 definition, 157, 473
crisps, 315 Pyrolysis, 63–64, 66 sesame seeds, 484
educational activities, 663–664, 674 Pyrraline, 82 walnut, 484
freeze-drying, 204, 349 Pyrroles, 68 wheat, 486–487
frying, 287, 315–316, 365, 366–367, 663–664 Pyruvic acid, 15 Robustness, recipes, 163–168
heat transfer, 383 Rosemary, 425, 845
instant mash, 205 Quantum theory, 305 Roux, 193
note by note (recipe), 773 Quercetin, 134, 314 Rum, 23
structure, 207 Quinic acids, 143, 146, 147, 477, 478
Potential compound release, 73–74, 378 Quinoa Sabayon, 499, 502
Poultry beers, 21–22 Saccharides see also polysaccharides; specific
ageing, 416 dolmas (recipe), 729–730 saccharides
cooking, 419 Quinones, 165 carrots, 75–77, 78
864 Index

chemistry of cooking, 159, 160 Saltiness alginates, 391, 392, 538, 689–690
coffee, 141–142, 147 perception, 7, 9–10, 436 carrageenans, 389
definition, 563, 631 sound effects, 571 chlorophylls, 153
Fischer reactions, 632 temperature effects, 586 hydrocolloids, 385
freeze-drying inhibition, 351, 352 umami effects, 622 phycogastronomy, 517–521
hexose dehydration, 563–566 Sambuca, 233 umami, 621
lecithin, 252 Sapiction, 628 see also taste Secoiridoids, 311
milk, 191 Saponification, 259 Sedimentation see creaming
pasteurization, 455 Sarcomeres, 415–416, 417 Seeds, plant-based drinks, 230
Saccharomyces boulardii, 330 Satiety, 385, 577 Seminars, 2, 721–723
Saccharomyces cerevisiae, 19, 60, 327 Sauces see also mayonnaise Semolina, durum wheat, 447–450
Saccharomyces exiguus, 58 apple, 505–514 Sensory effects see organoleptic properties
Saccharomyces pastorianus, 19 béarnaise, 241, 242, 499 Serine, 481
Saccharomyces spp., 331 béchamel, 241, 301 Sesame seeds
Safety “brilliancy,” 166–167 plant-based drinks, 230
cryogens, 172, 177–178, 179–180 classification, 495–497 roasting, 484
curing agents, 423 Codex Alimentarius, 241 tahini, 231, 666–667
essential oils, 275–279 custard, 241 Shallots
exudate gums, 399 Dispersed System Formalism (DSF), 495–497 capillarity, 93
microwave cooking, 431 educational activities, 674–675 kimchi, 321–324
milk, 191 “faraday,” 497 Shaping foods, cryo-shaping, 176–177
pasteurization, 453 glaze (stock), 284 Shelf-life
smoking, 529–530 history, 499 freeze-drying, 349
Safflower seeds, plant-based drinks, 230 hollandaise, 231, 241, 499–504 sous vide cooking, 531, 533–534
Saffron, 152 “kientzheim” (recipe), 262 Shikimic acid pathway, 275
Sage, 425, 427 oleosomes, 300 “Shitao” (recipe), 94
Salad dressings reduction, 283 Shortening, definition, 37
alginates, 395, 399 sous vide cooking, 532 Shrimps
gum tragacanth, 399 structure, 241–242 colour, 152
models for perception analysis, 286 tomato, 91, 313–314 freeze-drying, 356
Salad oil, 360 white, 241, 716 shrimp bisque, 160
Salads wine, 166–167 Sieve analysis, 524
cryo-sliced cucumber, 176 Sausages Singapore, 679–682
Greek salad (granita) (recipe), 727 bacterial starters, 425 Sitosterol, 144
vegetable salad (recipe), 785–787 cassoulet (recipe), 727–729 Size reduction, 523–525
Salep, 685 emulsion system, 232 Skatole, butter, 193
Saliva microwave cooking, 431 Slicing, cryo-slicing, 176
astringency, 134–136 mouthfeel, 581–584 Smell (perception), see also odour
flow rates, 8, 137 nitrates/nitrites, 423, 424, 425 irritation effects, 587
lemon word repetition effects, 628 pathogens, 426–427 language usage, 628–629
matrix effects, 345 production methods, 232 measurement, 286
sourness perception, 8–9, 10 salt content, 435 physiological process, 265
Salmiakki, 128 smoking, 530 temperature effects, 587
Salmon spoilage, 426 texture effects, 586
colour, 152 Save the Shark (recipe), 764–765, 769–770 viscosity effects, 587
smoking, 527, 529 Saveur, 628 see also taste Smoking, 527–530, 805
sous vide cooking, 533 Savory, essential oil, 427 Snacks, sales volume, 365
Salmonella, 426, 427, 453 Savouriness see umami Soap, 259, 262
Salt Scanning electron microscopy (SEM), 26–27, Sodium
cheese, 433 28–30 coffee, 145
drinking water, 436 Schiff, Hugo, 632 tea infusions, 436
emulsions, 245, 438 Schiff bases, 64, 67, 69, 82, 632 Sodium alginate, 385, 391–395, 537, 538,
kimchi, 321 Schmiederberg, M, 633 652–653, 689–700, 819
lowering of freezing point, 173 Schreiber test, 115–116 Sodium aluminium phosphate, 42
meat, 92, 435, 491–494 Science Sodium bicarbonate see also baking soda
note by note cooking, 843 goals, 1–2 drinking water, 436
osmosis, 443 relationship to cooking, 2 Sodium chloride see salt
pasta cooking, 716 use of term, 633 Sodium hydrogenocarbonate see baking soda
smoking, 529 Scientific method, 2, 633, 643–644, 647, 659 Sodium tartrates, 112
sous vide cooking, 532 SCOBY (Symbiotic COmmunity of Bacteria and Sodium tetraborate decahydrate (borax), 437
spherification, 394 Yeasts), 331 Sofrito, Mediterranean, 316
triglycerides oxidation, 454 Scurvy, 87 Solubility
tsukemono, 595–596 Seafood aqueous solutions, 282
vegetable pickles, 411 freeze-drying, 356 definition, 265
water saturation, 265 juice fractionation, 339 essential oils, 265
Salt content Sea of Change (recipe), 763, 767 Kelvin equation, 243
bitterness perception, 434–435 Searing Solutions, food matrix, 344
firmness perception, 311 cryo-searing, 176 Somatosensory system, 585
reduction, 433, 622 sous vide cooking, 534–535 “Sonic seasoning,” 570
sourness perception, 7, 9–10, 433–434 Seaweeds Sorbitan monolaurate, 229
Index 865

Sorbitan monostearate, 229 gastrophysics, 541–544 imaging methods, 409

Sorbitol, 229, 549, 557 imaging methods, 412 mineral ions effects, 437–438
Sorgum, 13 Stability Struvite, peas, 154
Sorokin, B, 632 emulsions, 227, 242–243, 244 Sublimation, 172, 349, 350–351, 353
Soufflés foams, 234, 235, 236, 242–243, Succinic acid, 478
crust thickness, 716 362–363 Suckling pigs, crisp skin, 165
culinary precisions, 166, 292 gellan gum gels, 398 Sucre à la crème, 545–547
definition, 164 gum arabic, 400 Sucroglycerides, 229
expansion, 291–292 meringues, 793–794 Sucrose
frozen EVOO soufflé (recipe), 739 microfiltration, 337 chocolate, 133, 137
Sound moisture content, 203 coffee, 476
seaweeds, 518 sauces, 241–242 confectionary crystallization, 545–547,
taste, 569–573 Turbiscan stability index (TSI), 400 557–560
Soundscapes, 570 Stabilizers erythritol mixtures, 557–562
Soups emulsions, 253, 500–502 fruitiness perception, 10
dilute emulsions, 243 foams, 357–358 glass transitions, 557–559
Save the Shark (recipe), 764–765, hydrocolloids, 386 hydrolysis, 76, 77, 283
769–770 ice cream, 194–195, 236 invertase, 685
Sour beers, 21 micro- and nanoparticles, 245 osmotic evaporation (OE), 338
Sourdough bread, 57–60 Stanford Introductory Science of Cooking course, pastry, 549–555
Sourness 673–678 state diagram, 352
irritation effects, 586 Staphylococcus carnosus, 424–425 structure, 476, 557
perception, 7–11, 683 Staphylococcus xylosus, 424–425, 426 yeast fermentation, 58
phosphates, 433–434 Starch Sugar (culinary ingredient) see sucrose
salt, 433–434 chemistry of cooking, 159, 160 Sugar beet pectin (SBP), 404
sound effects, 570–571 dough viscosity, 37 Sugar refining, 339
temperature effects, 586 emulsifiers, 229 Sugar replacers
viscosity perception, 586 flour, 58 chocolate, 127–128
Sous vide cooking, 452–453, 455, 531–535, 542, gelatinization, 53–55, 343, 448, 449 erythritol–sucrose mixtures, 557–562
674, 805 as gelling agent, 589 pastry, 549–555
Soy modified, as hydrocolloid, 402–403 Sugars see also saccharides
lecithin, 249 Ramsden emulsions, 263 gellan gum interactions, 395–396, 397–398
oleosomes, 299, 302 structure, 53 ice cream, 236
plant-based drinks, 230, 299 wastewater recovery, 339 note by note cooking, 843
protein concentration, 338 Statgels, 378–379 Sulfur dioxide fumigation, 13
soya lobster prototype 3D printing (recipe), Steam extraction, 270 Sulphates, drinking water, 436
735–736 Steaming, definition, 157 Sunflower oil, Pickering edible oil foams, 359,
tofu, 302 Steam ovens, 421 360, 362
Space food, 195, 205, 352 Stefan–Boltzmann law, 705 Sunflower seeds
Sparkling Meringue / Double Cream of Gruyère Sterilization, canning, 88–89 lecithin, 249, 255
(recipe), 777–778 Sterols, coffee, 144 plant-based drinks, 230
Specific thermal capacity, 382 Stevia, chocolate, 127 Surface tension, 258–259, 358, 704, 715
Spencer, Percy, 429 Stevioside, 549 Surfactants
Spheres, area and volume calculation, 704 Stigmasterol, 144 emulsions, 257–263, 716
Spherification, 392–395, 537–539, 652–653, 675, Stocks Gibbs–Marangoni effect, 245
689–700, 805–806, 819–823 bioactivity, 76 HLB (Hydrophilic Lipophilic Balance), 260,
Sphingomonas elodea, 395 consommés, 789 261
Sphingomyelin, 229 decantation, 752 micelles, 260–262
Spices definition, 284 Pickering edible oil foams, 358–359
cryo-grinding, 174, 175 evaporation, 283 Ramsden, 257–263
freeze-drying, 356 filtration, 789 soap, 259
grinding, 524 osmosis myth, 442 Surimi
nitrates, 423 recipe robustness, 168 cloud (recipe), 755–756
Spiciness, physiological process of tasting, 193 reduction, 284 soaking, 758
Spinach Stoke’s Law, 751 ultra (recipe), 758
capillarity, 92–93 Storage Sushis (note by note recipes), 847
chlorophylls, 153, 154 fats and oils, 308 Suspensions, 344, 507–509, 524, 716
nitrates, 423 fruit, 438 Sustainability, 657
Spin restriction, 305–306 meringues, 715 Sweet corn, 455
Spirits see distilled drinks sous vide cooking, 533 Sweeteners see sugar replacers
Spoilage, 285, 426, 533, 550 vegetables, 438 Sweetness
Spoons, musical, 570 Strawberry amino acids, 543
Spray drying, 204–205, 345 drinks taste perception, 10 chocolate, 133
Spreadability juice pasteurization, 455 ice cream, 236
butter, 194 Strawberry, Foie Gras, Pistachio (cocktail irritation effects, 586
cheese, 115–117 recipe), 828 perception, 8, 9–10
Spreads, 231, 300–301 Streptococcus spp., 331 sound effects, 569, 570
Squalene, 311 Streptococcus thermophilus, 114, 183 tea infusions, 436
Squid Stretching, cheese, 117–119 temperature effects, 586
confit, 543, 544 Structure texture effects, 286
866
umami effects, 622
viscosity perception, 586
Sweet potato, carotenes and carotenoids, 152
Swiss recipes, 777–778
Synaesthesia, 572
Syneresis, 111, 200, 391, 590
Syringol, 63–64, 66, 528
Tahini, hummus, 231, 666–667
Tannic acid, 316, 388, 403
Tannin–anthocyanin adducts, 14, 15
Tannins see also polyphenols
anthocyanidins formation, 14
astringency, 135–136, 587
chocolate, 133, 134
ellagitannins, 16
membrane filtration, 338
pears, 17
physiological function in plants, 276
Tart (recipe), 756–757
Tartaric acid see also cream of tartar
sources, 7
Taste see also flavour; mouthfeel
chocolate, 133
coffee, 147, 482
educational activities, 673–674
fat layers in pastry, 38
ice cream with olive oil, 317
infant responses, 10
irritation effects, 586
language usage, 628
measurement in the mouth, 286
mineral ions, 433–436
physiological process, 193, 283
seaweeds, 520
smoking, 529
sound, 569–573, 577
sourness perception, 7–11
temperature effects, 586
Temporal Dominance of Sensations (TDS),
575–578
Taste receptors, 628, 683
Taste sensation, 586
Taste-taste interactions, 9–10, 57
Tasting
Champagne, 99, 100, 102, 103
educational activities, 675
music effects, 570, 571–572, 577
Tea
carotenoids, 152
Earl Grey ice cream (recipe), 237
“Hot and Cold” (recipe), 398
lapsang souchong, 530
Nitro Poached Green Tea and Lime Mousse
(recipe), 172, 178
water mineral content, 436
Tea infusions, “Hot and Cold” (recipe), 802
Teeth, 585
Teff, 43
Temperature
colour effects, 154
eutectic point, 173
freeze-drying, 354
irritation effects, 587
meat cooking, 160
smell (perception) effects, 587
sous vide cooking, 531
taste effects, 586
viscosity effects, 587
Tempering, chocolate, 124, 125, 126, 132
Temporal Dominance of Emotions (TDE), 577
Temporal Dominance of Sensations (TDS),
575–578
Temporal Drivers of Liking (TDL), 577
Tenderness
meat, 415–417, 454, 470
sous vide cooking, 533
Tequila, 23
Terpenes, 844
Terpenic acids, 316
Terpenoids, carotenes and carotenoids, 152
Tetraborate, 437
Texture, see also consistency
analytical methods, 542
apple sauce, 512–514
auditory sensation, 569
definition, 587
dehydrated products, 203
flavour effects, 585–590
flavour perception, 286
ice cream, 236
ice crystals, 173, 178
identification of foods, 588
language usage, 629
matrix effects, 345
measurement, 512
mineral ions, 437–438
modification, 589–590
mouthfeel, 581–584
odour effects, 586–587
pasta, 448
pasteurization effects, 455
physiological process of tasting, 193
seaweeds, 519
smoking, 529
sous vide cooking, 532–533, 542
squid, 542
starch gelatinization, 53–55
sucrose in pastry, 549
temporal changes, 589, 590
tsukemono, 593, 595–597
Thawing, meat, 421
Thenard, Louis-Jacques, 222
Theobromine, 133, 134
Thermal processing methods, 157–158
Thermobacteriology, 88–89
Thermodynamics, 382
Thermo-functionality, 107, 119
Thermogravimetric analysis (TGA), 27
Thickeners, 599–604
Thiobarbituric acid (TBARS), 307, 454
Thiophens, 68
This, Hervé, 743–749, 809, 811
3D Printing, 605–617, 735–736
Threonine (Thr), 464, 465, 481
Thyme, 277
Tiger nuts, 230
Time–Intensity (TI) technique, 575
TMP (transmembrane pressure), 336
Toasting, 487
Tocopherols
coffee, 146
olive oil, 175, 311, 316
walnut, 484, 485
Tofu, 302
Tomatoes
Basic Spherification (recipe), 821–822
Cassoulet (recipe), 727–729
drying, 715
Greek salad (granita) (recipe), 727
mineral salts, 435–436
note by note (recipe), 773
Index
odour, 283
olive oil usage, 313–314, 316
pasteurization, 455, 456
sun-dried, 204
umami, 622
Tomato juice
clarification, 753
taste-taste interactions, 9
Tomato sauces, 91, 313–314
Tongue (human anatomy), 8, 585, 844
Tosaka-nori, 518
Touch, taste effects, 586
Toulmin’s Argumentation Pattern (TAP),
643–644, 646–647
Toxicity see also safety
essential oils, 275, 276
Traditions, 665 see also precisions
Training, cryogens safety, 179
Trans-2,trans-4-decadienal, 844
Trans-2-cis-6-nonadienal, 844
Trans fatty acid residues, 296
Transglutaminase, 302, 303, 849
Transmembrane pressure (TMP), 336
Triacylglycerols (TAGs)
chocolate, 122, 132
coffee, 144
crystallization, 296
fractionation, 296–297
function in fats and oils, 295
hydrogenation, 296
interesterification, 297
intersolubility, 296
milk, 181–182, 630
oils, 227
oxidation, 305–309
trans fatty acid residues, 296
Trigeminal lingual system, 7
Triglycerides
butter, 83, 192, 193
caramel, 546
cocoa butter, 122–123, 132
coffee, 144
emulsions, 227, 241, 243, 245, 259
foams, 359–363
lecithin, 251–252
note by note cooking, 809, 810
oxidation, 305–309, 454
pastry lamination, 37
polymorphism, 295, 296
vapour pressure, 281
Trigonelline, coffee, 143, 147, 475, 481
Trithioacetone, 844
Tryptophan (Trp), 464, 465
Tsukemono, 411, 520, 593–597
Tuna, 312–313
Turbiscan stability index (TSI), 400
Turkey, pasteurization, 454
Turmeric, 152–153
Tyrosol, 314, 316
Ultra (recipe), 758
Ultra-high temperature (UHT) treatment,
191
Ultrasound imaging, 409
Umami
amino acid sodium salts, 7
background noise effects, 569
dashi, 520
glutamic acid, 30
inosine monophosphate (IMP), 30
irritation effects, 586
Index 867

note by note cooking, 844 saliva, 135 cheese and wine tasting, 577–578
physiological process of tasting, 193, 619, 620 sauces, 242 cheese fondue, 112
salt effects, 435 smell (perception) effects, 587 clarification, 752
seafood, 543 sourness perception, 8 cryogenics, 174
seaweeds, 520 spherification, 395 fermentation, 328
smoking, 529 taste effects, 586 flavour extraction, 174
tea infusions, 436 temperature effects, 587 food matrix, 344
tongue movement and intensity, 586 yogurt, 185 food pairing, 348
Ursolic acid, 316 Visual appearance, seaweeds, 517–518 maceration, 174
Visual presentation, plating science, 459–461 meat marination, 92
Vacuum Vitamin C see ascorbic acid microfiltration, 337
vacuum frying, 366 Vitamin E, 308, 425 see also tocopherols music effects on tasting, 570, 571–572
vacuum pumps, freeze-drying, 350 Vitamins odour, 329
sous vide cooking, 452–453, 531–535 butter, 230 oxidation, 715
Valine (Val), 464, 465 lipid-soluble, 193 paraethylphenol, 809–810
Van Ermengem, Emile, 88 pasteurization effects, 456 pasteurization, 451
Vanilla, 268, 586–587 Vitrification, 351–352 production methods, 22
Vanillin, 528, 809 Vodka, 23 proteases, 469
Van’t Hoff, Jacobus Henricus, 441, 443 Volatility, 213–219, 265, 282–283 sauces, 166–167
Vaporization, 382, 383 Vortices, Champagne bubbles, 100–101 trends and innovations, 22–23
Vapour–֪liquid equilibrium (VLE), 282 waste filtration, 338, 339
Vapour pressure, 204, 265 Wakame, 518 Wood
Vauquelin, Nicolas, 223 Walnut, roasting, 484 barbecue, 63
Vauquelin eggs, 223, 225, 431 Water smoking, 529
Veal, fond, 166 boiling, 172 Wot, 43
Vegetables coffee/water ratio, 146–147 Würtz foams, 804
calcium texture effects, 437–438 desalination, 336, 444
dehydration, 411 evaporation, 281–287 Xanthan gum, 54, 229, 302–303, 385, 589
freeze-drying, 177, 355–356 extraction usage, 436 Xanthines, 481
imaging methods, 411, 412 freeze-drying, 349–356 X-ray diffraction (XRD), 25, 26
nutrients, 454–455 ice cube melting, 172 X-ray imaging, 409
pasteurization, 454–456 kefir, 331 Xtabentun, 233
pickles, 411, 520 microwave defrosting, 431 Xylitol, 127–128, 549, 557
“shitao” capillarity, 94 mineral ions effects, 436
sous vide cooking, 452, 455 moisture content, 203–204 Yeasts
storage firmness, 438 polarity, 227 odorant compound formation, 328–330
tsukemono, 411, 520, 593–597 surface tension, 258–259, 358 compared to chemical leaveners, 41
vegetable salad (recipe), 785–787 vitrification, 351–352 Ehrlich pathway, 328, 329
Vegetable stocks, bioactivity, 76 wastewater recovery, 336, 337, 339 fermentation, 19, 21, 327–330
Veri-green process, 155 water activity (WA), 285, 550 kefir, 331–333
Viennoiserie, 47–51, 824–825 Weissella spp., 321 probiotics, 330
Vinaigrette, 301 Welcome Coffee, 827–828 sourdough bread, 58
jellified EVOO vinaigrette (recipe), 738 Wheat Yersinia enterocolitica, 533
Vinegar Do, Re, Mi (recipe), 763, 765–767 Yogurt
educational activities, 683–684 grinding, 523 production, 183–184, 189, 683
taste-taste interactions, 9–10 plant-based drinks, 230 Reverse Frozen Spherification (recipe),
Vine leaves, dolmas (recipe), 729–730 roasting, 486–487 822–823
4-Vinylguaiacol, 528 Whey protein isolate (WPI) gels, 54–55 rheology, 185–186
Vinylphenol, 15 Whey proteins, 182, 186, 188, 229, 338, 468 taste preferences, 10
Viscosity Whipping, 361, 638–642, 716 Young’s equation, 357
aligot, 119 Whisk(e)y, 23, 530, 809–810 Youssef, Jozef, 460, 570
beer, 235 White sauce, 241, 716
carboxymethylcellulose (CMC), 600 Wine see also Champagne Zang, Auguste, 47
chocolate, 133 alcohol reduction, 337–338 Zein particles, 229
coffee, 147 anthocyanins, 14–15 Zinc, chlorophyll metallo-complexes,
heat transfer, 381 astringency, 134 154–155
hydrocolloids, 386 carotenoids, 152 Zingipain, 200
rheology, 185 category blurring, 23 Zucchini, 316, 443