You are on page 1of 1037

Handbook of Research

on Advances and
Applications in
Refrigeration Systems and
Technologies

Pedro Dinis Gaspar


University of Beira Interior, Portugal

Pedro Dinho da Silva


University of Beira Interior, Portugal

A volume in the Advances in Mechatronics and


Mechanical Engineering (AMME) Book Series
Managing Director: Lindsay Johnston
Managing Editor: Austin DeMarco
Director of Intellectual Property & Contracts: Jan Travers
Acquisitions Editor: Kayla Wolfe
Production Editor: Christina Henning
Development Editor: Caitlyn Martin
Cover Design: Jason Mull

Published in the United States of America by


Engineering Science Reference (an imprint of IGI Global)
701 E. Chocolate Avenue
Hershey PA, USA 17033
Tel: 717-533-8845
Fax: 717-533-8661
E-mail: cust@igi-global.com
Web site: http://www.igi-global.com

Copyright © 2015 by IGI Global. All rights reserved. No part of this publication may be reproduced, stored or distributed in
any form or by any means, electronic or mechanical, including photocopying, without written permission from the publisher.
Product or company names used in this set are for identification purposes only. Inclusion of the names of the products or
companies does not indicate a claim of ownership by IGI Global of the trademark or registered trademark.
Library of Congress Cataloging-in-Publication Data
Handbook of research on advances and applications in refrigeration systems and technologies / Pedro Dinis Gaspar and
Pedro Dinho da Silva, editors.
2 volumes cm
Includes bibliographical references and index.
ISBN 978-1-4666-8398-3 (set : hardcover) -- ISBN 978-1-4666-8399-0 (set : ebook) 1. Thermoelectric cooling. 2.
Food--Cooling--Energy consumption. 3. Refrigeration and refrigerating machinery. I. Dinis Gaspar, Pedro, 1974- editor. II.
Silva, Pedro Dinho da.
TP492.9.H36 2015
621.5’6--dc23
2015008067

This book is published in the IGI Global book series Advances in Mechatronics and Mechanical Engineering (AMME)
(ISSN: Pending; eISSN: pending)

British Cataloguing in Publication Data


A Cataloguing in Publication record for this book is available from the British Library.

All work contributed to this book is new, previously-unpublished material. The views expressed in this book are those of the
authors, but not necessarily of the publisher.

For electronic access to this publication, please contact: eresources@igi-global.com.


Advances in Mechatronics
and Mechanical Engineering
(AMME) Book Series
J. Paulo Davim
University of Aveiro, Portugal

ISSN: Pending
EISSN: pending

Mission
With its aid in the creation of smartphones, cars, medical imaging devices, and manufacturing tools,
the mechatronics engineering field is in high demand. Mechatronics aims to combine the principles of
mechanical, computer, and electrical engineering together to bridge the gap of communication between
the different disciplines.
The Advances in Mechatronics and Mechanical Engineering (AMME) Book Series provides in-
novative research and practical developments in the field of mechatronics and mechanical engineering.
This series covers a wide variety of application areas in electrical engineering, mechanical engineering,
computer and software engineering; essential for academics, practitioners, researchers, and industry leaders.

Coverage
• Manufacturing Methodologies
• Mechanisms and machines
• Bioengineering Materials IGI Global is currently accepting manuscripts
• Control Methodologies for publication within this series. To submit a pro-
posal for a volume in this series, please contact our
• Medical robotics
Acquisition Editors at Acquisitions@igi-global.com
• Nanomaterials and nanomanufacturing or visit: http://www.igi-global.com/publish/.
• Design and Manufacture
• Computer-Based Manufacturing
• Tribology and surface engineering
• Biologically Inspired Robotics

The Advances in Mechatronics and Mechanical Engineering (AMME) Book Series (ISSN Pending) is published by IGI Global, 701 E.
Chocolate Avenue, Hershey, PA 17033-1240, USA, www.igi-global.com. This series is composed of titles available for purchase individually;
each title is edited to be contextually exclusive from any other title within the series. For pricing and ordering information please visit http://
www.igi-global.com/book-series/advances-mechatronics-mechanical-engineering/73808. Postmaster: Send all address changes to above ad-
dress. Copyright © 2015 IGI Global. All rights, including translation in other languages reserved by the publisher. No part of this series may be
reproduced or used in any form or by any means – graphics, electronic, or mechanical, including photocopying, recording, taping, or informa-
tion and retrieval systems – without written permission from the publisher, except for non commercial, educational use, including classroom
teaching purposes. The views expressed in this series are those of the authors, but not necessarily of IGI Global.
Titles in this Series
For a list of additional titles in this series, please visit: www.igi-global.com

Engineering Creative Design in Robotics and Mechatronics


Maki K. Habib (The American University in Cairo, Egypt) and J. Paulo Davim (University of Aveiro, Portugal)
Engineering Science Reference • copyright 2013 • 366pp • H/C (ISBN: 9781466642256) • US $195.00 (our price)

Computational Methods for Optimizing Manufacturing Technology Models and Techniques


J. Paulo Davim (University of Aveiro, Portugal)
Engineering Science Reference • copyright 2012 • 395pp • H/C (ISBN: 9781466601284) • US $195.00 (our price)

701 E. Chocolate Ave., Hershey, PA 17033


Order online at www.igi-global.com or call 717-533-8845 x100
To place a standing order for titles released in this series, contact: cust@igi-global.com
Mon-Fri 8:00 am - 5:00 pm (est) or fax 24 hours a day 717-533-8661
Editorial Advisory Board
Mazyar Amin, Miami University, USA
Dragi Antonijević, University of Belgrade, Serbia
Ciro Aprea, University of Salerno, Italy
Marco Bortolini, University of Bologna, Italy
Federica Carluccio, University of Salento, Italy
Lin Chen, Peking University, China
Don Cleland, Massey University, New Zealand
Dana Dabiri, University of Washington, USA
Mani Sankar Dasgupta, Birla Institute of Technology and Science (BITS) – Pilani, India
Lorenzo De Pascalis, University of Salento, Italy
Judith Evans, London South Bank University, UK
Adriana Greco, University of Naples Federico II, Italy
Dileep Kumar Gupta, Birla Institute of Technology and Science (BITS) – Pilani, India
Moonyong Lee, Yeungnam University, Korea
Homayun Navaz, Kettering University, USA
Cezar Negrão, Federal University of Technology – Parana, Brazil
Kim Tiow Ooi, Nanyang Technological University, Singapore
Giuseppe Starace, University of Salento, Italy
Alison Subiantoro, Nanyang Technological University, Singapore


List of Contributors

Agrawal, Neeraj / Dr. B. A. Technological University Lonere, India................................................. 360


Aguado, Monica / National Renewable Energy Centre (CENER), Spain.......................................... 881
Amin, Mazyar / Miami University, USA............................................................................................ 543
Antonijevic, Dragi / Singidunum University, Belgrade, Serbia.......................................................... 752
Aprea, Ciro / Università di Salerno, Italy.......................................................................................... 295
Belman, Juan Manuel / University of Guanajuato, Mexico.............................................................. 641
Bergander, Marek J / Magnetic Development, Inc., USA & University of Hartford, USA................... 1
Bhattacharyya, Souvik / Indian Institute of Technology Kharagpur, India....................................... 360
Bortolini, Marco / University of Bologna, Italy................................................................................. 260
Butrymowicz, Dariusz Józef / Białystok University of Technology, Poland..................................... 155
Carluccio, Federica / University of Salento, Italy................................................................................ 36
Carvalho, Susana M.P. / Universidade Católica Portuguesa, Portugal............................................ 457
Casado, Sindia / National Renewable Energy Centre (CENER), Spain............................................. 881
Chen, Lin / Tohoku University, Japan & Peking University, China................................................... 106
Dabiri, Dana / University of Washington, USA.................................................................................. 543
Dasgupta, Mani Sankar / Birla Institute of Technology and Science, India..................................... 378
De Pascalis, Lorenzo / University of Salento, Italy.............................................................................. 36
Deuchande, Teresa / Universidade Católica Portuguesa, Portugal................................................... 457
do Nascimento, Samuel Mariano / University of Beira Interior, Portugal....................................... 590
Duarte, M. V. / University of Beira Interior, Portugal........................................................................ 207
Fikiin, Kostadin / Technical University of Sofia, Bulgaria................................................... 490,660,803
Gagan, Jerzy / Białystok University of Technology, Poland............................................................... 155
Galil, Mohamed Abdul / University of Belgrade, Serbia................................................................... 752
Gallegos, Armando / University of Guanajuato, Mexico................................................................... 641
Gamberi, Mauro / University of Padua, Italy.................................................................................... 260
Garcia-Naveda, Gabriel / National Renewable Energy Centre (CENER), Spain............................. 881
Garde, Raquel / National Renewable Energy Centre (CENER), Spain............................................. 881
Gaspar, P. D. / University of Beira Interior, Portugal........................................................................ 207
Gaspar, Pedro Dinis / University of Beira Interior, Portugal............................................................. 590
Graziani, Alessandro / University of Padua, Italy............................................................................. 260
Greco, Adriana / Università di Napoli Federico II, Italy................................................................... 295
Gupta, Dileep Kumar / Birla Institute of Technology and Science, India......................................... 378
Heidinger, Gustavo Galdi / Eletrofrio Refrigeração Ltda, Brazil...................................................... 590
Jimenez, Fernando / National Renewable Energy Centre (CENER), Spain...................................... 881
Karwacki, Jarosław / Institute of Fluid-Flow Machinery of PASc, Poland....................................... 155





Khan, Mohd Shariq / Yeungnam University, South Korea................................................................ 854


Komatina, Mirko S. / University of Belgrade, Serbia........................................................................ 752
Kosi, Franc Franc / University of Belgrade, Serbia........................................................................... 752
Larrigaudière, Christian / IRTA, Spain............................................................................................ 457
Lee, Moonyong / Yeungnam University, South Korea........................................................................ 854
Maiorino, Angelo / Università di Salerno, Italy................................................................................. 295
Manzini, Riccardo / University of Bologna, Italy.............................................................................. 260
Milovancevic, Uros Milorad / University of Belgrade, Serbia.......................................................... 752
Moerman, Frank Theo / Catholic University of Leuven, Belgium............................................. 490,660
Morosuk, Tatiana / Technische Universität Berlin, Germany.............................................................. 85
Navaz, Homayun K / Kettering University, USA............................................................................... 543
Ngoc, Cuong Ne / Chonnam National University, South Korea......................................................... 784
Oh, Jong-Taek / Chonnam National University, South Korea........................................................... 784
Ooi, Kim Tiow / Nanyang Technological University, Singapore........................................................ 720
Pires, L. C. / University of Beira Interior, Portugal............................................................................ 207
Santos da Silva, Paulo Ricardo / University of Vale do Rio dos Sinos – Unisinos, Brazil................ 433
Silva, P. D. / University of Beira Interior, Portugal............................................................................. 207
Silva, Pedro Dinho / University of Beira Interior, Portugal............................................................... 590
Śmierciew, Kamil Leszek / Białystok University of Technology, Poland........................................... 155
Stankov, Borislav / Technical University of Sofia, Bulgaria............................................................... 803
Starace, Giuseppe / University of Salento, Italy................................................................................... 36
Stefanovskiy, Vladimir / All-Russian Scientific Research Institute of Refrigerating Industry,
Russia............................................................................................................................................ 412
Subiantoro, Alison / TUM CREATE, Singapore................................................................................ 720
Tsatsaronis, George / Technische Universität Berlin, Germany.......................................................... 85
Vasconcelos, Marta W. / Universidade Católica Portuguesa, Portugal............................................ 457
Zivkovic, Branislav / University of Belgrade, Serbia......................................................................... 752
Table of Contents

Preface................................................................................................................................................. xxii

Volume I
Chapter 1
Ejector Refrigeration Cycles: Classification of Thermodynamic Cycles with Ejectors.......................... 1
Marek J Bergander, Magnetic Development, Inc., USA & University of Hartford, USA

Chapter 2
The Diffusion Absorption Refrigerator Operation and Performance.................................................... 36
Lorenzo De Pascalis, University of Salento, Italy
Giuseppe Starace, University of Salento, Italy
Federica Carluccio, University of Salento, Italy

Chapter 3
Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration
Systems.................................................................................................................................................. 85
Tatiana Morosuk, Technische Universität Berlin, Germany
George Tsatsaronis, Technische Universität Berlin, Germany

Chapter 4
New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration: Basic Principles and
System Designs.................................................................................................................................... 106
Lin Chen, Tohoku University, Japan & Peking University, China

Chapter 5
Ejection Refrigeration Cycles.............................................................................................................. 155
Dariusz Józef Butrymowicz, Białystok University of Technology, Poland
Kamil Leszek Śmierciew, Białystok University of Technology, Poland
Jerzy Gagan, Białystok University of Technology, Poland
Jarosław Karwacki, Institute of Fluid-Flow Machinery of PASc, Poland

Chapter 6
Current and Future Trends of Refrigerants Development.................................................................... 207
M. V. Duarte, University of Beira Interior, Portugal
 L. C. Pires, University of Beira Interior, Portugal



P. D. Silva, University of Beira Interior, Portugal


P. D. Gaspar, University of Beira Interior, Portugal

Chapter 7
Retrofitting of R404a Commercial Refrigeration Systems with R410a and R407f HFCs
Refrigerants.......................................................................................................................................... 260
Mauro Gamberi, University of Padua, Italy
Marco Bortolini, University of Bologna, Italy
Alessandro Graziani, University of Padua, Italy
Riccardo Manzini, University of Bologna, Italy

Chapter 8
Transcritical Carbon Dioxide Refrigeration as an Alternative to Subcritical Plants: An
Experimental Study.............................................................................................................................. 295
Adriana Greco, Università di Napoli Federico II, Italy
Ciro Aprea, Università di Salerno, Italy
Angelo Maiorino, Università di Salerno, Italy

Chapter 9
Capillary Tube as an Expansion Device in a CO2 (R744) Transcritical Heat Pump System.............. 360
Neeraj Agrawal, Dr. B. A. Technological University Lonere, India
Souvik Bhattacharyya, Indian Institute of Technology Kharagpur, India

Chapter 10
Transcritical CO2 Refrigeration System in Tropical Region: Challenges and Opportunities............. 378
Dileep Kumar Gupta, Birla Institute of Technology and Science, India
Mani Sankar Dasgupta, Birla Institute of Technology and Science, India

Chapter 11
Processes and Technological Systems for Freezing of Foodstuff........................................................ 412
Vladimir Stefanovskiy, All-Russian Scientific Research Institute of Refrigerating Industry,
Russia

Chapter 12
Study of Thermal and Microbiological Behavior of Foods Submitted to Evaporative Cooling
Process................................................................................................................................................. 433
Paulo Ricardo Santos da Silva, University of Vale do Rio dos Sinos – Unisinos, Brazil

Volume II
Chapter 13
Advances in Refrigerated and Controlled Atmosphere Storage of Fruits and Vegetables.................. 457
Teresa Deuchande, Universidade Católica Portuguesa, Portugal
Susana M.P. Carvalho, Universidade Católica Portuguesa, Portugal
Christian Larrigaudière, IRTA, Spain
Marta W. Vasconcelos, Universidade Católica Portuguesa, Portugal


Chapter 14
Guiding Principles for Hygienic Design of Evaporators to Mitigate Contamination-Related Risks
in Air Blast Freezing Systems.............................................................................................................. 490
Frank Theo Moerman, Catholic University of Leuven, Belgium
Kostadin Fikiin, Technical University of Sofia, Bulgaria

Chapter 15
Aerodynamic Isolation of Open Refrigerated Vertical Display Cases using Air Curtains.................. 543
Mazyar Amin, Miami University, USA
Dana Dabiri, University of Washington, USA
Homayun K Navaz, Kettering University, USA

Chapter 16
Experimental Analysis to Optimize the Performance of Air Curtains and Heat Exchangers:
Application to Refrigerated Display Cases.......................................................................................... 590
Samuel Mariano do Nascimento, University of Beira Interior, Portugal
Gustavo Galdi Heidinger, Eletrofrio Refrigeração Ltda, Brazil
Pedro Dinis Gaspar, University of Beira Interior, Portugal
Pedro Dinho Silva, University of Beira Interior, Portugal

Chapter 17
Implementation of Thermal and Energy Improvements in Domestic Refrigeration: Case Studies..... 641
Juan Manuel Belman, University of Guanajuato, Mexico
Armando Gallegos, University of Guanajuato, Mexico

Chapter 18
Effect of Hygienic Design and Operational Parameters on Frosting and Defrosting of Evaporators
in Refrigerated Food Processing and Storage Facilities...................................................................... 660
Frank Theo Moerman, Catholic University of Leuven, Belgium
Kostadin Fikiin, Technical University of Sofia, Bulgaria

Chapter 19
Expansion Power Recovery in Refrigeration Systems......................................................................... 720
Alison Subiantoro, TUM CREATE, Singapore
Kim Tiow Ooi, Nanyang Technological University, Singapore

Chapter 20
Cold Thermal Energy Storage............................................................................................................. 752
Franc Franc Kosi, University of Belgrade, Serbia
Branislav Zivkovic, University of Belgrade, Serbia
Mirko S. Komatina, University of Belgrade, Serbia
Dragi Antonijevic, Singidunum University, Belgrade, Serbia
Mohamed Abdul Galil, University of Belgrade, Serbia
Uros Milorad Milovancevic, University of Belgrade, Serbia


Chapter 21
Development of Heat Pump System Using Thermobank and Ejector for Heating Room and Cold
Storage................................................................................................................................................. 784
Jong-Taek Oh, Chonnam National University, South Korea
Cuong Ne Ngoc, Chonnam National University, South Korea

Chapter 22
Integration of Renewable Energy in Refrigerated Warehouses........................................................... 803
Kostadin Fikiin, Technical University of Sofia, Bulgaria
Borislav Stankov, Technical University of Sofia, Bulgaria

Chapter 23
Optimization of Natural Gas Liquefaction Process............................................................................. 854
Mohd Shariq Khan, Yeungnam University, South Korea
Moonyong Lee, Yeungnam University, South Korea

Chapter 24
Use of Hydrogen and Fuel Cells for Refrigerated Transport............................................................... 881
Raquel Garde, National Renewable Energy Centre (CENER), Spain
Sindia Casado, National Renewable Energy Centre (CENER), Spain
Fernando Jimenez, National Renewable Energy Centre (CENER), Spain
Gabriel Garcia-Naveda, National Renewable Energy Centre (CENER), Spain
Monica Aguado, National Renewable Energy Centre (CENER), Spain

About the Contributors................................................................................................................... xxvii

Compilation of References.................................................................................................................xlii

Index..................................................................................................................................................... cix
Detailed Table of Contents

Preface................................................................................................................................................. xxii

Volume I
Chapter 1
Ejector Refrigeration Cycles: Classification of Thermodynamic Cycles with Ejectors.......................... 1
Marek J Bergander, Magnetic Development, Inc., USA & University of Hartford, USA

This chapter describes a collaborative effort of US private companies and various departments of the US
Government to investigate the possibility of improving the efficiency of HVAC systems by use of one and
two-phase ejectors. It is anticipated that this technology, when fully developed will result in attractive,
energy saving products that significantly improve the performance of commercial and residential chiller/
air-conditioning systems, refrigeration plants, and heat pumps (geothermal and air-source). Although the
literature describing ejector applications in refrigeration dates back to the year of 1900, the ejector use
was always considered as controversial, because the previous research had resulted with only theoretical
results and without visible, commercial products. The research on the ejector application is consistent
with present directions in the HVAC industry and it will attract more attention and research funding in
the future. A classification of thermodynamic cycles where ejectors can be applied composed by three
distinctive “categories” is suggested.

Chapter 2
The Diffusion Absorption Refrigerator Operation and Performance.................................................... 36
Lorenzo De Pascalis, University of Salento, Italy
Giuseppe Starace, University of Salento, Italy
Federica Carluccio, University of Salento, Italy

This chapter focuses on the Diffusion Absorption Refrigerator (DAR) cycle and describes a new advanced
thermodynamic model which allows good predictions of the chiller performance in terms of efficiency
and cooling capacity, starting from a precise evaluation of the thermo-physical properties of the working
mixture at each point of the circuit. A steady state thermodynamic analytical model of the thermal pump
driving the DAR is also included. In addition, the experimental validation of the model, performed on
a prototype built coupling a domestic 750 W-magnetron with a small purposely modified commercial
DAR to activate the thermal pump, is here included: a maximum mismatch of 2.32% in the weak mixture
mass flow rate and lower than 5% in COP between the predicted and measured data were found.





Chapter 3
Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration
Systems.................................................................................................................................................. 85
Tatiana Morosuk, Technische Universität Berlin, Germany
George Tsatsaronis, Technische Universität Berlin, Germany

In the last decades an exergetic analysis became increasingly popular because this analysis identifies the
location, magnitude and sources of thermodynamic inefficiencies. A conventional exergetic analysis,
however, does not consider (a) the real potential for improving a system nor (b) the interactions among
the components of the system. The interactions among different components of the same system can
be estimated and the quality of the conclusions obtained from an exergetic evaluation can be improved,
when the exergy destruction (irreversibilities) within each system component are split into endogenous/
exogenous and avoidable/ unavoidable parts. We call this advanced exergetic analysis. The purpose
of this chapter is to demonstrate that the advanced exergetic analysis is a practical method that allows
engineers to extract useful information and conclusions and to develop ideas and solutions that cannot
be suggested by other methods. In this chapter the conventional and advanced exergetic analysis are
applied to an air refrigeration machine.

Chapter 4
New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration: Basic Principles and
System Designs.................................................................................................................................... 106
Lin Chen, Tohoku University, Japan & Peking University, China

Sublimation is one phase change mechanism which usually happens under low-to-moderate temperatures
and at the same time large amounts of latent heat is absorbed or released. Low temperature sublimation
has been proposed in a lot of applications as one useful fast cooling/refrigeration mechanisms, such as
medical cooling, food engineering, chemical synthesis, domestic cooling and many industrial sectors. In
this brief chapter, the basic mechanisms of static sublimation process and sublimation two-phase flows
are clarified and analyzed first, which covers the theoretical and physical problems of sublimation phase-
change. Then the previous studies are classified into numerical modeling and experimental verifications.
Representative refrigeration systems are also introduced and compared in this chapter, which may give
useful indications for future innovations in this field. Future research focuses are also summarized and
proposed in this chapter.

Chapter 5
Ejection Refrigeration Cycles.............................................................................................................. 155
Dariusz Józef Butrymowicz, Białystok University of Technology, Poland
Kamil Leszek Śmierciew, Białystok University of Technology, Poland
Jerzy Gagan, Białystok University of Technology, Poland
Jarosław Karwacki, Institute of Fluid-Flow Machinery of PASc, Poland

The chapter presents the development of ejector refrigeration technology that strongly reduces the
greenhouse gases emission by using natural refrigerants and also dramatically reduces the need for the
electric power. This is accomplished by using free or inexpensive heat – either solar or waste heat, as the
main source of energy instead of electricity. Nowadays, the thermal driven refrigeration system, especially
with low-temperature heat source became more and more popular. The operation of the ejection cycle
using low-temperature heat source can be considered as very attractive and the ejection cycles becomes
truly competitive in comparison with the absorption refrigeration systems.


Chapter 6
Current and Future Trends of Refrigerants Development.................................................................... 207
M. V. Duarte, University of Beira Interior, Portugal
L. C. Pires, University of Beira Interior, Portugal
P. D. Silva, University of Beira Interior, Portugal
P. D. Gaspar, University of Beira Interior, Portugal

In this chapter is addressed the thematic of refrigerants: its historical evolution; properties; legislation
applied in the area and future trends. The first refrigerant being marketed on a large scale was ethyl
ether (R610), in 1834. Since then, the evolution of the utilized refrigerants was stimulated, initially
due to constructive issues in the refrigeration system and later to environmental issues. This evolution
may be divided into four generations: 1st use of any fluid that worked; 2nd safety and durability of the
equipment; 3rd ozone layer protection and 4th increase of global warming concerns. During the process
of evolution many refrigerants were tested to understanding of their properties. Currently, environmental
concerns are taken as guide in the search for new refrigerants. The most promising refrigerants to be
used in future are the HFEs, HFOs and HFCs with low-GWP, natural refrigerants and blends between
(HCs/HFCs and HFCs/HFOs) refrigerants.

Chapter 7
Retrofitting of R404a Commercial Refrigeration Systems with R410a and R407f HFCs
Refrigerants.......................................................................................................................................... 260
Mauro Gamberi, University of Padua, Italy
Marco Bortolini, University of Bologna, Italy
Alessandro Graziani, University of Padua, Italy
Riccardo Manzini, University of Bologna, Italy

This chapter presents an experimental analysis about the retrofitting of two commercial stationary
refrigeration systems marketed by an Italian leading company of the sector. Such systems operate both at
medium temperature (MT) and low temperature (LT) and they are originally designed to work with the
high global warming potential (GWP) hydrofluorocarbon (HFC) R404a fluid (GWP = 3922). The purpose
is to investigate the performances of HFCs R410a (GWP = 2088) and R407f (GWP = 1825) chosen as
effective alternatives to HFC R404a, due to their compatibility, non-flammability and market availability.
Furthermore, such fluids meet the EU restrictions in force in the next future for high GWP HFCs. The
experimental analysis compares the performances, in terms of COP and cooling capacity, of R404a and
the two identified alternatives under different operating conditions, i.e. chamber and condenser inlet air
temperatures. In case of comparable performances, significant environmental benefits are introduced by
the adoption of R407f and R410a in the MT and LT refrigeration systems.

Chapter 8
Transcritical Carbon Dioxide Refrigeration as an Alternative to Subcritical Plants: An
Experimental Study.............................................................................................................................. 295
Adriana Greco, Università di Napoli Federico II, Italy
Ciro Aprea, Università di Salerno, Italy
Angelo Maiorino, Università di Salerno, Italy

Carbon dioxide (R744) is as a valid alternative to classical substances such as HFCs used in vapour
compression plants. A transcritical refrigeration cycle is needed because the critical temperature of carbon
dioxide is usually lower than the ambient temperature. In this chapter the performances of a transcritical


cycle have been evaluated with a prototype R744 system working as a classical spit-systems to cool air.
An experimental analysis has been carried out on the effect of: refrigerant charge, internal heat exchanger,
heat rejection pressure on the energetic performances of the transcritical plant. An experimental analysis
of a hybrid trans-critical refrigerator-desiccant wheel system has been carried out in order to improve
the COP. The experimental transcritical cycle has been examined in comparison with a classical vapour
compression plant working with the R134a.

Chapter 9
Capillary Tube as an Expansion Device in a CO2 (R744) Transcritical Heat Pump System.............. 360
Neeraj Agrawal, Dr. B. A. Technological University Lonere, India
Souvik Bhattacharyya, Indian Institute of Technology Kharagpur, India

Natural refrigerants which are ecologically safe and were in use extensively in the pre-CFC era are
witnessing a revival of CO2 (R744). Inherently being a low critical temperature (31.2 ○C) refrigerant,
the CO2 cycle based system operates in transcritical mode offers an opportunity to obtain an optimum
operating condition. Capillary tubes which are extensively used in small size vapour compression
systems work very differently in a CO2 transcritical heat pump system. In this chapter it is described
that installation of a capillary tube having an appropriately designed length replacing an expansion valve
will result in a natural adjustment of the gas cooler pressure, so that the system balance always shifts to
a favourable COP direction; this is contrary to the scepticism that exists on the capability of a capillary
tube to attain the optimal pressure operation. There is an optimal length of capillary tube for a given
diameter at which the heat pump runs optimally.

Chapter 10
Transcritical CO2 Refrigeration System in Tropical Region: Challenges and Opportunities............. 378
Dileep Kumar Gupta, Birla Institute of Technology and Science, India
Mani Sankar Dasgupta, Birla Institute of Technology and Science, India

Environmental concerns and enactments of Montreal and Kyoto Protocol for sustainable growth is a
welcome impetus for researchers towards a quest for ecologically safe and natural refrigerants and cost
effective designs of refrigeration systems. Carbon dioxide (CO2) is one such natural refrigerants that,
although was abandoned once due techno-economic reason, has been receiving tremendous attention
and is viewed as a strong candidate for long term alternative of synthetic refrigerants. The commercial
success of CO2 as a refrigerant and its universal acceptance, however demands cost effective and widely
accepted technology operable under various environmental conditions. In this chapter, the use of CO2
as refrigerant in trans-critical vapor compression system is discussed in detail along with its unique
challenges associated with operating in tropical region. Further the opportunities for using these systems
in tropical region with specific systematic modification are explored. Discussions on component design
and system level performance analysis are also included in the chapter.

Chapter 11
Processes and Technological Systems for Freezing of Foodstuff........................................................ 412
Vladimir Stefanovskiy, All-Russian Scientific Research Institute of Refrigerating Industry,
Russia

The principles and methods of holistic understanding of the foodstuff freezing process, as well as
structure, functioning and development of refrigeration systems are set out in this chapter. It is meant
for students, post-graduate students and experts working in the field of improvement of processes and


devices of refrigerating productions. This chapter represents a great interest and for science officers as
careful studying of the questions stated in it will relieve them of failures and repetition of the errors
which were taking place in Russia, and will help them to choose ways for effective research work in the
field of conservation of foodstuff by freezing.

Chapter 12
Study of Thermal and Microbiological Behavior of Foods Submitted to Evaporative Cooling
Process................................................................................................................................................. 433
Paulo Ricardo Santos da Silva, University of Vale do Rio dos Sinos – Unisinos, Brazil

Refrigeration is a widely used technique in the food industry to avoid microbiological spoilage of
products. Several variables can affect the performance of chilling process, including the shape and the
dimensions of the product, the thermophysical properties of the food and the cooling fluid. The evaporative
cooling increases the rate at which the temperature of the products is reduced. However, it makes the
refrigeration process more complex, since it involves the simultaneous mass and heat transfer. In this
context, numerical simulation is a useful tool to analyze the processes because it allows investigating
different operational conditions in a virtual environment, in a quick and inexpensive way. The goal of
this chapter was to study the influence of process variables on thermal and microbiological behavior
of minimally processed foods submitted to refrigeration process with and without evaporative cooling
effect, by means of two case studies.

Volume II
Chapter 13
Advances in Refrigerated and Controlled Atmosphere Storage of Fruits and Vegetables.................. 457
Teresa Deuchande, Universidade Católica Portuguesa, Portugal
Susana M.P. Carvalho, Universidade Católica Portuguesa, Portugal
Christian Larrigaudière, IRTA, Spain
Marta W. Vasconcelos, Universidade Católica Portuguesa, Portugal

Long term storage of a variety of crops as well as long-distance transport, has allowed meeting the
consumers’ expectations in the supply of many types of fresh fruits and vegetables throughout the year.
This is only possible with the use of several postharvest technologies. This chapter starts with a brief
historical context followed by an overview of the technologies used for fruits and vegetables storage,
including refrigerated and controlled atmosphere (CA) storage as well as the most recently developed
technologies for storing these produces. We also address the innovation requirements in the refrigeration
systems when integrating cold storage with CA, including the need for higher refrigeration capacity,
use of air tight storage chambers, CO2 scrubbers and atmosphere generators. The effects of these
methodologies on fruit physiology and quality during storage are further discussed. Finally, the current
recommendations for long term storage using ‘Rocha’ pear as a case study are presented.


Chapter 14
Guiding Principles for Hygienic Design of Evaporators to Mitigate Contamination-Related Risks
in Air Blast Freezing Systems.............................................................................................................. 490
Frank Theo Moerman, Catholic University of Leuven, Belgium
Kostadin Fikiin, Technical University of Sofia, Bulgaria

Evaporator coils, fins and drain pans in blast freezers may act as a source of microbial, physical and
chemical contamination (representing Critical Control Points within a HACCP study). For instance,
the evaporator might contaminate unpacked food products passing through the freezer with undesirable
ingredients remaining from a previous product being frozen. Drain pans below the refrigerating coils
and fans also require paramount attention, due diligence and professional care to ensure that both frozen
foods and freezer interior are kept at the highest possible hygienic level. This book chapter is, therefore,
focussing on knowledge-based hygienic design of evaporators in air blast freezing systems (e.g., proper
production and mounting of refrigerating coils, drain pans and fans, along with suitable materials of
construction to reduce frosting, microbial fouling and corrosion) with the ultimate aim of acquiring a
safe environment and extended shelf life of unwrapped products. An overview of the existing types of
air blast freezers is presented for that purpose. Furthermore, the today’s state of the art is introduced to
food chain operators who are not enough familiar with hygienic design. Typical contamination problems
and measures for contamination control are discussed as well.

Chapter 15
Aerodynamic Isolation of Open Refrigerated Vertical Display Cases using Air Curtains.................. 543
Mazyar Amin, Miami University, USA
Dana Dabiri, University of Washington, USA
Homayun K Navaz, Kettering University, USA

This chapter presents an experimental study on the effects of some of the most important variables that
affect the infiltration rate in open refrigerated vertical display cases at steady operation. The variables
were defined in dimensionless forms and were categorized in two groups. The goal was to examine the
cross-effects of the first group (primary variables) by performing tests on the permutation of the variables,
and to obtain some correction functions that contain the effects of the (secondary) variables thought
to have smaller impact on infiltration. Tracer gas method, a new means to measure infiltration rate, is
described and compared with other conventional methods. Most variables showed noticeable impacts
on infiltration either linearly or non-linearly, some of whom did not show very well-defined patterns
with infiltration, which indicates there are significant cross-effects among the variables. Others such as
temperature difference and relative humidity showed no or negligible effects on infiltration.


Chapter 16
Experimental Analysis to Optimize the Performance of Air Curtains and Heat Exchangers:
Application to Refrigerated Display Cases.......................................................................................... 590
Samuel Mariano do Nascimento, University of Beira Interior, Portugal
Gustavo Galdi Heidinger, Eletrofrio Refrigeração Ltda, Brazil
Pedro Dinis Gaspar, University of Beira Interior, Portugal
Pedro Dinho Silva, University of Beira Interior, Portugal

This chapter reports an overview about experimental studies concerning the thermal performance of air
curtains and heat exchangers installed in vertical open refrigerated display cases. The air curtain analysis
shows the influence on the thermal performance by varying the width of the discharge air grille and the
perforation density of the back panel by a mathematical model. The variation on the perforation density
of the back panel and the width of discharge air grille alter significantly the thermal entrainment factor
and the energy consumption of the equipment. Focusing the influence of environmental conditions on
the performance of the heat exchanger, a second mathematical model was developed to evaluate the
total heat load, its partial components and the condensate water mass. This analysis provides valuable
information to the design of the air curtain and heat exchanger based on in-store environmental conditions
and airflow efficiency.

Chapter 17
Implementation of Thermal and Energy Improvements in Domestic Refrigeration: Case Studies..... 641
Juan Manuel Belman, University of Guanajuato, Mexico
Armando Gallegos, University of Guanajuato, Mexico

This chapter concisely discusses two case studies experiences on domestic refrigeration. One of the
cases involves the theoretical and experimental analysis of the thermal profile in the compartments of a
refrigerator, all this under the modeling and simulation through CFD, thus obtaining interesting results
in terms of energy consumption. The second case is focus on the thermal and energy evaluation of new
material proposed as thermal insulator, which was developed in a conventional way in the laboratory
and presents convenient thermal features. Therefore, the guidelines for this chapter are aimed at finding
mechanism that streamline the domestic refrigeration systems, without modifying its cooling performance.

Chapter 18
Effect of Hygienic Design and Operational Parameters on Frosting and Defrosting of Evaporators
in Refrigerated Food Processing and Storage Facilities...................................................................... 660
Frank Theo Moerman, Catholic University of Leuven, Belgium
Kostadin Fikiin, Technical University of Sofia, Bulgaria

Proper control and performance of evaporators in food refrigeration facilities are vital to provide a suitable
temperature regime, safety, quality and wholesomeness of refrigerated products at minimum electricity
costs. When humid air passes along the surfaces of a low-temperature evaporator, frost is usually formed,
which decreases the heat transfer efficiency. Frosting and defrosting phenomena have been extensively
investigated for different industrial scenarios and extensive literature exists in the matter. However,
no studies have been published so far to address in a comprehensive way the methods and patterns of
evaporator defrosting as affected by hygienic design implications and criteria. This book chapter is


intended to fill in this gap by enforcing hygienic imperatives in the evaporator design. Various design
solutions and conditions of operation are considered as decisive in determining the amount, thickness
and structure of the frost build-up. Advantages and drawbacks of diverse defrost methods are outlined
with regards to contamination risks in refrigeration facilities.

Chapter 19
Expansion Power Recovery in Refrigeration Systems......................................................................... 720
Alison Subiantoro, TUM CREATE, Singapore
Kim Tiow Ooi, Nanyang Technological University, Singapore

Vapor-compression systems are the most popular method of refrigeration. However, the throttling loss at
the expansion valve is one of the “energy parasites” of such systems. This is especially acute in systems
with large operating pressure differences like the transcritical CO2 refrigeration systems. In this chapter,
a method to solve this issue by using an expander to recover the expansion energy of refrigeration
systems is explained. Relevant research works are then discussed to provide a general overview about
the state of the art technology. Various types of expander mechanisms, including reciprocating, rolling
piston, rotary vane, scroll, screw, turbine, swing piston and revolving vane, are discussed. Works on the
various aspects of expanders are also discussed. These include heat transfer, exergy analysis, expansion
process, internal leakage, lubrication, integration with refrigeration systems and the economic aspects.

Chapter 20
Cold Thermal Energy Storage............................................................................................................. 752
Franc Franc Kosi, University of Belgrade, Serbia
Branislav Zivkovic, University of Belgrade, Serbia
Mirko S. Komatina, University of Belgrade, Serbia
Dragi Antonijevic, Singidunum University, Belgrade, Serbia
Mohamed Abdul Galil, University of Belgrade, Serbia
Uros Milorad Milovancevic, University of Belgrade, Serbia

The chapter gives an overview of cold thermal energy storage (CTES) technologies. Benefits as well as
classification and operating strategies of CTES are discussed. Design consideration and sizing strategies
based on calculated load profile for design day is presented. Some recommendation concernig designing of
CTES equipment are given. Special attention was paid to the analysis of specific features of heat transfer
phenomena in ice storage tank including the assessment of the duration and the rate of ice formation
and melting. The methodology of sizing components of the ice thermal storage system included in an
air conditioning system for a office building situated in hot wet and dry climate are presented. Based on
hourly cooling load calculation that was carried out using Carrier’s Hourly Analysis Program, sizing of
ice thermal storage system for different operating strategies included full, chiller priority and ice priority
storage operation for the design day are presented. Finally, an analysis of some operational characteristics
of the system are analyzed.


Chapter 21
Development of Heat Pump System Using Thermobank and Ejector for Heating Room and Cold
Storage................................................................................................................................................. 784
Jong-Taek Oh, Chonnam National University, South Korea
Cuong Ne Ngoc, Chonnam National University, South Korea

In this study, heat pump system has been developed by CFD simulation and experimental investigation.
It studies the thermal behavior of a thermobank and COP on heat pump system and cold storage. The
thermobank stores the wasted heat of during refrigeration cycles and its energy is used for defrosting
process and heating room. It also reduces defrosting time and condensation load so that the temperature
regulation in the cold storage is constant. The system is investigated experimentally and CFD simulated under
thermobank. The dimension L×W×H of cold storage is 3×1,6×1,4 m, thermobank is 600×300×400mm.
The temperature of ambient on CFD simulation process is 150C.This heat pump system can be used to
keep preservation of agricultural products in cold storage warehouse together with floor panel heating for
room in winter. The ejector is used in system which the purpose increases in coefficient of performance
(COP) and decrease in compressor displacement. As the calculated results, its COP is increased about
38% when using thermobank and ejector in heat pump system.

Chapter 22
Integration of Renewable Energy in Refrigerated Warehouses........................................................... 803
Kostadin Fikiin, Technical University of Sofia, Bulgaria
Borislav Stankov, Technical University of Sofia, Bulgaria

Refrigerated warehouses are large energy consumers and account for a significant portion of the global
energy demand. Nevertheless the opportunity for integration of renewable resources in the energy
supply of large cold storage facilities is very often unjustifiably neglected, whereas the employment of
renewable energy for many other industrial and comfort applications is actively promoted and explored.
In that context, the purpose of this chapter is to bridge the existing gap by raising the public awareness
of stakeholders, researchers, practicing engineers and policy makers about the availability of a number
of smart engineering solutions and control strategies to exploit renewables of different nature (solar,
wind, geothermal, biogas, etc.) in the food storage sector, as well as by calling the readers’ attention to
the specialised knowledge in the matter, which has been published so far.

Chapter 23
Optimization of Natural Gas Liquefaction Process............................................................................. 854
Mohd Shariq Khan, Yeungnam University, South Korea
Moonyong Lee, Yeungnam University, South Korea

This chapter provides a brief review of the developments in the optimization of Natural gas (NG)
liquefaction techniques since 2001. NG liquefaction is energy intensive and small improvements in
liquefaction efficiency brings huge cost benefits thus optimization is needed. To tackle the NG liquefaction
optimization problem, two different optimization philosophies, i) deterministic and ii) stochastic, have been
adopted. The limitations of the deterministic approach have paved the way for derivative-free stochastic
approaches. Although both techniques work well for the reported problem, their application is limited to
the specific problems and generalization is quite difficult. Therefore to overcome this problem, a third of
the so called knowledge-inspired class have been evolved for NG liquefaction optimization. Thus, this
chapter covers the major development that took place in NG liquefaction area and after reviewing the
trends future research directions are given.


Chapter 24
Use of Hydrogen and Fuel Cells for Refrigerated Transport............................................................... 881
Raquel Garde, National Renewable Energy Centre (CENER), Spain
Sindia Casado, National Renewable Energy Centre (CENER), Spain
Fernando Jimenez, National Renewable Energy Centre (CENER), Spain
Gabriel Garcia-Naveda, National Renewable Energy Centre (CENER), Spain
Monica Aguado, National Renewable Energy Centre (CENER), Spain

Benchmark refrigeration systems in the road transportation sector are powered by diesel, having operation
costs of up to 6,000 €/y with the consequent increase of the goods cost. This chapter presents an alternative
refrigeration system based on fuel cells (FC) and hydrogen as fuel, with higher efficiency, reduced costs
and independent of diesel price fluctuations. Examples of the energy load profiles impact on the FC
sizing, H2 consumption and system autonomy are presented as well as a description of the FC model
and performance simulation results. The economical feasibility of this new refrigeration system linked
to renewable energies is also analyzed and an economical assessment for different scenarios is presented.

About the Contributors................................................................................................................... xxvii

Compilation of References.................................................................................................................xlii

Index..................................................................................................................................................... cix
xxii

Preface

Due to the growing political, economical and social concerns about energy consumption, in particular
with the need to reduce greenhouse gas emissions and fossil fuel consumption, taking into account the
sharp rise of energy costs in recent years, sustainability and competitiveness of industries depend, to
some extent, on the ability to use energy rationally and efficiently. The refrigeration systems are widely
used in domestic, commercial and industrial applications. Simultaneously, these systems are large energy
consumers, specifically of electrical energy, besides presenting a global negative environmental impact.
Therefore, there is a need to develop numerical and experimental research in order to promote the energy
efficiency and to reduce the carbon footprint.
In terms of food cold chain, the sustainability and food safety are a major concern of consumers in
recent years. There has been an increasing demand for foods as well as a change in consumption habits.
To deal with this new trend, there is a greater availability and variety of food with longer shelf life. In
this context, refrigeration plays an important role, not only for its ability to preserve the products with
the maintenance of its physical, chemical, nutritional and organoleptic characteristics, but also because
it is an indispensable tool for processing, distributing and storage of perishable foods.
In general, processes involving food products are carried out in artificial environments with strict
control of air temperature, humidity and velocity, resulting in the use of systems which generate cold
and heat, although cooling systems are the main responsible by the energy consumption.
To improve the refrigeration sustainability is necessary to pursuit new technological and scientific
advances on both processes and equipments, through their innovative designs. These designs configure
improved vapor compression cycles and efficient heat and mass transfer processes that simultaneously
can minimize the refrigerant charge. Commercial and industrial refrigeration require processing, trans-
portation and storage in refrigerated environments. The improvement of the refrigeration process in these
environments require a better understanding of the cold production itself, as well as the facilities design
optimization, proper selection of devices and equipments, and the development and use of new refrigerants.
These procedures allow ensuring food quality and safety both in commercial and industrial sectors, and
the quality of specific products produced in the latter sector. Thus, research and development (R&D) in
refrigeration is towards the development of sustainable technologies in order to reduce carbon footprint,
mitigate climate change and improve energy efficiency. So, it includes the development of innovative
and/or alternative cooling technologies and their novel concepts, associated with improved control and
monitoring systems based on advanced techniques and technologies such as artificial intelligence and
energy harvesting, respectively. These studies follow two main paths: experimental or numerical studies.
The experimental studies involve the use of several equipment and techniques for measuring relevant



Preface

parameters variation. The other path require modeling and predicting the parameters variation through
computational tools. These results of R&D in refrigeration can provide guidelines for new legislations,
policies and strategies related to the energy efficiency, equipment safety and performance.
The Handbook of Research on Advances and Applications in Refrigeration Systems and Technolo-
gies identifies the main issues involving refrigeration processes and systems. It provides an introduction
to refrigeration processes and systems and their advance due to the ongoing research, being not limited
to these topics but exploring a wide span of applications, their technologies, equipments, procedures,
environmental impact and energy concerns. Moreover, it aims to unveil trends and opportunities for the
improvement of the sector’s efficiency and its sustainability. The handbook reviews the state of the art
and recent advances in several application areas of refrigeration systems. It covers specific subjects,
from the difficulties that arise from the frost formation on the surface area of evaporators to the carbon
footprint triggered from refrigerants use. Nevertheless, it also provide a board insight of refrigeration,
focusing general aspects likewise optimization, sustainability and technology innovation in cold chains.
At the same time, it is aimed to be a reference text presenting true implications of the use of refrigeration,
its devices operation and research efforts to improve processes and systems efficiency.
The Handbook of Research on Advances and Applications in Refrigeration Systems and Technologies
is organized in such a way that highlights both specific and general topics in order to provide its use to
a large audience with refrigeration interest, from undergraduate students to researchers. It can be used
in didactic context by under-graduate and graduate students, and in research context by post-graduate
students and researchers. The practicing engineers can find in this book several topics that can be an
added value to the selection, design and retrofit of refrigeration systems, equipments and facilities.
Although some topics are placed at a level that does not require considerable previous expertise in
the technical details of specific areas, others are somehow knowledge demanding due to the complex
research discussion.
The Handbook of Research on Advances and Applications in Refrigeration Systems and Technologies
is organized into twenty-four chapters. The chapters present the recent research concerning refrigeration
systems and applications and survey the state of the art in this field. The contributors review advances
in processes and equipments design, efficient heat and mass transfer processes, optimization of real op-
erating systems, modeling and predictive computational tools, innovative technologies, among others.
A brief description of each of the chapters follows:
Chapter 1 describes a collaborative effort of US private companies and various departments of the US
Government to investigate the possibility of improving the efficiency of HVAC systems by use of one
and two-phase ejectors. The author anticipates that this technology, when fully developed, will result in
attractive, energy saving products that significantly improve the performance of commercial and residen-
tial chiller/air-conditioning systems, refrigeration plants, and heat pumps (geothermal and air-source).
Chapter 2 presents the Diffusion Absorption Refrigerator (DAR) together with a new advanced
thermodynamic model. The authors perform the validation of the model on a prototype built coupling a
domestic 750 W-magnetron with a small purposely modified commercial DAR to activate the thermal
pump.
Chapter 3 applies a conventional and advanced exergetic analysis to an air refrigeration machine. The
authors argue that a conventional exergetic analysis does not consider the real potential for improving a
system nor the interactions among the components of the system.

xxiii
Preface

Chapter 4 reviews the basic mechanisms of static sublimation process and sublimation two-phase
flows. The authors start to classify the previous studies into numerical modeling and experimental
verifications and introduce and compare representative refrigeration systems. Future research focuses
is then established.
Chapter 5 presents the development of ejector refrigeration technology. Authors suggest that this
technology strongly reduces the greenhouse gases emission by using natural refrigerants and also dra-
matically reduces the need for the electric power. Authors contend that this is accomplished by using
free or inexpensive heat – either solar or waste heat, as the main source of energy instead of electricity.
The authors suggest that the operation of the ejection cycle using low-temperature heat source can be
considered as very attractive and the ejection cycles becomes truly competitive in comparison with the
absorption refrigeration systems.
Chapter 6 presents a review of refrigerants and its historical evolution. The authors present the main
characteristics and properties that characterize the refrigerants and discussed the different replacement
possibilities for the currently most used refrigerants. The authors address the issue of the most promis-
ing refrigerants to be used in future.
Chapter 7 presents an experimental analysis about the retrofitting of two commercial stationary
refrigeration systems marketed by an Italian leading company of the sector. The authors examine the
performance of R410a and R407f as effective alternatives to R404a due to their compatibility, non-
flammability and market availability.
Chapter 8 examines the performance of a transcritical cycle that has been evaluated with a proto-
type R744 system working as classical split-systems to cool air. The authors argue that Carbon dioxide
(R744) is a valid alternative to classical substances such as HFCs used in vapor compression plants. The
experimental transcritical cycle has been examined in comparison with a classical vapor compression
plant working with the R134a.
Chapter 9 addresses the issue of the use of a capillary tube as an expansion device, with particular
reference to CO2 (R744) transcritical heat pump system. The authors contend that capillary tubes which
are extensively used in small size vapor compression systems work very differently in a CO2 transcriti-
cal heat pump system. The authors suggest that there is an optimal length of capillary tube for a given
diameter at which the heat pump runs optimally.
Chapter 10 examines the challenges and opportunities of the use of transcritical CO2 refrigeration
system in tropical region. The authors present a discussion on component design and system level per-
formance analysis.
Chapter 11 presents the principles and methods of holistic understanding of the foodstuff freezing
process, as well as structure, functioning and development of refrigeration systems. The author examines
the modern classification of frozen products, new methods of research of freezing process as well as selec-
tion, accumulation and systematization of general principles of refrigeration technology implementation
Chapter 12 analyses the influence of process variables on thermal and microbiological behavior of
minimally processed foods submitted to refrigeration process with and without evaporative cooling ef-
fect. The author ground his arguments by means of two case studies.
Chapter 13 presents a brief historical context followed by an overview of the technologies used for
fruits and vegetables storage, including refrigerated and controlled atmosphere storage as well as the
most recently developed technologies for storing these produces. The authors present the current recom-
mendations for long term storage using ‘Rocha’ pear as a case study.

xxiv
Preface

Chapter 14 identifies the existing challenges on the proper design of evaporators to mitigate contam-
ination-related risks in air blast freezing systems. The authors identifies the importance of evaporator
coils, fins and drain pans in blast freezers as a source of microbial, physical and chemical contamina-
tion. The authors discuss contamination problems specific to the key components of blast freezers and
measures for contamination control are discussed.
Chapter 15 presents an experimental study on the effects of some of the most important variables
that affect the infiltration rate in open refrigerated vertical display cases at steady operation. The authors
also analyze and compare methods to measure the air infiltration rate.
Chapter 16 presents an overview about experimental studies concerning the thermal performance of
air curtains and heat exchangers installed in vertical open refrigerated display cases.
The authors identify the role of the discharge air grille and the perforation density of the back panel
on the thermal performance of the air curtain.
Chapter 17 examines two experimental case studies on domestic refrigerators without modifying its
cooling performance. In the first case, the authors analyze the thermal profile in the compartments of a
refrigerator. In the second case, the authors propose a new material as thermal insulator.
Chapter 18 addresses the issue of hygienic imperatives in the evaporator design, with particular refer-
ence to the frosting and defrosting phenomena. The authors address the methods and patterns of evaporator
defrosting as affected by hygienic design implications and criteria. The authors present the advantages
and drawbacks of diverse defrost methods with regards to contamination risks in refrigeration facilities.
Chapter 19 discusses the use of expander mechanisms to recover the expansion energy of refrigera-
tion systems. The authors present and discuss an array of techniques that help in achieving this purpose.
Chapter 20 addresses the issue of cold thermal energy storage. The authors discuss the benefits as
well as the classification and operating strategies of this technology. The authors paid special attention
to the analysis of specific features of heat transfer phenomena in ice storage tank and present the sizing
of ice thermal storage system for different operating strategies.
Chapter 21 presents a high efficiency heat pump system using for cold storage and heating. Based
on experimental and numerical results, the author contends that in order to improve the coefficient of
performance, one can use a thermobank that stores the wasted heat during refrigeration cycles and an
ejector to decrease the compressor displacement.
Chapter 22 presents and analysis the availability of a number of smart engineering solutions and
control strategies to exploit the employment of renewable energy in the food storage sector. The authors
argue that it is possible to reduce the energy consumption of refrigerated warehouses and hence the
global energy demand, if adequate importance is placed on the integration of renewable resources in
the energy supply of large cold storage facilities.
Chapter 23 identifies the existing challenges in the natural gas liquefaction techniques. After review-
ing the trends in the natural gas optimization area, the authors identify future research directions.
Chapter 24 presents an alternative refrigeration system based on fuel cells. The authors give a de-
scription of the fuel cell model and present examples of the energy load profiles impact on the fuel cells
sizing, hydrogen consumption and system autonomy. The authors examine the economical feasibility
of this new refrigeration system linked to renewable energies and present an economical assessment for
different scenarios.
In conclusion, the Handbook of Research on Advances and Applications in Refrigeration Systems
and Technologies, accounting with the contribution of 60 authors from 16 countries, intends to be a
reference book for all interested parties, from undergraduate students, practicing engineers, researchers

xxv
Preface

and decision-makers. This Handbook aims to cover more refrigeration aspects, presenting the results
of ongoing research and defining paths for the future evolution of refrigeration equipment and systems.
The main output of the Handbook is related to the improvement of energy efficiency in the refrig-
eration sector, and simultaneously contributing for its sustainability. Additionally, it gives particular
relevance to nowadays topics such as innovative technologies, energy efficiency, environmental impact,
sustainability, legislation and policy.
We hope that you enjoy the reading.
The Editors,

Pedro Dinis Gaspar


University of Beira Interior, Portugal

Pedro Dinho de Silva


University of Beira Interior, Portugal

xxvi
1

Chapter 1
Ejector Refrigeration Cycles:
Classification of Thermodynamic
Cycles with Ejectors

Marek J Bergander
Magnetic Development, Inc., USA & University of Hartford, USA

ABSTRACT
This chapter describes a collaborative effort of US private companies and various departments of the US
Government to investigate the possibility of improving the efficiency of HVAC systems by use of one and
two-phase ejectors. It is anticipated that this technology, when fully developed will result in attractive,
energy saving products that significantly improve the performance of commercial and residential chiller/
air-conditioning systems, refrigeration plants, and heat pumps (geothermal and air-source). Although
the literature describing ejector applications in refrigeration dates back to the year of 1900, the ejec-
tor use was always considered as controversial, because the previous research had resulted with only
theoretical results and without visible, commercial products. The research on the ejector application is
consistent with present directions in the HVAC industry and it will attract more attention and research
funding in the future. A classification of thermodynamic cycles where ejectors can be applied composed
by three distinctive “categories” is suggested.

INTRODUCTION

It is estimated by US Department of Energy that 40% of all electricity usage in commercial buildings is
related to HVAC systems. This 40% provides an attractive target for reducing future energy needs. Im-
proved methods of heating, refrigeration, and air-conditioning that can be implemented cost effectively
could provide a significant breakthrough in green energy efficiency. This describes the investigations
conducted by the author during last 10 years to improve the efficiency of HVAC systems by use of one
and two-phase ejectors.
There is an extensive body of previous knowledge and a large literature base regarding ejector ap-
plications in refrigeration dating back to the year of 1900 (Bergander, 2012). In spite of this, the ejector
use in refrigeration systems was always considered as controversial at best, mainly because a lot of
DOI: 10.4018/978-1-4666-8398-3.ch001

Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

Ejector Refrigeration Cycles

research has been conducted for many years with only theoretical results and without visible, commer-
cial products. However, just recently (Sept 2009) it was disclosed by Denso Corp. of Japan that they
developed an ejector-based, air conditioning (A/C) system to be installed in all Toyota cars and trucks
starting with 2010 models outside of the US (Takeuchi, 2009). Further, this author’s investigations dem-
onstrated another example of effective use of ejectors in A/C systems: we were able to demonstrate the
ejector-based air conditioning system for residential dwellings, which was able to considerably lower
the ambient temperature, while using low-grade heat (solar, waste, etc.) as a main source of energy and
further, to completely eliminate the compressor from the cycle (Butrymowicz & Bergander, 2010) . It
is extremely encouraging that two very different approaches using different working fluids and heat
sources have demonstrated the practical use of ejectors to provide economical cooling.
The research on the ejector application is consistent with present directions in the HVAC industry,
which are: 1) improving the efficiency, 2) reducing the refrigerant charge and 3) reducing the footprint
(size) of the refrigeration units. We believe that the ejector, when considered as a “technology platform”
is capable to significantly contribute to all three above goals and therefore, it will attract more attention
and research funding in the future. World demand for HVAC equipment is projected to increase over
5% per year according to the Fredonia Group’s new World HVAC Equipment study. Demand in the U.S.
for HVAC equipment is forecasted to increase 3.2 percent per year to $15 billion annually in 2011.
Sales will be driven by a projected recovery in residential construction along with ongoing strength in
residential remodeling. Because of the large amount of HVAC equipment in place, the demand attribut-
able to replacements and improvements makes up about 70% of total demand. The increasing interest in
energy-efficient building systems and energy conservation will continue to drive growth in the replace-
ment sector. This is well aligned with our value proposition.
While numerous publications describe the ejectors and their use, none of them to our best knowledge
had attempted to classify various types of thermodynamic cycles where those ejectors can be applied.
Consequently, the author proposes such classification, and he recognizes three distinctive “categories”
of cycles with ejectors. This chapter discussed these categories in more details. The Category 1 (ejector
installed on compressor suction), receives somewhat less attention because it has been well known and
the intellectual property belongs to others. Most detailed discussion and research results are presented
for the case where the ejector is installed on compressor discharge (Category 2), as this is the main area
of the author’s expertise and involvement. As a result of his research in this field, he possesses the intel-
lectual property in form of mentioned above US Patent that controls this specific area of application.
There seems to be a confusion of terminology in the literature, where both terms “ejector” and “injec-
tor” are used interchangeably. We believe that this is incorrect, since the “injector” is a device, also known
as a “jet pump”, intended to increase the pressure of a flow without the input of work, while an “ejector”
is an equivalent device, but with the intended purpose of creating suction from a reservoir (Babcock &
Wilcox Co., 2005). Therefore, the proper term throughout this book should rather be “injector”, however
since historically and traditionally the term “ejector” has been used, we decided to stay with that conven-
tion. To further justify such an approach, there are certain cases, where the purpose of an ejector is not
that straightforward and it may be disputable whether its first priority is pressure increase or suction.

2

Ejector Refrigeration Cycles

BACKGROUND: BASIC THEORY AND APPLICATION OF EJECTORS

Types of Ejectors

In general, an ejector can be described by the five basic components shown in Figure 1 (Chunnanond
& Aphornratana, 2004). The primary nozzle inlet - section P to ii, is composed of a de Laval nozzle
(converging-diverging) used to accelerate and expand the high pressure flow of the primary, or motive,
stream into the suction chamber at a supersonic velocity. The suction inlet, point S, is the inlet for the
suction, or secondary stream into the mixing chamber. The mixing chamber, section ii to iv, is the loca-
tion of mixing between the motive and suction streams. The constant area section, shown by section
iv to vi, is typically where the shock, (position v), will occur to induce the pressure rise. The diffuser,
section vi to vii, decreases the velocity of the flow in order to convert the kinetic energy into a further
increase in pressure at the outlet of the injector.
A correctly designed ejector, with proper inlet conditions, has the ability to take advantage of the
shock that occurs in the constant area section in order to get a pressure rise across the entire ejector. Under
these conditions, the pressure at the outlet of the ejector is greater than both of the inlets (Sun, 1996).
Ejectors can be divided into two classes based upon the geometrical location of the primary nozzle
inlet. The first type is the constant area mixing ejector, shown in Figure 2 and henceforth referred to as
a “constant area ejector”, which places the outlet of the primary nozzle inlet within the constant area
section of the ejector. For the constant area ejector the mixing chamber and constant area sections are
equivalent as the primary and secondary flows mix completely within the constant area region.

Figure 1. Constant pressure ejector


(Chunnanond & Aphornratana, 2004)

3

Ejector Refrigeration Cycles

Figure 2. Constant area ejector


(Chunnanond & Aphornratana, 2004)

The second type is the constant pressure mixing ejector, shown in Figure 3 and henceforth referred to
as a “constant pressure ejector”, places the outlet of the primary nozzle inlet within the suction chamber.
This ejector theory, developed by Keenan et al., defines the mixing between the primary and secondary
flows at the exit of the primary nozzle inlet to be at an equivalent constant pressure (Keenan et al., 1950).
It has been shown in the literature that the constant pressure ejector is more complicated to design, as the
distance between the outlet to the primary nozzle inlet and the constant area section will have a signifi-
cant effect on the efficiency of the device (Aphornratana & Eames, 1997). The mixing section and the
constant area section will be separate for this type of ejector, as all mixing occurs in the mixing chamber.

Single Phase Ejector Theory

Beyond the geometrical considerations, an ejector can be classified into two more categories. The first is
the single phase ejector, which uses a working fluid in its vapor state for both the primary and secondary
inlet conditions. Keenan et al. analytically and experimentally investigated the use of both the constant
pressure and constant area single phase ejectors using air as the working fluid (Keenan et al., 1950).
The analysis of the ejector involved the use of the momentum, energy, and continuity equations with
the assumption that ideal gas relations are valid. Figures 4 and 5 show the analytical results for varying
pressure and area ratios, respectively.
The analytical results show that the constant area ejector consistently provides pressure ratios lower
than that of constant pressure ejector for equivalent entrainment ratios. The entrainment ratio, given by
U, is the ratio of the mass flow rate of the secondary flow to the mass flow rate of the primary flow.
The pressure is measured at the outlet of the diffuser, p vii in Figure 1, and divided by the pressure at the

Figure 3. Constant pressure ejector


(Chunnanond & Aphornratana, 2004)

4

Ejector Refrigeration Cycles

Figure 4. Analytical results for constant area and pressure ejectors with varying p S /p P ratio
(Keenan et al., 1950)

Figure 5. Analytical results for constant area and pressure enjectors with varying A iv /A i ratio
(Keenan et al., 1950)

5

Ejector Refrigeration Cycles

inlet of the secondary flow, p S , to normalize. The square root of each axis was taken to improve the
distribution over the chart, which makes it easier to see that the constant pressure ejector is a more ef-
ficient design as it always gives a higher pressure ratio than the constant area ejector for equivalent
entrainment ratios.
When compared against the experimental results, the constant area ejector analysis proved to be an
appropriate approximation when the value of the entrainment ratio is not near the maximum (Keenan et
al., 1950). As U approaches the maximum value, the calculated pressure at the end of the constant area
section diverges significantly from the measured value. Munday & Bagster (1977) furthered the theory of
constant pressure mixing by determining that the motive stream stays independent of the suction stream
while fanning out upon exiting the nozzle inlet. After exiting the primary nozzle inlet, the motive stream
fans out into a converging duct that decreases the pressure of the primary flow while accelerating the
secondary flow. The primary flow will continue to fan out until the secondary flow chokes, after which
the primary and secondary flows are at equivalent pressures and then constant pressure mixing occurs.
The location at which choking occurs is defined as the effective area (Munday & Bagster, 1977), shown in
Figure 1. It had been determined that injector devices were constant capacity (Blatchley, 1961), meaning
that the back pressure has no effect on the entrainment ratio until the critical pressure is reached. Once
the critical pressure is reached the entrainment ratio will suddenly drop to zero as shown in Figure 6.

Figure 6. Entrainment ratio for varying back pressure


(Keenan et al., 1950)

6

Ejector Refrigeration Cycles

While the constant capacity phenomenon had been documented, it had not previously been explained.
Munday and Bagster determined that their theory of independent streams after the nozzle inlet was
consistent with the observed constant capacity theory (Munday & Bagster, 1977). They determined that
constant capacity arises because when the secondary flow is choked at the effective area location, it will
have a pressure lower than the backpressure. Lower pressures downstream are not propagated upstream
and thus do not effect the flow rate and cause constant capacity (Keenan et al., 1950).
Huang et al. elaborated further on Munday and Bagster’s theory by hypothesizing that the effective
area, given by section yy in Figure 7, is located within the constant area region (Huang et al., 1999).
If the effective area were to be in the constant area section, then it would mean that all mixing occurs
only in the constant area region. Previously the constant pressure theory had stipulated that the mix-
ing occurred only in the mixing chamber (Takeuchi, 2009). Huang et al. (1999) also established three
operational modes of an injector based on the back pressure, given by p... The first operational mode is
termed the double-choking or critical mode, which is when choking occurs in the primary fluid in the
nozzle inlet and the secondary fluid in the mixing chamber.
In order for this mode to occur, p c would have to be less than or equal to the critical pressure dis-
cussed in Figure 6. The second operational mode is termed the single-choking or subcritical mode, which
is when choking occurs only within the primary nozzle inlet. Under this condition, p c has to be greater
than the critical pressure and less than the value for which the entrainment ratio is zero. The entrainment
ratio will vary for this mode of operation as it occurs on the slanted lines shown in Figure 6. The third
mode is termed the back-flow or malfunction mode in which neither the primary or secondary flows are
choked within the ejector. Under this condition, p c is greater than or equal to the value for which the
entrainment ratio is zero, thus forcing the entrainment ratio to be less than or equal to zero. For the ejec-
tor to operate most efficiently it would have to be operating in the critical mode.
By considering the effective area to be within the constant area region, gas dynamic and isentropic
relations are able to be utilized in order to predict the area and entrainment ratios for the ejector, shown
in Figure 8 (Huang et al., 1999). These results show that the method followed by Huang et al. has the
ability to predict the flow in single phase 1-D ejectors accurately.

Figure 7. Constant pressure mixing condensing ejector


(Provided by Huang et al., 1999)

7

Ejector Refrigeration Cycles

Figure 8. Analytical and experimental comparison for A 3 /A t and ω


From Huang et al. (1999)

Two-Phase Ejector Theory

The second category of ejectors is the two phase ejector, which utilizes the sub-cooled liquid and vapor
phases of a single working fluid as the primary and secondary flow inlets. The large temperature dif-
ferences between the primary and secondary flow, coupled with the high relative velocity establishes
a high rate of heat transfer (Keenan et al., 1950). Under the proper inlet conditions, the mixing of the
primary and secondary flows causes a condensation shock to occur, within which all of the vapor flash
condenses into the liquid exit state (Keenan et al., 1950). For this reason, two phase ejectors are com-
monly referred to as condensing ejectors (CE). Due to the phase interactions in the form of mass, mo-
mentum, and energy transfers, the theories created for single phase ejectors cannot be directly applied
to two phase ejectors (Trela et al., 2008). For this study, the vapor phase is assumed to be the secondary
fluid unless otherwise noted.
As the principle of CE relies on such concepts as “two-phase flow”, “critical flow”, “condensation
shock”, an extensive study of the literature related to this subject was conducted.
The advantages of critical flow are apparent—the independence of flow rate from the discharge
conditions (downstream pressure and geometry), in a broad range of counter-pressures. We believe that
further understanding of two-phase flow phenomena could provide the fundamentals for the design of
hydraulic equipment with higher efficiency, for example pumps with flat characteristics, heat exchang-
ers, ideal mixers and proportioners. Similarly, many ejector designs for refrigeration applications rely on
the condition of sustained critical flow. Our overall conclusion from the literature study was that even
though many researchers present various results and data from their studies, the governing parameters
associated with critical flow in two-phase medium, such as gas content, bubble radius, pressure ratios,
flow ratios, etc. remain to be investigated.
It is known that gas-liquid flow often has a non-equilibrium character, which demonstrates itself by
occurrence of pulsation of pressure, density, velocity and temperatures for both phases. Under certain
conditions, these processes cause undesired effects, such as water hammer, vibrations of pipelines,
perturbation of circulation modes and heat exchange, while under different conditions, occurrence of

8

Ejector Refrigeration Cycles

two-phase flow improves the heat exchange and increases the efficiency for many components of energy
systems. Therefore, various theoretical schemes were considered in which the dynamic properties of
two-phase stream or “bubbly liquids” could be profitably used in the design of propulsion devices. The
underlying idea is that the expansion of a compressed gas bubble-liquid mixture might be an efficient
way to produce the momentum necessary for compression.
Models exist for the prediction of unsteady two-phase flashing flows in variable cross section ducts
and valves, including the Homogeneous Equilibrium Model (HEM) (Wallis, 1980) and the Homoge-
neous Relaxation Model (HRM) (Downar-Zapolski et al., 1996). One of fundamental publications on
the subject, by Wallis (1980) acknowledges the presence of critical flow but simultaneously confirms
that the mathematical model for such conditions is extremely complicated since it has to consider not
only criticality in one location, but also may include parts of the upstream system. Concluding, it is
possible to state that a two-phase mixture, if it is sufficiently homogeneous, has completely different
properties than each of its components. The most important is the reduction of speed of sound as first
observed in 1941 (Van Wijngaarden, 1972), which brings the possibility of supersonic flow at moderate
velocities: 20-50 m/s.
The newer research shows that the two-phase medium is more compressive than the gas and con-
sequently, the speed of sound in two-phase mediums is much lower than that in the homogeneous gas
or liquid (Fisenko, 1987). By starting from the volumetric content of gas in a gas-liquid mixture and
introducing certain assumptions, i.e. no slip between phases and isothermal nature of the flow, Van
Wijngaarden (1972) derives his fundamental formula for the speed of sound in a two-phase mixture:

p
a2 = (1)
ρf β (1 − β )

where a is the speed of sound, p is the pressure, ρf is the density of the liquid phase and β is the volume
occupied by the gas in a unit volume of the mixture.
The formula contains the result that, unless β is very close to either zero or unity, the speed of sound
in the two-phase mixture is lower than the speed of sound in pure gas. A minimum exists for β = 0.5,
in which case, at a pressure of 1 bar, a mixture of air and water has a sound velocity of 20 m/s. Similar
results were obtained for a two-phase homogeneous mixture, such as water-steam, liquid refrigerant-
vapor refrigerant, etc. These results were confirmed by a number of researchers in USA and abroad.
Consequently, a handful of books articles (Fisenko, 1987) published in the 1980’s and 90’s present
diagrams for the speed of sound vs. β for a two-phase mixture. The general character of such diagrams
is shown in Figure 9 below.
Van Wijngaarden (1972) describes also the process when the vapor is accelerated in a jet device to
a velocity at or slightly above the sonic velocity. The mixing of two phases, vapor and liquid, leads to
the decrease in local sound velocity according to the diagram of Figure 9 and the concomitant creation
of a “shock wave” with a consequent increase of pressure. The pressure ratio achieved by such a shock
wave can be calculated from the formula given by Campbell & Pitcher (1958) for the isothermal process:

p2
= M2 (2)
p1

9

Ejector Refrigeration Cycles

Figure 9. Speed of sound in two-phase bubbly flow

where M is the Mach number.


For an adiabatic process, the formula given by Fisenko (1987) takes a slightly different form:

p2
= 1 + k βM 2 (3)
p1

where k is the adiabatic coefficient: k = cp / cv


The above theory brings about the possibility of obtaining the supersonic flow in Laval nozzles and
this can be considered in propulsion devices. Among others, Witte (1969) investigated the efficiency of
a propulsion device based on the injection of compressed air bubbles in the throat section of a nozzle
and observed the pressure jump associated with the supersonic flow. Indeed, this was further confirmed
by both computer modeling and laboratory experiments.
In summary, the state-of-the-art study on two-phase critical flow in ejectors has concluded the fol-
lowing:

1. Critical flow in two-phase media such as air-water and steam-water were observed by a handful
of researchers (including this author), however the theoretical models are still under development.
Under critical flow conditions the flow rate from the system is independent of the conditions in
the receiver (especially pressure).

10

Ejector Refrigeration Cycles

2. The speed of sound in a two-phase mixture is much lower than in any of its components, thus the
supersonic flow can exist in such mixtures at relatively low velocities, i.e. 20-50 m/s.
3. The supersonic flow in Laval nozzles produces “shock waves”, thus increasing the outlet pressure
proportionally to the square of the Mach number achieved in the nozzle. Shock waves of this nature
were observed in experiments and called either “mixing shocks” or “condensing shocks”.

Most comprehensive experiments for a constant area CE conducted by Levy and Brown determined
the criteria for a condensation shock to occur (Brown & Levy, 1972). The first variable that was deter-
mined to effect the condensation shock was the back pressure valve. With the back pressure valve open
(BPVO), a condensation shock could not occur no matter the inlet conditions due to the liquid stream
creating a cylindrical jet in the center and the vapor creating an annular region at the wall with little to
no mixing occurring. As the back pressure valve was closed (BPVC) a condensation shock occurred
under proper inlet conditions due to the mixing of the flows. In order to gain insight into other factors
effecting condensation shock, further experiments were conducted mainly with the back pressure valve
open. With the BPVO the flow of the constant area CE could be divided into three distinct regimes,
shown in Figure 10 (Brown & Levy, 1972).

Figure 10. Effect of inlet liquid velocity on axial wall static pressure in a constant area ejector (BPVO)
(Brown & Levy, 1972)

11

Ejector Refrigeration Cycles

Regime I, (runs A,B,C, and D), is termed the High Inlet Liquid Velocity Flow Regime. Within this
regime the liquid jet keeps a defined cylindrical jet until x = 5, shown in Figure 11.
Once that point is reached, the liquid jet breaks up, typically between x = 5 and x = 9, resulting in a
two phase cylindrical jet region with the annular vapor phase region still present. Line M in Figure 11
shows the vapor Mach number across the ejector, specifically noting that it is supersonic in the annular
region and unity at the primary nozzle exit plane.
Regime II, given by runs E, F, G, and H, is termed the Intermediate Inlet Liquid Velocity Regime.
Under these inlet conditions, the liquid jet breakup length was between x = 1 and x = 5, but the vapor
Mach number dropped to a subsonic value at approximately x = 3.
Regime III, given by run I, is termed the Low Inlet Liquid Velocity Regime. In this region the liquid
jet breakup is just after the primary nozzle exit plane, where the Mach number of the vapor region is
subsonic for the entirety of the injector. These regimes show that the liquid jet breakup is accompanied
by a sharp decrease in the vapor Mach number resulting in a subsonic value (Brown & Levy, 1972).
Based on the experiments conducted by Levy and Brown the following conditions are required for a
condensation shock to occur (Brown & Levy, 1972):

1. The back pressure valve must be near BPVC conditions.


2. The inlet conditions must be such that the flow will be in Regime I, as it is the only regime capable
of producing a condensation shock. Regimes II and III do not produce a condensation shock due
to early liquid jet breakup that causes the vapor Mach number to become subsonic.
3. The vapor flow must be supersonic. As the back pressure valve is closed, an initial length of the
ejector, the influence length, is independent of back pressure. Once the back pressure valve is closed

Figure 11. Two phase flow in a constant area ejector obtained from impact pressure probes
(Brown & Levy, 1972)

12

Ejector Refrigeration Cycles

Figure 12. Experimental results (BPVO) for constant area ejector for Regime I, top, and Regime II, bottom
(Levy, 1967)

sufficiently, then wall static pressures begin to increase within the influence length. Therefore, since
the supersonic region of the vapor is within the influence region, the vapor must be supersonic in
order to produce a condensation shock.

Levy and Brown were able to establish computational methods for both the constant pressure and
constant area ejectors, (Levy, 1967). The methods utilized the conservation of mass, momentum, and
energy equations coupled with the rate of heat transfer from the liquid-vapor interface to the liquid core
to relate to the condensation rate.

13

Ejector Refrigeration Cycles

Background of Ejector Application in Refrigeration Cycles

Vapor compression refrigeration systems typically utilize expansion valves or other throttling devices to
lower the pressure of liquid refrigerant and deliver it to the evaporator. In a typical refrigeration cycle,
the expansion valve lowers the refrigerant pressure by 5-7 times. The reason for lowering the pressure
is to allow the refrigerant to evaporate at certain desired low temperature. However, the process of
throttling is isenthalpic, which means that the kinetic energy produced during the pressure reduction is
dissipated and eventually wasted. Therefore, it is desirable to recover this kinetic energy to increase the
efficiency of the entire refrigeration cycle. The literature search has revealed that the principal method
to accomplish this task in the past was to use the ejector instead of the throttling valve.
The velocity increase in the throat of the ejector device is used to entrain the refrigerant exiting the
evaporator by momentum exchange. The following diffuser section of the ejector re-compresses the
refrigerant by slowing down the mixed stream. Through the action of an ejector, the compressor suction
pressure is therefore higher than it would be in a standard cycle. This results in less compression work,
thus improving in cycle efficiency.
The first theoretical principles of the ejector were elaborated by Parsons (Bergander, 2012) in the
early 1900’s while the first prototype was built by Leblanc (Bergander, 2012). Further improvements
were introduced by Gay in 1931 (Gay, 1931). Ejectors were first applied for refrigeration cycles by
Heller in 1955 for absorption systems and by Badylkes in 1958 for vapor compression systems (Ber-
gander, 2012). In the USA, the first application was reported by Kemper in 1966 (Kemper et al., 1966),
but only the patent is in existence while no experimental or theoretical background has been published.
Following up on this early work, Kornhauser (1999) has conducted a theoretical analysis and presented
a certain ideal ejector cycle showing that it can produce a 21% efficiency benefit when compared with
a standard vapor compression cycle. The prototype unit was built, however its performance was much
less than ideal and reached a maximum of only 5% using working fluids CFCs/ HCFCs/ HFCs. This
was attributed to shortcomings in the design of the ejector, and that an overly-simplified two-phase flow
model was assumed in the design.
Recent work on ejectors has concentrated on using them in transcritical CO2 systems where high
pressures allow for better recovery of the kinetic energy. Detailed investigations were presented in par-
ticular a constant pressure mixing model for the superheated vapor ejector that was established and the
thermodynamic analysis of the ejector expansion for transcritical CO2 was performed. It was found that
the COP (Coefficient of Performance) of such transcritical CO2 cycle with an ejector can be improved by
as much as 16% over the basic CO2 cycle for typical A/C operation conditions. However, only a theoreti-
cal model is presented in the subject reference with no supporting practical experiments.
While ejectors have been shown in the past to be capable of increasing the efficiency of refrigeration
cycles, they have also been shown to be complicated devices. There are many factors to consider when
designing an ejector. First it must be determined which flavor of ejector to be used, either constant area
or constant pressure mixing and either single or two phase flows. Once the flavor is chosen, the entrain-
ment ratio and inlet conditions of the flows must be determined. If the inlet conditions do not put the
ejector into Regime I, then the device will not create a condensation shock. The area ratios between the
constant area section and primary nozzle inlet also must be calculated. Furthermore if it is a constant
pressure mixing ejector, then the distance between the primary nozzle inlet and the constant area section
will be an additional term to calculate. Once the design is determined, it must then be checked that the
pressure at the outlet is equal to the pressure at the inlet of the motive stream if the intended purpose is

14

Ejector Refrigeration Cycles

within a refrigeration cycle. While it would be possible for an ejector to cause an increase in pressure in
an open loop, if a closed loop cycle does not have a net change of zero for the pressure, then the continual
build up of pressure within the cycle causes instabilities. The literature on this subject has been able to
analytically predict and experimentally show the behavior of the flow within a given ejector to a certain
degree, but thus far have been unable to determine the most efficient configuration.

CLASSIFICATION OF EJECTOR REFRIGERATION CYCLES

Categories of Ejector Refrigeration Cycles

The discussion below is limited to vapor compression refrigeration cycles. Such a typical cycle is com-
posed of an evaporator, a condenser, a compressor and an expansion valve. A compressor is a component
that consumes electrical energy in order to increase the pressure of the refrigerant and therefore is a
major contributor to the efficiency of the cycle. As previously determined, an ejector has the ability to
induce a pressure rise. Due to this ability, ejectors have long been sought for applications in a refrigera-
tion cycle to “help” the compressor and to either take a portion of the compression task or even take
over the entire compression. An ejector placed in a refrigeration cycle allows the compressor to reduce
its pressure ratio and consequently, to reduce the work a compressor needs to perform – this increases
cycle efficiency. In the course of our research we studied previous ejector applications in refrigeration
cycles and found more than 200 papers available in the open literature. A good overview of historical
and recent developments regarding refrigeration and air-conditioning systems that use ejectors is given
by Elbel & Hrnjak (2008) with special emphasis on transcritical CO2 (R744) cycles. All these studies
helped to evaluate the past solutions and established trends.
Based on a literature search and our own research, we identified three various configurations where
ejectors can be applied in refrigeration, A/C systems, and in heat pumps, for efficiency improvement. The
first two arrangements help the compressor to lower its pressure ratio, while the third one, eliminates the
compressor completely. For the purpose of this publication, we called those arrangements – “categories”
and they are as follows:

Category 1: Ejector installed as a first step of compression, increasing the suction pressure of the com-
pressor, consequently, helping a compressor and improving the efficiency of the cycle.
Category 2: Ejector installed as a second step of compression, decreasing the discharge pressure from
a compressor, consequently taking up to 1/3 of total compression, thus improving the efficiency
of the cycle.
Category 3: Ejector completely replacing the compressor and taking over an entire vapor compression.
In this case, additional compression is done on a liquid phase of the refrigerant by a pump, elimi-
nating electric power needs by almost 90% (because, thermodynamically, it takes much less energy
to compress liquid than vapor). The major power source in such cycles is a low-temperature/low
quality heat, which can be either cheap (waste heat) or free (solar heat).

The remainder of this chapter is devoted to a more detailed discussion of the three above types.
However, only limited information on Category 1 is given due to the fact that this is traditionally the
most common application and a large literature base exists on this subject. Further, the great majority

15

Ejector Refrigeration Cycles

of intellectual property to this solution belongs to the Japanese company Denso Corp. and this poses
the major constrains in further development of this technology without the risk of patent infringement.
Therefore, only the principle of operation will be presented here.
Category 2 is a solution proposed by this author and patented by him in the USA. Category 3 is a very
promising and attractive concept of eliminating a compressor from the refrigeration cycle and powering
the system by low-grade heat. The author had performed numerous research programs on this subject,
financed by the US Government and private industry with selected results shown later in this publication.

Category 1: Ejector on the Suction of a Compressor

This ejector application can be considered from two points of view: the first, that it increases the suction
pressure of the compressor, lowering the required pressure ratio and decreasing the compression work.
The second position treats the ejector as a device in which pressure drops without the throttling process. It
is widely recognized that the highest energy loss in the refrigeration cycle is caused by throttling, during
which the kinetic energy is converted to friction and eventually lost. Therefore, the most effective way
of reducing this loss is the recovery of the kinetic energy of the fluid, produced as an effect of pressure
drop. One method to achieve it would be a thermodynamic process with isentropic expansion rather
than isenthalpic throttling. Such a process is provided by a two-phase ejector installed in the liquid line
between the evaporator and liquid separator as shown in Figure 13 below. Other arrangements within
this category are shown by Bohdal et al. (2003) specifically, a system with two evaporators, and another
system with two throttling valves.

Figure 13. Schematics of the Category 1 ejector refrigeration cycle


(Bergander, 2012)

16

Ejector Refrigeration Cycles

Liquid refrigerant is the motive fluid in this ejector. Liquid refrigerant expands in the motive nozzle
and partly evaporates. The two-phase flow consists of the droplet in the vapor phase flows out of the
nozzle. This two-phase mixture, still as a motive fluid, causes suction of the vapor from the evaporator.
The mixture of these streams flows from the diffuser of the ejector to the liquid separator. Then the liquid
phase flows to the evaporator via the throttling valve and the vapor phase is sucked by the compressor.
The physical phenomena occurring in the motive nozzle, suction chamber, mixing chamber, and the
diffuser of the discussed ejector are extremely complicated and the knowledge about the maximum
performance and energy effectiveness of such an ejector is still very limited.
The T-s and p-h graphs for the Category 1 refrigeration cycle of Figure 13 is shown in Figure 14. It is
important to note that mass flow rates of the refrigerant are different for different processes of the cycle.
The ideal liquid expansion in the motive nozzle proceeds at constant entropy – the process 3-4s. In the
case of the throttling process during which pressure drops from condensation pressure pk to evaporation
pressure po, this process is isenthalpic 3-4d. The actual process occurring in the motive nozzle of the
ejector proceeds along the curve 3-4. The pressure drop in the suction line from the evaporator to the
suction chamber of the ejector can be omitted as negligible, so p5 ≈ po. The refrigerant is partly com-
pressed in the mixing chamber due to mixing shock, so the pressure increases from evaporation pressure
po to the pressure at the outlet of the mixing chamber p6. The ideal compression of the two-phase flow in
the diffuser proceeds at constant entropy 6-7s. The actual compression proceeds along the line 6-7. The
suction pressure of the compressor is the inter-stage pressure pm and is higher then evaporation pressure
po, so the specific work of the compression lt is lower in comparison with a pure vapor compression
refrigeration device.
The most comprehensive state-of-the-art and history of this development is presented by Butry-
mowicz (2013). While many researchers showed that from a theoretical point of view it is possible to

Figure 14. Theoretical cycle of compressor-ejector refrigeration Category 1 device shown in Figure 13:
(a) in coordinates T-s, b) in coordinates p-h
(Bergander, 2012)

17

Ejector Refrigeration Cycles

achieve more than 20% increase of the COP for most of the refrigerants HCFC and HFC in comparison
with classic vapor compression refrigeration devices, only as much as 4% of COP was achieved so far
for refrigerant R-12.
The most important reason for such a moderate result is a very low energy efficiency of the two-
phase ejector. The processes of the momentum transfer in the suction and mixing chambers between the
motive liquid and sucked vapor phase is extremely difficult for modelling. Theoretical and experimental
investigations of these processes are in the early stage so far, so the upper limit of the efficiency is not
known yet. On the basis of Butrymowicz model (Butrymowicz, 2013) of the compressor-ejector cycle it
may be concluded that a significant increase of the cycle efficiency could be achieved by the application
of the two-phase ejector. The model of the ejector requires however further developing since the phase
change occurring in the ejector should be taken into account.

Category 2: Ejector on the Discharge of a Compressor

In this Category, the two-phase ejector is installed on the compressor discharge and it assumes up to 1/3
of total compression thus decreasing a compression ratio and the compressor work. Therefore, the amount
of mechanical energy required by a compressor can be significantly reduced. The typical system is shown
in Figure 15 below and the corresponding T-s and p-h graphs are in Figure 16. This configuration was
originally proposed and patented by Bergander (2009). Vapor compressed in the mechanical compressor
is suctioned by the ejector. The motive fluid in the ejector is a liquid delivered by a mechanical pump.
The presented configuration is a totally novel approach for improving the efficiency of refrigeration
systems by means of a two-phase liquid-vapor ejector.

Figure 15. Schematic of Category 2 ejector refrigeration system

18

Ejector Refrigeration Cycles

Figure 16. Category 2 Compression-ejection cycle of the system presented in Figure 15 in coordinates p-h
(Bergander, 2012)

The cycle is composed of the main piping circuit containing the evaporator (5), a compressor (1),
an ejector device (7,8), a condenser (4), a liquid receiver tank, and an expansion valve. The circulation
of a liquid phase of the working medium is provided by the additional liquid line and a pump (6). The
evaporator absorbs the heat from source Q0, while the condenser is connected to the heat sink—high
temperature heat receiver Qk. It should be noted that the device described above can be used also for
heating and in this capacity it can operate as a heat pump.
The thermodynamic reason for the efficiency improvement is an effect of difference in compression
work for liquid and vapor phases. Therefore, increasing the efficiency of the standard single-stage vapor
compression cycle is enabled through a reduction of mechanical compression at the expense of harness-
ing kinetic energy of vapor in the ejector device.
The Category 2 compression-ejection cycle in p-h coordinates is presented in Figure 16. The suction
point 1 is located at the saturation line. However, for the purpose of our analysis, the suctioned vapor can
be also considered as superheated. The isentropic compression process in the mechanical compressor
was assumed in these charts, however the internal efficiency of the mechanical compressor has been
included in our analysis as well. The important feature of the system are two additional pressures, i.e.
inter-stage pressure pm and corresponding inter-stage saturation temperature Tm, as well as pump motive
pressure pp and liquid phase temperature Tp. Therefore the discharge of the mechanical compressor in
the discussed system is represented by point 2 while in the classic one-stage system the discharge will
be located at the point 2’.
No subcooling was taken into account in the above charts and therefore the liquid state is represented
by point 4 located at the liquid saturation line. Nevertheless the liquid subcooling may be considered in
the further analysis. For the liquid compression, an isentropic process was assumed in Figure 16. If the

19

Ejector Refrigeration Cycles

expansion process of the liquid phase occurs isentropically then the outlet of the ejector motive nozzle
is represented by the point 7s. Taking an actual efficiency of the motive nozzle into account – the real
outlet is located at the point 7. In most practical cases wet vapor may be expected at the outlet of the
motive nozzle as the flashing process will occur. The quality of the vapor discharged from the motive
nozzle depends on the nozzle efficiency as well as possible liquid subcooling for the given operation
pressures in the system.
It has to be noted that a good representation of the operation of the ejector at the thermodynamic
charts is difficult as a combination of compression and mixing processes occur at the mixing chamber
simultaneously. In Figure 16, a common approach was applied: the mixing process and compression
process due to momentum exchange were separated. Moreover, the mixing process is assumed to be
an isobaric process of the discharged vapor from mechanical compressor 2 and fluid expanded in the
motive nozzle 7. Therefore the mixing process of 7-8-2 was assumed in Figure 16. The compression
process due to momentum and energy transfer between motive fluid and secondary fluid (vapor dis-
charged from the mechanical compressor) is represented by compression line 8-3s for the ideal case of
isentropic compression and 8-3 for real compression in the mixing chamber and the diffuser. Location
of point 8 as well as point 3 depends on the entrainment ratio of the ejector, which is not visible in the
thermodynamic charts.
Broad-based laboratory experiments were conducted to determine the efficiency improvement which
is offered by this cycle and this research is still continuing.The efficiency improvement anywhere from
3 to 16% was demonstrated (Bergander, 2005), however it strongly depends on the type of refrigerant
used. It appears that higher efficiency gains can be achieved for refrigerants with a low-density ratio for
liquid and vapor phase, such as CO2 and ammonia. Consequently, we have recently switched our interest
to those refrigerants and applications, i.e. large ammonia refrigeration units and possibly automobile
A/C using CO2 as a refrigerant. In the latter case, the cycle becomes transcritical and consequently
condebser is replaced by a gas cooler. The layout for such a system using transcritical CO2 (R-744) and
the corresponding pressure-enthalpy diagram are shown in Figure 17.

Figure 17. Category 2 cycle for transcritical CO2 refrigeration system [37]

20

Ejector Refrigeration Cycles

Despite that the Category 2 cycle has been introduced only a few years ago (patented in 2009) it
already has a quite broad literature base, both from the standpoint of theoretical modeling (Bergander,
2005; Bergander et al., 2010; Colarossi et al., 2012) and practical applications/results of laboratory
experiments (Bergander et al., 2006, 2009; Li & Groll, 2005).

Category 3: Ejector Replacing a Compressor

This is potentially a very attractive application that eliminates a compressor and uses low-grade heat, such
as solar, waste, geothermal, etc. as a main source of energy. An ejector is at the heart of this system, as it
compresses the refrigerant to a required condensation pressure without expanding any mechanical energy.
By eliminating the mechanical compressor from a cycle, the cycle can provide the same amount of refrig-
eration (cool) with using only a fraction of electrical energy (for a pump) as compared to conventional
vapor compression cycle (VCC) systems. This is due to a thermodynamic principle that compressing
the liquid requires substantially less work than compressing the gas (vapor). Less electricity results in
a considerable reduction in emission of carbon dioxide and other greenhouse gases. Additionally, the
cycle offers the possibility of using natural refrigerants, such as isobutane, propane, methanol, etc. and
this leads to eliminating ozone layer depletion effects unlike the case with commonly used refrigerants
belonging to CFC and HCFC groups.
The schematics of the Category 3 ejection refrigeration system together with its thermodynamics
cycle in p-h coordinates are presented in Figure 18. Liquid refrigerant is passed through the pump to
the generator (point 8). The liquid is then heated in the generator by either energy from solar panels or
waste heat. The first stage of heating produces saturated vapor, which is then heated further and leaves
the generator in a superheated condition (point 1). The degree of vapor superheat is the function of the
generator capacity as well as the mass flow rate. Such superheated vapor enters the motive nozzle of the
ejector and undergoes the expansion from the generator pressure pg to the lower pressure, which is an
evaporation pressure pe (point 2). The ejector suctions vapor flowing from the evaporator (point 7), and
mixes it with expanded vapor (point 2) and in consequence, the mixed vapor in state 3 is obtained. The
pressure of the working fluid initially rises slightly as a result of the momentum exchange, and then it

Figure 18. Schematic diagram of Category 3 ejector refrigeration system (left) and thermodynamic cycle
in pressure-enthalpy co-ordinates (right)

21

Ejector Refrigeration Cycles

rises more in the diffuser up to point 4, achieving the level of the condensation pressure pc. Compressed
vapor enters the condenser, where it condenses and may also subcool depending on the conditions of
the cooling in the condenser. The working fluid leaves the condenser in the liquid state (point 5). It is
then divided to two parts: one part flows to the generator through the circulating pump, meanwhile the
remaining part flows to the evaporator through the expansion (throttling) valve, in which it is throttled to
the evaporation pressure of pe, achieving the condition of wet vapor (point 6). Through the boiling in the
evaporator, a working fluid absorbs cooling capacity Qo from the refrigerated medium, which may be the
air circulating in an air-conditioned room, or ice water, which in turn will cool the air-conditioned room.
The main difference between this cycle and the conventional refrigeration cycle (reverse Rankine
cycle) besides the elimination of a compressor, is that it requires three heat sources at different tempera-
tures rather than two, namely at the generator level (8-1) which is the temperature of the heat source,
a condensing level (4-5), which is the ambient temperature and the evaporator temperature (6-7). The
ejector is the key component of this system as it can increase the pressure without consuming mechanical
energy. Due to these characteristics, applying an ejector may be the simpler and safer technologically
than applying mechanical devices for pressure increase.
This technology is still under development but important milestones have been reached by the au-
thor and his research team as well as by some others. The technical background has been established
for designing a prototype for heat-driven air-conditioners and the best refrigerants have been selected
(Bergander, 2005). Further, we completed the model of ejector refrigeration cycles and proposed the
initial geometry for an ejector for this application.
The model ejector was fabricated and installed on a laboratory test stand working on isobutane as a
refrigerant. We achieved sustainable operation of the system and were able to lower the ambient tem-
perature by 20C, utilizing heat at temp. 80C. Further, we concluded that the amount of heat required
to deliver a certain cooling load is reasonable and achievable with solar or waste heat sources (3.5 kW
of refrigeration capacity required 2-3 solar panels, which is reasonable). We believe this was the first
practical demonstration of a working ejector-based, heat-driven cooling system in the world (Butrymo-
wicz & Bergander, 2010).
The results obtained to-date have confirmed many positive claims related to this technology, and
especially that it:

1. Reduces electric energy consumption by approx. 90% when compared to today’s systems,
2. Eliminates the compressor, and hence results in an A/C system having a much lower initial cost
(50-60% less than today’s systems)
3. Is scalable from a small capacity in the range of few Watts for computer cooling up to tens of kW
for residential and commercial A/C,
4. Can be used with natural refrigerants and hence has no net GHG emissions.

ANALYSIS OF THE CONDENSING EJECTOR FOR


CATEGORY 2 REFRIGERATION CYCLE

Theoretical Analysis of Condensing Ejectors

As shown previosuly in this chapter, a special version of two-phase ejector, called a “condensing ejec-
tor” is at heart of the Category 2 refrigeration system and despite of fair amount of previous research it

22

Ejector Refrigeration Cycles

is still the least understood component. Therefore, one of primary objectives of our investigations was
to develop a flow model that would predict the performance of such ejector operating as a second step
compression in a refrigeration cycle (reversed Rankine cycle).
A condensing ejector (CE) is a form of a jet pump operating with two-phase flow, which can produce a
discharge liquid stream with a stagnation pressure higher than either of the two inlet stagnation pressures.
Figure 19 presents the predicted behavior of a condensing ejector based on the results of literature study
and our initial computer simulation. All previous studies were performed for water-steam mixtures while
our research is the first attempt to define the condensing ejector operation for vapor-liquid refrigerant as
a working medium. The high relative velocity between vapor and liquid streams produces a high value
of heat transfer and the rate of condensation of the vapor phase is high. This condensed vapor adds to
the momentum flux of liquid stream. The remaining vapor and liquid jet (original plus condensed vapor)
enter the constant area mixing section where a condensation shock may occur such that a completely
liquid state exists downstream of the shock. The pressure rises due to lowering the velocity of the stream
and the mass flow balance is maintained by a sudden change of density (condensation).

Modeling Condensing Ejectors

The basic question that needed to be answered was, given the inlet conditions, what will be the exit pres-
sure from the condensing ejector. A conventional mathematical analysis, which consists of equations of
mass, momentum and energy, results in the determination of the discharge pressure; however it does not

Figure 19. Condensing ejector processes and pressure distribution

23

Ejector Refrigeration Cycles

explain the character nor the location of a pressure shock. Therefore in order to gain the full understanding
of these phenomena, we were using computer simulation together with experiments. In order to capture
the essential physics of condensation shocks, a multi-dimensional simulation capability was developed
using the Homogeneous Relaxation Model (HRM). The experimental ejector for R22 refrigerant was
designed using the basic fluid dynamics theory, fabricated and installed in the existing laboratory stand.
The first objective of experiments was to investigate the possibility of the pressure jump in the ejector
and to define the inlet conditions at which such pressure shock occurs. Consequently, the inlet pressures
and temperatures of both vapor and liquid refrigerant were controlled within the range permitted by the
laboratory stand that we had to our disposal. The back-pressure was controlled by the valve installed on
the discharge from the condensing ejector. The key scientific objective was to obtain the pressure jump
on the ejector and such condition was indeed observed during our experiments.
Two modeling approaches were pursued: a conventional control-volume approach was first employed
to offer quick, general guidelines for operating conditions for the condensing ejector. This approach can
provide general information for system optimization and operating constraints. The second approach
was to construct a multi-dimensional CFD model of the condensing ejector. The multi-dimensional
CFD model requires the construction of an entirely new solver that was customized for this application.

Control Volume Analysis

A control volume representing the condensing ejector and is shown in Figure 20. Note that the number-
ing of locations differs from Figure 19. The following assumptions were taken: 1) only inlet and outlet
parameters were considered while the processes that take place inside of an ejector were not examined,
2) liquid is incompressible, 3) a full condensation takes place in a constant-area mixing chamber and
4) heat loss to environment is neglected. From this control volume analysis, the continuity, momentum,
and energy equations (modified Bernoulli equation) are obtained:
Continuity equation:

−m v − m L + m 3 = 0 (4)

Figure 20. Control Volume for CE analysis. Point 2 (not shown in the picture) relates to the constant-
area mixing chamber - between converging and diverging areas

24

Ejector Refrigeration Cycles

Momentum conserv:

−m vwv − m LwL + m 3w 3 − pv Av − pLAL + p3A3 +WPF = 0 (5)

Energy conservation:

 w 2   w 2   w 2 
−m v hv + v  − m L hL + L  + m 3 h3 + 3  = 0 (6)
 2   2   2 

.
where the subscripts v, L, 3 signify vapor, liquid, and outlet, m is the mass flow rate, U is axial veloc-
ity, p is pressure, A is cross-sectional area, h is enthalpy, and WPF is the wall pressure force. WPF
is evaluated by assuming the pressure is constant and equal to the vapor phase pressure at the inlet and
the end of the convergent mixing section:

WPF = pv (A1 − A2 ) (7)

where A1 is a sum of inlet areas for vapor and for liquid: A1= Av + AL
By re-arranging formulas (4) through (6) above, and calculating nozzle efficiency using generally
accepted calculation methodology, the following formulas for final enthalpy and pressure were obtained:

 2  2 2 2
h3 =
mv h + wv  + mL h + w L  − w 3 + z w 3 (8)
 v  
m  2  m  L 2  2 2

AL A mw
p3 = pL + pv v + L L +
A3 A3 A3
(9)
mvwv m 3w 3 pv (A1 − A2 ) p2 (A2 − A3 )
+ − − −
A3 A3 A3 A3

where z is a nozzle efficiency coefficient.


The developed Control Volume model allows analyzing the influence of various inlet conditions on
the final enthalpy and pressure. Such analysis was done but it is too voluminous to be presented here.
For example, it was concluded that the final pressure and enthalpy can be increased by increasing the
ratio of liquid-to-vapor mass flow. In the refrigeration system we can only achieve this up to certain
values, and from this analysis it would appear that the optimum would be to have this ratio above 0.53.
The effect of inlet pressures and nozzle geometry were also investigated.
In addition to the above equations of motion, the system must satisfy the second law of thermodynam-
ics, i.e. entropy generation must be greater than or equal to zero. In an adiabatic process with no work
applied, the entropy balance equation becomes:

25

Ejector Refrigeration Cycles

−m vsv − m LsL + m 3s 3 ≥ 0 (10)

where s is the entropy at the specified location.


Knowing the input entropies and mass flow rates and the exit mass flow rate from (4), the inequal-
ity sign is replaced with an equals sign, and (10) can be solved for s 3 . Once s 3 is known, and using h3
from (6), the output pressure p3 can be obtained and compared with the output pressure when entropy
is not taken into account. These equations can be solved using an iterative method with the Engineering
Equation Solver (EES) or any other equivalent software.
In order to experimentally check the developed model, the typical inlet conditions as per Table 1
were plugged into above developed formulas.
For the above operating conditions, the ratio of the theoretical output pressure to the input vapor
pressure, p3/pv, vs. the ratio of the vapor to liquid mass flow rates, mv /mL was investigated. The analysis
indicates that the second law is an active constraint in the design of condensing ejectors and suggests
that in a perfect system with zero entropy generation, the ratio mv /mL must be less than approximately
1.25 to obtain a gain in output pressure. The quality of R22, x= mg /(mg + mL), at the output of the con-
trol volume should be liquid for maximum efficiency. In order for the output to be a liquid phase at the
operating temperature shown in Table I, the ratio mg /mL must be less than 0.15.
While observing an experimental run, the extreme caution had to be taken to prevent the liquid phase
inlet pump from cavitating. Upon further inspection it was noticed that the condensate temperature was
very close to the vapor temperature for R22. This prompted an investigation to see what effect the input
liquid phase temperature has on the output pressure. In order to examine this, mv /m L was set to a typical
operating value of 0.3, Tv was set to 64°C, and the liquid phase temperature was varied from 10-33oC.
The results of this analysis are shown in Figure 21, where a large increase in output pressure is seen
when TL is decreased.
The location of the current operating temperature is marked with an “X” and is seen to be very close
to the right edge where the performance of the CE is weakest. This result indicates that the condensate
should be cooled before being used in the condensing ejector.
The foregoing discussion is a sample of the analysis achieved with the control-volume assessment of
the condensing ejector. The low computational cost and the convenience of this type of calculation have
greatly informed our operation of the condensing ejector. For example, we have identified optimum mass
flow ratios of liquid to vapor and the optimum area ratio of inlet nozzle to throat area.

Table 1. Condensing ejector operating conditions

Tv = 63.7 oC pv = 1.574MPa

TL = 32.4 oC pL = 1.641MPa

m L = 0.0487kg / s Fluid = R22

26

Ejector Refrigeration Cycles

Figure 21. Effect of liquid temperature on output pressure, as constrained by the second law. The dashed
line represents an upper bound on pressure ratio.

COMMERCIAL POTENTIAL OF CATEGORY 3 CYCLE


FOR AIR-CONDITIONING PURPOSES

Anticipated Benefits

Cheaper, more efficient, and less polluting solutions are required for future air conditioning (A/C) and
refrigeration needs because they account for 15-25% (Bergander, 2012) of electric energy consumption
across the world and create two sources of environmental pollution and global warming: 1) the ozone-
depletion effect of traditional refrigerants and 2) the emission of greenhouse gases connected with the
generation of electricity. The Category 3 ejector refrigeration cycle offers a novel approach for A/C
systems for residential and commercial buildings that, completely eliminates the ozone depletion ef-
fect and decreases the need for electric power by 90%. This technology explores an intriguing concept
- capturing the heat from either the sun or from other low-temperature heat sources to produce cooling.
Indeed, the cycle as described in the previous chapter uses free or inexpensive heat, either solar or waste,
as the main source of energy instead of electricity.
The main characteristic of the Category 3 cycle is that it uses an ejector for non-mechanical com-
pression. In this system, a liquefied refrigerant is pressurized by a simple pump, then heat is added to
evaporate it and finally the refrigerant is re-compressed in an ejector without any mechanical energy
spent. Further, the system is well suited for natural and synthetic refrigerants, instead of highly destructive
Freons. However, the adaptation of this technology to residential/commercial A/C is not that straightfor-
ward and it requires resolving many challenging technical problems that were never addressed before.
It is expected that this development will eventually emerge as an attractive, energy saving product with
potential worldwide applications. The system is best suited for southern areas with plenty of sun exposure.

27

Ejector Refrigeration Cycles

We made an estimate for Southern US states and came up with the result that if only 10% of houses in
those states adopt this technology for air-conditioning, the electricity savings may reach $2.65Bln per
year and the emission of greenhouse gases can be decreased by 36 Million Tons/year.
This technology offers other attractive applications for cooling field hospitals, command centers and/
or data centers. There is also a great potential to provide refrigeration and A/C in Third World countries
and in remote areas where electricity is unavailable. Small, solar-based units can be developed for the
storage of medicines, perishable food, and to cool field clinics.
It has to be admitted that ejector use in refrigeration and A/C systems was considered as controversial
in the past, mainly because a lot of research has been conducted for many years with only theoretical
results and without visible, commercial products. However, recent successes in either commercializing
ejector-based, A/C systems (Denso Corp) (Bergander, 2012) or in obtaining favorable practical results
(this author) (Takeuchi, 2009) are extremely encouraging in demonstrating that ejector-based systems
are indeed capable to provide economical cooling.

Marketable Products

• Current Solutions to Customer Needs: The problem of reducing both the cost and environmen-
tal effects of commonly used refrigerants is very well recognized by many governments in the
world, especially in the USA and European Union. Specifically, the US Energy Independence and
Security Act of 2007 created funding for new solar air-conditioning R&D programs. These activi-
ties should develop and demonstrate multiple new technology innovations and mass production
economies of scale. Solar air-conditioning will therefore play an increasing role in zero energy
and energy-plus building design. Specific technologies promoted under this program include: so-
lar A/C using desiccants, passive solar cooling, solar thermal cooling (absorption), photovoltaic
(PV) solar cooling, geothermal cooling and others. We believe that the solar ejector–based system
and its subsequent engineering development as presented here is on the forefront of this program.
The specific application of ejector refrigeration offers major benefits quickly because of the rela-
tively short time required for full commercialization.
• Residential Solar Cooler: The Category 3 cycle offers an opportunity to develop an air-condi-
tioning system that 1) utilizes natural refrigerants and therefore operates without any ozone deple-
tion effects, 2) utilizes solar or waste heat energy as a main source of energy, and 3) eliminates a
mechanical compressor, which is a main user of electricity. Further development can lead to at-
tractive, energy saving products that significantly reduce the electric energy use for residential and
commercial chiller/air-conditioning systems. The author’s work is concentrated on one specific
application, which is a central A/C residential system and consequently, the marketable product is
an A/C unit without a compressor and driven by waste and/or solar heat. It is expected that the line
of A/C units will be developed and eventually manufactured in various sizes for residential use,
i.e. from 1 to 5 Tons capacity (1 Ton of refrigeration = 12,000 Btu/hr or 3.517 kW). A schematic
layout of a home A/C system based upon this new generation of cooling technology is shown in
Figure 22.

Since solar heat is relatively unreliable, especially in certain climates, and is not available at night, there
is the need to investigate the operation of the system during those times without the availability of solar
heat. Figure 22 shows already two possible solutions to this problem; specifically, storage of either hot

28

Ejector Refrigeration Cycles

Figure 22. Schematics of the residential solar cooler

or cold water in hot or cold tanks, respectively, in order to use the heat/cold during the night hours.
The feasibility of these solutions was indeed investigated and this analysis showed that storage tanks of
reasonable size will be able to provide adequate cooling during night, at least in certain geographical
areas. Other options also exist and could be incorporated into a commercial system, namely adding an
electric heater or an auxiliary compressor.

• Value Proposition: Since there is a large potential for substantial energy savings, the Category
3 ejector cycle technology can contribute on the international scale to the preservation of natural
resources and reducing our dependence on foreign oil. There are many economical benefits that
this technology offers, but the most important is the opportunity to decrease significantly electric-
ity use in air-conditioning. At the same time, the initial price of those new A/C units can be sub-
stantially lower than for a comparable size conventional unit because the mechanical compressor
is eliminated. We anticipate that the price may be lower even by 50-60%.. The new unit will be
especially attractive to those homeowners who have already installed solar panels for domestic hot
water. In this case, the A/C unit will be a natural extension of the solar-based system and it will
produce a bulk of required A/C almost free-of-charge.

The payback period for the solar residential cooler and its comparison to leading competitors’ prod-
ucts is shown below for specific US conditions. It is expected that future government regulations will
either mandate or promote energy conservation and provide financial incentives for energy-efficient
equipment. One past example of this is the US Energy Act of 2005.
The main advantage of the solar A/C is that heat rather than electricity is the main source of energy.
Further, the driving heat is either very cheap (waste heat) or free (solar heat). However, an economic
life cycle analysis has to consider the initial installation costs, which are higher for the new system than

29

Ejector Refrigeration Cycles

for conventional VCC air-conditioning in the case when brand-new solar panels have to be installed.
However, the cost of the new A/C unit itself is much lower than for a typical VCC unit, since it eliminates
the compressor, a most expensive and trouble-prone component. Table 8.1 below shows the estimated
costs of installation and operation over 10 and 20 yrs for a conventional A/C with an average SEER (13)
and for three different variations of the solar cooler: (i) using unglazed solar panels, (ii) using glazed
panels and (iii) using solar vacuum tubes. The installation cost information was taken from the DOE
EERE website and from discussions with local installers.
As shown, the solar A/C system based on Category 3 cycle offers a considerable savings of electri-
cal energy. While a 1 Ton size compressor uses approx 1 kW of electrical energy per hour for 2800 hrs
annually (data for Florida), the new system cuts the electric energy use to just 100W - the power needed
to run two pumps, one for the solar circuit and another for the A/C system. This is actually a very con-
servative estimate; as thermodynamic calculations for isobutane shows that the energy consumed by the
pump is 1-2% of heat energy needed to drive the complete A/C cycle. Based upon the data in the Table
2, one concludes that the new A/C system will pay for itself in less than 10 years. It may be especially
attractive for those homeowners who have already installed solar panels for other applications (hot water,
pool, hot tub, etc.). Consideration of unglazed panels may be worthwhile to consider. These panels are
very inexpensive to buy and install and they are especially well suited for warmer climates. They are
commonly used for swimming pool heating and domestic hot water. Unglazed panels are also extremely
popular in Third World countries for all domestic hot water needs.

Table 2. Cost comparison for solar A/C vs conventional VCC (vapor compression cycle) system for unit
size: 1 ton of refrigeration (3.5 kW). All costs in US$.

Conventional VCC, SEER 13 Solar Solar Solar


Unglazed Panels Glazed Panels Vacuum Tubes
Installation cost for A/C 3,500 2,000 2,000 2,000
Installation cost for N/A 2,500 4,500 6,000
solar panels
Annual maintenance, 100(1) N/A but panel repairs/ 50 50
and repairs replacement after 10 yrs:
2,000
Electricity cost for 10 4,600(2) 450(3) 450 450
yrs operation
Electricity cost for 20 12,200 1,200 1,200 1,200
yrs operation
Total cost: 9,000 4,900 7,400 8,900
10 yrs 17,600 7,150 8,650 10,150
20 yrs
Payback period N/A 2-3 yrs 7-8 yrs 9-10 yrs
Tons of CO2 emission (4)
39 4 4 4
over 20 yrs operation:
Note (1) Compressor maintenance, refrigerant additions, etc.
Note (2) Assuming 2,800 hours of operation per year (data for Florida) at 0.12$/kWh, and electricity price increases 5% per year.
Note (3) The cost for running 2 pumps, one for the solar system and another for the A/C system, Total electric power 100W for 2,800 hrs
annual operation.
Note (4) CO2 emission calculated from EPA data: 1kWh produces 1.34 lbs of CO2

30

Ejector Refrigeration Cycles

The total operating cost over the entire life cycle of 20 yrs is considerably lower for solar A/C systems.
In the worst case scenario, that of the vacuum tubes as the most expensive solar panels, the 20-year cost
is still only slightly more than a half of the cost of the running a traditional VCC air-conditioner. How-
ever, for the nation and the world, the most dramatic feature is the reduction in carbon dioxide and other
greenhouse gas emissions. According to EPA estimates, the production of 1 kWh of electricity emits
on average 1.34 lbs of greenhouse gases. Therefore, for a one Ton-size unit, CO2 reduction over the life
cycle of 20 yrs reaches 35 Tons. If we consider just the state of Florida with 9 Mln residential units, and
further assuming the average size of an A/C system to be 3 Tons, the use of ejector refrigeration systems
could achieve savings of greenhouse gas emissions over the next 20 years approaching 1 Billion Tons.

• Other Opportunities: Besides the obvious applications in the US, European Union and other
developed economies, there is great potential of using Category 3 cycle for refrigeration and
air-conditioning purposes in Third World countries and in remote areas where electric energy is
unavailable. Specifically for this purpose, an author’s company, Magnetic Development, Inc. of
Madison, CT, USA has previously developed an inexpensive mobile solar unit, shown in Figure
23, which together with the proposed cooler can produce about 1/3 Tons of refrigeration (1kW)
without any need for outside electricity if the pump is run by a small, attached, solar photovoltaic
panel with only 2 sq. ft surface area.

TOWARDS SUCCESSFUL COMMERCIALIZATION OF PRODUCTS


BASED ON CATEGORY 3 CYCLE: FUTURE RESEARCH DIRECTIONS

The discussion in this chapter has demonstrated that the solar cooler technology meets and exceeds the
criteria imposed by many governments for successful new renewable energy projects and specifically:

1. It reduces electric energy consumption by approx. 90% when compared to today’s systems,
2. It eliminates the compressor, and hence results in an A/C system having a much lower initial cost
(50-60% less than today’s systems),
3. It is scalable from a small capacity in the range of few Watts for computer cooling up to tens of
kW for residential and commercial A/C,
4. It can be used with natural refrigerants and hence has no net GHG emissions.

Therefore, the continuation of research toward a successful commercialization of a product is per-


fectly justified. The main technical objective for the commercialization phase will be to build a working
prototype and to prove its validity by installing it in a single-family house and monitoring its operation
and energy usage for a period of at least a full year. In addition to technical reasons, such a prototype
would serve as a demonstration for the engineering community and prospective investors/licensees.
Even though the achievements of our previous research have provided a solid technical foundation, the
design and fabrication of an environmentally friendly air-conditioning system still requires solving many
difficult problems, ones never approached before. Specifically, the most critical issues include: 1) study
of the system operation over a variety of climatic conditions - solar radiation, ambient temperatures and

31

Ejector Refrigeration Cycles

Figure 23. A mobile solar unit developed by author’s company, Magnetic Development, Inc.

required cooling capacities, 2) options for night operation when solar heat is not available, 3) defining
requirements for a control system, which considers all the above variations of operational conditions.
In order to demonstrate the commercial viability of this technology, we have to ask one basic question:
are the results of the research done to-date adequate for developing the design concept for a working
prototype? If so, is this solution economically viable?
If the answers to above questions are positive, then near-commercial version of the unit should be built
and tested both in the laboratory and in the field. Additionally, the issues associated with manufacturing
in large volumes at competitive prices will be also addressed.
The successful completion of this major objective requires a number of intermediate steps for the
realization of full potential of a Category 3 refrigeration cycle. Specifically, the following major techni-
cal problems should be investigated and further applied in the design of a prototype:

1. Continuation of the analysis of the thermodynamic cycle,


2. Further CFD simulation to produce the optimal geometry of an ejector for a given operation
conditions,
3. Selection of heat transfer equipment
4. Development of a control system
5. Design the prototype refrigeration cycle by utilizing all results of previous investigation
6. Decide the best solution for cooling needs at night

It is anticipated that such research will be able to provide the background to further improvements in
technology. If this development is successful, then other applications of similar thermodynamic cycles
might be pursued, for example the concept of power generation utilizing low temperature heat sources.
In addition, some components, especially ejectors/jet compressors might sustain improvements enhanc-
ing their commercial value.

32

Ejector Refrigeration Cycles

REFERENCES

Aphornratana, S., & Eames, I. W. (1997). A small capacity steam-ejector refrigerator: Experimental
investigation of a system using ejector with movable primary nozzle. International Journal of Refrigera-
tion, 20(5), 352–358. doi:10.1016/S0140-7007(97)00008-X
Bergander, M. (2005). New Regenerative Cycle for Vapor Compression Refrigeration, U.S. Dept. of
Energy, OSTI I.D. 850491, Final Report, Aug.
Bergander, M. (2006). Refrigeration Cycle with Two-Phase Condensing Ejector, 11th Intern. Refrigera-
tion and Air-Conditioning Conference at Purdue, Paper No. R008, Purdue University, July.
Bergander, M. (2009). Refrigerant Pressurization System with a Two-Phase Condensing Ejector, US
Patent No. 7,559,212.
Bergander, M. (2012). Ejector Refrigeration Cycles. Krakow: AGH University Press.
Bergander, M., Butrymowicz, D., & Karwacki, J. (2009). Improvement of Refrigeration Efficiency by
means of Two-Phase Ejector as Second Stage Compressor, International Seminar on Ejector/Jet Pump
Technology and Application, Louvain-an-Neuve, Belgium, September.
Bergander, M. J., Butrymowicz, D., Smierciew, K., & Karwacki, J. (2010). Refrigeration Cycle With
Ejector for Second Step Compression, 13th Intern. Refrigeration and Air-Conditioning Conference at
Purdue, Paper No. R2211, Purdue University, July.
Blatchley, C. G. (1961). Selection of Air Ejectors. Chemical Engineering Progress, 57, 114–144.
Bohdal, T., Charun, H., & Czapp, M. (2003). Urzadzenia Chlodnicze Sprezarkowo Parowe. Warszawa:
Wydawnictwa Naukowo-Techniczne. (in Polish)
Brown, G. A., & Levy, E. K. (1972). Liquid-vapour interactions in a constant area condensing ejector.
Journal of Basic Engineering, 3, 169–180.
Butrymowicz, D. (2013). Improvement of Compression Refrigeration Cycle by means of Two-Phase
Ejector, Int. Congress of Refrig. 2003, Washington, DC, paper No. ICR0310.
Butrymowicz, D., & Bergander, M. (2010). Ejector-Based Air Conditioner Utilizing Natural Refrigerants,
13th Intern. Refrig. and Air-Conditioning Conference at Purdue, Paper No. R2210, Purdue University, July.
Campbell, I. J., & Pitcher, A. S. (1958). Shock Waves in a Liquid Containing Gas Bubbles. Proceedings
of the Royal Society of London. Series A, Mathematical and Physical Sciences, 243(1235), 534–545.
doi:10.1098/rspa.1958.0018
Chunnanond, K., & Aphornratana, S. (2004). Ejectors: Applications in refrigeration technology. Renew-
able & Sustainable Energy Reviews, 8(2), 129–155. doi:10.1016/j.rser.2003.10.001
Colarossi, M., Trask, N., Schmidt, D., & Bergander, M. (2012). Multidimensional Modeling of Con-
densing Two-Phase Ejector Flow. International Journal of Refrigeration, 35(2), 290–299. doi:10.1016/j.
ijrefrig.2011.08.013

33

Ejector Refrigeration Cycles

Downar-Zapolski, P., Bilicki, Z., Bolle, L., & Franco, J. (1996). The Non-Equilibrium Relaxation Model
for One-Dimensional Flashing Liquid Flow. International Journal of Multiphase Flow, 22(3), 473–483.
doi:10.1016/0301-9322(95)00078-X
Elbel, S., & Hrnjak, P. (2008). Ejector Refrigeration: An Overview of Historical and Present developments
with an Emphasis on Air-Conditioning Applications, 12th Intern. Refrigeration and Air-Conditioning
Conference at Purdue, Paper No. R2350, Purdue University, July.
Fisenko, V. V. (1987). S’zhimayemost Teplonosityela i Effektivnost Raboty Konturov Tsirkulatsiy Ya.E.U.,
(book in Russian). Moscow: Energoatomizdad.
Gay, N. H. (1931). Refrigerating System, US Patent No. 1,836,318.
Huang, B. J., Chang, J. M., Wang, C. P., & Petrenko, V. A. (1999). A 1-D analysis of ejector performance.
International Journal of Refrigeration, 22(5), 354–364. doi:10.1016/S0140-7007(99)00004-3
Keenan, H., Neumann, E. P., & Lustwerk, F. (1950). An investigation of ejector design by analysis and
experiment. J. Appl. Mech. Trans. ASME, 72, 299–309.
Kemper, C. A., Harper, G. F., & Brown, G. A. (1966). Multiple-phase ejector refrigerating system, US
Patent 3,277,660.
Kitto, J. B., & Stultz, S. C. (2005). Steam – Its Generation and Use (41st ed.). Barberton, OH: The
Babcock and Wilcox Company.
Kornhauser, A. A. (1999). The use of an Ejector as a Refrigerant Expander, Proc. of 1990 USNCR/IIR-
Purdue Refrigeration Conference, West Lafayette, IN. p.10-19.
Levy, E. (1967). Investigation of liquid-vapor interactions in a constant area condensing ejector. (SCD
thesis). Dept. of Mech. Engg., MIT, Cambridge, MA.
Li, D., & Groll, E. A. (2005). Transcritical CO2 refrigeration cycle with ejector-expansion device. In-
ternational Journal of Refrigeration, 28(5), 766–773. doi:10.1016/j.ijrefrig.2004.10.008
Munday, J. T., & Bagster, D. F. (1977). A new theory applied to steam jet refrigeration. Industrial &
Engineering Chemistry Process Design and Development, 16(4), 442–449. doi:10.1021/i260064a003
Sun, D. W. (1996). Variable geometry ejectors and their applications in ejector refrigeration systems.
Energy, 21(10), 19–29. doi:10.1016/0360-5442(96)00038-2
Takeuchi, H. (2009). World’s First Ejector Cycle for Mobile Refrigerator to Stop Global Warming.
International Seminar on Ejector/Jet Pump Technology and Application, Louvain-an-Neuve, Belgium,
September.
Trela, M., Kwidzinski, R., & Butrymowicz, D. (2008). Exergy analysis of losses in a two-phase steam-
water injctor. Chemical and Process Engineering, 29, 452–464.
Van Wijngaarden, L. (1972). One-Dimensional Flow of Liquids Containing Small Gas Bubbles. Annual
Fluid Mechanic, 4(1), 369–396. doi:10.1146/annurev.fl.04.010172.002101

34

Ejector Refrigeration Cycles

Wallis, G. B. (1980). Critical Two-Phase Flow. International Journal of Multiphase Flow, 6(1-2), 97–112.
doi:10.1016/0301-9322(80)90041-5
Witte, J. H. (1969). Mixing Shocks in Two-phase Flow. Journal of Fluid Mechanics, 36(04), 101–139.
doi:10.1017/S002211206900190X

KEY TERMS AND DEFINITIONS

Air Conditioning: A process of altering the temperature and humidity of air to more comfortable
conditions.
Ejector: A pump-like device without moving parts.
Global Warming: A climate change observed as a rise in the average temperature of the Earth’s.
Greenhouse Gas Emission: Rejection of gases that trap heat in the atmosphere (especially CO2)
by various industrial processes.
Jet Pump: A device where pumping effect is obtained by nozzle using a Venturi principle.
Natural Refrigerants: Chemicals, occuring in nature that can be used as refrigerants. Examples
being ammonia, CO2, propane, isobutene, etc. They do not deplete the ozone layer and make only a
negligible contribution to global warming.
Ozone Depletion: A steady decline in the total volume of ozone in earth stratosphere.
Refrigeration: A process in which work is done to move heat from one location to another.
Renewable Energy: An energy that comes from resources which are naturally replenished such as
sunlight, wind, geaothermal heat, ocean tides, waves, biomass.
Solar Energy: A radiant light and heat from the sun that can be harnessed using various technolo-
gies, either for production of heat or electric power.
Thermodynamics: A branch of physics concerned with heat and temperature and their relation to
energy and work.
Vapor Compression Cycle: A thermodynamic cycle used in most refrigerators and air conditioners
relying on the properties of compressed gases.
Waste Heat: Heat that is normally rejected by machines and other processes that use energy.

35
36

Chapter 2
The Diffusion Absorption
Refrigerator Operation
and Performance
Lorenzo De Pascalis
University of Salento, Italy

Giuseppe Starace
University of Salento, Italy

Federica Carluccio
University of Salento, Italy

ABSTRACT
This chapter focuses on the Diffusion Absorption Refrigerator (DAR) cycle and describes a new advanced
thermodynamic model which allows good predictions of the chiller performance in terms of efficiency
and cooling capacity, starting from a precise evaluation of the thermo-physical properties of the working
mixture at each point of the circuit. A steady state thermodynamic analytical model of the thermal pump
driving the DAR is also included. In addition, the experimental validation of the model, performed on a
prototype built coupling a domestic 750 W-magnetron with a small purposely modified commercial DAR
to activate the thermal pump, is here included: a maximum mismatch of 2.32% in the weak mixture mass
flow rate and lower than 5% in COP between the predicted and measured data were found.

THE DIFFUSION ABSORPTION REFRIGERATOR CYCLE FUNDAMENTALS

The Diffusion Absorption Refrigerator (DAR) cycle was introduced by Von Platen and Munters (1928)
and uses a binary mixture of ammonia (refrigerant) and water (absorbent) as working fluids in solution
with either hydrogen or helium as auxiliary inert gas. The latter plays its role to reduce the partial pres-
sure of the refrigerant in the evaporator according to Dalton’s law and to allow the fluid to evaporate
and to produce the cooling effect.

DOI: 10.4018/978-1-4666-8398-3.ch002

Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

The Diffusion Absorption Refrigerator Operation and Performance

The main characteristic of the DAR is of having no moving parts: a thermally driven pump (also
referred to as the thermal pump) acts to circulate the fluid instead of a mechanical device. Therefore, the
refrigerator is also both reliable and quiet. Nowadays, it is commercially available only in small cooling
capacities up to 100 W where heat is supplied to the thermal pump by electrical heating cartridges or
natural gas burners. Due to the noiseless operation and to the possibility of using natural gas as a source
of energy, the system is nowadays used as hotel and domestic refrigerator, portable freezer and ice-maker,
caravan and boat coolbox, wine cellar and so on.
The cycle can be described (Figure 1) starting from the thermal pump, where the heat power Q H is
supplied to the strong mixture to reach the state 2. The hot refrigerant and absorbent vapor mix arises
in the riser tube and, at point 4, is separated from the remaining weak liquid mixture that returns through
the shell of the riser tube (state 3). The vapor at state 5 goes through the rectifier and is cooled down to
separate most of the absorbent fluid. The remaining near-pure ammonia vapor at state 6 moves towards
the condenser where condenses (7). Then it flows to the evaporator entrance (9) after a cooling in the
gas heat exchanger (also named as the gas HX). The uncondensed refrigerant flows to a reservoir through
the gas bypass. At the evaporator entrance, the liquid refrigerant reduces its partial pressure as it mixes
with the auxiliary inert gas arriving from the absorber through the gas HX. The resulting mixture leaves
the evaporator and the following pre-heating in the gas HX allows the refrigerant to become a saturated
vapor (state 10). Then, the gas and vapor mixture passes through the reservoir, entering the absorber
coil from the bottom by flowing upward in a counter-current arrangement to the weak mixture, that
enters the absorber coil from the top (state 8) through the liquid mixture heat exchanger (liquid HX).
The refrigerant vapor is absorbed in the weak mixture. The resulting strong mixture flows to the reser-
voir. Then it leaves the reservoir (11) towards the thermal pump. The inert gas is not absorbed and returns
to the evaporator.
The cycle here described makes the DAR small and simple to be built (no particular manufacturing
technologies have to be used), but not really efficient with very low COPs (defined as the ratio between
the cooling power removed at the evaporator and the heat power supplied to the thermal pump) and,
therefore, with high energy consumptions. This is mainly due to:

• The presence of the auxiliary gas inside the evaporator, that reduces the refrigerant mass flow rate
evaporating and thus the cooling capacity;
• The rectifier, which dissipates towards the ambient a not negligible amount of the supplied heat.

The above issues explain the reason why the DAR cycle is widely employed in the refrigeration and
not in the HVAC applications; here single and double stage absorption systems filled either with water
and lithium bromide, or ammonia and water as working mixtures are the most appropriate and provide
the highest performance and efficiencies.

OVERVIEW

This chapter contains a review of the state of the art and the research lines about the DAR technology
as well as a description of the latest results found by the authors, in order to create an up-to-date and
complete source of information and knowledge about the DAR.

37

The Diffusion Absorption Refrigerator Operation and Performance

Figure 1. Schematic diagram of the DAR cycle


(Adapted from Dometic, 2004)

After a deep description of the thermo-physical characteristics of the working mixture (Carluccio,
Starace, & Bongs, 2014) useful to implement thermodynamic models of all the absorption applications,
as well as to predict the behaviour of each sub-device, the authors provide the advanced thermodynamic
model of the DAR cycle they developed, with no pure ammonia as refrigerant assumptions and with
the consideration of the thermal pump heat losses towards the ambient (Starace & De Pascalis, 2012).
Comparing this model with others available in literature, it results in a more flexible and realistic
analysis of the cycle, and in a more suitable description of the real operation of the cycle; this was mainly
due to the introduction of a a steady state analysis based on the momentum balances of the two-phase

38

The Diffusion Absorption Refrigerator Operation and Performance

flow inside the thermal pump (Starace & De Pascalis, 2013). During the design process the use of the
model of the DAR circuit can help to optimize each component in terms of mixture flow rates, of heat
and mass transfer and of energy performance (both cooling power and efficiency).
The model is implemented in a Labview v.8.6 (National Instruments Academic Site License, 2009)
code and also allows to predict the behavior of the flow in the riser tube as a function of the geometry
of the thermal pump.
Several runs of the model were carried out to investigate its potential and to perform a sensitivity
analysis to the main input parameters.
In addition, the operation of a commercially available DAR and its thermodynamic conditions in
each section of the circuit were tested varying the heat power supplied to the thermal pump, in order to
validate the model. The measured values, compared with the model outputs, are presented here.
The experimental campaign was carried out in a steady state operation, thanks to a high heat power
supplied to the cycle, that allowed to acquire a number of measurements adequate for the model accuracy
to be verified in a reasonable testing period.
To perform the campaign, the circuit of the above said commercial DAR was modified integrating
its original thermal pump activation section with the magnetron of a domestic microwave oven. The
microwaves, guided to the working mixture, energize the water molecules. The ammonia remains heated
by the water molecules, the mixture enthalpy increases and refrigerant vaporization occurs. The vapor
rises the riser tube and drags the liquid water as it occurred in the circuit before the change in the heating
process. Anyway, the higher energy density (in terms of joule per cubic meter) supplied to the mixture
causes the reduction of the separation time between water and ammonia. In addition, thermal insulation
of the thermal pump walls was highly increased. The more uniform heating process inside the reservoir
allows to reduce the low temperature stagnation areas of the mixture and to decrease the time needed to
reach a suitable two-phase flow in the riser tube. The actual power of the chosen common magnetron
(750 W) led to an increase in the refrigerant mass flow rate and a reduction of the auxiliary gas draining
time at the condenser.
Therefore, the experimental results show that the designed microwaves activated thermal pump pro-
duces a distinct reduction in the starting transient time of the refrigerator and for this reason, with the
necessary engineering optimization, it can be proposed as a valid candidate to be used as a rapid cooler
in small refrigeration applications.
In addition, in terms of the performance of the DAR cycle, a sensitivity analysis (performed using
the developed thermodynamic model) shows that, as the thermal pump temperature increases, the strong
mixture mass flow rate reduces more than that of the refrigerant one and COP decreases. This happens
as the absorption temperature increases as well, due to the lower absorption of ammonia in the weak
mixture. COP is found to increase when the ammonia concentration decreases. In addition, for lower heat
fluxes supplied to the thermal pump, the heat dissipation towards ambient dominate and COP becomes
small; when the heat flux overtakes a threshold, the evaporator reaches its maximum heat transfer, that
remains constant for any further increase of heat supplied. COP presents its highest values over a limited
range of the thermal pump’s operating conditions.

BACKGROUND

Since the 1930s the DAR cycle was studied and many authors proposed papers on its performance and
optimization. Up to now, a large number of commercial DARs have been investigated. Results showed

39

The Diffusion Absorption Refrigerator Operation and Performance

COPs of 0.2-0.3 and cooling powers between 16 W and 62 W at about 255 to 267 K obtained with tem-
peratures of heat supplied to the thermal pump in the range of 433 and 503 K (Keizer, 1979; Bourseau,
Mora, & Bugarel, 1987; Gutierrez, 1988; Jakob, Eicker, Schneider, Taki, & Cook, 2008). Some methods
to graphically and/or numerically calculate or to experimentally evaluate the properties of the working
fluids circulating in the system were carried out by Reistad (1968), Kouremenos & Stegou-Sagia (1988),
Chen et al. (1996), Srikhirin & Aphornratana (2002), Al-Shemmeri & Wang (2003) and Maiya (2003).
Some of these works present models based on very limiting assumptions and their results frequently
diverge from real DAR behavior. Actually, when authors compare and fit the analytical predictions with
the experimental data in order to model the performance of the refrigerator, the results of their models
can be only referred to those particular cases of DAR circuits.
Zohar et al. (2005) proposed a complex and complete thermodynamic model which explored the
system performance. He and his co-authors found that the maximum COP values reachable with a DAR
cycle could be obtained when the ammonia concentrations in strong and weak mixtures were respectively
0.25-0.30 and 0.10, and the cycle activation temperature was approximately 473 K. The assumption was
that the vapor leaving the rectifier is pure ammonia and no traces of water are present. This is rarely true,
due to the not full efficiency of the rectifier in the process of separating the water from the ammonia.
In addition, it must be pointed out that the temperature of the saturated liquid mixture strongly depends
on ammonia concentration. For instance, at 25 bar a reduction from 1.00 to 0.98 of the ammonia molar
concentration in the liquid mixture causes an increase of the liquid saturation temperature from 331.10 to
332.13 K, higher temperatures at the condenser and a DAR coefficient of performance (COP) reduction.
Moreover, in literature, up to now, several components and circuit improvements were proposed.
Steirlin and Ferguson (1990) carried out a study on the thermal pump optimization. Benhmidene et al.
(2014) developed a two-phase flow model to predict the ammonia and water mixing process inside the
thermal pump. The results show a strong influence of the riser tube diameter on the optimal heat power
supplied to the system in order to maximize the COP values of the cycle. Jacob et al. (2008) improved
the auxiliary gas piping between the evaporator and the absorber. Wang & Herold (1992) experienced
the use of helium as auxiliary gas to replace hydrogen and to reduce the flammability risk.
Despite all these attempts resulted in many advancements for the DAR steady state operation, the
starting transient remains very high, due to the thermodynamic cycle intrinsic behavior. Actually, when
the external heat power is supplied to the thermal pump, time is needed to:

• Heat the working mixture;


• Vaporize the refrigerant;
• Start up the two-phase flow inside the riser tube;
• Drain the auxiliary gas from the condenser.

Even if the refrigerant vapor enters the condenser, it cannot condensate there, due to the presence
of the auxiliary gas. The vapor bypasses the evaporator and is directed towards the absorber, with no
cooling effect. So, more time is needed to generate an adequate refrigerant mass flow rate able to drain
the gas from the condenser and to move it to the evaporator and the absorber. Only after all these steps,
the cycle can reach its steady state operation. So, the DAR total starting time has to be seen as the sum
of the times listed above.

40

The Diffusion Absorption Refrigerator Operation and Performance

The depicted background shows that high performance in terms of cooling capacity and COP can be
reached when the DAR works continuously with a low number of interruptions and small variations in
the thermodynamic conditions. This is the reason why a new testing prototype was developed coupling
a 750 W-domestic magnetron to the thermal pump.

THE AMMONIA-WATER WORKING MIXTURE

The properties of ammonia and water are suitable to be used as working mixture in absorption cycles.
They are soluble in each other and easily separable. Ammonia and water are both strongly polar substances
of about the same molecular structure and mass. Both substances also show hydrogen bondings, even
though stronger in water. They are inexpensive and extensively used in the industry field. The boiling
point of ammonia is substantially lower than that of water, and this makes it possible to recover low-
temperature waste heat with a low temperature boiling. In addition, ammonia-water mixture is a non-
azeotropic mixture: temperature and composition continuously change during a constant pressure boiling.
Thorin et. al (1998) presented a review on the correlations for thermodynamic properties of ammonia-
water mixtures found in literature. The correlations can be divided into seven groups according to the way
they are derived: cubic equations of state, virial equations of state, Gibbs excess energy, corresponding
states method, perturbation theory, group contribution method and polynomial functions.
The thermodynamic properties of mixture are usually obtained combining equations for the pure
substances that the mixture consists of. The mixture of ammonia and water is non-ideal. The mixing
effects in liquid are substantial. The liquid volume of mixing decreases in the order of up to 20%. The
enthalpy of mixing can be as great as –4 kJ mol-1 (Tillner-Roth & Friend, 1998).
The correlations used here were presented by Xu and Goswami (1999). These correlations are based
on the work presented by Ziegler & Trepp (1984) and El-Sayed & Tribus (1985). Ziegler & Trepp (1984)
have modified the correlations presented by Schulz (1971). By these modifications, the correlations are
valid over a pressure range from 0.2 to 50.0 bar. Different equations are used for vapor and liquid phases.
The vapor is supposed to be an ideal mixture of real gases, while the properties of the liquid phase are
corrected by a term calculated from the Gibbs excess energy.

Gibbs Free Energy

The mathematical formulation of the thermodynamic properties of the ammonia-water mixture is based
on the Gibbs free energy method.
The fundamental equation of the Gibbs energy is given in an integral form as (Ziegler & Trepp, 1984;
Xu & Goswami, 1999):

T p T
cp
G = h0 −Ts 0 + ∫ cpdT + ∫ vdp −T ∫ dT (1)
T0 p0 T0
T

The correlations for the volume V and the heat capacity at constant pressure cp for the liquid and vapor
phases of both pure water and pure ammonia are assumed to be empirically correlated by the following
expression (Ziegler & Trepp, 1984):

41

The Diffusion Absorption Refrigerator Operation and Performance

v l = A1 + A2 p + AT
3
+ AT
4
2
(2)

RT c c c p2
vg = + c1 + 23 + 311 + 4 11 (3)
p T T T

cpl = b1 + b2T + b3T 2 (4)

p
 ∂2v 
cp g = D1 + D2T + D3T 2 −T ∫  dp (5)
0 
 ∂T 2 

The values of v and cp for the liquid state obtained from Equation (2) and Equation (4) and for vapor
state obtained from Equation (3) and Equation (5) are substituted in Equation (1) to get for Gibbs free
energy for liquid state and vapor state.
Liquid state:

B  B 
GRl = hRl,0 −TRsRl,0 + B1 (TR −TR,0 ) +  2  TR2 −TR2,0 +  3  TR3 −TR3,0 +
( ) ( )
 2   3 
 T  B 

−B1TR ln  R  − B2TR (TR −TR,0 ) −  3 TR TR2 −TR2,0 +
( ) (6)
TR,0   2 
 A  2
(A1 + AT3 R
+ AT 2
4 R ( pR )
− pR, 0 ) +  2  R
 
(
 2  p − p 2
R, 0 )

Vapor state:

D 
GRg = hRg,0 −TRsRg,0 + D1 (TR −TR,0 ) +  2  TR2 −TR2,0 +
( )
 2 
D   T 

+  3  TR3 −TR3,0 − D1TR ln  R  − D2TR (TR −TR,0 ) +
( )
 3   TR,0 
D   p 

−  3 TR TR2 −TR2,0 + TR ln  R  + C 1 (pR − pR,0 ) +
( ) (7)
 2   pr ,0 
 p T  p p 
p  TR
+C 2  R3 − 4 R3,0 + 3pR,0 4R  + C 3  R11 − 12 R11,0 + 11PR,0 12  +

T T T  T T T
R R, 0 R, 0 R R, 0 R, 0 

C   p 3 p3 TR 
+  4   R11 − 12 R11,0 + 11pR3 ,0 12 
 3  TR TR,0 TR,0 

42

The Diffusion Absorption Refrigerator Operation and Performance

where the reduced thermodynamic properties are TR = T/TB, pR = p/pB, G = G/RTB, hR = h/RTB, sR = s/R,
vR = vpB/RTB, R,TB and pB are 8.314 kJ mol-1 K-1, 100 K and 10 bar respectively and the coefficients Ai,
Bi, Ci, Di, hR,0L, hR,0g, sR,0L, sR,0g, TR,0, pR,0 are reported in Table 1.
The molar specific enthalpy h, the entropy s and the volume v of a pure component can be related to
Gibbs free energy in terms of reduced variables as:

 ∂ G 
h = −RTBTR2   R  (8)
 
 R TR  pR
∂T

 ∂G 
s = −R  R  (9)
 ∂TR 
R
p

RTB  ∂G 
v=  R  (10)
pB  ∂p 
 R T
R

Ammonia-Water Liquid Mixture

According to the analysis given by Ziegler & Trepp (1984), the liquid mixture of ammonia and water
deviates from an ideal solution mixture. This is taken into account through Gibbs excess energy. The
Gibbs excess energy for a liquid mixture is calculated from the following empirical relations:

Table 1. Coefficients of Equations (6) and (7)

Ammonia Water Ammonia Water


A1 +3.971423·10 -2
+2.748796·10 -2
C4 –3.045352·10 3
+0.000000·100
A2 –1.790557·10-5 –1.016665·10-5 D1 +3.673647·100 +4.019170·100
A3 –1.308905·10-2
–4.452025·10-3
D2 +9.989629·10-2
–5.175550·10-2
A4 +3.752836·10-3 +8.389246·10-4 D3 +3.617622·10-2 +1.951939·10-2
B1 +1.634519·101 +1.214557·101 hr,0L +4.878573·100 +2.182114·101
B2 –6.508119·10 0
–1.898065·10 0
hr,0 g
+2.646888·10 1
+6.096506·101
B3 +1.448937·100 +2.911966·10-1 sr,0L +1.644773·100 +5.733498·100
C1 –1.049377·10-2
+2.136131·10 -2
sr,0 g
+8.339026·10 0
+1.345343·101
C2 –8.288224·100 –3.169291·101 Tr,0 +3.225200·100 +5.070500·100
C3 –6.647257·102 –4.634611·104 pr,0 +2.000000·100 +3.000000·100

43

The Diffusion Absorption Refrigerator Operation and Performance

 2
GrE = x (1 − x ) F1 + F2 (2x − 1) + F3 (2x − 1)  (11)
 

where

E5 E6
F1 = E1 + E 2 pR + (E 3 + E 4 pR )TR + + (12)
TR TR2

E11 E12
F2 = E 7 + E 8 pR + (E 9 + E10 pR )TR + + (13)
TR TR 2

E15 E16
F3 = E13 + E14 pR + + (14)
TR TR 2

Coefficients for Equations (12), (13) and (14) are given in Table 2.
The excess specific enthalpy, entropy and volume for the liquid mixtures are calculated as follow:

 ∂ G E 
h = −RTBT 
E 2  R  (15)
 
 ∂TR  TR  p ,x
R

 ∂G E 
s = −R  R 
E
(16)

 ∂Tr p ,x
R

RTB  ∂G E 
vE =  R  (17)
pB  ∂p 
 R T ,x
R

Table 2. Coefficients of Equations (12), (13) and (14)

E1 –41.733398 E5 +63.608967 E9 +0.387983 E13 –3.553627


E2 +0.024140 E6 –62.490768 E10 +0.004772 E14 +0.000904
E3 +6.702285 E7 +1.761064 E11 –4.648107 E15 +24.361723
E4 –0.011475 E8 +0.008626 E12 +0.836376 E16 –20.736547

44

The Diffusion Absorption Refrigerator Operation and Performance

Therefore, enthalpy, entropy and volume of a liquid mixture become:

hm l = xhNH
l
+ (1 − x )hHl O + h E (18)
3 2

sm l = xsNH
l
+ (1 − x )sHl O + s E + s l mix (19)
3 2

s l mix = −R x ln(x ) + (1 − x ) ln(1 − x ) (20)


 

vm l = xvNH
l
+ (1 − x )vHl O + v E (21)
3 2

Ammonia-Water Vapor Mixture

The vapor phase is considered as an ideal solution, so the specific enthalpy, the entropy and the volume
are given by:

hm g = yhNH
g
+ (1 − y )hHg O (22)
3 2

sm g = shNH
g
+ (1 − y )sHg O + s g mix (23)
3 2

s g mix = −R y ln(y ) + (1 − y ) ln(1 − y ) (24)


 

vm g = yvNH
g
+ (1 − y )vHg O (25)
3 2

Bubble Point and Dew Point Temperature

The condensation and boiling processes for a binary mixture normally take place over a range of tem-
perature at a specified pressure. Hence, for the binary mixture of ammonia and water, it is essential to
determine the bubble point and the dew point temperatures to evaluate the thermodynamic properties of
the mixture corresponding to a given ammonia mass fraction at a given pressure. Xu & Goswami (1999)

45

The Diffusion Absorption Refrigerator Operation and Performance

proposed the use of the equation presented by El-Sayed & Tribus (1985) in order to determine the bubble
point and the dew point temperatures, instead to solve the iterative equation of the component fugacity.

i
7 10    p 
Tb = Tc − ∑ C i + ∑ C ij x j  ln  c  (26)
i =1 
    p 
j =1
 

i

6 4   p  
  ln  c  
j
Td = Tc − ∑ ai + ∑ Aij  ln (1.0001 − x )   p   (27)
  
i =1 
 j =1    

where

4
Tc = Tcw − ∑ ai x i (28)
i =1

 8 
pc = pcw exp ∑ bi x j  (29)
 i =1 

with p expressed in [psia] and T in [°F]. Coefficients for Equations (26), (27), (28) and (29) are given
in Table 3.

Patek and Klomfar Correlations


Patek & Klomfar (1995) presented a set of five equations describing the vapor/liquid equilibrium prop-
erties of the ammonia-water mixture necessary for absorption cycle design. These correlations were
obtained by making use of thirteen papers as source of vapor/liquid equilibrium data for water-ammonia
mixtures. The set comprises the dependences needed for the construction of an enthalpy-concentration
diagram; in particular: T(p, x), T(p, y), y(p, x), hl(T, x) and hv(T, y). These equations are as an example
of fast approximation functions, because they avoid iterative calculations.
The bubble point temperature is calculated with Equation (30). The parameters needed for this cal-
culation are given in Table 4.

n
  p  i
T (p, x ) = T0 ∑ ai (1 − x )
mi
 ln  0  (30)
  p 
i   

The dew point temperature is calculated from Equation (31). The parameters used in this equation
are reported in Table 5.

46

The Diffusion Absorption Refrigerator Operation and Performance

Table 3. Coefficients of Equations (26), (27), (28) and (29)

ai a1 +205.8889 a2 +280.930556 a3 –317.0138889 a4 +263.194444


Ai A1 +153.170553460 A2 –11.7705687461 A3 –1.78126355957 A4 +0.647385455059
A5 –0.07199507519 A6 +0.002854239508
Ai,j A1,1 +194.793913463 A1,2 +74.236124188 A1,3 +9.84103819552 A1,4 +0.436843852745
A2,1 –74.3508283362 A2,2 –33.2941879809 A2,3 –4.78866918581 A2,4 –0.225416733476
A3,1 +13.0175447367 A3,2 +6.1586564117 A3,3 +0.789740337141 A3,4 +0.032151083496
A4,1 –0.90857587517 A4,2 –0.356752691147 A4,3 +0.023806727550 A4,4 +0.00495593934
A5,1 –0.000718635741 A5,2 –0.025102638353 A5,3 –0.019166461330 A5,4 –0.001701425387
A6,1 +0.00195441703 A6,2 +0.002805333499 A6,3 +0.001389943656 A6,4 +0.000116422612
bi b1 +0.368105523897 b2 –3.6679548875 b3 +46.6000470809 b4 –262.921061996
b5 +732.99536936 b6 –1076.0613489 b7 +797.948078048 b8 –235.903904222
Ci C1 +153.634521459 C2 –13.0305543892 C3 –1.14845282991 C4 +0.550358094447
C5 –0.075345014843 C6 +0.004811166627 C7 –0.000120433757
Ci,j C1,1 –462.460321366 C1,2 +23739.9986309 C1,3 –194504.35292 C1,4 +639383.528867
C1,5 –523748.057636 C1,6 –2328271.47551 C1,7 +7562418.53499 C1,8 –9668295.89504
C1,9 +5922081.87086 C1,10 –1432405.52125 C2,1 +421.443122208 C2,2 –14560.354925
C2,3 +53051.4495633 C2,4 +382763.793582 C2,5 –3583589.86875 C2,6 +12243265.3815
C2,7 –22307970.0156 C2,8 +22896656.8499 C2,9 –12483324.8091 C2,10 +2813311.71633
C3,1 –248.783804168 C3,2 +4807.07241098 C3,3 +13565.1003309 C3,4 –466407.780832
C3,5 +2827083.44764 C3,6 –8469715.15799 C3,7 +14459588.8962 C3,8 –14281087.5331
C3,9 +7596403.59678 C3,10 -1684002.64482 C4,1 +126.965580728 C4,2 –2090.45270574
C4,3 +1993.17101166 C4,4 +100706.510396 C4,5 –687388.808612 C4,6 +2132412.46959
C4,7 –3699199.65914 C4,8 +3688365.22546 C4,9 –1975122.39296 C4,10 +440201.446068
C5,1 –33.5343446156 C5,2 +601.878586689 C5,3 –3064.82070658 C5,4 +71.7954752052
C5,5 +51780.666659 C5,6 –209714.899856 C5,7 +405011.985355 C5,8 –428310.461566
C5,9 +238153.698326 C5,10 –54497.0973336 C6,1 +3.97454953787 C6,2 –77.026846469
C6,3 +541.19105807 C6,4 –1696.60270972 C6,5 +1713.45942707 C6,6 +4019.01019872
C6,7 –14844.7928004 C6,8 +19481.0094551 C6,9 –12107.0794501 C6,10 +2966.92804386
C7,1 –0.170806170177 C7,2 +3.48182859299 C7,3 –27.7957587743 C7,4 +113.762064546
C7,5 –258.750496922 C7,6 +311.002585218 C7,7 –123.917993454 C7,8 –123.480627492
C7,9 +154.375042114 C7,10 – 48.50838287

n
mi   p  i
T (p, y ) = T0 ∑ ai (1 − y ) 4 ln  0  (31)
  p 
i   

The vapor phase composition is calculated using Equation (32) and the parameters are given in Table 6.



m
  p  i ni 
y (p, x ) = 1 − exp ln (1 − x ) ∑ ai  ln  0  x 3  (32)
   p  

i    

47

The Diffusion Absorption Refrigerator Operation and Performance

Table 4. Parameters of Equation (30)

i mi ni ai i mi ni ai
1 0 0 +0.322302⋅10 1
8 1 2 +0.106154⋅10-1
2 0 1 –0.384206⋅100 9 2 3 –0.533589⋅10-3
3 0 2 +0.460965⋅10 -1
10 4 0 +0.785041⋅101
4 0 3 –0.378945⋅10-2 11 5 0 –0.115941⋅102
5 0 4 +0.135610⋅10 -3
12 5 1 –0.523150⋅10-1
6 1 0 +0.487755⋅100 13 6 0 +0.489596⋅101
7 1 1 –0.120108⋅100 14 13 1 +0.421059⋅10-1
T0 = 100 K; p0 = 20 bar

Table 5. Parameters of Equation (31)

i mi ni ai i mi ni ai
1 0 0 +0.324004⋅10 1
10 3 0 –0.201780⋅102
2 0 1 –0.395920⋅100 11 3 1 +0.110834⋅101
3 0 2 +0.435624⋅10 -1
12 4 0 +0.145399⋅102
4 0 3 –0.218943⋅10-2 13 4 2 +0.644312⋅100
5 1 0 –0.143526⋅101 14 5 0 –0.221246⋅101
6 1 1 +0.105256⋅10 1
15 5 2 –0.756266⋅100
7 1 2 –0.719281⋅10-1 16 6 0 –0.135529⋅101
8 2 0 +0.122362⋅102 17 7 2 +0.183541⋅100
9 2 1 –0.224368⋅10 1

T0 = 100 K; p0 = 20 bar

Table 6. Parameters of Equation (32)

i mi ni ai i mi ni ai
1 0 0 +1.98022017⋅101 8 3 2 –3.42198402⋅103
2 0 1 –1.18092669⋅101 9 4 3 +1.19403127⋅104
3 0 6 +2.77479980⋅101 10 5 4 –2.45413777⋅104
4 0 7 –2.88634277⋅101 11 6 5 +2.91591865⋅104
5 1 0 –5.91616608⋅10 1
12 7 6 –1.84782290⋅104
6 2 1 +5.78091305⋅102 13 7 7 +2.38419434⋅101
7 2 2 –6.21736743⋅100 14 8 7 +4.80310617⋅103
p0 = 20 bar

48

The Diffusion Absorption Refrigerator Operation and Performance

The saturated liquid phase enthalpy is given by Equation (33). Table 7 shows the parameters of this
equation.

m
T  i n
hl (T , x ) = h0 ∑ ai  − 1 x i (33)
i T 0 

The saturated vapor phase enthalpy is given from Equation (34). Table 8 shows the parameters.

m
 T  ni 4
i

hv (T , y ) = h0 ∑ ai 1 −  x (34)
i  T 0 

In the above equations, T [K] is the temperature, p [bar] is the pressure, x and y are the molar concen-
trations of ammonia respectively in the liquid vapor mixtures, hl [kJ kg-1] and hv [kJ kg-1] represent the
enthalpies respectively of the saturated liquid and of the saturated vapor mixtures. Equations (30) and
(31) give an absolute error of less than 1.5 K, while Equations (32) and (33) of less than 2.5 kJ kg-1 for
temperatures between –80 and +180 °C and pressures between 0.2 and 20 bar. Outside of these ranges,
relationships do not guarantee the correctness of the results within the accuracy range indicated.

THE DAR MODEL

Referring to the Figure 1, the analytical thermodynamic model presented here is based on the following
assumptions:

• Pressure drops along the pipes are negligible;


• Hydrostatic pressures are negligible;

Table 7. Parameters of Equation (33)

i mi ni ai i mi ni ai
1 0 1 –0.761080⋅101 9 2 1 +0.284179⋅101
2 0 4 +0.256905⋅102 10 3 3 +0.741609⋅101
3 0 8 –0.247092⋅103 11 5 3 +0.891844⋅103
4 0 9 +0.325952⋅103 12 5 4 –0.161309⋅104
5 0 12 –0.158854⋅10 3
13 5 5 +0.622106⋅103
6 0 14 +0.619084⋅102 14 6 2 –0.207588⋅103
7 1 0 +0.114314⋅102 15 6 4 –0.687393⋅101
8 1 1 +0.118157⋅10 1
16 8 0 +0.350716⋅101
h0 = 100 kJ kg-1; T0 = 273.16 K

49

The Diffusion Absorption Refrigerator Operation and Performance

Table 8. Parameters of Equation (34)

i mi ni ai i mi ni ai
1 0 0 +0.128827⋅10 1
10 1 3 +0.164508⋅102
2 1 0 +0.125247⋅100 11 2 3 –0.936849⋅101
3 2 0 –0.208748⋅10 1
12 0 4 +0.842254⋅101
4 3 0 +0.217696⋅101 13 1 4 –0.858807⋅101
5 0 2 +0.235687⋅10 1
14 0 5 –0.277049⋅101
6 1 2 –0.886987⋅101 15 4 6 –0.961248⋅100
7 2 2 +0.102635⋅102 16 2 7 +0.988009⋅100
8 3 2 –0.237440⋅10 1
17 1 10 +0.308482⋅100
9 0 3 –0.670155⋅101
h0 = 1000 kJ kg-1; T0 = 324 K

• The liquid mixture (state 3) and the vapor bubbles (state 4) exit the riser tube and leave the thermal
pump at the same temperature (i.e., T3 = T4);
• The weak mixture (state 3+5) is not re-heated in the thermal pump before entering the liquid HX
because it passes through a separated pipe;
• The thermal pump is not completely thermally insulated, so a part Q H ,d of the supplied heat, as a
function of the variable heat exchanger efficiency, dissipates towards the ambient;
• In the state 6 the whole refrigerant flow rate is condensed (i.e., no flow occurs in the gas bypass)
and becomes a saturated liquid (i.e., no sub-cooling is realized in the condenser);
• The refrigerant and the strong mixture leave respectively the condenser and the reservoir at the
same temperature, as the cooling medium is ambient air (i.e., T7 = T11);
• The refrigerant and inert gas mixing at the entrance of the evaporator is adiabatic;
• The refrigerant leaves the gas HX in state 10 as a saturated vapor;
• No absorption takes place inside the reservoir.

To extend the model applicability to most of the real situations, no assumptions are made on the
pureness of ammonia vapor in state 6 (i.e., the mixture does contain water vapor).
In order to determine the state of the working mixtures at each section of the cycle, the relations
among temperature, pressure, concentration and enthalpy described by Pátek & Klomfar (1995) are used.
In Table 9 the state properties of the working fluids are resumed.
The input parameters to the model are:

• The temperature of the strong mixture entering the thermal pump, T1;
• The temperature reached by the strong mixture after the heating process, T2;
• The temperature of the weak mixture and the vapor leaving the thermal pump, T3 = T4;
• The temperature of the refrigerant leaving the rectifier, T6;
• The temperature of the refrigerant leaving the condenser, T7;
• The temperature of the refrigerant and of the auxiliary gas leaving the gas HX, T10 = T10,ig;
• The operating pressure, pC;

50

The Diffusion Absorption Refrigerator Operation and Performance

Table 9. State properties of working fluids

State Variable Value


1 p
p1 = pC

h h1 = hl (T1, x 1 )

x T1 = T (pa , x 1 ) ⇒ x 1 (by iterations)

2 p
p2 = pC

x x 2 = x 12
3 p
p3 = pC

h h3 = hl (T3, x 3 )

x T3 = T (pC , x 3 ) ⇒ x 3 (by iterations)

4 p
p4 = pC

h h4 = hv (T4 , y 4 )

y T4 = T (pC , y 4 ) ⇒ y 4 (by iterations)

5 p
p5 = pC

T T5 = T (pC , x 5 )

h h5 = hl (T5, x 5 )

x y 4 = y (pC , x 5 ) ⇒ x 5 (by iterations)

6 p
p6 = pC

h h6 = hv (T6, y6 )

x T6 = T (pC , y6 ) ⇒ y6 (by iterations)

continued on following page

51

The Diffusion Absorption Refrigerator Operation and Performance

Table 9. Continued

State Variable Value


7 p
p7 = pC

h h7 = hl (T7 , x 7 )

x
x 7 = y6
8 p
T8 = T (p8 , x 8 ) ⇒ p8
(by iterations)

T h8 = hl (T8 , x 8 ) ⇒ T8
(by iterations)

x
x8 = x3
8,ig p
p8,ig = pC

h h8,ig = cp,ig ⋅T8,ig


9 p
p9 = pe

T T9 = T (p9, x 9 )

h h9 = hl (T9, x 9 )

x
x 9 = y6
10 p
T10 = T (p10, y10 ) ⇒ p10 = pe
(by iterations)

h h10 = hv (T10, y10 )

y
y10 = y6
10,ig p
p10,ig = pC − pe

h h10,ig = cp,ig ⋅T10,ig

continued on following page

52

The Diffusion Absorption Refrigerator Operation and Performance

Table 9. Continued

State Variable Value


11 p
m r x r RNH + (1 − x r ) RH O 
pa =  3 2 
pC
m ig Rig + m r x r RNH + (1 − x r ) RH O 

 3 2 

h h11 = hl (T11, x 11 )

x
x 11 = x 1

• The heat power supplied to the thermal pump, Q H ;


• The temperature of the air inside the refrigerated chamber, Tair;
• The refrigerated air mass flow rate, m air .

The condition T10 = T10,ig allows to determine the temperature of the refrigerant (as a saturated liquid)
and the auxiliary gas inside the evaporator, T9 = T9,ig.
A large number of input variables is necessary to describe the real working operations of the DAR
cycle. Most of previous models use fewer input values and, despite the easiness of use, results remain
affected by too many assumptions. A great number of input variables, on the contrary, allows to run the
model here proposed according with the real DAR cycle operation and to perform better predictions.
In Figure 2, detailed schemes of the evaporator and of the gas HX are shown: hatched contours in-
dicate the presence of the thermal insulation.
For all the heat exchangers shown in Figure 1, considering positive the heat supplied to the working
mixture and referring to the scheme of the evaporator of Figure 2, the total mass, the ammonia mass and
the energy balances can be written as:
Generator:

m = m + m
 1 3 4
m x = m x + m y (35-37)
 1 1 3 3 4 4
 
QH − QH ,d = m 3 h3 + m 4 h4 − m 1h1
 T3 =T2 T4 =T2

Rectifier:

m 4 = m 5 + m 6

 y − x5
m 6 = 4 m (38-40)
 y6 − x 5 4
 
QI ,r = m 5h5 + m 6h6 − m 4h4

53

The Diffusion Absorption Refrigerator Operation and Performance

Figure 2. Scheme of the evaporator and of the gas HX

Condenser:

Q I ,c = m 6 (h7 − h6 ) (41)

Evaporator + Gas HX:

m 7 = m 9 = m 10

Q L = m 9 (h10 − h7 ) + m ig (h10,ig − h8,ig ) if Q L ≤ Q L,max (42-44)

Q L = Q L,max if Q L > Q L,max

Liquid HX:

m = m
 11 1

m 8 = m 3 + m 5 (45-47)

m 3h3 + m 5h5 − m 8h8 = m 1 (h1 − h11 )


54

The Diffusion Absorption Refrigerator Operation and Performance

Absorber:

Q = m h − m h − m h +
 I ,a 11 11 10 10 8 8

 +mig h8,ig − mig h10,ig if Q L ≤ Q L,max
 
Q = m h − m h − m h + (48-49)
 I ,a 111 11 10 10 8 8

  
+m ig h8,ig − m ig h10,ig + if QL > QL,max

 + (Q L − Q L,max )

where

m 1 = m 2 = m 11 = m r (50)

m 8 = m p (51)

m 6 = m 7 = m 9 = m 10 = m refr (52)

x 1 = x 2 = x 11 = x r (53)

x3 = x8 = xp (54)

y6 = x 7 = x 9 = y10 = yrefr (55)

pC − pe yrefr RNH 3 + (1 − yrefr ) RH 2O


m ig = m refr (56)
pe Rig

(
Q H ,d = m 3 h3
T3 <T2
− h3
T3 =T2 ) + m (h
4 4 T <T
4 2
− h4
T4 =T2 ) (57)

Q L,max = C p,min ⋅ ∆Tmax (58)

55

The Diffusion Absorption Refrigerator Operation and Performance

The terms hi and hi indicate that enthalpy is calculated respectively at a temperature Tj =


Tj =Tk Tj <Tk

Tk and at Tj < Tk. In Equation (56) the auxiliary gas mass flow rate is estimated imposing the Dalton’s
law in the mixing zone (9) of the evaporator. The calculation of Q H ,d (as the heat dissipated towards the
ambient in the thermal pump), using Equation (57), considers the temperature reduction of the working
mixture changing its state from 2 to 3 and 4. This is a measure of the imperfect insulation of the thermal
pump. In Equation (58),

C p,min = min (m air ⋅ cp,air ; m refr ⋅ cp,refr )

is the minimum value between heat capacities of the refrigerated air and the refrigerant in the evaporator,
and ΔTmax = Tair – T9 is the maximum temperature difference.
The performance of the DAR cycle can be calculated in terms of COP and mass flow rates ratio f, in
order to evaluate the strong mixture mass flow rate needed to produce 1 kg s-1 of refrigerant mass flow
rate in the evaporator:

Q
COP =  L (59)
QH

m r
f = (60)
m refr

THE INFLUENCE OF THE THERMAL PUMP ON THE DAR PERFORMANCE

As stated in the introduction, the thermal pump performance plays a primary role on the DAR opera-
tion. The existence of a minimum heat power allowing the separation of a sufficient refrigerant mass
flow rate in order to activate the two-phase flow inside the riser tube was experimented by Chen et al.
(1996). Below the minimum power, the refrigerant vapor phase cannot drag the weak mixture liquid
phase; thus, the absorption process does not take place as in the absorber no weak mixture is present.
Exceeding this value, the cycle starts and an increase of the heat power leads to a growth in refriger-
ant and weak mixture mass flow rates evolved in the thermal pump; the cooling capacity and the COP
values raise consequently. Anyway, when the heat power becomes too high, the two-phase flow in the
riser tube turns out in a regime where the vapor phase reduces its potential to drag the liquid phase and
the COP starts to reduce (Delano, 1998).
Referring to the six two-phase flow regimes in vertical tubes recognized by Thome (2010) and showed
in Figure 3, in the bubbly flow the gas phase is dispersed in the form of small sized spherical bubbles in
the liquid phase. Increasing the gas-to-liquid mass ratio (also referred to as void fraction), the bubbles
grow in a bullet shape and reaching the diameter of the riser tube. The so-called slug flow starts.

56

The Diffusion Absorption Refrigerator Operation and Performance

Figure 3. Two-phase flow regimes in vertical tube as recognized by Thome (2010)

When the velocity of the fluids increases, the flow becomes unstable and the phases go up and down
across the tube in an oscillatory way, but with a net upward direction. This behavior is generated by the
quite parity of the gravity and the shear forces acting in opposition on the bubbles liquid contours. The
regime is called churn flow.
The next annular flow occurs at higher fluids velocities: the liquid phase concentrates in the central
axis of the riser tube, surrounded by the gas phase. A further increase of the flow rate causes an entrap-
ment of some gas bubbles in the central liquid flow and the annular gas phase decrease in its thickness
(wispy annular flow).
At highest flow rates, the gas phase dominates and the liquid phase appears as spherical bubbles in
the fluids core (mist flow).
Chisholm (1983) shows experimentally that in vertical tubes the slug flow is the thermal pump re-
gime with the best efficiency in terms of refrigerant mass flow rate produced per unit of strong mixture
mass flow rate processed. In this regime the riser tube and the bubbles of the vapor phase have a similar
diameter. These so-called Taylor bubbles have a bullet shape and each bubble is separated from the oth-
ers and from the riser tube wall by a thin layer of liquid phase.
Therefore, as recognized by Benhmidene et al. (2014), the diameter of the riser tube, together with
its length and the height of the liquid head in the reservoir, plays a fundamental role in the design of
the thermal pump.

57

The Diffusion Absorption Refrigerator Operation and Performance

THE THERMAL PUMP ANALYTICAL DESCRIPTION

Adopting the thermal pump schematization of Figure 4, together with the assumptions previously made,
the proposed model of the device works with the following ones:

• The frictional losses at the entrance to the riser tube are negligible;
• The change in momentum throughout the riser tube is negligible.

In order to determine the state of the working mixture at each section of the thermal pump, the rela-
tions in M. Conde Engineering (2006) were added to the Pátek & Klomfar (1995) ones. Referring to
Figure 5, the pressure in the point 11 can be calculated as:

2
w11
p11 = pC + ρl gH − ρl (61)
2

The momentum equation in the direction of the flow and the mass balance between sections 1 and
11 are, respectively:

Figure 4. Scheme of the thermal pump

58

The Diffusion Absorption Refrigerator Operation and Performance

Figure 5. Comparison between the present and the Zohar et al. (2009) models in terms of mixture con-
centrations vs. T2 at constant heat power supplied to the thermal pump

p1 = p11 − ρl w11 (w1 − w11 ) (62)

ρl A11w11 = ρl A1w1 (63)

where A11 = A1 = πdi2/4; so, w11 = w1 and p11 = p1. The momentum equation between sections 1 and 2 is:

p2 = p1 − ρhw1 (w2 − w1 ) (64)

After section 2, the strong mixture is heated up by Q H and starts boiling. Here, ρh, the homogeneous
density of the two-phase flow, has to be used because the liquid and the vapor velocities are similar and
the latter is limited in mass. Imposing the mass balance between sections 1 and 2, it results:

ρldi2w1
ρh = (65)
dc2w2

59

The Diffusion Absorption Refrigerator Operation and Performance

Calling jv, jl and j respectively the superficial velocities of the vapor and of the liquid phases and the
total superficial velocity, the following expressions can be written:

Vv V V +Vv
jv = , jl = l and j = jl + jv = l (66)
A2 A2 A2

Thus:

Vl A
w1 = w11 = = jl 2 (67)
A1 A1

Vv +Vl
w2 = = j (68)
A2

Defining the void fraction ε as

Av A − Al
ε= = 2 (69)
A2 A2

and referring to the control volume between 2 and the exit of the riser tube, the pressure at section 2
can be calculated as:

(ρ j + ρv jv )  L 
2

  + ρl Lg (1 − ε)
l l
p2 = pC + fTP (70)
ρTP dc 

where fTP and ρTP = ρvε + ρl(1 – ε) are, respectively, the two-phase friction factor and the two-phase
density of the mixture inside the riser tube. In Equation (70) the frictional pressure drop term should
be calculated using ρTP instead of ρh since now there is slip between the two phases and the flow cannot
be considered homogeneous. Actually, after section 2 the vapor phase is completely developed and the
buoyancy phenomena let the vapor velocity increase more than the liquid one.
Combining Equations (61), (62), (64) and (70), a general equation for the submergence ratio H/L
can be obtained as:

d   d  
2 2
4
 d   
jl2  c  jl ρh    j − jl   
c c

(ρl jl + ρv jv )
2
 di    di  
H  di   
= fTP + + + (1 − ε) (71)
L 2gdc ρl ρTP 2gL ρl gL

60

The Diffusion Absorption Refrigerator Operation and Performance

To evaluate the friction factor, Bettie and Whalley (1982) recommended the use of the Colebrook
equation based on the hypothesis of homogeneous two-phase flow:

 
1  ε 9.35 
= 3.48 − 4 log10 2 R +  (72)
*
fTP  di *
ReTP fTP 
 

where f*TP = fTP/4, εR is the riser tube surface roughness and the two-phase Reynolds number is defined as:

ReTP =
(ρ j
v v
+ ρl jl )di
(73)
µTP

with

µTP = εh µv + µv (1 − εh )(1 + 2.5εh ) (74)

and

X
εh = (75)
ρ
X + l (1 − X )
ρv

In order to evaluate the void fraction, the drift flux model (Zuber & Findlay, 1965) provides a
method to account for the effects of the local relative velocity between the phases as well as the effects
of non-uniform phase velocity and distributions. The model was improved by Chexal et al. (1997) and
expresses the average vapor void fraction as a function of the superficial velocities of the two phases, of
the distribution parameter C0 and of the drift velocity wvj = wv – j. In particular:

jv
ε= (76)
C 0 ( jl + jv ) + wvj

Many authors provide empirical correlations to evaluate C0 and wvj. Since the thermal pump operates
more efficiently in the slug regime, White (2001) compared her experimental results with those obtained
using the correlations in vertical tubes and comparable conditions provided by Nicklin et al. (1962),
Nicklin & Davidson (1962), de Chachard & Delhaye (1996), Reinemann et al. (1990) and Chexal et al.
(1992). White found that the de Chachard & Delhaye’s (1996) correlation fits better the experimental
results. In particular, the following values and expressions have to be used:

C 0 = 1.2 (77)

61

The Diffusion Absorption Refrigerator Operation and Performance

 0.01N f   3.37−Bo 
  
wvj = 0.345 1 − e 0.345  1 − e m  gdi

(78)
   

where

ρl (ρl − ρv ) gdi3
N = 2
f
(79)
µl2

Bo =
(ρ − ρ ) gd
l v i
2

(80)
σ

m = 10 if N f > 250


 −0.35
m = 69N f if 18 ≤ N f ≤ 250 (81)

m = 25 if N f < 18

The equations here reported were included in the model described in The DAR model section in
order to calculate the effective flow rates exiting the thermal pump ( m 3 and m 4 in Figure 2), respec-
tively from jl and jv calculated by Equation (71), taking into account the geometrical thermal pump pa-
rameters.
In particular, the calculation of m 3 and m 4 cannot be performed measuring the temperature T1 (the
section 1 is where the strong mixture enters the thermal pump): due to the configuration of the liquid
HX, T1 measure cannot be direct; that must be derived from the measure of the temperature on the sur-
face of the external tube and no modelling of the heat transfer across the region occupied by the weak
mixture can be considered sufficiently affordable. At the operating conditions of Table 10, the differ-
ences between the temperature measured on the surface of the external tube of the liquid HX and T1,
calculated using the thermal pump description, are always in the range of –3 K and –1 K. This shows
that the thermal pump description gives a better evaluation of the effective T1 of the strong mixture inside
the liquid HX.
In the developed model, the calculation of m 3 and m 4 comes out from the mathematical description
of the thermal pump riser tube flow with its real geometry. In Table 11, the weak mixture mass flow
rate m 8 = m 3 + m 5 , the COP values and the differences between the experimental and calculated results
are reported in correspondence to the increasing values of the supplied heat power Q at the operating
H
conditions of Table 10. The maximum errors made between measuring T1 and using the thermal pump
description are respectively 1.41% and 3.65% in terms of m 8 , and 2.99% and 5.93% in terms of COP.
Experimental values were compared to calculated ones not affected by the thermocouple errors. Then,
the model enhanced with the thermal pump description shows to be more accurate.

62

The Diffusion Absorption Refrigerator Operation and Performance

Table 10. DAR steady state operating conditions depending on heat power supplied

Q H T1 T2 T4 T6 T7 T10 Tcb pC

[W] [K] [K] [K] [K] [K] [K] [K] [bar]

120 357 441 439 409 300 298 295 15.0

200 359 444 441 410 300 299 295 15.3

250 362 446 442 413 302 299 295 15.4

300 365 447 444 414 302 298 295 15.5

325 365 448 444 415 303 298 295 15.6

350 368 449 445 416 304 298 295 15.7

375 369 450 446 417 305 298 295 15.8

400 371 451 447 417 305 298 295 15.8

425 371 452 447 417 305 298 295 15.9

450 371 453 448 418 306 298 295 16.0

500 375 454 449 418 306 298 295 16.1

525 377 455 450 418 306 298 295 16.2

550 379 456 451 418 306 298 295 16.3

COMPARISON WITH ZOHAR ET AL. (2009) MODEL

The analytical model was set up with Labview v.8.6 (National Instruments Academic Site License, 2009).
In order to test the model consistency, a comparison with the results provided by Zohar et al. (2009)
was performed. The differences between the assumptions of the model here developed and the Zohar’s
one are reported in Table 12.
To guarantee a correct comparison, the same operating conditions and working fluids have been
adopted:

• Fluids: ammonia and water;


• Auxiliary inert gas: hydrogen;
• Operating pressure: pC = 15.5 bar;
• Maximum evaporator cooling capacity: Q L,max ≥ Q L ;
• Temperature of the working mixture leaving the thermal pump: T3 = T4 = T2 – 5 °C;
• Condensing temperature: T7 = 40 °C;
• Evaporator inlet temperature: T9 = –5 °C.

The comparisons between weak and strong mixture concentrations are depicted in Figure 5 as tem-
perature T2 goes from 423 to 433 K; the COP variations are shown in Figure 6.

63

The Diffusion Absorption Refrigerator Operation and Performance

Table 11. Comparison between m 8 and COP calculated with the original (O) and the model with the
thermal pump description (TPD), varying Q H at the operating conditions of Table 10

Q H m 8 ×10 5 Err (m 8 ) COP Err (COP )

meas. eval. O TPD O TPD eval. O TPD O TPD


[W] [kg s-1] [kg s-1] [kg s-1] [%] [%] [-] [-] [-] [%] [%]

120 2.94 2.86 2.95 +2.82 –0.32 0.236 0.242 0.234 –2.48 +0.86

200 4.78 4.82 4.86 –0.81 –1.62 0.257 0.259 0.263 –0.77 –2.28

250 6.09 6.03 6.15 +1.05 –0.92 0.254 0.270 0.259 –5.93 –1.93

300 7.60 7.33 7.55 +3.65 +0.63 0.258 0.260 0.256 –0.77 +0.78

325 8.17 8.07 8.07 +1.24 +1.24 0.239 0.233 0.233 +2.58 +2.58

350 9.00 8.90 8.97 +1.11 +0.33 0.175 0.179 0.173 –2.24 +1.16

375 9.63 9.54 9.56 +0.98 +0.77 0.143 0.142 0.141 +0.70 +1.42

400 10.53 10.40 10.50 +1.24 +0.25 0.125 0.126 0.125 –0.79 +0.00

425 10.73 11.00 10.70 –2.52 +0.27 0.112 0.115 0.112 –2.61 +0.00

450 10.94 11.10 10.90 –1.46 –0.33 0.098 0.100 0.099 –2.00 –1.01

500 12.90 13.00 13.10 –0.65 –1.41 0.065 0.067 0.067 –2.99 –2.99

525 14.20 14.00 14.30 +1.16 –0.96 0.052 0.053 0.052 –1.98 –0.10

550 15.20 14.80 15.30 +2.83 –0.53 0.051 0.052 0.052 –2.91 –2.91

Table 12. Differences between the assumptions of the present model and the Zohar’s one

Issue Zohar Model Present Model Advantages


Thermal pump’s riser Assumption of a fixed reduction in T2 and T3 = T4 as input variables Flexible and better description
tube the temperature of the fluid along of the real operation of the cycle
the tube
Temperature of the Input value The evaporator and the gas HX None
gas mixture in the are treated as a single control
evaporator volume (Figure 2): T10 = T10,ig is
the input value
Thermal insulation of No heat losses towards the ambient More realistic conditions
the thermal pump Heat losses ( Q H ,d ) are taken into
account with some assumptions of
heat transfer towards the ambient

Fluid composition at the Pure ammonia Mixture of water and ammonia More realistic conditions
rectifier outlet
Weak mixture entering Equilibrium state Not necessarily in the equilibrium More flexible model
the absorber (state 8) state
(Zohar et al., 2009)

64

The Diffusion Absorption Refrigerator Operation and Performance

Figure 6. Comparison between the present and the Zohar et al. (2009) models in terms of COP vs. T2 at
constant heat power supplied to the thermal pump

The maximum differences in results between the Zohar et al. (2009) model and the present one are
8.5% in terms of xp, 6.1% in terms of xr and 2.0% in terms of COP. They are due to the pure ammonia
as refrigerant (i.e., yrefr = 1) assumption in the model of Zohar et al. (2009).
The model here presented, actually, shows an ammonia molar concentration at the vapor refrigerant
exit from the rectifier different from 1 (yrefr = 0.985).
Considering pure ammonia as refrigerant, therefore, leads to no significant errors when the thermal
pump temperatures are approximately up to 445 K; errors increase with T2, due to a greater water evapo-
ration rate in the thermal pump.

EXPERIMENTAL SETUP

As anticipated in the Introduction section, a purposely designed microwaves activated thermal pump was
coupled to the F400 DAR model produced by Dometic (2004). A scheme of the prototype is showed in
Figure 7, where dc is the diameter of the riser tube, di is the diameter of the liquid HX inner tube, L is
the length of the riser tube, H is the height of the absorber reservoir liquid head, de is the diameter of
the evaporator pipe and Ae,air is the heat transfer surface of the evaporator.
Referring to Figure 7, the external heat power Q H is supplied directly to the working mixture in the
thermal pump reservoir and the weak mixture leaving the outer tube of the riser is not reheated before

65

The Diffusion Absorption Refrigerator Operation and Performance

Figure 7. Schematic diagram of the DAR circuit coupled to the microwaves activated thermal pump
Adapted from Dometic (2004)

entering the liquid HX because it flows in a separated insulated pipe. Zohar et al. (2008) experimented
that, for a fixed heat power (Q H = 160 W), the thermal pump configuration of Figure 8(b) allows to
generate the maximum refrigerant mass flow rate and to realize the highest COP values. In particular,
in the configuration of Figure 8(a), the riser tube is completely separated from the weak mixture return
pipe, in the configuration of Figure 8(b) the separation is only in the section of heat power supplying
and the configuration of Figure 8(c) is a shell and tube heat exchanger. Due to the better performance,
the Figure 8(b) configuration was chosen for the prototype, replacing the original Figure 8(c) configura-
tion of the F400 system.
The microwaves activated thermal pump (Figure 9) consists of a metal waveguide coupled on a steel
flange. The magnetron is fixed on the waveguide and the flange is screwed to the thermal pump stainless
steel reservoir. The thermal pump is instrumented with a 0-40 bar manometer and a K-type thermocouple
(accuracy: ±1 K). The temperatures in the other sections of the circuit are measured using another K-type
thermocouple in contact with the metallic surfaces of the pipes.

66

The Diffusion Absorption Refrigerator Operation and Performance

Figure 8. Different thermal pump configurations: (a) completely separated, (b) partially separated, (c)
shell and tube
(Zohar et al., 2008)

A plate is fixed between the reservoir and the flange in order to separate the microwaves generation
zone (at atmospheric pressure) from the refrigerator circuit (at working pressure). The plate is made of
sintered alumina, a material with the required mechanical and dielectric behavior to resist to the stress
produced by the circuit pressure and to be transparent to the microwaves. In Figure 10 is shown the
built prototype.
The purposely modified DAR circuit was filled with a 70.1% H2O, 27.9% NH3 and 2.0% Na2CrO4
(a corrosion inhibitor salt) mixture and pressurized with helium up to 11.5 bar, following a rigid filling
protocol and using a system that lets the operator to remove efficiently the air trapped in the circuit, with
no performance penalization in the DAR operation.

VALIDATION AT STEADY STATE

The measured parameters were: H/L, T2, T3, T6, T7, T10, Tcb, pC, Q H (also used as model inputs) and T8
(a control quantity). Tcb is the temperature in the evaporator cold box of the refrigerator. As a direct

67

The Diffusion Absorption Refrigerator Operation and Performance

measure of the flow rate is not possible due to its very small values, the measurement of T8 becomes
necessary to evaluate the weak mixture mass flow rate m 8,eval as a function of the heat power supplied
to the thermal pump.
The validation process, resumed using a flowchart in Figure 11, followed the steps listed here:

1. Measure of T8;
2. Calculation of x8, h8, h1, h3, h5, h11, m 1 , m 3 and m 5 ;
3. Calculation of m 8,eval ;
4. Determination of the error between m 8,eval and the calculated weak mixture mass flow rate m 8,calc .

As the parameters listed below and shown in Figure 11 are or depend on measured or input values
respectively, and parameters derived by them come out from the mass and energy balances written for
the liquid HX, the difference between m 8,eval and m 8,calc can be interpreted as an estimation of the errors
made by the model. Actually:

• The m 3 (m 3 + m 5 ) and m 5 (m 3 + m 5 ) ratios depend on the geometrical configuration of the
thermal pump and on both the power Q H supplied and the power Q H ,d dissipated towards the
ambient (independent from T8);
• The ammonia concentration x8 is calculated from T3 (a measured value);
• h1 and h11 are calculated respectively from H/L (that brings to T1) and x1, and from T11 and x11.

In the previous Table 10, the steady state operating conditions reached by the prototype are listed
increasing Q H values. In Table 13 the control temperature T8, the weak mixture mass flow rate and the
COP values, respectively measured and calculated by the model, are reported.
Figure 12 shows the estimated errors of the model in terms of T8, m 8 and COP. The bar amplitude
in correspondence to each value of Q accounts for the thermocouple uncertainty. The model is af-
H

fected by a maximum absolute error of 1.33% in terms of T8, of 2.32% in terms of m 8 and of less than
5% in terms of COP, proving an adequate accuracy to predict the real DAR operating conditions and
performance.
However, the values of COP are very low. The reason is in the poor optimization of the prototype.
The original F400 system is designed to work with a power input of 120 W. The system is modified
only in its thermal pump configuration and not in the other components’ (such as evaporator, condenser,
absorber, etc.) size and layout. This leads to a reduction in the performance of the cycle.

SENSITIVITY ANALYSIS

In Figure 13, the decrease of mass flow rates ratio f with T2 is shown. This happens because the reduc-
tion of the strong mixture mass flow rate is more intense than that of the refrigerant mass flow rate; at
higher temperatures this yields to a higher quantity of produced refrigerant per mass of strong mixture.
Anyway, the reduction of the refrigerant mass flow rate leads to the COP decrease shown in Figure 14
for different temperatures of the refrigerant leaving the rectifier.

68

The Diffusion Absorption Refrigerator Operation and Performance

Figure 9. Scheme of the microwaves activated thermal pump

As expected, increasing T6 leads to a reduction of the purity of ammonia in the refrigerant and COP
increases as a consequence. Thus, a raise in the rectifier temperature causes an increase in COP, an
increase in the water content of the refrigerant, a decrease of the evaporation partial pressure and an
increase of the temperature glide in the evaporator (Grossman, DeVault, & Creswick, 1995). This hap-
pens because the latent heat of water is quite higher than that of ammonia.
The DAR cycle performance is influenced by the absorption temperature T11 as well. Assuming fixed
geometry and layout of the absorber, T11 results strongly influenced by the ambient temperature (Figure
15): higher ambient temperatures lead to lower absorption rates of the refrigerant in the absorbent. This,

69

The Diffusion Absorption Refrigerator Operation and Performance

Figure 10. The built prototype

in turn, leads to a decrease of the ammonia concentration in the strong mixture entering the thermal pump
and, therefore, to a decrease of the refrigerant mass separating from the liquid mixture that produces a
cooling effect at the evaporator. COP values decrease as well.
The heat power Q H supplied to the thermal pump, as well as the thermal pump and absorption tem-
peratures, influences the DAR cycle performance. COP actually depends on Q for three different
H
reasons:

1. The thermal insulation of the thermal pump is not perfect (so, a part of Q H is dissipated towards
the ambient);
2. A minimum value of Q H is needed to realize the two-phase slug flow regime in the riser tube of
the thermal pump, i.e. bubbles begin to form at the inner surface of the tube walls only when a
certain value of heat power supplied is reached (Jakob et al. 2008);
3. The dimensional and geometrical characteristics limit the maximum evaporator heat transfer.

70

The Diffusion Absorption Refrigerator Operation and Performance

Figure 11. The m 8,eval indirect measurement and the derivation of the model error

Table 13. Measured and calculated control temperature, evaluated and calculated weak mixture mass
flow rate and COP depending on heat power supplied at the operating conditions of Table 10

Q H T8 m 8x105 COP

Meas. Calc. Eval. Calc. Eval. Calc.

[W] [K] [K] [kg s-1] [kg s-1] [-] [-]

120 304 305 2.94 2.95 0.236 0.234

200 302 305 4.78 4.86 0.257 0.263

250 303 304 6.09 6.15 0.254 0.259

300 304 303 7.60 7.55 0.258 0.256

325 305 303 8.17 8.07 0.239 0.233

350 305 305 9.00 8.97 0.175 0.173

375 306 305 9.63 9.56 0.143 0.141

400 306 305 10.53 10.50 0.125 0.125

425 306 305 10.73 10.70 0.112 0.112

450 306 306 10.86 10.90 0.098 0.099

500 305 307 12.92 13.10 0.065 0.067

525 305 307 14.16 14.30 0.052 0.052

550 306 307 15.22 15.30 0.050 0.052

71

The Diffusion Absorption Refrigerator Operation and Performance

Figure 12. m 8 and COP errors as functions of Q H

Reasons no. 1 and no. 2 imply that the working mixture in the thermal pump need a minimum value
of heat power Q H ,d supplied to start the cycle. The reason no. 3 implies that the physical configuration
of the evaporator allows the refrigerant to exchange up to a certain Q L,max
. Therefore, when this limit is
reached, an increase Q H + ∆Q H of the heat power supplied to the thermal pump will not correspond
to an increase of energy stored in the fluid; the exceeding power ∆Q will be dissipated towards the
H
ambient through the absorber coils. Then, the COP begins to decrease.
In Figure 16, COP and Q L are depicted as functions of Q H . The heat power Q L is calculated using
Equation (9) up to Q and then using Equation (10). The left field of the graph represents the op-
L,max
eration of DAR dominated by the heat dissipation and the flow regime behaviors; the right field that by
the evaporator heat transfer limit.
These behaviors are important in the design process of a DAR cycle and circuit. The designer has to
carefully take into account the influence of the heat supplied to the generator and the physical configura-
tion of the evaporator. Consequences will be evident on the DAR performance.

72

The Diffusion Absorption Refrigerator Operation and Performance

Figure 13. Mass flow rates ratio f and mass flow rates m r , m refr as functions of the thermal pump
temperature T2

Figure 14. COP as a function of the thermal pump temperature T2 for different ammonia concentrations
in refrigerant

73

The Diffusion Absorption Refrigerator Operation and Performance

Figure 15. COP as a function of the absorption temperature T1

Figure 16. COP and Q L as functions of the heat Q H supplied to the thermal pump

74

The Diffusion Absorption Refrigerator Operation and Performance

THE PROTOTYPE TRANSIENT OPERATION

A second set of experiments carried out on the tested DAR circuit (filled with 70.1% H2O, 27.9% NH3,
2.0% Na2CrO4 and pressurized with He) clearly shows a reduction in the time needed to reach a steady
state operation. Using the original commercial system (Q H = 120 W) and starting from an ambient
temperature of 289 K, the time to reach a temperature T9 (on the evaporator surface) of 273 K was about
7.5 hours. At the same operating conditions, a heat power of Q H = 600-750 W reduces this time to about
2.0 hours. In Figure 17 and in Figure 18, the temperatures T9 and T2 trends with time are shown when
Q H = 600 W and Q H = 750 W, and the system pressure pC at 289 K is respectively 14.5 bar and 16.0
bar.
In the first case the transient time is lower of about 45 minutes than in the second one. A cause is the
imperfect insulation of the thermal pump: the Q H decrease produces the lowering of the ratio between
the heat power dissipated towards the ambient and the supplied heat power, with a benefit in the evapo-
rator exchange process that comes next.

Figure 17. Temperature versus time with Q H = 600 W

75

The Diffusion Absorption Refrigerator Operation and Performance

Figure 18. Temperature versus time with Q H = 750 W

In addition, increasing the supplied heat power, the evaporator geometrical and physical exchange
limits become more influent on the heat transfer. At Q H = 750 W, the mixture temperature T2 in the
thermal pump reservoir is higher than that at Q = 600 W. Reaching higher temperatures, the two-phase
H
flow configuration in the riser tube varies consequently. In this case a reduction of T2 may lead to the
generation of a two-phase flow regime characterized by a higher refrigerant mass flow rate per unit of
strong mixture mass flow rate in the thermal pump.
These first results suggest a use of the purposely modified DAR circuit coupled with a microwaves
activate thermal pump as a rapid cooler, suitable for domestic and commercial applications. However,
more experimental campaigns and circuit optimizations have to be planned to obtain a stronger reduction
of the transient times consistent with the hypothesized use.

76

The Diffusion Absorption Refrigerator Operation and Performance

CONCLUSION

A deep description of the Diffusion Absorption Refrigeration operation and performance was given aware
that this technology can in a next future overcome its limitations mainly due to its low efficiencies. The
extended description of the properties of the ammonia-water working mixture was also given as it is
fundamental to find its exact properties at each stage of the DAR cycle and to simulate it.
An advanced thermodynamic analytical model of the system developed by the authors was then
described, with no assumptions made on the composition of the refrigerant leaving the rectifier. The
model uses the mass flow rates and energy balances imposed between the enter and exit section of each
component of the circuit (Starace & De Pascalis, 2012) and a steady state analytical description of the
thermal pump performance based on its physical and geometrical features (Starace & De Pascalis, 2013).
In particular, as the thermal pump allows to move the fluid through the DAR circuit and to separate the
refrigerant from the absorbent, its design parameters play a primary role in the system performance. As
the slug flow is recognized as the most suitable regime for the two-phase flow to be actuated (Chisholm,
1983), the right choices of the riser tube length and diameter, together with the absorber reservoir height
of the liquid head are crucial for a performing operation. So, the model of thermal pump here proposed
also helps in the design of the riser tube geometry to predict the fluid flow characteristics.
The model was compared with that proposed by Zohar et al. (2009). The differences between the
predictions vary in the range from 2.0% to 8.5% and are certainly due to the main difference in the as-
sumption concerning the ammonia content in the refrigerant flow. This was assumed as yrefr = 1 by the
Zohar et al. (2009) and, in the present model, comes out form calculations (for the case here reported,
yrefr = 0.985).
Differences become much less consistent when the thermal pump temperatures are low and the mis-
matches in results increase with T2 due to a higher water evaporation rate in the thermal pump.
Compared with the other models in literature (Reistad, 1968; Kouremenos & Stegou-Sagia, 1988;
Chen, Kim, & Herold, 1996; Srikhirin & Aphornratana, 2002; Al-Shemmeri & Wang, 2003; Maiya, 2003;
Zohar et al., 2005; Zohar et al., 2009), the one presented here uses a higher number of input variables and
this lets to a more realistically description of the working operations in a wider range of the DAR cycle.
An experimental campaign was then carried out to validate the model and the results show that the
high number of input values leads to an increase of model’s accuracy but, on the other hand, to a stronger
complexity in use.
In the experimental campaign, a purposely designed cycle activated by a magnetron and coupled to a
specifically modified commercial DAR circuit (the Dometic F400) was used. The design of the modified
refrigerator allowed a correct steady state operation after a significant reduction of the transient times.
Filled the refrigerator with a 70.1% H2O, 27.9% NH3 and 2.0% Na2CrO4 mixture and pressurized with
helium up to 11.5 bar, the system was used for the experimental validation, varying the heat power ac-
tivating the circuit. The measured and calculated mass flow rate were compared and the errors in terms
of weak mixture mass flow rate ( m 8 ) and coefficient of performance (COP) were estimated. The
maximum absolute errors were found of 1 K on T8, of 2.32% on m 8 and up to 5% on COP, thus proving
an adequate accuracy.

77

The Diffusion Absorption Refrigerator Operation and Performance

Furthermore, the model was also used to carry out a sensitivity analysis of the DAR cycle perfor-
mance. The results showed that, as the thermal pump temperature increases, the strong mixture mass
flow rate decreases more than the refrigerant mass flow rate. This leads to a decrease of the mass flow
rates ratio and to a reduction in COP.
The pureness of ammonia in refrigerant was demonstrated to be another leading parameter that in-
fluences the DAR cycle performance: a higher water content in refrigerant causes a higher COP value.
Some problems, on the other hand, could occur in the circuit with water freezing in the evaporator when
a too low evaporation pressure is reached.
On the contrary, the COP decreases as the absorption temperature increases because of the reduction
in the absorption rate of the ammonia in the weak mixture.
The DAR performance are influenced as well by the heat power supplied to thermal pump: for low
heat powers, the heat dissipation towards the ambient and the reduced fluid mass flow rates in the ther-
mal pump dominate and COP values become very low. Vice-versa, if Q H exceeds a certain value that
allows the evaporator to exchange its maximum power, Q remains constant and COP begins to reduce.
L
As a result, the COP shows its maximum values over a limited range of the heat power supplied to
the thermal pump.
A second set of experiments on the purposely designed system clearly showed a reduction in the time
needed to reach a steady state operation, suggesting a use of the DAR circuit coupled with a microwave
generator to activate thermal pump as a rapid cooler, suitable for domestic and commercial applications.

REFERENCES

Al-Shemmeri, T., & Wang, Y. (2003). Theoretical Investigation and Parameter Study of a Diffusion
Absorption Refrigeration System. In Proceedings of the 21st IIR International Congress of Refrigera-
tion (pp.17-22). Washington D.C.
Benhmidene, A., Chaouachi, B., & Gabsi, S. (2014). Effect of Tube Diameter on the Performance of
the Bubble Pump. International Journal of Multiphase Flow, 58, 195–204. doi:10.1016/j.ijmultiphase-
flow.2013.09.007
Bettie, D. R. H., & Whalley, P. B. (1982). A Simple Two-Phase Frictional Pressure Drop Calculation
Method. International Journal of Multiphase Flow, 8(1), 83–87. doi:10.1016/0301-9322(82)90009-X
Bourseau, P., Mora, J. C., & Bugarel, R. (1987). Coupling of Absorption-Diffusion Refrigeration
Machine and a Solar Flat-Plate Collector. International Journal of Refrigeration, 10(4), 209–216.
doi:10.1016/0140-7007(87)90054-5
Carluccio, F., Starace, G., & Bongs, C. (2014). Modeling and Simulation of Gas Absorption Heat Pump. In
Proceedings of the International Sorption Heat Pump Conference (ID 106). Washington D.C.: Publisher.
Chen, J., Kim, K. J., & Herold, K. E. (1996). Performance Enhancement of a Diffusion Absorption Re-
frigerator. International Journal of Refrigeration, 19(3), 208–218. doi:10.1016/0140-7007(96)87215-X
Chexal, B., Lellouche, G., Horowitz, J., & Healzer, J. (1992). A Void Fraction Correlation for General-
ized Applications. Progress in Nuclear Energy, 27(4), 255–295. doi:10.1016/0149-1970(92)90007-P

78

The Diffusion Absorption Refrigerator Operation and Performance

Chexal, B., Merilo, M., Maulbetsch, M., Horowitz, J., Harrison, J., Westacott, J., & Scmidt, H. et al.
(1997). Void Fraction Technology for Design and Analysis. Palo Alto: Electric Power Research Institute.
Chisholm, D. (1983). Two-Phase Flow in Pipelines and Heat Exchangers. New York: George Goodwin.
Conde, M. (2006). Thermophysical Properties of NH3+H2O Mixtures for the Industrial Design of Ab-
sorption Refrigeration Equipment. Zurich: M. Conde Engineering.
de Cachard, F., & Delhaye, J. M. (1996). A Slug-Churn Model for Small-Diameter Airlift Pumps. Inter-
national Journal of Multiphase Flow, 22(4), 627–649. doi:10.1016/0301-9322(96)00003-1
Delano, A. D. (1998). Design Analysis of the Einstein Refrigeration Cycle (Ph.D. thesis). Atlanta: Georgia
Institute of Technology.
Dometic©, formerly known as Electrolux in Europe. (2004). Available from: http://www.dometic.com
El-Sayed, Y., & Tribus, M. (1985). Thermodynamic Properties of Water-Ammonia Mixtures; Theoretical
Implementation for Use in Power Cycle Analysis. ASME Special Publication, AES-1, 89–95.
Grossman, G., DeVault, R. C., & Creswick, F. A. (1995). Simulation and Performance Analysis of an
Ammonia-Water Absorption Heat Pump Based on the Generator-Absorber Heat Exchange (GAX) Cycle.
ASHRAE Transactions, 101(1), 1313–1323.
Gutiérrez, F. (1988). Behavior of a Household Absorption-Diffusion Refrigerator adapted to Autonomous
Solar Operation. Solar Energy, 40(1), 17–23. doi:10.1016/0038-092X(88)90067-9
Jakob, U., Eicker, U., Schneider, D., Taki, A. H., & Cook, M. J. (2008). Simulation and Experimental
Investigation into Diffusion Absorption Cooling Machines for Air-Conditioning Applications. Applied
Thermal Engineering, 28(10), 1138–1150. doi:10.1016/j.applthermaleng.2007.08.007
Keizer, C. (1979). Absorption Refrigeration Machine driven by Solar Heat. In Proceedings of the 15th
IIR International Congress of Refrigeration (pp. 861-868). Venice.
Kouremenos, D. A., & Stegou-Sagia, A. (1988). Measuring the Evaporation of NH3 in Triple-Fluid
Gas Absorption Units. International Journal of Refrigeration, 11(3), 153–158. doi:10.1016/0140-
7007(88)90029-1
Maiya, M. P. (2003). Studies on Gas Circuit of Diffusion Absorption Refrigerator. In Proceedings of the
21st IIR International Congress of Refrigeration. Washington D.C.
National Instruments. (2009). Labview v.8.6. NI Academic Site License, Spring 2009. Available from:
http://www.ni.com/labview/
Nicklin, D. J., & Davidson, J. F. (1962). The Onset Instability in Two-Phase Slug Flow. In Proceedings
of the Symposium on Two-Phase Flow (paper #4). London: Institute of Mechanical Engineers.
Nicklin, D. J., Wilkes, M. A., & Davidson, J. F. (1962). Two-Phase Flow in Vertical Tubes. Transactions
of the Institution of Chemical Engineers, 40, 61–68.

79

The Diffusion Absorption Refrigerator Operation and Performance

Pátek, J., & Klomfar, J. (1995). Simple Function for Fast Calculations of Selected Thermodynamic
Properties of Ammonia-Water System. International Journal of Refrigeration, 18(4), 228–234.
doi:10.1016/0140-7007(95)00006-W
Reinemann, D. J., Parlange, J. Y., & Timmons, M. B. (1990). Theory of Small-Diameter Airlift Pumps.
International Journal of Multiphase Flow, 16(1), 113–122. doi:10.1016/0301-9322(90)90042-H
Reistad, B. (1968). Thermal Conditions in Heat Driven Refrigerating Units for Domestic Use. Särtryck
ur Kylteknisk Tidskrift, 3.
Schulz, S. (1971). Eine Fundamentalgleichung für das Gemisch aus Ammoniak und Wasser und die
Berechnung von Absorptionskältemaschinen-Prozessen. Habilitationsschrift, Abt. für Maschinenbau
und konstruktiven Ingenieurbau. Ruhr Universtität Bochum.
Srikhirin, P., & Aphornratana, S. (2002). Investigation on a Diffusion Absorption Refrigerator. Applied
Thermal Engineering, 22(11), 1181–1193. doi:10.1016/S1359-4311(02)00049-2
Starace, G., & De Pascalis, L. (2012). An Advanced Analytical Model of the Diffusion Absorption Refrig-
erator Cycle. International Journal of Refrigeration, 35(3), 605–612. doi:10.1016/j.ijrefrig.2011.11.007
Starace, G., & De Pascalis, L. (2013). An Enhanced Model for the Design of Diffusion Absorption Re-
frigerators. International Journal of Refrigeration, 36(5), 1495–1503. doi:10.1016/j.ijrefrig.2013.02.016
Steirlin, H., & Ferguson, J. R. (1990). Diffusion Absorption Heat Pump (DAHP). ASHRAE Transac-
tions, 96(1), 3319–3328.
Thome, J.R. (2010). Engineering Data Book III. Decatur: Wolverine Tube, Inc.
Thorin, E., Dejfors, C., & Svedberg, G. (1998). Thermodynamic Properties of Ammonia-Water Mixtures
for Power Cycles. International Journal of Thermophysics, 19(2), 501–509. doi:10.1023/A:1022525813769
Tillner-Roth, R., & Friend, D. G. (1998). Survey and Assessment of Available Measurements on Ther-
modynamic Properties of the Mixture Water + Ammonia. Journal of Physical and Chemical Reference
Data, 27(1), 45–61. doi:10.1063/1.556014
Von Platen, B. C., & Munters, C. G. (1928). Refrigerator. U.S. Patent 1,685,764.
Wang, L., Herold, K.E. (1992). Diffusion-Absorption Heat Pump. Annual Report to Gas Research In-
stitute, GRI-92/0262.
White, S. J. (2001). Bubble Pump Design and Performance (M.Sc. thesis). Georgia Institute of Technol-
ogy, Atlanta, GA.
Xu, F., & Goswami, D. Y. (1999). Thermodynamic Properties of Ammonia-Water Mixtures for Use in
Power Cycles. Energy, 24(6), 525–536. doi:10.1016/S0360-5442(99)00007-9
Ziegler, B., & Trepp, C. (1984). Equation of State for Ammonia-Water Mixtures. International Journal
of Refrigeration, 7(2), 101–106. doi:10.1016/0140-7007(84)90022-7

80

The Diffusion Absorption Refrigerator Operation and Performance

Zohar, A., Jelinek, M., Levy, A., & Borde, I. (2005). Numerical Investigation of a Diffusion Absorp-
tion Refrigerator Cycle. International Journal of Refrigeration, 28(4), 515–525. doi:10.1016/j.ijre-
frig.2004.11.003
Zohar, A., Jelinek, M., Levy, A., & Borde, I. (2008). The Influence of the Generator and Bubble Pump
Configuration on the Performance of Diffusion Absorption Refrigeration (DAR) System. International
Journal of Refrigeration, 31(6), 962–969. doi:10.1016/j.ijrefrig.2008.01.009
Zohar, A., Jelinek, M., Levy, A., & Borde, I. (2009). Performance of Diffusion Absorption Refrigera-
tion Cycle with Organic Working Fluids. International Journal of Refrigeration, 32(6), 1241–1246.
doi:10.1016/j.ijrefrig.2009.01.010
Zuber, N., & Findlay, J. (1965). Average Volumetric Concentration in Two-Phase Flow Systems. Journal
of Heat Transfer, 87(4), 453–468. doi:10.1115/1.3689137

KEY TERMS AND DEFINITIONS

Absorption Refrigeration: Refrigerating technology based on the ability of a fluid to be absorbed


in another in order to produce the cooling effect.
Coefficient of Performance (COP): Ratio of the cooling effect to the supplied energy.
Diffusion Absorption Refrigerator (DAR): Type of absorption chiller using an inert gas to reduce
the pressure of the refrigerant and producing its evaporation.
Magnetron: high-power vacuum tube producing non-coherent microwaves.
Sensitivity Analysis: Study of the effects produced on outputs by the variation in the inputs of an
analytical model.
Thermal Pump: Device used to circulate a fluid mixture supplying heat and realizing a two phase flow.
Thermo- and Fluid-Dynamic Modeling: Mathematical and analytical description of the thermo-
and fluid-dynamic behavior of the mixture.

81
The Diffusion Absorption Refrigerator Operation and Performance

APPENDIX: NOMENCLATURE

A = area, m2
Bo = Bond number
C0 = distribution parameter
COP = coefficient of performance
cp = specific heat at constant pressure, J kg-1 K-1
C p,min = minimum heat capacity, J K-1 s-1
d = diameter, m
Err(z) = error of the model on the parameter z
f = friction factor
f* = Fanning friction factor
G = Gibbs free energy, kJ kg-1
g = acceleration of gravity, m s-2
h = specific enthalpy, J kg-1
H = height of the liquid head in the reservoir, m
H/L = submerge ratio
j = superficial velocity, m s-1
L = length of the riser tube, m
m = constant
m = mass flowrate, kg s-1
Nf =viscous effects parameter
p = static pressure, bar
pcw = water critical pressure, bar
Q = heat transfer, W
R = gas constant, J kg-1 K-1
Re = Reynolds number
s = specific entropy, kJ kg-1
T = temperature, K
Tb = bubble point temperature, K
Td = dew point temperature, K
Tcw = water critical temperature, K
U = global heat transfer coefficient, W m-2 K-1
v = specific volume, m3 kg-1
V = volumetric flow rate, m3 s-1
w = velocity, m s-1
x = molar concentration of ammonia in liquid mixture
X = vapor quality
y = molar concentration of ammonia in vapor mixture
ΔTmax = maximum temperature difference between refrigerated air and refrigerant in the evaporator, K
ε = void fraction
εR = pipe roughness, m
λ = wavelength, m

82
The Diffusion Absorption Refrigerator Operation and Performance

μ = fluid viscosity, kg m-1 s-1


ψ = magnetron rotation frequency, Hz
ρ = density, kg m-3
σ = surface tension, N m-1

Subscripts

1…11 = states of working mixture


a = absorption conditions
air = refrigerated air in the evaporation chamber
c = riser tube
calc = calculated value
cb = cold box
C = condensation conditions
e = evaporation conditions
eval = evaluated (starting from a measured value)
h = homogeneous conditions
H = highest temperature in the thermal pump
H,d = heat dissipated at the highest temperature in the thermal pump
H2O = water
i = inner tube of the liquid HX
ig = auxiliary inert gas
I,a = intermediate temperature in the absorber
I,c = intermediate temperature in the condenser
I,r = intermediate temperature in the rectifier
m = mixture
meas = measured value
NH3 = ammonia
l = saturated liquid state
L = lowest temperature in the evaporator
L,max = maximum evaporator heat transfer
p = weak mixture
r = strong mixture
R = reduction state
refr = refrigerant
TP = two-phase
v = saturated vapor state
vj = drift

83
The Diffusion Absorption Refrigerator Operation and Performance

Superscripts

E = excess energy
v = vapor condition
l = liquid condition

84
85

Chapter 3
Advanced Exergetic Analysis
is a Modern Tool for
Evaluation and Optimization
of Refrigeration Systems
Tatiana Morosuk
Technische Universität Berlin, Germany

George Tsatsaronis
Technische Universität Berlin, Germany

ABSTRACT
In the last decades an exergetic analysis became increasingly popular because this analysis identifies
the location, magnitude and sources of thermodynamic inefficiencies. A conventional exergetic analysis,
however, does not consider (a) the real potential for improving a system nor (b) the interactions among
the components of the system. The interactions among different components of the same system can be
estimated and the quality of the conclusions obtained from an exergetic evaluation can be improved,
when the exergy destruction (irreversibilities) within each system component are split into endogenous/
exogenous and avoidable/ unavoidable parts. We call this advanced exergetic analysis. The purpose
of this chapter is to demonstrate that the advanced exergetic analysis is a practical method that allows
engineers to extract useful information and conclusions and to develop ideas and solutions that cannot
be suggested by other methods. In this chapter the conventional and advanced exergetic analysis are
applied to an air refrigeration machine.

INTRODUCTION

Exergy-based methods is a general term that includes the conventional and advanced exergetic, exer-
goeconomic, and exergoenvironmental analyses and evaluations (Tsatsaronis & Morosuk, 2009). It is
well known that conclusions obtained through the evaluation of an energy-conversion system using
exergy-based methods are unique and cannot be obtained by using other methods. However, conventional
exergy-based analyses do not consider (Tsatsaronis, 1999)

DOI: 10.4018/978-1-4666-8398-3.ch003

Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

• The real potential for improving a system nor


• The interactions among the components of the system.

The interactions among different components of the same system can be estimated and the quality
of the conclusions obtained from an exergetic evaluation can be improved, when the exergy destruction
within each (important) system component is split into endogenous/exogenous and avoidable/ unavoid-
able parts. The most important publications up to date in the field of the advanced exergy-based methods
are summarized and generalized in (Tsatsaronis & Morosuk, 2008, 2009).
Conventional and advanced exergy-based analyses have already been applied to different energy-
conversion systems (power, cogeneration, and refrigeration), for example by (Tsatsaronis & Morosuk,
2008, 2009, 2010, 2011; Petrakopoulou, Tsatsaronis, & Morosuk, 2011; Morosuk & Tsatsaronis, 2008,
2009a, 2009b, 2011). The interconnections among components of these systems are usually not very
strong and the advantages of the advanced exergy-based methods could not have been fully demonstrated.
In this paper conventional and advanced exergetic analyses are applied to an air refrigeration machine.
We selected this kind of a refrigeration machine as an academic example of an energy-conversion system
with very strong interconnections among components.
The purpose of the chapter is to emphasize that

• Exergy-based methods are practical methods that allow engineers to extract useful conclusions,
and
• The informations provided by advanced exergy-based analyses help engineers to develop ideas
and solutions that cannot be suggested by other methods.

CASE STUDY

Figure 1 shows the air refrigeration machine that is analyzed here. The machine consists of the compres-
sor (CM) driven by expander (EX) and electrical motor (EM, ηEM =0.9), the heat exchanger (HE) where

Figure 1. Schematic of a simple air refrigeration machine

86

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

the working fluid is cooled by water, the expander, and the refrigerator (R) where the working fluid is
heated by air. The refrigeration capacity of the machine is assumed to be Qcold =100kW. The compressor
and the expander are turbomachines with a theoretical pressure ratio p3 / p1 =5. The assumed operation
conditions of the air refrigeration machine are given in the column “Real process” of Table 1.
The thermodynamic data (mass flow rate, temperature and pressure) obtained from the simulation
are given in Table 2. The first column of Table 3 shows the results of the energetic analysis, where
Wcycle = W EX −WCM . Note that the energetic efficiency of an air refrigeration machine (COP ) is very
low in general, and the value of COP=0.25 for the machine analyzed here is a realistic one.

CONVENTIONAL EXERGETIC ANALYSIS

After many years of continuous development and applications, an exergetic analysis is considered to-
day to be the most effective method for evaluating the quality of energy carriers and energy-conversion
processes, and the rational use of energy. An exergetic analysis can be applied to any energy-conversion
system or chemical process.
An exergetic analysis identifies the location, the magnitude and the causes of thermodynamic inef-
ficiencies, assists in better understanding the operations in an energy conversion process, and enhances
the creativity of engineers dealing with the improvement of such systems and processes (Bejan, Tsat-
saronis, & Moran, 1996).
The exergy concept provides the following information which is useful for improving an exergy
conversion system: (a) the true thermodynamic value of an energy carrier, (b) the real thermodynamic
inefficiencies in a system, and (c) variables that unambiguously characterize the performance of a system
(or one of its components) from the thermodynamic point of view.

Table 1. Values of parameters assumed for the different operation conditions of the air refrigeration
machine considered in this paper (Figure 1)

Component Parameter, Unit Theoretical Process with Unavoidable Thermodynamic Real Process
Process Inefficiencies

CM ηCM [-] 1 0.95 0.8

HE 0 1 10
∆THE [K]

∆pHE [%] 0 1 5

EX ηEX [-] 1 0.95 0.8

R 0 3 20
∆TR [K]

∆pR [%] 0 1 5

87

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

Table 2. Thermodynamic data for the air refrigeration machine at real operation conditions

Stream Material Stream T p eT eM e


m [°C] [bar] [kJ/kg] [kJ/kg] [kJ/kg]
[kg/s]

1 Air 4.198 -30 1.00 5.82 0 5.82

2 Air 4.198 154 5.25 21.97 141.90 163.87

3 Air 4.198 35 5.00 0.15 137.70 137.85

4 Air 4.198 -54 1.05 12.73 4.17 16.90

11 Air 9.968 -10 1.00 2.24 0 2.24

12 Air 9.968 -20 1.00 3.80 0 3.80

21 Water 8.015 25 1.50 0 0.05 0.05

22 Water 8.015 40 1.50 1.53 0.05 1.58

Table 3. Results obtained from the energetic analysis and definition of the exergy of product and exergy
of fuel for the exergetic analysis for the components of the air refrigeration machine (Figure 1)

Component Heat Rates and Power (kW) Exergetic Analysis


CM
WCM = 776.7 E F ,CM = WCM + E 1T
(
E P ,CM = E 2M − E 1M + E 2T )
EM
Wcycle = 402.8 E F ,EM = W EM
W = 447.6
EM
E P ,EM = Wcycle

HE
Q HE = 502.8 ( ) (
E F ,HE = E 2T − E 3T + E 2M − E 3M )
E P ,HE = E 22 − E 21

EX
W EX = 373.9 (
E F ,EX = E 3M − E 4M + E 3T )
E P ,EX = W EX + E 4T

R
Q R = Qcold =100 ( ) (
E F ,R = E 4T − E 1T + E 4M − E 1M )
E P ,R = E 12 − E 11

Overall system
Q E F ,tot = W EM
COP = cold = 0.25
WEM E P ,tot = E P ,R
EL,tot
= E P ,HE

88

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

Early developments of an exergetic analysis were based on the concepts of input exergy and output
exergy (for example, Szargut & Petela, 1965; Kotas, 1985; Szargut, Morris, & Steward, 1988),

E in ,k = E out ,k + E D ,k (1a)

and

E in ,tot = E out ,tot + ∑ E D ,k + E L,tot (1b)


n

and the exergetic efficiency of kth component was defined as the ratio between these two variables

E
εk = out ,k (2)
Ein ,k

This definition of exergetic efficiency is, however, not consistent with the general definition of ef-
ficiency

Desired result
Efficiency = (3)
Resources spent to generate this result

because not all incoming streams to the kth component are associated with resources, neither do all
outgoing streams belong to the desired result. This deficiency was corrected by the general concept of
exergy of product (the desired result achieved by the kth component expressed in exergy terms) and
exergy of fuel (exergetic resources expended to provide the exergy of product), which was initially pro-
posed by Tsatsaronis (1984), further discussed by Tsatsaronis & Winhold (1994), and finally generalized
by Lazzaretto & Tsatsaronis (2006). Using the exergy rates associated with fuel and product, E F and
E , the exergetic balances for the kth component and for the overall system, are respectively
P

E F ,k = E P ,k + E D ,k (4a)

and

E F ,tot = E P ,tot + ∑ E D ,k + E L,tot (4b)


n

The exergetic efficiencies for the kth component and for the overall system, are respectively (Bejan,
Tsatsaronis, & Moran, 1996)

89

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

E
εk =  P ,k (5a)
E F ,k

and

E
εtot =  P ,tot (5b)
E F ,tot

The concept of fuel and product is also applicable to the definition of energetic efficiencies. In this
way, energetic and exergetic efficiencies can be compared.
The exergy destruction E D (or internal exergy loss) is the exergy destroyed due to the irreversibil-
ity within the kth component or within a system. The exergy loss E (or external exergy loss) is the
L
exergy transfer from the system to its surroundings.
For practical applications of the exergy concept, it is important to know how to interpret the exergetic
variables, and how to use the obtained information, in order to improve the overall system.
The strengths and limitations of a conventional exergetic analysis have been discussed by Tsatsaronis
(1999):

• The only variable that characterizes the performance of a component from the thermodynamic
viewpoint is the exergetic efficiency, which also should be used to compare the performance of
similar components operating under similar conditions in the same system or in different systems.
• The only variable that may be used for comparing dissimilar components is the exergy destruction
ratio

E
yk =  D ,k (6)
E F ,tot

The value of the exergy destruction ratio calculated as

E
y *k =  D ,k (7)
E D ,tot

can be used only for apportioning the total exergy destruction among the components, and for illustrating
the distribution of thermodynamic inefficiencies within the overall system.
The definitions of E F and E P for the kth component and for the overall system (Bejan, Tsatsaronis,
& Moran, 1996; Lazzaretto & Tsatsaronis, 2006) are given in the last column of Table 3.
Since no chemical reactions occur within the components of the analyzed air refrigeration machine,
only the physical exergy e j associated with the j th material stream is considered

90

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

e j = m j (h j − h0 ) −T0 (s j − s 0 ) (8)


 

The subscript 0 refers to the property values of the same mass flow rate at temperature T0 and pres-
sure p0 of the reference state. Here we assumed T0 = T21 =25ºC and p0 =1 bar.
In all components (with the exception of HE) of the air refrigeration machine, either all temperatures
are below the reference temperature T0, or this temperature is crossed during component operation. In
either case, the physical exergies of all states should be split into their thermal (eT ) and mechanical (
e M ) exergy components according to the approach presented by Morosuk & Tsatsaronis (2005)

e j = (h j − h j ,X ) −T0 (s j − s j ,X ) +


  p =const

eT (9)
(h − h ) −T (s − s )
 j ,X j ,0  T0 =const

j ,0 0 j ,X

eM

In Equation 9 the point X (T0, p ) is defined at the given pressure p of the j th stream and the tem-
perature T0 of the environment. The values of eTj , e jM and e j are given in last three columns of Table
2.
Table 4 shows the results of the conventional exergetic analysis of the air refrigeration machine. The
exergetic efficiency of the air refrigeration machine is extremely low and equal to 3.5%. Based on the
values of E Dreal
,k
we can conclude that the expander and the compressor are the most important components
from the thermodynamic viewpoint. The improvement of the evaporator cannot significantly affect the
improvement of the air refrigeration machine.
Now we want to conduct a sensitivity analysis and consider improvement options for each component,
in order to confirm the conclusions obtained from the conventional exergetic analysis. Let us conduct
an analysis by assuming that the COP value of the air refrigeration machine should be increased in all

Table 4. Conventional exergetic analysis of the air refrigeration machine (Figure 1)

Component
E F,k
real
[kW] E P,k
real
[kW] E D,k
real
[kW] µk [%] yk [%]

CM 801.20 687.90 113.30 85.9 25.3

EM 447.56 402.80 44.76 90.0 10.0

HE 109.10 12.24 96.86 11.2 21.6

EX 561.20 427.30 133.90 76.1 29.9

R 46.56 15.51 31.05 33.3 6.9


Overall system
( E L,tot = 12.24 kW) 447.6 15.51 419.90 3.5 93.7

91

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

calculations from 0.25 to 0.27. We can achieve this increase by changing only one design parameter in
each component at a time: The isentropic efficiency of the compressor from 0.8 to 0.84, the isentropic
efficiency of the expander from 0.80 to 0.82, the minimum temperature difference in the heat exchanger
from 10 K to 6 K, or the minimum temperature difference in the refrigerator from 20 K to 17 K. Pres-
sure drops around 5% have not a significant effect on the exergy destruction within the heat exchanger
and refrigerator. Therefore, only the temperature difference in varied in the sensitivity analysis for these
components. Through each one of these changes we obtain a COP value of 0.27. Table 5 summarizes
the assumptions and the results of this sensitivity analysis.
The results of the analysis (last four columns in Table 5) show, that, in order to increase by the same
amount the overall thermodynamic efficiency (COP) of the air refrigeration machine for 10% (from 0.25
to 0.27), one of the following options can be used:

• Increase the isentropic efficiency of the expander from 0.8 to 0.82, which is a realistic and easy to
implement change.
• Increase the isentropic efficiency of the compressor from 0.8 to 0.84, which is a quasi-realistic
change.
• Decrease the temperature difference in the refrigerator from 20K down to 17K, which is a realistic
and easily implementable change.
• Decrease the temperature difference in the heat exchanger from 10K down to 6K which is only
quasi-realistic change.

Table 5. Data assumed for the sensitivity analysis (conventional exergetic analysis) of the air refrigera-
tion machine and obtained results

Component Parameter, Real Sensitivity Analysis


Unit Process
CM HE EX R

CM ηCM [-] 0.80 0.84 0.8 0.8 0.8

HE 10 10 6 10 10
∆THE [K]

∆pHE [%] 5 5 5 5 5

EX ηEX [-] 0.80 0.8 0.8 0.82 0.8

R 20 20 20 20 17
∆TR [K]

∆pR [%] 5 5 5 5 5

Overall system COP [-] 0.25 0.27

εtot [%] 3.5 3.8

E D ,tot [kW] 419.90 379.7 377.6 373.1 380.0

92

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

A quasi-realistic change means here that the design of the component should be modified significantly.
Only the expander and the refrigerator can be improved from the engineering point of view without a
significant change in the design.
Thus, the conclusions obtained from the conventional exergetic analysis are only partially confirmed
by the sensitivity analysis.

ADVANCED EXERGETIC ANALYSIS

As already mentioned, a conventional exergetic analysis cannot evaluate the mutual interdependencies
among the system components nor the real potential for improving each component and the overall sys-
tem. This becomes possible in an advanced exergetic analysis, in which the exergy destruction in each
component is split into different parts.

Definitions

Endogenous exergy destruction is the part of exergy destruction within a component obtained when all
other components operate ideally and the component being considered operates with the same efficiency
as in the real system. The exogenous part of the variable is the difference between the value of the vari-
able within the component in the real system and the endogenous part:

E D ,k = E DEN,k + E DEX,k (10)

This splitting of the exergy destruction gives engineers an opportunity to estimate the interconnec-
tions among the system components. Another splitting of the exergy destruction (into unavoidable and
avoidable parts) shows us the real potential for improving the kth component.
The unavoidable ( E DUN,k ) exergy destruction cannot be further reduced due to technological limitations
such as availability and cost of materials and manufacturing methods. The difference between total and
unavoidable exergy destruction for a component is the avoidable exergy destruction ( E DAV,k ) that should
be considered during the improvement procedure

E D ,k = E DUN,k + E DAV,k (11)

By combining the above splitting procedures we can calculate

• The avoidable endogenous exergy destruction ( E DAV,k ,EN ), which can be reduced by improving the
kth component from the thermodynamic point of view, and
• The avoidable exogenous exergy destruction ( E DAV,k ,EX ) that can be reduced by a structural im-
provement of the overall system, or by improving the efficiency of the remaining components, and
always of course by improving the efficiency in the kth component.

93

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

To better understand the interactions among components, the exogenous exergy destruction as well
as the exogenous avoidable exergy destruction within the kth component are split further as:

n
E DEX,k = ∑ E DEX,k ,r + E Dmexo
,k
(12)
r =1
r ≠k

The value of E DEX,k ,r represents the part of the exogenous exergy destruction within the kth component
that is caused by the irreversibilities occurring within the rth component. The remaining part E mexo we
D ,k
call mexogenous exergy destruction within the kth component.
The methodology of splitting has been described in detail for refrigeration machines by Morosuk &
Tsatsaronis (2008, 2009b).
In order to identify priorities for improving components based on information obtained from an
advanced exergetic analysis, the variable E DAV,k ,Σ should be used. This variable presents the sum of the
avoidable endogenous exergy destruction within the kth component and the avoidable exogenous exergy
destructions caused by the kth component within the remaining components (Morosuk & Tsatsaronis,
2007)

n
E DAV,k ,Σ = E DAV,k ,EN + ∑ E DAV,r ,EX ,k (13)
r =1
r ≠k

METHODOLOGY AND RESULTS

For splitting the exergy destruction into unavoidable and avoidable parts, we need to conduct a simula-
tion of the refrigeration machine with so-called “unavoidable irreversibilities”, that is taking into account
the technological limitations (e.g., manufacturing methods) which cannot be overcome in future for heat
exchangers having air as a secondary working fluid, and for reciprocating compressors. The column
“Process with unavoidable thermodynamic inefficiencies” of Table 1 contains the parameter values used
to calculate the unavoidable inefficiencies within all components of the air refrigeration machine. The
value of the unavoidable exergy destruction for the kth component E DUN,k is calculated from

UN
 E 
E DUN,k = E Preal  D  (14)
,k   
 E P 
k

using the methodology presented by Tsatsaronis & Park (2002), Cziesla, Tsatsaronis, & Gao (2006),
UN
 E 
 D
and Tsatsaronis & Morosuk (2008). The term   is the ratio between exergy destruction and en-
 E P 
k

94

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

Table 6. Selected data for conducting the advanced exergetic analysis

Component Unavoidable Operating Conditions Hybrid Operating Conditions - 1

 E 
UN
E P,k
EN
 D,k 
 E  [kW]
 P,k 

CM 0.0364 224.30

HE 6.0272 4.06

EX 0.0666 196.80

R 1.1464 15.51

ergy of product for the k th component at the operating conditions where only the entire unavoidable
exergy destructions occur (from the simulation of the process with “unavoidable thermodynamic irre-
UN
 E 
 D
versibilities”). The value of   are given in Table 6.
 E P 
k

The values of E UN and E AV are given in Table 7. For the turbomachines E UN << E AV , and for the
D ,k D ,k D ,k D ,k

heat exchanger E DUN,k >> E DAV,k . For the refrigerator E DUN,k > E DAV,k , but these values are comparable. The
expander has the highest potential of improvement ( E AV is the highest among all E AV values). The
D ,EX D ,k

compressor has the second highest E DAV,k value. The heat exchanger has a relative low potential for im-
provement ( E AV =23.09 kW, that is 24% of the total exergy destruction within the heat exchanger).
D ,HE
Again we conclude that the potential for improvement of the evaporator is very low. Note that if all
components would operate only with unavoidable irreversibilities, then the efficiency of the air refrig-
eration machine would be: COPUN=1.147 and εUN = 17%.
The interconnections among the components within the system can be estimated through splitting
the exergy destruction into endogenous and exogenous parts. For this splitting, we need a set of special
simulations of the system.
The first step is to simulate a “theoretical process”. The assumptions used for the simulation of the
theoretical process are given in the corresponding column of Table 1. More detailed explanations can
be found in publications by Morosuk & Tsatsaronis (2008, 2009a, 2009b) and Tsatsaronis & Morosuk
(2010; 2011).
The second step is to simulate a “hybrid processes 1” (Morosuk & Tsatsaronis, 2008; 2009a, 2009b;
Tsatsaronis & Morosuk, 2010; 2011): The number of these hybrid processes is equal to the number of
components E DAV,k within the system. In a hybrid process only the component being considered is simu-
lated with its real inefficiency, while all other components correspond to the theoretical process. The
exergy destruction calculated for the k th component (with real irreversibility) is equal to the endogenous
exergy destruction within this component.
The values of E DEN,k and E DEX,k are given in Table 7. For the compressor and the heat exchanger, the
exogenous exergy destruction dominates ( E EN < E EX ); for the expander these values are comparable.
D ,k D ,k

95
96
Table 7. Splitting the exergy destruction within the components of the air refrigeration machine at real operation conditions E Dreal
,k

Component UN AV real
EN EX E D,k E D,k E D,k
E D,k E D,k [kW] Splitting [kW]
[kW] [kW]
[kW]
UN AV
E D,k [kW] E D,k [kW]

UN,EN UN,EX AV,EN AV,EX


E D,k E D,k [kW] E D,k E D,k [kW]
[kW] [kW]

CM 35.09 78.21 HE 4.11 25.05 88.25 8.16 16.89 HE 1.11 26.93 61.32 HE 3.00
EX 14.56 EX 3.39 EX 11.17
R 15.13 R 2.77 R 12.36
mexo 44.41 mexo 9.62 mexo 34.79
HE 31.59 65.27 CM 17.53 73.77 23.09 24.47 49.30 CM 8.55 7.12 15.97 CM 8.98
EX 15.65 EX 12.15 EX 3.50

R -1.32 R 1.88 R -3.20


mexo 33.41 mexo 26.72 mexo 6.69
EX 62.24 71.66 CM 0 28.48 105.42 13.11 15.37 CM 0 49.13 56.29 CM 0

HE 9.56 HE 2.05 HE 7.51


R 40.76 R 8.76 R 32.00
mexo 21.34 mexo 4.56 mexo 16.78
R 31.05 0 CM 0 17.78 13.27 17.78 0 CM 0 12.76 0 CM 0

HE 0 HE 0 HE 0

EX 0 EX 0 EX 0
mexo 0 mexo 0 mexo 0
Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems


Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

This means that the interconnections between the components are very strong. A particular case is the
refrigerator where E DEX,R =0. This means that the exergy destruction within the refrigerator can be de-
creased only through improvement of the component itself and that the exergy destruction within this
component is independent of the irreversibilities within the remaining components of the air refrigera-
tion machine.
Subsequently the value of unavoidable endogenous exergy destruction E DUN,k ,EN is calculated from

UN
 E 
E UN ,EN
= E EN  D  (15)
D ,k P ,k  E 
 P k

and the values of E DUN,k ,EX , E DAV,k ,EN and E DAV,k ,EX are easily estimated.
To understand how one component affects other components, we need to split the exogenous exergy
destruction E DEX,k according to Equation 12. For this splitting, we need again new simulations, i.e. a set
of “hybrid processes 2”: Two components are simulated with their real efficiency while all other com-
ponents correspond to the theoretical process. Not only the exogenous exergy destruction E DEX,k should
be split, but also the terms E UN ,EX and, particularly, E AV ,EX . All data are given in Table 7.
D ,k D ,k

The following information can be obtained by splitting the value of E DEX,k :

• The values of E Dmexo


,k
are relative large for the compressor, the heat exchanger and the expander.
This fact confirms again the strong interconnections among the components.
• The refrigerator is a significant contributor to the exogenous exergy destruction within the tur-
bomachines. The negative value of E DEX,HE,R demonstrates, in general, that “the worse the perfor-
mance of the refrigerator, the better the performance of the heat exchanger”. This conclusion is
mainly associated with changes in state 1.

The total splitting of the avoidable exergy destruction (last three columns of Table 7) shows the in-
terconnections between the components and the potential for improving the components:

• The values of E DAV,k ,EX are higher than E DAV,k ,EN for the compressor, the heat exchanger and the
expander.
• For improving the compressor, the expander and refrigerator play a significant role. For improving
the heat exchanger, the compressor is very important, while the refrigerator is the most important
component affecting the exergy destruction within the expander.

The decision about the priority for improving the overall system by improving the single components
can be made based on the value of E DAV,k ,Σ , (Equation 13): E DAV,CM,Σ =35.91 kW, E DAV,HE,Σ =17.63 kW, E DAV,EX,Σ
=63.80 kW, and E AV ,Σ =53.92 kW. But also a detailed analysis is very informative. Based on the data
D ,R
obtained from the advanced exergetic analysis, we can conclude the following:

97

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

• The overall air refrigeration machine can be significantly improved (from the thermodynamic
point of view) through decreasing the irreversibilities within the expander and the refrigerator.
• The improvements of the compressor and the heat exchanger are of secondary important.

Thus, the conclusions obtained from the advanced exergetic analysis obtain a very good confirmation
by the sensitivity analysis (Table 8).
An advanced exergetic analysis is now conducted for each one of the three steps (realistic from en-
gineering point of view) used to improve the air refrigeration machine (Tables 7 and 8):

Step 1: ηEX =0.81,


Step 2: ∆TR =18K, and
Step 3: ηEX =0.81 and ∆TR =18K.

( )
Only the values of the unavoidable exergy destruction E DUN,k remain constant in each one of these

( )
calculations. The values of endogenous E DEN,k and unavoidable endogenous exergy destructions E DUN,k ,EN( )
depend on whether the characteristic variables for a component remain unchanged (for the compressor
and the heat exchanger), or change (for the expander and refrigerator). The values of E DEN,k and E DUN,k ,EN
remain constant in the former cases and change in the latter.
In order to demonstrate, how the exergy destruction within the components as well as the avoidable
endogenous, avoidable exogenous exergy destructions and the sum of the avoidable exergy destructions
change during the improvement of the air refrigeration machine, selected data are presented in Table 8.
From Table 8 we conclude the following:

• (
The total exergy destructions E Dreal
,k )
within the expander and the refrigerator are decreased from
133.90 kW to 112.80 kW and from 31.05 kW to 29.15 kW, respectively. The improvement of the
expander affects significantly the decrease in the value of E Dreal
,EX
as well as the value of E DAV,r ,EX ,EX
(effect of the expander on the other components, as shown by the effect of the expander on the
exogenous exergy destruction of the rth component). The improvement of the refrigerator leads to
a relatively small decrease of the exergy destruction within the refrigerator itself (value of E Dreal ,R
).
However, it significantly decreases the values of E AV ,EX ,
D ,r
R
(effect of the refrigerator on the other
components here indicated by the index r).
• (
The total exergy destruction E Dreal
,k )
within the compressor and the heat exchanger are decreased
from 113.30 kW to 100.20 kW and from 96.86 kW to 89.33 kW, respectively. Note that these two
components are improved “indirectly”, i.e. by improving the expander and/or the refrigerator.

ANOTHER APPROACH TO THE ADVANCED EXERGETIC ANALYSIS

The effect of the inefficiencies within one component on the other components can be determined us-
ing the approach that was proposed initially by Valero & Torres (1988) and applied to the structural

98

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

Table 8. The data are given: for the base case in bold underline; step 1 ( ηEX =0.81) – in bold; step 2 – (
∆TR =18K) – in italic, and step 3 – ( ηEX =0.81 and ∆TR =18K) in “regular” fonts

Component
E D,k
real
E D,k
AV,EN
E D,k
AV,EX
E D,k
AV,£

[kW] [kW] [kW] [kW]

CM 113.30 26.93*) 61.32 HE 3.00*) 35.91*)


108.20 56.22
EX 11.17
104.50 52.52
10.39
100.20 48.22
11.17
10.37
R 12.36
12.36
10.92
10.92
mexo 34.79
30.47
27.43
23.91
HE 96.86 7.12 15.97 CM 8.98*) 17.63
92.54 11.65 16.80
EX 3.50
93.18 12.29 17.63
3.24
89.33 8.44 16.80
3.50
3.24
R -3.20
-3.20
-3.02
-3.02
mexo 6.69
2.63
2.83
-0.76
EX 133.90 49.13 56.29 CM 0*) 63.80
121.90 44.97 48.45 58.60
HE 7.51
123.50 49.13 45.89 63.80
6.68
112.80 44.97 39.35 58.58
7.51
6.68
R 32.00
23.31
27.06
24.00
mexo 16.78
13.46
11.32
8.67
R 31.05 12.76 0*) CM 0 53.92
31.05 12.76 45.74
29.15 11.88 HE 0 46.33
29.15 11.88 43.27
EX 0

mexo 0
*)
The value remains unchanged.

99

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

theory and to the thermoeconomic diagnosis by Lozano, Bartolomé, Valero, & Reini (1994) as well as
by Valero et al. (2004). This approach is called malfunction/dysfunction analysis. Many papers related
to the theoretical background of the structural theory and thermoeconomic diagnosis as well as the ap-
plication of these theories to the different energy conversion and chemical systems have been published
during last two decades.
Any additional irreversibilities (irreversibilities which are not included in the design case of an en-
ergy conversion system) within the kth component results in additional fuel being supplied to the over-
all system. According to the malfunction/dysfunction analysis, the additional fuel being supplied to the
overall system (∆E F ,tot ) is distributed to the system components as additional fuel for each component
(∆E ) . This additional fuel for each component is caused by a malfunction (produced by an increase
F ,k

of the specific consumption in the component itself) and dysfunction (introduced in a component being
considered by malfunctions of other components). The malfunction/dysfunction analysis is based on the
exergy balances written for the k th components (Equation 4a) and for the overall system (Equation 4b).
For this kind of advanced exergetic analysis we need to know the design conditions and the real operat-
ing conditions for an existing energy conversion system.
It is necessary to note, that the malfunction/dysfunction analysis and the splitting of the exergy de-
struction into endogenous/exogenous parts are completely different approaches. The main differences
between these approaches are the following: Malfunction/dysfunction analysis means splitting only the
additional part of the fuel in the kth component ∆E F ,k , which should be known. The main assumptions
for the analysis of malfunction/dysfunction include:

1. The product of the previous component becomes the fuel of the subsequent component. This as-
sumption is possible only for a theoretical system, for example, discussed in detail by Morosuk &
Tsatsaronis (2008), and
2. Each component of the energy conversion system is analyzed in isolation from the overall system,
therefore the effect of the mass flow rate m k is neglected. The importance to consider the effect
of the mass flow rate to the results are also discussed in detail by Morosuk & Tsatsaronis (2008).

In this way, the malfunction/dysfunction analysis can be used only for the diagnosis of an energy
conversion system where the design conditions and the real operation conditions are not so far from each
other. On the other hand, the theory of splitting the exergy destruction into endogenous/exogenous parts
is more general and can be applied to the diagnosis as well as to the design analysis and optimization
of an energy conversion system. A comparison between the malfunction/dysfunction analysis and the
splitting of the exergy destruction into endogenous/exogenous parts is reported by Kelly, Tsatsaronis,
& Morosuk (2009) with the application to an open-cycle gas-turbine system.

CONCLUSION

Advanced exergetic analysis is based on a splitting of exergy destruction. Through this technique our
understanding of energy conversion processes, of the interactions among system components is greatly
improved. The application of advanced exergetic analysis to an air refrigeration machine and the com-

100

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

parison with the results obtained from sensitivity analyses clearly demonstrate the advantages of this
analysis in understanding and improving the overall system. Also the advantages of the advanced exer-
getic analysis over the conventional one have been demonstrated.
The advanced exergetic analysis uses some more than less arbitrary assumptions selected by the
designer to extract useful conclusions with respect to the improvement of the overall system. Important
information on the potential for improvement of components and on the interactions among components
is obtained through the advanced analysis. This information used carefully can assist the creativity of
the designer to provide improved solutions.

REFERENCES

Bejan, A., Tsatsaronis, G., & Moran, M. (1996). Thermal Design and Optimization. New York, NY:
John Wiley & Sons.
Cziesla, F., Tsatsaronis, G., & Gao, Z. (2006). Avoidable thermodynamic inefficiencies and costs in
an externally fired combined cycle power plant. Energy –. International Journal (Toronto, Ont.), 31,
1472–1489.
Kelly, S., Tsatsaronis, G., & Morosuk, T. (2009). Advances exergetic analysis: Approaches for splitting
the exergy destruction into endogenous and exogenous parts, Energy –. International Journal (Toronto,
Ont.), 34(3), 384–391.
Kotas, T. J. (1985). The exergy method of thermal plant analysis. Florida, USA: Krieger Publishing
Company.
Lazzaretto, A., & Tsatsaronis, G. (2006). SPECO: A systematic and general methodology for calculating
efficiencies and costs in thermal systems. Energy –. International Journal (Toronto, Ont.), 31, 1257–1289.
Lozano, M. A., Bartolomé, J. L., Valero, A., & Reini, M. (1994). Thermoeconomic diagnosis of energy
systems. Proceedings of the Florence World Energy Research Symposium, E. Carneval (Ed.), Florence,
Italy, 149-156.
Morosuk, T., & Tsatsaronis, G. (2005). Graphical models for splitting physical exergy. Proceedings of
ECOS-2005 “Shaping our future energy systems”, S. Kjelstrup, J.E. Hustad, T. Gundersen, A. Rosjorde,
& G. Tsatsaronis (Eds.), (Vol. 1), 377-384.
Morosuk, T., & Tsatsaronis, G. (2008). New approach to the exergy analysis of absorption refrigeration
machines. Energy –. International Journal (Toronto, Ont.), 33, 890–907.
Morosuk, T., & Tsatsaronis, G. (2009a). Advanced Exergetic Evaluation of Refrigeration Machines Using
Different Working Fluids. Energy –. International Journal (Toronto, Ont.), 34, 2248–2258.
Morosuk, T., & Tsatsaronis, G. (2009b). Advanced Exergy Analysis for Chemically Reacting Systems
– Application to a Simple Open Gas-Turbine System. International Journal of Thermodynamics, 12(3),
105–111.
Morosuk, T., & Tsatsaronis, G. (2011). Comparative evaluation of LNG-based cogeneration systems
using advanced exergetic analysis. Energy –. International Journal (Toronto, Ont.), 36(3), 3771–3778.

101

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

Petrakopoulou, F., Tsatsaronis, G., & Morosuk, T. (2011). Exergoeconomic Analysis of an Advanced
Zero Emission Plant. Journal of Engineering for Gas Turbines and Power, 133, 113001-1 - 113001-12.
Szargut, J., Morris, D. R., & Steward, F. R. (1988). Exergy Analysis of Thermal, Chemical, and Metal-
lurgical Processes. Berlin, Germany: Springer-Verlag.
Szargut, J., & Petela, R. (1965). Egzergia. Warsaw, Poland: Wydavnictwa Naukowo-Techniczne.
Tsatsaronis, G. (1984). Combination of Exergetic and Economic Analysis in Energy-Conversion Pro-
cesses. Proceedings of the European Congress “Energy Economics and Management in Industry”,
Algarve, Portugal (Vol. 1), Pergamon.
Tsatsaronis, G. (1999). Strengths and limitations of exergy analysis. In A. Bejan & E. Mamut (Eds.),
Thermodynamic Optimization of Complex Energy Systems (pp. 93–100). Aarhus, Denmark: Kluwer
Academic Publishers. doi:10.1007/978-94-011-4685-2_6
Tsatsaronis, G., & Morosuk, T. (2008). A general exergy-based method for combining a cost analysis
with an environmental impact analysis. Proceedings of the ASME International Mechanical Engineering
Congress and Exposition, Boston, Massachusetts, USA, files IMECE2008-67218 & IMECE2008-67219.
Tsatsaronis, G., & Morosuk, T. (2009). In A. Boyano (Ed.), Advances in exergy-based methods for
improving energy conversion systems. “Optimization using exergy-based methods and computational
fluid dynamics”, G.Tsatsaronis & A (pp. 1–10). Clausthal-Zellerfeld, Germany: Papierflieger Verlag.
Tsatsaronis, G., & Morosuk, T. (2010). Advanced exergetic analysis of a novel system for generating
electricity and vaporizing liquefied natural gas. Energy –. International Journal (Toronto, Ont.), 35,
820–829.
Tsatsaronis, G., & Morosuk, T. (2011). Advanced exergetic analysis of a refrigeration system for liq-
uefaction of natural gas. International Journal of Energy and Environmental Engineering, 1(1), 1–18.
Tsatsaronis, G., & Park, M. H. (2002). On avoidable and unavoidable exergy destructions and invest-
ment costs in thermal systems. Energy Conversion and Management, 43(9-12), 1259–1270. doi:10.1016/
S0196-8904(02)00012-2
Tsatsaronis, G., & Winhold, M. (1985). Exergoeconomic Analysis and Evaluation of Energy Conversion
Plants. Part I - A New General Methodology. Energy –. International Journal (Toronto, Ont.), 10(1),
69–80.
Valero, A., Correas, L., Zaleta, A., Lazzaretto, A., Verda, V., Reini, M., & Rangel, V. (2004). On the
thermoeconomic approach to the diagnosis of energy system malfunctions. Part 2. Malfunction defini-
tions and assessment. Energy –. International Journal (Toronto, Ont.), 29, 1889–1907.
Valero, A., & Torres, C. (1988). Algebraic thermodynamic analysis of energy systems. Proceedings of
approaches to the design and optimization of thermal systems AES 7 (pp. 13–23). New York: ASME.

102

Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

KEY TERMS AND DEFINITIONS

Avoidable Exergy Destruction: Difference between the total exergy destruction and the unavoidable
exergy destruction within one component.
Endogenous Exergy Destruction: Exergy destruction within one component when all other system
components are assumed to operate with exergetic efficiencies of 100%, and the total system produces
the same product.
Exergy Destruction: Refers to the exergy destroyed due to irreversibilities within a component (system).
Exergy Loss: Transfer of exergy from the overall system to its surroundings. This exergy transfer is
not further used in this or another system.
Exergy of Fuel: Represents the exergetic resources expended to provide the exergy of product.
Exergy of Product: Desired result achieved by a component (system) expressed in exergy terms.
Exergy: Maximum theoretical useful work obtainable from a system as this is brought into complete
thermodynamic equilibrium with the exergy reference environment while interacting only with this
environment.
Exogenous Exergy Destruction: Difference between the total exergy destruction and the endogenous
exergy destruction within one component.
Unavoidable Exergy Destruction: Part of the total exergy destruction within a component that can-
not be reduced due to technological limitations associated with the availability and costs of materials
and manufacturing methods regardless of the amount of investment.

103
Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

APPENDIX: NOMENCLATURE

COP = Coefficient of Performance [-]


E = exergy rate [W]
e = specific exergy [J/kg]
j = j th stream
k = k th component
m = mass flow rate [kg/s]
p = pressure [bar]
Q = heat rate [W]
r = r th component
T = temperature [ºC]
W = power [W]

Greek Symbols

ε = exergetic efficiency [%]


η = isentropic efficiency [-]

Superscripts

AV = avoidable
EN = endogenous
EX = exogenous
M = mechanical
mexo = mexogenous
T = thermal
th = theoretical
UN = unavoidable
Σ = sum of exergy destructions caused by the component being considered

Subscripts

D = exergy destruction
F = exergy of fuel
k = kth component
L = exergy loss
P = exergy of product
tot = overall system
0 = thermodynamic environment (reference state)

104
Advanced Exergetic Analysis is a Modern Tool for Evaluation and Optimization of Refrigeration Systems

Abbreviations

CM = compressor
EM = electrical motor
EX = expander
HE = heat exchanger
R = refrigerator

105
106

Chapter 4
New Mechanisms for
Cryogenic Solid-Gas
Sublimation Refrigeration:
Basic Principles and System Designs

Lin Chen
Tohoku University, Japan & Peking University, China

ABSTRACT
Sublimation is one phase change mechanism which usually happens under low-to-moderate temperatures
and at the same time large amounts of latent heat is absorbed or released. Low temperature sublimation
has been proposed in a lot of applications as one useful fast cooling/refrigeration mechanisms, such as
medical cooling, food engineering, chemical synthesis, domestic cooling and many industrial sectors.
In this brief chapter, the basic mechanisms of static sublimation process and sublimation two-phase
flows are clarified and analyzed first, which covers the theoretical and physical problems of sublima-
tion phase-change. Then the previous studies are classified into numerical modeling and experimental
verifications. Representative refrigeration systems are also introduced and compared in this chapter,
which may give useful indications for future innovations in this field. Future research focuses are also
summarized and proposed in this chapter.

INTRODUCTION

In recent years, sublimation flow and heat transfer has been proposed and utilized in real applications.
Compared with traditional vapor-compression thermodynamic refrigeration cycles, using solid-gas
sublimation flow can achieve more stable operation and probably higher heat recovery capacity, due to
the relative high latent heat of sublimation process. Therefore, recently the investigation of solid-gas
sublimation flows and its related application system designs has attracted a lot of research groups and
engineer from both scientific world and industrial sectors (Robertson, 1932; Nelson, 1942; Lester &
Somorjai, 1968; Eisenbraun et al., 1995; Michaelides & Lasek, 1987).

DOI: 10.4018/978-1-4666-8398-3.ch004

Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Indeed, sublimation is one of the phase change mechanisms happening everyday in the world. As
shown in Figure 1 (Aoki et al., 2002), dry ice sublimation is shown specifically for two different states
when immersed in liquid tank. Early studies around 1930s have begun to investigate the basic thermo-
physical nature (Robertson, 1932; Nelson, 1942). At that stage, the main topics were set around the
sublimation rate from the viewpoint of chemical reaction/engineering or chemical physics. Later, more
groups have studied the thermal equilibrium and near-equilibrium sublimation process, which may as-
sume the ideal condition of vacuum sublimation in order to obtain reasonable results with experiments.
As the sublimation substances such as Iodine, Naphthalene and Camphor were more and more used in
domestic and engineering field, more studies came out since 1970s (Somorjai, 1968; Davy & Branton,
1970). However, the sublimation mechanism and physics behind what is seen is still unknown. At the
same time, more studies has focused on the different factors that affect the sublimation rate of a crystal
or particle, where the structural and chemical arrangements are analyzed from surface vaporizing to
sublimation. Many groups have reported theoretical and experimental results measured for atomic crystal,
molecular crystal, ionic crystal and others, under congruent or non-congruent sublimation process (Lester
and Somorjai, 1968). Until recent years, there are still studies based on this general understanding of
sublimation physics and the main focus on theoretical analysis and methods of numerical developments
can be found (Schinzer & Kinzel, 1998; Smilauer & Vvedensk, 1995; Zhu et al., 2007; Krishnamurthy
et al., 1990; Latyshev et al., 1996). The current review study will focus on the basic historic development
of sublimation field and related cryogenic/refrigeration oriented applications in recent years.
However, the majority of those theories and experiments are based on strict assumptions of crystal
structure and surface vaporization/sublimation laws (Somorjai & Lester, 1967). Large deviations are
found between experiments and theoretical predictions, or from experiment to experiment. Since 1980s,
with the development of social technology, sublimation flow related scientific and application studies

Figure 1. Sublimation patterns of dry ice in liquid. (a) film-state sublimation (water of 25 ºC); (b)
nucleate-state sublimation (ethanol of 17 ºC)
(Aoki et al., 2002)

107

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

became activated again (Gershanik & Zeiri, 2010; Goldstein & Cho, 1995; Mendes, 1991). New ap-
plication demand from industry sectors, such as process cooling, refrigeration, domestic cooling, food
engineering, medical cooling, drying and engineering chilling/cleaning, has triggered a lot of new de-
velopments in this field. With the aid of recently developed precise experimental methods and advanced
computational method and capacity, a lot of new results have been reported (Eisenbraun et al., 1995).
First, new results for the basic sublimation process and physics of specific working fluids have been
accumulated. New thermo-dynamic cycles and mechanisms have been proposed for cryogenics. Sublima-
tion related cooling, drying, food engineering, medical engineering and other field are now experiencing
dramatically change with various innovations of apparatus manipulation and energy utilization. The de-
velopment of phase change flows will possibly change the future of cryogenic refrigeration/drying field.
Second, with the development of computation capacity, large scale and precise simulation of sublima-
tion process and the coupled solid-gas two-phase behaviors have become possible. Third, the physics of
particle sublimation and channel flow sublimation have been explored by asymptotic analysis. Fourth,
the related thermo-physical characteristics and various effects on the sublimation process can be clari-
fied. Fifth, new developments are made by using multi-scale analysis, which can cover the sublimation
process from macro scale particle to micro scale sublimation, so as to give the general picture during
the whole sublimation process (Garvin et al., 2007a, 2007b).
Major developments have also been made in other relevant fields and applications (Spur et al., 1999;
Gigiel, 1985; Gigiel et al., 1985). A lot of research groups have also proposed novel techniques for real
cryogenic refrigeration system by solid-gas sublimation flows. Some of them have been utilized in real
process in industry. In addition, it should be noted that recently the heat transfer aspect during sublima-
tion flow and the two-phase flow control has also become hot topics, giving rise to future development/
overlaps with heat transfer engineering and flow dynamics. The working fluid now mostly proposed for
sublimation design is dry ice (solid carbon dioxide) (Spur et al., 1999; Sherman, 2007). Indeed, dry ice
has a long history in cryogenic applications. Basically, some cryogenic refrigeration/cooling processes are
proposed and tested around the triple point of CO2 (Ttr = -56.6°C, Ptr = 0.518MPa). By low temperature
expansion process, cryogenic solid-gas two-phase flow of CO2 can be obtained. By sublimation of the
flowing solid-gas two-phase mixture, refrigeration in specific geometries and systems can be achieved
(Zhang & Yamaguchi, 2011). This kind of phase change heat transport in the ‘evaporator’ can maintain
the cooling temperature below -56.6°C with relative high heat recovery/chilling rate by the latent heat of
dry ice. In addition, the heat transfer characteristics of cryogenic solid-gas sublimation flow have been
reported in recent years (Zhang & Yamaguchi, 2011; Yamaguchi et al., 2008; Yamaguchi & Zhang, 2009;
Yamaguchi et al., 2011). Applications in food storage, medical cooling, drying, jet flow cleaning and
others can also be found, using the relative stable cooling capacity and environmental benign nature of
related fluids. Indeed, the sublimation flow and its performances in application systems have been one
promising topic in engineering field. Besides those applications mentioned above, it should be noted
that the solid-gas sublimation system are still found in astrophysics (Baer & Ambrosio, 1961; Malin
et al., 2011). It is another important application of sublimation system of CO2 (dry ice system) for the
climate change and planetary evolution of Mars and others, which may contribute to the understanding
of planet and geophysics.
Indeed, both the solid-gas sublimation mechanisms and application designs are now under intense
study. A lot of critical problems still need more systematic research efforts. The overall picture of
sublimation flow related areas are drafted in Figure 2, where the above-mentioned development and
potential applications are included. Following the basic outline from general theories to applications as

108

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 2. Basic picture of related topics, methods and applications of solid-gas sublimation flow

shown Figure 2, the current review study is mainly focused the major important topics and basic status
of each related methods/applications. Following Figure 2, the specific development of each branch/re-
lated topics and applications are discussed and compared in this review study. In the following sections,
the basic sublimation physics and solid-gas two-phase flows are introduced, and then recent numerical
and experimental methods are reviewed, especially for particle sublimation and related thermo-physical
analysis in recent years. It should be noted here that as dry ice is the one of the most popular material
now in use, this review study is more focused on it (and we also include some other sublimation materials
when necessary). Later, major developments in multi-scale characteristics and experimental apparatus of
sublimation flows are summarized. Application studies of solid-gas sublimation flow are also classified
in this paper, centering on cryogenics and cooling systems. In the summary part, the major developments
are generalized and possible future research topics are discussed.

BASIC PHYSICS OF SUBLIMATION FLOW DYNAMICS

Physics of Sublimation

Sublimation phenomena are often seen in nature and widely used in our society. The phase change
directly from solid states to gas states can be found when the temperature and pressure fall lower than
the triple point. As shown in Figure 3 (IIchi-Ghazaani & Parvin, 2011), phase change along the subli-
mation line (between the sublimation point and triple point) will directly happen between solid phase
and gas phase. From fast expanding of liquid states below -20ºC can also generate solid gas two-phase
flow if properly controlled below the triple point. This feature of CO2 fluid allows the possibility of

109

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 3. Phase diagram of CO2 fluid


(IIchi-Ghazaani and Parvin, 2011)

cryogenic operation or refrigeration below the triple point. Similarly, for other fluids, sublimation flow
also happens. Indeed, the solid to gas phase change is one simple change of substance physical state of
aggregation. From the physics viewpoint, the changes of distance between molecular or ions may lead
to thermo-dynamic changes, or deviations from balanced states, and if the molecules have velocities
larger than some critical values, the crystal bonds break and the whole structure will collapse into a new
phase. That process from solid state to gas state generally involves energy transportation between the
crystal and the ambient as the sublimation molecules have to overcome some activation energy values
before they go to form a new phase (Somorjai, 1968; Davy & Branton, 1970; Somorjai & Lester, 1967).
Based on this understanding, later theoretical analysis of such sublimation process then conveniently
assumed the activation ‘energy well’ and energy balance models. Such models took in the conception
of Arrhenius reaction mechanism and former researchers have established the basic models of sublima-
tion (Gershanik & Zeiri, 2010; Goldstein & Cho, 1995; Mendes, 1991; Eisenbraun et.al., 1995; Garvin
et al., 2007a, 2007b).
Sublimation (or vaporization) of solids has been the subject of many experimental and theoretical
investigations since late 19th century (Hertz, 1899). Later in the first half of 20th century, the study of
equilibrium sublimation and non-equilibrium sublimation were both carried out by many groups (Rob-
ertson, 1932; Nelson, 1942). The basic scientific discussions on sublimation process were reported,
where the basic physic process of sublimation phase change was introduced. Other literatures can be also
available for accounting of thermodynamic constants, vapor pressures, etc. The status of art for related
early developments can be found in review books of Pitzer & Brewer (1961) and Stull & Sinke (1956).
For early stage of sublimation study, heat transfer is considered only one of the factors that affect the
rate, like the crystal nature, imperfections, dislocations, surface conditions, non-equilibrium sublima-
tion, etc. As early as 1967, Somorjai & Lester (1967) have published their famous paper on the physical

110

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

description of general solid sublimation. From the chemical engineering viewpoint, they took the solid
evaporation process as one kind of chemical reaction from solid surface to another gas phase. In that
research, the principles of vaporization kinetics, evaporation experiments and different solids effects
and chemical reaction step analysis were made. The sublimation or so-called solid vaporization process
is viewed as one kind of chemical reactions and the vacuum sublimation rate is also expressed in the
Arrhenius equation of

(
J v =k[A]s =k 0[A]s exp -E * RT )
where k is the rate constant for vaporization, k0 is a constant related to the frequency of attempted motion
of vaporizing molecules over the energy barrier E*, R is the gas constant and T represents the temperature.
For the description of real sublimation rate and ideal dynamic equilibrium sublimation rate (the upper
limit for given temperature) one ratio coefficient is given as

α (T ) = J v (T ) J max (T ) .

For some substances that rate α can be close to 1.0 while for some others it may be orders below.
Somorjai (1968) has made a systematic study on the mechanism of sublimation, where the sublima-
tion rates for four kinds of solids with different crystal structures (sodium chloride, silver, arsenic and
cadmium sulfide) are discussed. It is also indicated that the crystal imperfections, surface conditions
and dislocations in crystals may affect the sublimation rate a lot, leading to large deviations between
theoretical predictions and real experiments.
For specific solid structures, many studies have focused on the effect of dislocations of crystals. Lester
and Somorjai (1968) have reported the sublimation rate of surface (100) into vacuum under different
dislocation densities, where they reported basic equations as:

J exp ≅ J ther =Peq (2πmkt )


-1 2
.

When large etch pit densities (greater than 1.0×107/cm2) had Jexp equal to maximum Jther while small
etch pit densities (~1.0×106/cm2) had rates lower by a factor of 2 to 2.5.
Charwat (1964) reported the effect of surface evaporation kinetics on the sublimation into a leading
edge boundary layer and found the coupling effects of high-speed boundary flow and sublimation effect.
After that, Charwat (1965) developed the coupling effects for effect of surface evaporation kinetics and
gave the basic relationships for leading edge sublimation process. Davy & Branton (1970) developed
the method of freeze-etch technique (FET) for surface morphology of solids that sublime below room
temperature. The FET method was dependent on electron microscopic examination of the replica of an
etched fracture plane in a frozen specimen. As very little was known about the morphological changes
that occur on sublimation surfaces during sublimation, that method provided new details for later inves-
tigations (as shown in Figure 4 (Davy & Branton, 1970).
The study of solid sublimation mechanisms from surface analysis and morphological viewpoint is
now still a hot topic as in the manufacturing of semi-conductor materials (such as cadmium telluride

111

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 4. Electron micrographs of ice surface (30,000). (a) as cleaved; (b) basal face, etched 1 minute at
-100ºC; (c) prism face, etched 1 minute at -100ºC; (d) prism face, 10 seconds at -90 ºC; (e) prism face,
3 seconds at -85ºC. (shadows are white; shadow direction from bottom to top in this figure)
(Davy & Branton, 1970)

and silicium) (Schinzer & Kinzel, 1998; Smilauer & Vvedensky, 1995; Zhu, et. al., 2007; Krishnamur-
thy et al., 1990; Latyshev et al., 1996). Those studies still follows the surface modeling method (or
so-called ‘solid on solid (SOS)’ model) to model molecular bean expiry process (Pagonabarraga et al.,
1994; Pimpinelli & Villain, 1994). The sublimation of surfaces is distinguished to have three regimes:
layer-by-layer, free step flow and hindered step flow. Some major models in this field may include basic
Burton-Cabrera-Frank model and others, which are based on the solid surface morphologies. Similar
with the method used in early 20th century, the surface condition and crystal structure are the main focus
in large numbers of recent publications (Tatarenko et. al., 1995; Shalev & Shtein, 2013; Curteanu et al.,
2014; Hikal & Weeks, 2014). However, in recent years, more detailed and model revisions are made
due to the greatly improved computational resources available.
Later, some experimental studies on sublimation have been developed. Love et al. (1963) reported
one apparatus for molecular distillation and vacuum sublimation. Johnson (1967) proposed several use-
ful techniques for micro and macro sublimation by capillary tube visualization design. Walter (1968)
established one apparatus for large-scale sublimation and measured the sublimation rate. After nearly
30 years, developments in large-scale evaporation and distillation still continued. For example, in 1995,
Eisenbraun et al. (1995) reported improved apparatus for evaporative-distillation and large-scale subli-
mation process. The use of high-resolution mass spectrometry in the measurement of thermodynamic
properties of metallic systems has also been proposed (Moore et al., 1975).
Other factors during sublimation have also been studied and go to more details in recent years. For
example, Krasuk & Plachco (1974) studied the effect of porous media solid sublimation structure with

112

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

high thermal conductivity. In more recent years, numbers of new developments for the study of sublima-
tion can be found in literature. For sublimation energy determination (Chen et al., 2009; Picciochi et al.,
2008; Brown & Feldman, 1966); for specific material sublimation investigations (Giustini et al., 1998;
Tricot et al., 2010; Gershanik & Zeiri, 2010; Goldstein & Cho, 1995; Mendes, 1991); other sublimation
models (Wu & Zhang, 2005) can also be found in literature.

Sublimation of Solid-Gas Particle Flow

Basic Study of Particle Sublimation and Flow Systems

Recent study of basic particle sublimation has its specific route different from traditional chemical re-
action viewpoint. The major focus of this review study is the dry-ice sublimation and its applications.
Spalding & Christie (1964) proposed one experimental study of the sublimation rate of solid carbon
dioxide at an axisymmetric stagnation point in a stream of air (airflow from 80ft/s to160ft/s, temperature
ranges from 68°F to 762°F). The experimental sublimation rates lay about 16% below the laminar mass
transfer theoretical predictions of Evans (1962) and Stewart & Prober (1962). Such kind of systems have
used airflow to form stagnation point, and the heat transfer with ambient cannot be clearly measured;
therefore the results show differences with predictions. In addition, it may be quite difficult if the heat
transfer condition becomes the measuring target.
In addition, some groups used immersed dry ice sublimation, leaving room for buffer liquid control.
Dhir group (Dhir et al., 1976; Dhir, 1980) discussed the sublimation behaviors of dry ice slab in water,
where stable film boiling and turbulent nature of gas film are found for the major control factors of
sublimation rate and heat transfer process. Other studies have also been carried out on CO2 thin film
sublimation processes by high-resolution laser method (Weida et al., 1996). Later, Aoki et al. (2002)
reported theoretical and experimental analysis of cylinder dry ice sublimation rate when rested in liquid
environment. It is found that the sublimation process can be classified to be Film-state and Nucleate-stage
(as shown in Figure 1). For Film-state sublimation, they set up two-layer-flow model and conducted
numerical simulation. It is found that dry ice gas and water contact surface temperature reach its lowest
value at the bottom point (or stagnation point) of the dry ice cylinder, where the water is also firstly
condensed to ice. Such kind of studies can be applied to the fast reactor of immersed cryogenic cool-
ing systems. Some of related system design will also be reviewed in the application section. The basic
sublimation model of K. Aoki’s study can be seen in Figure 5 (Aoki et al., 2002), where the vapor film
and transition boundary are noted. In Figure 5, it is also clearly shown that local non-equilibrium of
sublimation particle may affect the sublimation rate a lot for different locations. Detailed radial distribu-
tion of parameters can be found in Figure 6. Detailed analysis of local behaviors across different particle
scales should be considered in future studies.
For the effect analysis of ambient conditions, Iwasaka (2009) has reported one study using change-
able gradient magnetic field in dry ice sublimation. It is found that the magnetic field on the order of a
Tesla can possibly enhance the release of CO2 gas from dry ice. Also, it is found if other gases such as
CO2 or O2 are added in the system, the sublimation rate will be greatly changed (as shown in Figure 7,
which results may also contribute to the study of planetary evolution.
Sovershenny & Tirsky (1961) set up analytical equations for the analysis of static solid particle-gas
contact face sublimation process. In that model, solid particles are subjected to laminar gas flows; the
model is solved with vapor-gas mixture diffusion equations with the assumption of constant sublima-

113

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 5. Schematic diagram show the sublimation of dry ice in water


(Aoki et al., 2002)

tion rate. This kind of simplification model then became on typical model of particle sublimation flow
configurations. The effect of Prandtl number and gas flow Reynolds number are analyzed for different
initial conditions. However, in this model the propagation of sublimation interface and solid particle tem-
perature are also prescribed, leading to some doubts on the reasonability of the results. Further, Charwat
(1965) used near-equilibrium assumptions in sublimation process and focused on the thermodynamic
evolution of solid-gas interface. In this analytical model, the solving of continuous boundary layers and
boundary layer sublimation description are two of the major developments. Also in recent years, some
studies have touched the modeling of transient sublimation process of solids (Sahin & Dincer, 2000),
providing general analytical method for both the analysis of sublimation and melting process. It is stressed
that even the boundary thickness and boundary gradient is prescribed, such consideration of boundary
effects are advantageous compared to the previous methods.

Solid-Gas Sublimation Flow Dynamics

For the flow dynamics study of sublimation flow, models are generally based on Navier-Stokes equations
and phase coupling methods. For early stage studies, equilibrium or near-equilibrium conditions have
been mainly focused. Also in literature, major particle sublimation flow dynamics studies are based on
the flow dynamics of multi-phase flow researches (Ding et al., 2008; Dong et al., 2008a, 2008b; Subra-
maniam, 2013; Bi et al., 2000; Zaichik & Pershukov, 1996; Wang et al., 2011; Tsuji, 2007; Cangialosi et

114

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 6. Sublimation interface (a) and results comparison (b)


(Aoki et al., 2002)

al., 2006; Wu et al., 2006). Indeed particle sublimation flows will include the particle deformation and
mass transfer, momentum and energy transportation between phases (or the formation of new phase).
Therefore, the development of sublimation flow should be dependent on the basic methods in solid-gas
two-phase flows. Some major review articles of this field may be found in refs (Wang et al., 2010; van
der Hoef et al., 2006), where the multi-scale methods are also very indicative for the future development
of multi-scale sublimation flow system analysis. And here we may focus on some benchmark studies
in the sublimation flow field.

115

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 7. Effects of additional gas flow (He, O2, and CO2) under 4–5T gradient magnetic fields. The top
panel (table) shows a qualitative evaluation for a relative change in dry ice sublimation by providing
additional gas flows
(Iwasaka, 2009)

Michaelides group (Michaelides & Lasek, 1987) set up mathematical model for solid-gas sublima-
tion flows. Their model considered the non-equilibrium status of thermal and mechanical process in
sublimation flows and respectively wrote the continuous equation, momentum and energy equations.
By numerical calculation, they found that sublimation rate will increase with the increase of solid wall
temperature, or decrease of particle size. Also in that model analysis, it is found that with the increase
of gas temperature, the relative velocity between particle and gas phase will go up quickly. Further, the
increase of relative velocity, or the increase of turbulence inside the two-phase flow will in turn faster
the sublimation process. Michaelides & Liang (1992) reported the study of turbulence effect on particle
phase-change flows. Different from previously discussed equilibrium states (Michaelides & Lasek, 1987;
Michaelides, 1991), this study was focused on non-equilibrium transportation of mass and heat between
two phases. Basic particle change model and fluid dynamic correlations are set up between the solid and
gas phases so as to capture the detailed behaviors of sublimation flows inside channels. It is found in
that study that turbulence plays an important role in the phase change process and the evaporation rate
is increased by 25% when turbulence level is doubled with respect to normal condition.
Such kind of Navier-Stokes based fluid dynamic analysis has become one of the major models for
multi-phase flow simulations in recent years (Pomeroy & Male, 1991; Korovin, 1975). For low con-
centration flows with particle volume fraction less than 1%, very good agreement was found. In those
studies, the main target is sublimation rate prediction, leaving the related heat transfer problem less
discussed. In addition, those studies indeed have put forward another important topic of size effect of
the sublimation particles, which has already become one major challenging field of multi-scale analysis
of sublimation study.
Later, some further studies discussed the single particle sublimation flow inside channels. S.S. Sadhal
and his colleagues (Chung et al., 1984; Asavatesanupap & Sadhal, 2009) used perturbation method in
the numerical simulation of channel particle sublimation flow. In their model the radial flow is treated

116

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

as the leading order, while axial Pioseuille flow and particle motion is treated as higher order pertur-
bation (with nonlinear interactions of viscous and inertia forces considered). By perturbation method,
such nonlinear interactions can be simplified to linear equations and solved for approximate analytical
results. It is shown that when the particle size is large, friction factor will increase quickly with that of
Reynolds number; when the particle size is small, the friction is found equal to laminar Poiseuille flow.
Still, such model assumed that the particle only moves along with the centerline of the channel, and
the model assumed that the particle would never reach the vertical walls and no heat transfer happens
between them. Gan et al. (2003) also proposed one theoretical model for the analysis of single particle
sublimation evolution under channel gas flow conditions. Typical particle sublimation model can be
found in Figure 8. In Figure 8, vertical set channel geometry is assumed. The results of sublimation rate
(or particle changes) are plotted in Figure 9, which is typical evolution mode for the circular particle
geometry. That study systematically presented the effect of channel effect, particle geometry and Reyn-
olds number, Grashof number effects.
More recently, Naiman et al. (2010) used Large Eddy Simulation (LES) method in the simulation of
high-speed aircraft contrail problem and found the ice formation/sublimation process in related simula-
tion area. In the simulation, they used Lagrangian tracking method and found highly turbulent flow will
lead to higher contrail ice formation/sublimation.
Indeed, solid gas sublimation flow can be categorized to one kind of multi-phase flow condition
with sublimation heat and mass transfer. Though there already exists numbers of studies on multi-phase
flow, here we do not include the detailed review of them as they can easily be found in literature. The
inclusion of particle heat and mass transfer with the gas phase indeed brought about new challenges of
multi-phase flows. Until recently, less study has reported results on related analytical models for subli-
mation flows. Some of them have been listed in above sections. However, from general sublimation to
heat transfer calculations, the size effect, the interaction between solid particle and solid wall, between
solid particle and gas, and among solid particles, especially during transient mass diffusion and trans-
portation from particle to gas phases, have raised new critical problems in the modeling of sublimation
flows. Basically, if the model of single or several particle sublimation inside one channel is considered,
how is the interactions, accompanying with mass and heat changes? And if the particle goes through
mini-scale to vanishing small size, micro or nano particle for example, how is the sublimation rate and
heat transfer condition? Such kind of questions are now still under discussion and the method of multi-
scale analysis is in great demand.

Multi-Scale Analysis of Two-Phase Sublimation Flows

As reviewed, the series studies of E.E. Michaelides group (Michaelides & Lasek, 1987; Michaelides &
Liang, 1992) and others (Kucherov, 2000; Asavatesanupap & Sadhal, 2009) have used both the macro-
fluid dynamics models and solid particle size function. This kind of models should consider both the
macro-geometry of container or flow field and the particle volume changes, which may change from
mini-meter to micro or even in nanoscale. Indeed, this direction should be dependent on both the devel-
opment of physical sublimation theory and related precise experimental methods. The basic solid-gas
two-phase flow system analysis, especially multi-scale studies as discussed in section 2.2.2, then can
be seen as the fundamental stage for further development of sublimation flows. In multi-scale systems,
much more complex effects of scaling and transient transportation phenomena should be considered.
During such size changes, the sublimation rate of particle and heat transfer surface/energy demand will

117

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 8. Schematic model of particle sublimation flow configurations


Basic analytical model of Gan et al. (2003)

change a lot. At the same time, it has been shown that the local surface between solid particle and gas
phase experiences non-equilibrium process when mass transportation is considered. The local tempera-
ture and pressure field, most often with some kind of fluctuations at the surface and solid volume can
be seen, will greatly affect the sublimation process. Such effects will still be different from particle to
particle. Therefore, the current problem becomes one typical multi-scale problem, which must be solved
by using coupled methods that both satisfy the macro conservation laws and micro-scale transient evolu-
tions. Assumptions generally used in traditional studies, uniform particle temperature and diameter for
example, should be changed in future models.
Multi-scale analysis of sublimation process and its adverse process are both very useful for sublimation
flow studies. Besides the theoretical studies already mentioned in former parts, there are more and more
application-oriented investigations in recent developments. For example, several studies have focused

118

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 9. Nusselt number evolution of particle surface


(Gan et al., 2003)

on the local thermo-physical changes and parameter behaviors of sublimation system (Siddiqi et al.,
2009). Numerically, Garvin et al. (2007a, 2007b) have studied the methodology and pushing engulfment
transition of from dynamics of particle solidification process (as shown in Figure 10), which basically
used Navier-Stokes equations and the phase change from dynamics is systematically incorporated in
the numerical model. In Figure 10, the basic force field are shown. In Figure 11 respective numerical
parameters are indicated. Such models showed capability of incorporating the surface effect between
different phases. The pressure evolution near the contact surface is shown in Figure 12, which indicates
that in sublimation systems the local behaviors should be treated more carefully. Therefore the detailed
particle sublimation models are now clearly presented, which is quite different from previous mathemati-
cal model of simple two-phase flows (Gan et al., 2003). Many later studies followed a similar way of
phase change modeling. Still, some groups used molecular dynamics and hybrid modeling method and
discussed the solid-liquid surface transition in multiscale channels from nanometer to micrometer (Xu
and Li, 2007) (see Figure 13), which provide hybrid computation scheme and scaling effects from nano-
to micron. Still, there are studies on the inverse process of condensation/sublimation analysis, and the
knowledge of numerical methods can be referred (Nam & Song, 2007; Kochs et al., 1991, 1993), where
fixed grid systems are used, and volume-averaged properties and scheme are utilized. Other method

119

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 10. (a) Schematic of the overall system. (b) Schematic of the zoomed-in box in (a) illustrating the
different mechanisms involved in the particle-solidification front interaction
(Garvin et al., 2007a)

Figure 11. (c) Schematic of a solidification front moving towards and interacting with a particle. (d) A
zoomed in view of the solidification front in (c) interacting with the particle. The rectangular grid in (d)
represents the mesh. The bold line on the right denotes the “matching plane”
(Garvin et al., 2007a)

by Lattice Boltzmann Method (LBM), Molecular Dynamics, Mont Carlo simulation, or coupled Finite
Volume Method (FVM) and LBM method have also been proposed in similar multiscale phase change
problems (Zhou et al., 2010; Luan et al., 2011; Pisipati et al., 2011; Ruan et al., 2012; Picciochi et al.,
2008; Götzendorfer et al., 2005; Fartaria et al., 2002). For more information, some review article on
recent developments in multiscale modeling of heat transfer and fluid flow systems can be referred (Tao
& He, 2009).

120

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 12. Pressure distribution in the gap for a 1μm particle approaching a plane wall at 500 μ/s at
dmin/Rp= 0.02 (no disjoining effects included)
(Garvin et al., 2007a)

HEAT TRANSFER CHARACTERISTICS OF SOLID GAS SUBLIMATION FLOW

Heat Transfer Characteristics and Challenges of Sublimation Flow

For early stage of sublimation study, heat transfer is considered only one of the factors instead of one
target. Later, the heat transfer effect, or the heat transfer behaviors during sublimation flow has been
more and more studied.
Kern (1956) has reported one study into the basic process of sublimation heat of carbon sublimation.
In early 1960s, Baer and Ambrosio (1961) reported one study of heat conduction in a semi-infinite slab
with surface sublimation model study for planetary climate evolution. Later, Vitovec (1975) reported
the study of sublimation heat transfer surface in a fluidized bed configuration. Chickos (1975) also
proposed one equilibrium equation for determining of sublimation heat. Similar studies of sublimation
heat transfer in porous media can be found in literature (Fey and Boles, 1987, 1988), where analytical
models were developed for dried region and frozen region of the operation ranges. Indeed, for sublima-
tion conditions, the heat transfer behaviors have been found much different from traditional temperature
regions (Sumarokov et al., 2007).
Also there are studies that focused on the thermo-dynamic aspect of sublimation flows. Glass and
Klein (1959) have reported one study of third law thermo-dynamic study into sublimation process. And
also the entropy or enthalpy analyses of solid-gas sublimation systems have been reported from the
viewpoint of thermal dynamics and micro scale analysis (Madsen & Cotterill, 1981; Fernandes et al.,
2004; Perlovich & Raevsky, 2010; Puri et al., 2002; Monte et al., 2010; Gharagheizi et al., 2013; Aqra,
2014). Based on those understandings, the lower end of sublimation curves can be adjusted as discussed
by Diu et al. (2002) and others. Smolik & Vitovec (1983) analytically and experimentally analyzed the
transient heat mass transfer of naphthalene (solid)-He (gas) system. In their model, one-dimensional

121

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 13. The Couette flow geometry and the domain decomposition
(Xu and Li, 2007)

naphthalene (solid) - He (gas) system with ideal gas condition is assumed, only gravity, temperature and
concentration effects are considered. The model results show good agreement with previous experimental
data (when the mass transfer is controlled in a small range).
In recent years, more application oriented studies have been published as the sublimation technologies
has become more and more demanded in real applications. The basic heat transfer coefficient of a snow
bag has been studied by Veiga & Meyer (2002) (as shown in Figure 14). That experimental research is
focused on the prediction of heat transfer coefficient and then the amount/size of snow bag for a given
size of heat load. It is reported that the heat transfer coefficient of CO2 snow ranges from 5.15 W/m2·K
to 2.73 W/m2·K, the geometry and outside moist air may affect this value.
Zhang & Yamaguchi (2011) proposed one solid-gas sublimation system and tested it around the triple
point of CO2 (Ttr = -56.6°C, Ptr = 0.518MPa). Using a low temperature expansion process, cryogenic

122

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 14. Cross section of the snow bag


(Veiga and Meyer, 2002)

solid-gas two-phase flow of CO2 can be obtained. By sublimation of the flowing solid-gas two-phase
mixture, refrigeration can be achieved. This kind of phase change heat transport in the ‘evaporator’ can
maintain the cooling temperature below -56.6°C with relative high heat recovery rate by the latent heat
of dry ice. Recently, related refrigeration/cooling processes have been presented in some proposals or
simple real applications.
In Figure 15, a schematic of the CO2 refrigeration principle is shown, where refrigeration is achieved
by liquid CO2 expanded by throttling into two-phase fluid consisting of gas and solid, as shown for the
process a-b shown in Figure 15(a). By the CO2 expansion process, the CO2 solid-gas flow is obtained,
which is below -56.6°C. Figure 15(b) indicates the refrigeration flow by CO2 solid sublimation heat
transfer when flowing through a pipe, which is also shown in the b-c process in CO2 P-h diagram in Figure
15(a). Then by using the above methods, solid gas two-phase flow is visualized as shown in Figure 16,
where stable cryogenic environment is achieved by sublimation flow. In Figure 17, the horizontal heat
transfer results are shown for the test part. More application related studies of heat transfer behaviors
during sublimation flow will be discussed later in this paper.

Visualization and Other Experimental Studies

Yamaguchi & Zhang (2009) also developed visualization study of sublimation two-phase flow inside
channels. The basic visualization system is shown in Figure 18 and the basic results can be found in
Figure 19. Such systems can not only operate stably for single visualization process, but it also can be
utilized in cryogenic closed refrigeration systems as the dry ice sublimation flow was identified promis-
ing for low temperature heat transfer.
Multi-scale experiments of solid-gas sublimation process can be also found in other related studies.
Pitchimani et al. (2007) proposed Atomic Force Microscope for quantitative analysis of sublimation process.
In their experiments, the micro scale thermodynamic analysis can be made with detailed information on
solid evaporation process obtained (see Figure 20). It is clearly shown in Figure 20 that the particle and
surface evolution during the sublimation process, making it convenient for tracking of sublimation rate.
The effects of temperature and energy barriers are also analyzed in this study. Neumann et al. (2009)
have carried out experimental analysis for snow sublimation rate and heat transfer measurements. Their
experimental sublimation rates fall around 0.89 to 1.03 times that of theoretical predictions.

123

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 15. Schematic showing the principle of the new refrigeration method. (a) P-h diagram for CO2.
A-b process represents liquid CO2 expanding into solid-gas two-phase fluid flow, which goes through-
out the CO2 triple point of -56.6 °C. b-c is the sublimation process of the CO2 solid-gas fluid and (b)
schematic of the new evaporator using CO2 solid-gas two-phase fluid flow
(Zhang and Yamaguchi, 2011)

For the analysis of sublimation structure of solid, Varela et al. (2009) have made one experimental
study into the detailed formation and sublimation ice. Recently, Hong and Song (2007) have developed
one optical method for analyzing sublimation process of naphthalene. The basic sample plate and optical
system can be seen in Figure 21. By surface profile and weight analysis, the detailed evolution process
of sublimation then can be captured and explained. Different from the formerly introduced electron
microscope method (Davy & Branton, 1970), the current Atomic Force Microscope (Pitchimani et al.,
2007) and optical measurement system (Hong & Song, 2007) can provide clearer and more detailed

124

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 16. Schematic of the test section in the experimental system. (a) Diagram of the test section and
data acquisition; (b) Details of the test section and its measurement. Here, the test section is a copper-
made horizontal circular tube
(Zhang and Yamaguchi, 2011)

information of sublimation process. Other more direct measurement systems have also been introduced
and the basic information of sublimation and flow behaviors can be properly visualized (Zhang and
Yamaguchi, 2011).
In addition, sublimation related experimental systems have also been developed and tested by many
groups from physics and engineering fields in recent years. For example, the Field-Effect Transistors
(Roberson et al., 2005), situ-TEM (Yim et al., 2009), ultrahigh vacuum reflection electron microscopy
(Latyshev et al., 1996), are also developed for the study of sublimation systems.

125

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 17. Variations of the measured data of the outside wall temperature and Nusselt number along
the horizontal length x under the different heat fluxes, for the case of the condensation temperature of
-25 °C and the opening of the expansion valve of 15 mm. (a) temperature; (b) Nusselt number
(Zhang and Yamaguchi, 2011)

Also, for microscopic and macroscopic evaluation of sublimation heat transfer and related explora-
tions, based on the optical system, Kim & Song (2002) have also made a series of studies into the flow
field and thermo-dynamic aspect of naphthalene sublimation process. It is concluded that the boundary
surface layer can be most important factor for controlling of sublimation process and there exists optimal
region for sublimation local mass transfer. However, judging from the current status, more sophisticated
systems are still needed in the future for more details on both macroscale flow dynamics study and for
transient interface transportation visualizations. In one recent important study of Tang et al. (2006)
low energy electron microscopy was utilized to study the sublimation of atomic surface of Chromium,

126

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 18. Schematic of the experimental set-up made to investigate the feasibility of liquid CO2 expand-
ing into solid-gas two-phase flow by an expansion valve
(Yamaguchi et al., 2008)

where activation energy analysis and vacancy island evolutions are identified during the sublimation
process. It should be noted here that such kind of micro scale visualization and analyzing systems can
play especially important roles in the future development in this field.

APPLICATIONS OF DRY ICE SUBLIMATION FLOW

Thermal Cycles and Power Engineering

Yamaguchi & Zhang (2009) have proposed one cascade refrigeration cycles based on dry ice solid-gas
two-phase flow as shown in Figure 22. The cascade system is consisted of one high temperature cycle
and one low temperature cycle, in the low temperature side, stable refrigeration below -56.6°C can be
achieved. The proposed cascade systems are shown in Figure 4. In their series of studies, cascade refrigera-
tion system using this sublimation cooling process can successfully achieve a refrigeration environment
below -56.6 °C and in the high-pressure cycle, thermal energy above 100.0°C can be harvested. Based

127

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 19. Photographs of CO2 solid–gas two-phase fluid flow in the visualization section. CO2 solid–
gas flow at a CO2 flow velocity measured at (a) 2.50 m/s and (b) 0.78 m/s. For each, the left photograph
was taken at 4500 fps and the right photograph at 13,500 fps
(Yamaguchi et al., 2008)

on the thermodynamic calculation and real experiment, the averaged COP in low-pressure cycle is 2.45,
which show promising of such solid-gas flow based systems. According to the thermodynamic curves as
shown in Figure 23 more operation parameters can be tested and optimized. Later, Huang et al. (2008) put
forward similar models and theoretically analyzed the system feasibility and efficiency. They concluded
that such kind of refrigeration cycles may improve as high as 50% compared with traditional ways of
cryogenic refrigeration. Still, the need for temperature control in effusion experiments (and application
to heat of sublimation determination) has also been put forward since 1975 (Ginkel et al., 1975). The
major challenge for optimization of the dry ice sublimation refrigeration system is the solid-gas two-phase
flow control. Under some conditions, dry ice blockage phenomena occur. This dry ice sedimentation
or blockage will lead to system failures during operation and greatly lower the performance and system
safety. For example, in the study of Yamaguchi et al. (2011) the dry ice sedimentation and blockage
process are seen. Further development and analysis on the control (by expansion valves, pre-heating,
pressure control, heat supply rate, pipe diameter and heating length, etc.) of the heat exchanger in the
cooling side are needed, in order to achieve stable performance and better thermal efficiency.

128

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 20. Atomic force microscopy height images of (a) PETN film 24 h after depositing at room tem-
perature (b-i) after annealing at 30, 35, 40, 45, 50, 55, 60, and 65 °C (each for 12 h). All images are in
50 μm × 50 μm dimension
(Pitchimani et al., 2007).

Dry-Ice Particle Flow Fast Cooling

As early as 1920s, Dunn (1929) has reported the study of dry ice as one laboratory refrigerant. Later,
dry ice has also been proposed in tissues freezing (Dun, 1931). One typical cooler system design by dry
ice sublimation in a container is shown in Figure 24.
Eisenbraun & Lucas (1995) has developed one cooling design by tube sublimation process, good cool-
ing behaviors are identified for tube cooling areas. Later, the solid-gas two-phase flow inside channels has
also been tested for movable refrigerating/chilling applications. For example, IIchi-Ghazaani & Parvin
(2011) experimentally proposed one dry ice based sublimation heat exchanger. In the heat exchanger,
liquid phase CO2 can be directly converted into solid-gas flow inside the glycol bath and serve as the
refrigeration source. Operation tests show that high speed chilling can be achieved by controlling the
flow rate and pipe designs. The apparatus of the above system are shown in Figure 25. In that experiment,
the operation temperature, and flow rate and off design parameters can be tested. Relative good results
were obtained and shown in Figure 25(b), indicating more future innovations in related system designs.

129

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 21. Optical study of naphthalene sublimation. (a) naphthalene coating; (b) schematic diagram
of the optical experimental system
(Hong & Song, 2007)

Before the design of refrigeration cycles proposed by Zhang & Yamaguchi as discussed in Section
2.4 and 4.1, solid-gas sublimation cooler unit has already been tested by Dvornitsyn et al. (2004). In that
study, basic CO2 solid-gas sublimation flow for cooling systems are proposed and analyzed by visualiza-
tion and measuring of sublimation rate in a cubic chamber. The basic channel sublimation flow model
is shown in Figure 26 and in Figure 27 the basic chamber sublimation cooling behaviors are presented.
Such local behavior analysis in fast sublimation systems can be critical as in real applications the situa-
tions are usually non-homogeneous. Also it is found that the solid particles usually have micro-porous
structure and can be self-restructured during the sublimation flow. Therefore, it is concluded that such
flows can achieve stable flow even in closed refrigeration cycles. This research indicated the idea of
incorporating such pipe sublimation cooling into refrigeration cycles. This idea was later fulfilled in

130

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 22. Schematic of the CO2 cascade refrigeration system


(Yamaguchi & Zhang, 2009)

the developments of Zhang & Yamaguchi group (Zhang & Yamaguchi, 2011; Yamaguchi et al., 2008;
Yamaguchi & Zhang, 2009; Yamaguchi et al., 2011). Such experimental studies still found large deviations
from theoretical predictions, which is due to the perfect solid assumption and homogeneous assumption
in theoretical analysis. This series of studies concluded that solid sublimation mechanism can bring about
new opportunities and challenges in cryogenic refrigeration systems. Similar cooler systems by dry ice
sublimation flow can be also found in several other studies, such as the bottle-sublimation cycle cooler
(Dvornitsyn et al., 2006) (see Figure 28).

Medical Engineering and Other Related Jet Flow Applications

In medical engineering field, high-speed chilling and cooling technologies are also in great demand. This
category mainly involves the study of comfortable environment maintenance, surgery, medical instrument
cooling and others. Those applications may be based on previously discussed chilling processes. Dry ice
jet flow cleaning is also one promising kind of dry ice related technologies. This technology basically
sued pressurized air to blow the small size dry ice particles to the surface of substances the need to be
cleaned. By sudden jet flow with high momentum, cryogenic sublimation heat transfer, sudden expand-
ing (or so-called ‘micro explosion’ when dry ice particle enters into the gap of dirt parts, it’s volume
expanding to 600 times of solid state within only around 10-4s), this dry ice jet flow is very effective
for mechanical cleaning (Spur et al., 1999; Sherman, 2007). The dry ice particle generally used has the
particle diameter around 2-5 mm or smaller. This method is very simple, convenient and will not cause
secondary pollution or surface damages (Spur et al., 1999; Sherman, 2007). It has been widely used
in vacuum deposition, aerospace engineering, electric engineering, electronic process, biological and
medical process, small particle process (Edmund, 1994; Satoh et al., 2010), etc.

131

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 23. Graph of thermodynamic cycle of cryogenic CO2 for the experimental results
(Yamaguchi et al., 2011)

Food Engineering and Drying

In food engineering or drying process, dry ice is also widely used. In this field, dry ice has long being
used with simple designs or direct cooling. In recent years, some more sophisticated systems have been
developed. Indeed, for a long time, dry ice has been used for long distant fishing and food keeping
processes. Related research can be found in a lot of food conservation corporations around the world.
Scientific publication can be also easily found in literature. For example, the method of dry ice chilling
for hot boned meat was proposed in 1985 by Gigiel group (Gigiel, 1985; Gigiel et al., 1985), where the
author systematically arranged experiments on chilling process and the weight changed of boned meat
during storage and post-processing. For slab-shaped food products, based on the flat bed/plate sublima-
tion studies, new conjugate heat and mass transfer model has been developed for sublimation freeze
drying/keeping of food (Nam and Song, 2007). Also, the convective-diffusive CO2 protection (from dry
ice sublimation) of stored grain bulks has been reported by Alagusundaram et al. (1996).

132

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 24. Cooling system by dry ice sublimation


(Aoki et al., 2002)

For drying process, recent models can be found for fluidized beds using dry ice particle sublimation
process (Schlichthaerle & Werther, 2000). In that study, the influences of particle size, inner bed mate-
rial, temperature and superficial gas velocity are discussed into detail.
Indeed, As early in 1950s, sublimation from frozen state has been proposed in the dehydration of heat
sensitive materials, the important features of drying rates and commercial design were also discussed
(Beckett, 1951; Flosdorf, 1956; Greaves, 1946) (see Figure 29 and Figure 30). Theoretical analysis of
dehydration in continuum, transition and free molecular flow conditions has also been made by Hill &
Sunderland (1971).
Later in 1990s, several new designs of drying system by sublimation have been proposed. Kochs et
al. (1991, 1993) developed one channel flow freeze drying system and tested low Reynolds flow dry-
ing conditions. The concentration distribution and sublimation from surface are also monitored in this
experimental study.
In addition, since 1970s some studies also proposed direct dry ice personal cooling. For example,
Konz et al. (1974) tested the effect of dry ice conduction and sweat evaporation cooling for human skin,
where dry ice condition, shape and insulation are discussed. They found around 79 kcal/hr of cooling
benefit to human skin. Also, dry ice has been proposed in the chemical treatment of the Fight against
Ceratitis capitata (Velazquez-martin et al., 2006). In that study one system for insect storage and used
an external flow of air passing through a bed of solid carbon dioxide sublimated at -79°C, supplying
cooling and suitable storage environment for sterilized male insects and lead them to compete with wild
flies in mating.

133

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 25. Study of integrated dry ice chiller with direct feeding of liquid CO2 into the glycol bath through
the nozzles. (a) schematic of the experimental apparatus; (b) basic results (Minimum achievable water
temperature versus various VR for loadings of 10, 12.5 and 15 kg)
(IIchi-Ghazaani & Parvin, 2011)

134

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 26. Phase transition chain in non-adiabatic flow (L-liquid; S-solid; V-vapor)
(Dvornitsyn et al., 2004)

Carbon Capture and Sequestration (CCS) and Cold Energy Utilization

Sublimation flow process is also often seen in Carbon Capture and Sequestration (CCS) related projects,
as in the storage and transportation systems large pressure changes often happen. In June 2010, the UNEP
(United Nations Environment Programme) international conference on the Montreal Protocol was held in
Geneva, during which the problem of reducing hydrofluorocarbons (HFCs) and develop new generation
working fluids/systems acquired a new urgency. Basic CCS project drafts and transportation pipelines
and reception centers can be found in Figure 31.
Indeed, major procedures of the capture and sequestration are closely related with cryogenics and
sublimation/condensation process. In recent years, some groups have used the sublimation flow nature of
CO2 in post combustion CO2 capture process. For example, Tuinier et al. (2010) proposed one effective
CO2 separation method using the differences of gases in dew and sublimation points. In their system,
both H2O and CO2 can be separated from a flue gas simultaneously, circumventing costly pretreatment
steps or chemical absorbent or pressurize process. Later, Tuinier et al. (2011) developed the basic sys-
tems for continuous operating cryogenically cooled packed beds by numerical and experimental study.
Other groups (Schach et al., 2011; Mazzoldi et al., 2008; Song et al., 2012) have also proposed
similar system of CO2 capture by anti-sublimation process, so as to lower down the energy demand and
cost. For the transportation of captured CO2, there are also some studies that discuss the possible dry
ice formation and sublimation process under possible leakage conditions.

Planetary Study and Climate Evolution Research

Other major studies of dry ice sublimation generally focus on planetary climate evolutions of Mars.
Baer & Ambrosio (1961) reported one study of heat conduction in a semi-infinite slab with surface
sublimation model study for planetary climate evolution. Later some more observational studies and
analytical researches have also been published in literature (Harrison et al., 1968). In recent years, this

135

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 27. Phase evolution inside the sublimation chamber. (a) Liquid, monolithic (HS) and porous
solid refrigerant (HCPS) in low-temperature chamber (H); (b) Schematic of the solid phase advance
(Dvornitsyn et al., 2004)

sublimation dominated climate change mode is still a hot topic. For example, Byrne et al. (2003) reported
one sublimation model for Martian south polar ice features and presented the seasonal condition of CO2
sublimation on Mars. In 2011, Malin et al. (2011) reported observational evidence for an active surface
reservoir of dry ice on Mars and discussed its sublimation rate evolutions.
Detailed studies on the gas transport flows during sublimation of dry ice on Mars have also been
analyzed by several groups (Skorov et al., 2002; Thomas et al., 2011; Mangold, 2011; Brown et al.,
2012). Some of those studies have provided simple thermal-physical model analysis, however only the

136

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 28. Schematic diagram of cooler


(Dvornitsyn et al., 2006)

major observational results are generally accepted. Also, the dry ice pellets for cloud seeding or models
of large-scale climate changes due to sublimation particle transitions have also been proposed in literature
(Kochtubajda & Lozowski, 1985; Scarchilli et al., 2010; Volkov, 2012). Some review studies of astro-
physical evolution of solid-gas sublimation process may be of help in the understanding of the applica-
tions in those related fields (Fray and Schmitt, 2009). Due to the limitation of direct observation method,
the study in this field and large-scale sublimation system should demand more attention in the future.

FUTURE RESEARCH DIRECTIONS

For future studies, several directions need special attention as they are in urgent demand of innovations:

1. Clarification of new thermo-dynamic cycles and mechanisms proposed for cryogenics; Sublimation
related cooling, drying, food engineering, medical engineering and other techniques; though many
new systems are coming out in recent years, there still need systematic analysis and development
for such basic cooling under cryogenics;

137

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 29. Centrifuge assembly with ampoules containing frozen material


(Beckett, 1951)

2. Theoretical analysis and multi-scale study into the mechanisms of particle sublimation processes;
hybrid models and multi-scale analysis methods are in urgent demand;
3. Third, the physics of particle sublimation and channel flow sublimation has been explored by
asymptotic analysis.
4. Fourth, the related thermo-physical characteristics and various effects on the sublimation process;
5. Innovations in experimental system design and incorporation of disciplinary technologies, which
can help cover the sublimation process from macro scale particle to micro scale sublimation, so as
to give the general picture during the whole sublimation process.

CONCLUSION

In this brief review study, the basic status of art in solid-gas sublimation flow systems is analyzed. Several
critical problems involved in theoretical, numerical and experimental methods have been clarified. It is

138

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 30. Basic system design of a sublimation drying apparatus


(Beckett, 1951)

shown that from late 19th century, solid gas sublimation systems have been brought into daily use. Then
due to the limitation of computational capacity, the research basically focused on half-practical analysis
of sublimation rates and mechanisms. With the help of modern techniques, more detailed information on
sublimation process can be revealed in recent half century. Both numerical and experimental progresses
show that the solid-gas sublimation system is quite different from other phase change processes. Though
sublimation systems are now widely used in society and industry, the basic theoretical understanding of
the multi-scale problem is still limited. In addition, the explorations of new applications have already
introduced new challenging problems, when making the visualization and analyzing process more con-
venient than before. Indeed this topic would include multi-scale and multi-disciplinary investigations
in the future, considering the challenges in industrial world and also the interesting physics behind it.
Future studies and research challenges may include multi-scale system analysis and the specific stable
system design, so as to meet the ever-increasing demand from the industry.

139

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Figure 31. Schematic diagram of CCS projects


(IPCC 2005)

ACKNOWLEDGMENT

The support of National Science Foundation of China (No.51276001) and the Common Development
Fund of Beijing are gratefully acknowledged. The author also gives thanks for the warm and insightful
discussions with Prof. Xin-Rong Zhang in Peking University.

REFERENCES

Algusundaram, K., Jayas, D. S., Muir, W. E., & White, N. D. G. (1996). Convective-diffusive transport
of carbon dioxide through stored-grain bulks. Transactions of American Society -. Agricultural Engi-
neering, 39, 1505–1510.
Aoki, K., Sawada, M., & Akahori, M. (2002). Freezing due to direct contact heat transfer including sub-
limation. International Journal of Refrigeration, 25(2), 235–242. doi:10.1016/S0140-7007(01)00084-6
Aqra, F. (2014). Correlations for calculating the surface tension and enthalpies of sublimation of alkali
halides. Physica B, Condensed Matter, 441, 54–57. doi:10.1016/j.physb.2014.02.014

140

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Asavatesanupap, C., & Sadhal, S. S. (2009). Fluid dynamical analysis of a particle with large vapor
transport in poiseuille flow. Interdisciplinary Transport Phenomena, 1161, 268–276. PMID:19426325
Baer, D., & Ambrosio, A. (1961). Heat conduction in a semi-infinite slab with sublimation at the surface.
Planetary and Space Science, 10, 436–446. doi:10.1016/0032-0633(61)90150-7
Beckett, L. G. (1951). Freeze drying-vacuum sublimation. Journal of Scientific Instruments, 40(S1),
66–68. doi:10.1088/0950-7671/28/S1/319
Bi, H. T., Ellis, N., Abba, I. A., & Grace, J. R. (2000). A state-of-the-art review of gas-solid turbulent
fluidization. Chemical Engineering Science, 55(21), 4789–4825. doi:10.1016/S0009-2509(00)00107-X
Brown, J. S., & Feldman, J. L. (1966). Isotopic differences in the 0 °K volume and sublimation ener-
gies of solid neon and argon. Proceedings of the Physical Society, 89(4), 993–1001. doi:10.1088/0370-
1328/89/4/322
Brown, R. H., Lauretta, D. S., Schmidt, B., & Moores, J. (2012). Experimental and theoretical simulations
of ice sublimation with implications for the chemical, isotopic, and physical evolution of icy objects.
Planetary and Space Science, 60(1), 166–180. doi:10.1016/j.pss.2011.07.023
Byrne, S., & Ingersoll, A. P. (2003). A sublimation model for Martian south polar ice features. Science,
299(5609), 1051–1053. doi:10.1126/science.1080148 PMID:12586939
Cangialosi, F., Liberti, L., Notarnicola, M., & Stencel, J. (2006). Monte Carlo simulation of pneu-
matic tribocharging in two-phase flow for high-inertia particles. Powder Technology, 165(1), 39–51.
doi:10.1016/j.powtec.2006.03.009
Charwat, A. F. (1964). The effect of surface-evaporation kinetics on the sublimation into a boundary
layer, Memorandum of the Rand Corporation for United States Air Force. Project No. AF 49(638)-700.
Charwat, A. F. (1965). The effect of surface-evaporation kinetics on sublimation near the leading edge.
International Journal of Heat and Mass Transfer, 8(3), 383–394. doi:10.1016/0017-9310(65)90001-3
Chen, H. (2009). A new method for determination of sublimation energy. Chinese Physics Letters, 26(4),
046401. doi:10.1088/0256-307X/26/4/046401
Chickos, J. S. (1975). A simple equilibrium method for determining heats of sublimation. Journal of
Chemical Education, 52(2), 134. doi:10.1021/ed052p134
Chung, J. N., Ayyaswamy, P. S., & Sadhal, S. S. (1984). Laminar condensation on a moving drop: Singular
perturbation technique. Journal of Fluid Mechanics, 139(-1), 105–130. doi:10.1017/S0022112084000288
Curteanu, S., Smarandoiu, M., Horoba, D., & Leon, F. (2014). Naphthalene sublimation. Experiment
and optimisation based on neuro-evolutionary methodology. Journal of Industrial and Engineering
Chemistry, 20(4), 1608–1611. doi:10.1016/j.jiec.2013.08.007
Davy, G., & Branton, D. (1970). Subliming ice surfaces: Freeze-etch electron microscopy. Science,
168(3936), 1216–1218. doi:10.1126/science.168.3936.1216 PMID:17843591
Dhir, V. K. (1980). Sublimation of a horizontal slab of dry ice: An analog of pool boiling on a flat plate.
ASME Journal of Heat Transfer, 102(2), 380–382. doi:10.1115/1.3244297

141

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Dhir, V. K., Castle, J., & Catton, I. (1976). Role of Taylor instability on sublimation of a horizontal slab
of dry ice. Journal of American Society of Mechanical Engineers, 76, 12.
Ding, Y., He, Y., Cong, N. T., Yang, W., & Chen, H. (2008). Hydrodynamics and heat transfer of gas-
solid two-phase mixtures flowing through packed beds-a review. Progress in Natural Science, 18(10),
185–1196. doi:10.1016/j.pnsc.2008.03.023
Diu, B., Guthmann, C., Lederer, D., & Roulet, B. (2002). The lower end of the sublimation curve. Eu-
ropean Journal of Physics, 23(2), 205–212. doi:10.1088/0143-0807/23/2/314
Dong, W., Wang, W., & Li, J. (2008a). A multiscale mass transfer model for gas-solid riser flows: Part
I - Sub-grid model and simple tests. Chemical Engineering Science, 63(10), 2798–2810. doi:10.1016/j.
ces.2008.01.038
Dong, W., Wang, W., & Li, J. (2008b). A multiscale mass transfer model for gas-solid riser flows: Part
II – Sub-grid simulation of ozone decomposition. Chemical Engineering Science, 63(10), 2811–2813.
doi:10.1016/j.ces.2008.01.039
Dun, F. L. (1931). The use of solid carbon dioxide (Dry-Ice) in the freezing of tissues for microtome
sectioning. The Journal of Laboratory and Clinical Medicine, 16, 627–628.
Dunn, S. (1929). The use of ‘dry ice’ or solid carbon dioxide as a laboratory refrigerant. Science, 69(1787),
359–360. doi:10.1126/science.69.1787.359 PMID:17784462
Dvornitsyn, A., Naer, V., & Rozhentsev, A. (2004). Autostructurizing solid phase of a refrigerant as a
multi-functional low-temperature unit of a cooler. International Journal of Refrigeration, 27(5), 484–491.
doi:10.1016/j.ijrefrig.2004.03.014
Dvornitsyn, A., Naer, V., & Rozhentsev, A. (2006). Experimental investigation of a bottle-sublimation
cooler. International Journal of Refrigeration, 29(1), 101–109. doi:10.1016/j.ijrefrig.2005.05.013
Edmund, F. (1994). Cryogenesis-the art of cleaning with dry ice. Asia Engineer, 22, 14–15.
Eisenbraun, E. J., Cagle, M. D., Lucas, J. M., & Denton, T. T. (1995). Large-scale sublimation and
evaporative distillation using an improved apparatus. Journal of Chemical Education, 72(5), 476.
doi:10.1021/ed072p476
Eisenbraun, E. J., & Lucas, J. M. (1995). Cooling-through condenser for sublimation tubes. Journal of
Chemical Education, 72(11), 1056. doi:10.1021/ed072p1056
Evans, H. L. (1962). Mass transfer through laminar boundary layers. International Journal of Heat and
Mass Transfer, 5(5), 373–407. doi:10.1016/0017-9310(62)90027-3
Fartaria, R. P. S., Fernandes, F. M. S. S., & Freitas, F. F. M. (2002). Monte Carlo Simulation of the Phase
Diagram of C60 Using Two Interaction Potentials. Enthalpies of Sublimation. The Journal of Physical
Chemistry B, 106(39), 10227–10232. doi:10.1021/jp026170a
Fernandes, F. M. S. S., Freitas, F. F. M., & Fartaria, R. P. S. (2004). Phase Diagram and Sublima-
tion Enthalpies of Model C60 Revisited. The Journal of Physical Chemistry B, 108(26), 9251–9255.
doi:10.1021/jp049399u

142

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Fey, Y. C., & Boles, M. A. (1987). An analytical study of the effect of convection heat transfer on the
sublimation of a frozen semi-infinite porous media. International Journal of Heat and Mass Transfer,
30(4), 771–779. doi:10.1016/0017-9310(87)90207-9
Fey, Y. C., & Boles, M. A. (1988). Analytical study of vacuum-sublimation in an initially partially filled
frozen porous medium with re-condensation. International Journal of Heat and Mass Transfer, 31(8),
1645–1653. doi:10.1016/0017-9310(88)90277-3
Flosdorf, E. W. (1956). Freeze Drying. New York: Reinhold Publishing Corporation.
Fray, N., & Schmitt, B. (2009). Sublimation of ices of astrophysical interest: A bibliographic review.
Planetary and Space Science, 57(14-15), 2053–2080. doi:10.1016/j.pss.2009.09.011
Gan, H., Feng, J. J., & Hu, H. H. (2003). Simulation of the sedimentation of melting solid particles.
International Journal of Multiphase Flow, 29(5), 751–769. doi:10.1016/S0301-9322(03)00035-1
Garvin, J. W., Yang, Y., & Udaykumar, H. S. (2007a). Multiscale modeling of particle-solidification
from dynamics, Part I: Methodology. International Journal of Heat and Mass Transfer, 50(15-16),
2952–2968. doi:10.1016/j.ijheatmasstransfer.2006.12.031
Garvin, J. W., Yang, Y., & Udaykumar, H. S. (2007b). Multiscale modeling of particle-solidification
from dynamics, Part II: Pushing-engulfment transition. International Journal of Heat and Mass Transfer,
50(15-16), 2969–2980. doi:10.1016/j.ijheatmasstransfer.2006.12.022
Gershanik, A. P., & Zeiri, Y. (2010). Sublimation Rate of TNT Microcrystals in Air. The Journal of
Physical Chemistry A, 114(47), 12403–12410. doi:10.1021/jp105168h PMID:21050019
Gharagheizi, K., Ilani-Kashkouli, P., Acree, W. E. Jr, Mohammadi, A. H., & Ramjugernath, D. (2013). A
group contribution model for determining the sublimation enthalpyof organic compounds at the standard
reference temperature of 298 K. Fluid Phase Equilibria, 354, 265–285. doi:10.1016/j.fluid.2013.06.046
Gigiel, A. J. (1985). Chilling hot boned meat with solid carbon dioxide. International Journal of Refrig-
eration, 8(2), 91–96. doi:10.1016/0140-7007(85)90079-9
Gigiel, A. J., Swain, M. V. L., & James, S. J. (1985). Effects of chilling hot boned meat with solid carbon diox-
ide. International Journal of Food Science & Technology, 20(5), 615–622. doi:10.1111/j.1365-2621.1985.
tb01820.x
Ginkel, van C.H.D, Kruif, de C.G., & Waal, de F.E.B (1975). The need for temperature control in effu-
sion experiments (and application to heat of sublimation determination). Journal of Physics E: Scientific
Instruments, 8, 490.
Giustini, A., Brunetti, B., & Piacente, V. (1998). A sublimation study of Lindane. Journal of Chemical
& Engineering Data, 43(3), 447–450. doi:10.1021/je970264+
Glass, S. J., & Klein, M. J. (1959). Sublimation and the third law of thermodynamics. Physica A, 25,
277–280.
Goldstein, R. J., & Cho, H. H. (1995). A review of mass transfer measurements using naphthalene subli-
mation. Experimental Thermal and Fluid Science, 10(4), 416–434. doi:10.1016/0894-1777(94)00071-F

143

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Götzendorfer, A., Kreft, J., Kruelle, C. A., & Rehberg, I. (2005). Sublimation of a Vibrated Granular
Monolayer: Coexistence of Gas and Solid. Physical Review Letters, 95(13), 135704. doi:10.1103/Phys-
RevLett.95.135704 PMID:16197150
Greaves, R. I. N. (1946). Preservation of proteins by drying. Special Reports Series, No. 258. London:
H.M. Stationery Office.
Harrison, H., Scattergood, D. M., & Shupe, M. R. (1968). The condensation and sublimation of CO2
with H2O: Carbonic acid on Mars? Planetary and Space Science, 16(4), 495–499. doi:10.1016/0032-
0633(68)90163-3
Hertz, H. R. (1899). The Principles of Mechanics Presented in a New Form. London: Macmillan Press.
Hikal, W. M., & Weeks, B. L. (2014). Sublimation kinetics and diffusion coefficients of TNT, PETN, and
RDX in air by thermo gravimetry. Talanta, 125, 24–28. doi:10.1016/j.talanta.2014.02.074 PMID:24840410
Hill, J. E., & Sunderland, J. E. (1971). Sublimation-dehydration in the continuum, transition and
free-molecular flow regimes. International Journal of Heat and Mass Transfer, 14(4), 615–638.
doi:10.1016/0017-9310(71)90011-1
Hong, K., & Song, T. H. (2007). Development of optical naphthalene sublimation method. International
Journal of Heat and Mass Transfer, 50(19-20), 3890–3898. doi:10.1016/j.ijheatmasstransfer.2007.02.017
Huang, D., Ding, G., & Quack, H. (2008). New refrigeration system using CO2 vapor-solid as refrigerant.
Frontiers of Energy and Power Engineering in China, 2-4(4), 494–498. doi:10.1007/s11708-008-0070-x
Ilchi-Ghazaani, M., & Parvin, P.IIchi-Ghazaani. (2011). Characterization of a dry ice heat exchanger.
International Journal of Refrigeration, 34(4), 1085–1097. doi:10.1016/j.ijrefrig.2011.02.015
IPCC. (2005). IPCC Special Report of Carbon Dioxide Capture and Storage. Cambridge, UK: Cam-
bridge University Press.
Iwasaka, M. (2009). Effects of gradient magnetic fields on CO2 sublimation in dry ice, J. Physics: Con-
ference Series - 3rd Int. Workshop on Materials Analysis and Processing in Magnetic Fields (MAP3),
156, 012029.
Johnson, B. L. (1967). Useful techniques for micro and macro sublimation. Journal of Chemical Educa-
tion, 44(10), 608. doi:10.1021/ed044p608
Kern, D. M. (1956). The heat of sublimation of carbon. Journal of Chemical Education, 33(6), 272.
doi:10.1021/ed033p272
Kim, J. Y., & Song, T. H. (2002). Microscopic phenomena and macroscopic evaluation of heat transfer
from plate fins/circular tube assembly using naphthalene sublimation technique. International Journal
of Heat and Mass Transfer, 45(16), 3397–3404. doi:10.1016/S0017-9310(02)00047-9
Kochs, M., Korber, C., Heschel, I., & Nunner, B. (1993). The influence of freezing process on vapor
transport during sublimation in vacuum-freeze-drying of macroscopic samples. International Journal
of Heat and Mass Transfer, 36(7), 1727–1738. doi:10.1016/S0017-9310(05)80159-0

144

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Kochs, M., Korber, C., Munner, B., & Heschel, I. (1991). The influence of the freezing process on vapor
transport during sublimation in vacuum-freeze-drying. International Journal of Heat and Mass Transfer,
34(9), 2395–2408. doi:10.1016/0017-9310(91)90064-L
Kochtubajda, B., & Lozowski, E. P. (1985). The sublimation of dry ice pellets used for seeding. Climate
Journal of Climate Meteorology, 24(6), 597–605. doi:10.1175/1520-0450(1985)024<0597:TSODIP>
2.0.CO;2
Konz, S., Hwang, C., Perkins, R., & Borell, S. (1974). Personal cooling with dry ice. American Indus-
trial Hygiene Association Journal, 35(3), 137–147. doi:10.1080/0002889748507015 PMID:4821302
Korovin, V. Y. (1975). Acts of laser radiation on ice plate crystals. Proceedings of IEM, Obninsk, Kaluga
Region, 11(54), 34-40.
Krasuk, J. H., & Plachco, F. P. (1974). Sublimation from a porous solid of high thermal conductivity.
Chemical Engineering Science, 29(1), 181–189. doi:10.1016/0009-2509(74)85043-8
Krishnamurthy, S., Berding, M. A., Sher, A., & Chen, A. B. (1990). Semiconductor surface sublima-
tion energies and atom-atom interactions. Physical Review Letters, 64(21), 2531–2534. doi:10.1103/
PhysRevLett.64.2531 PMID:10041736
Kucherov, A. N. (2000). Sublimation and vaporization of an ice aerosol particle in the form of thin cylinder
by laser radiation. International Journal of Heat and Mass Transfer, 43(15), 2793–2806. doi:10.1016/
S0017-9310(99)00321-X
Latyshev, A. V., Minoda, H., Tanishiro, Y., & Yagi, K. (1996). Dynamical Step Edge Stiffness on the
Si(111) Surface. Physical Review Letters, 76(1), 94–97. doi:10.1103/PhysRevLett.76.94 PMID:10060442
Lester, J. E., & Somorjai, G. A. (1968). The effect of dislocations on the vaporization rate of NaCl single
crystals. Applied Physics Letters, 12(6), 216–217. doi:10.1063/1.1651959
Love, B., Snader, K. M., & Kormendy, M. F. (1963). Apparatus for molecular distillation and vacuum
sublimation. Journal of Chemical Education, 40(8), 426. doi:10.1021/ed040p426
Luan, H. B., Xu, H. X., Chen, L., Sun, D.-L., He, Y.-L., & Tao, W.-Q. (2011). Evaluation of the coupling
scheme of FVM and LBM for fluid flows around complex geometries. International Journal of Heat
and Mass Transfer, 54(9-10), 1975–1985. doi:10.1016/j.ijheatmasstransfer.2011.01.004
Madsen, J. U., & Cotterill, R. M. J. (1981). The Isochoric Entropies of Melting and Sublimation. Physica
Scripta, 24(6), 959–960. doi:10.1088/0031-8949/24/6/009
Malin, M. C., Caplinger, M. A., & Davis, S. D. (2011). Observational evidence for an active surface
reservoir of solid carbon dioxide on Mars. Science, 294(5549), 2146–2148. doi:10.1126/science.1066416
PMID:11768358
Mangold, N. (2011). Ice sublimation as a geomorphic process: A planetary perspective. Geomorphology,
126(1-2), 1–17. doi:10.1016/j.geomorph.2010.11.009

145

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Mazzoldi, A., Hill, T., & Colls, J. J. (2008). CO2 transportation for carbon capture and storage: Sublima-
tion of carbon dioxide from a dry ice bank. International Journal of Greenhouse Control, 2(2), 210–218.
doi:10.1016/S1750-5836(07)00118-1
Mendes, P. R. S. (1991). The naphthalene sublimation technique. Experimental Thermal and Fluid Sci-
ence, 4(5), 510–523. doi:10.1016/0894-1777(91)90031-L
Michaelides, E. E., & Lasek, A. (1987). Fluid-solids flow with thermal and hydrodynamic non-equilibrium.
International Journal of Heat and Mass Transfer, 30(12), 2663–2669. doi:10.1016/0017-9310(87)90148-7
Michaelides, E. E., & Lasek, A. (1991). Particulate flow with sublimation or evaporation and with
thermal and hydrodynamic non-equilibrium. International Journal of Heat and Mass Transfer, 34(3),
601–609. doi:10.1016/0017-9310(91)90109-R
Michaelides, E. E., Liang, L., & Lasek, A. (1992). The effect of turbulence on the phase change of drop-
lets and particles under non-equilibrium conditions. International Journal of Heat and Mass Transfer,
35(9), 2069–2076. doi:10.1016/0017-9310(92)90052-T
Monte, M. J. S., Sousa, C. A. D., Fonseca, J. M. S., & Santos, L. M. N. B. F. (2010). Thermodynamic
Study on the Sublimation of Anthracene-Like Compounds. Journal of Chemical & Engineering Data,
55(11), 5264–5270. doi:10.1021/je100850z
Moore, R. H., Robinson, D., & Argent, B. B. (1975). The use of high resolution mass spectrometry in
the measurement of thermodynamic properties of metallic systems. Journal of Physics. E, Scientific
Instruments, 8(8), 641–648. doi:10.1088/0022-3735/8/8/010
Naiman, A. D., Lele, S. K., Ham, F., Wilkerson, J. T., & Jacobson, M. Z. (2010). Large eddy simulation
of persistent contrails. DLR Deutsches Zentrum fur Luft Conference Paper 10, 191-196.
Nam, J. H., & Song, C. S. (2007). Numerical simulation of conjugate heat and mass transfer during
multi-dimensional freeze drying of slab-shaped food produces. International Journal of Heat and Mass
Transfer, 50(23-24), 4891–4900. doi:10.1016/j.ijheatmasstransfer.2007.08.004
Nelson, O. A. (1942). An improved sublimation apparatus. Industrial & Engineering Chemistry Re-
search, 14(2), 153.
Neumann, T. A., Albert, M. R., Engel, C., Courville, Z., & Perron, F. (2009). Sublimation rate and the
mass-transfer coefficient for snow sublimation. International Journal of Heat and Mass Transfer, 52(1-
2), 309–315. doi:10.1016/j.ijheatmasstransfer.2008.06.003
Pagonabarraga, I., Villain, J., Elkinani, I., & Gordon, M. B. (1994). Lattice effects in crystal evapora-
tion. Journal of Physics. A. Mathematical Nuclear and General, 27(6), 1859–1876. doi:10.1088/0305-
4470/27/6/013
Perlovich, G. L., & Raevsky, O. A. (2010). Sublimation of Molecular Crystals: Prediction of Sublimation
Functions on the Basis of HYBOT Physicochemical Descriptors and Structural Clusterization. Crystal
Growth & Design, 10(6), 2707–2712. doi:10.1021/cg1001946

146

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Picciochi, R., Lopes, J. N. C., & Diogo, H. P. (2008). Experimental and molecular dynamics simulation
study of the sublimation energetic of Cyclopentadienyltricarbonylmanganese (Cymantrene). The Journal
of Physical Chemistry A, 112(41), 10429–10434. doi:10.1021/jp805607d PMID:18800776
Pimpinelli, A., & Villain, J. (1994). What does an evaporating surface look like? Physica A, 204(1-4),
521–542. doi:10.1016/0378-4371(94)90446-4
Pisipati, S., Geer, J., Sammakia, B., & Murray, B. T. (2011). A novel alternate approach for multiscale
thermal transport using diffusion in the Boltzmann transport equation. International Journal of Heat
and Mass Transfer, 54(15-16), 3406–3419. doi:10.1016/j.ijheatmasstransfer.2011.03.046
Pitchimani, R., Burnham, A. K., & Weeks, B. L. (2007). Quantitative thermodynamic analysis of sub-
limation rates sing an atomic force microscope. Journal of Chemical Physics B, 111(31), 9182–9185.
doi:10.1021/jp073516e PMID:17630792
Pitzer, K. S., & Brewer, D. F. (1961). Thermodynamics. New York: McGraw-Hill Companies.
Pomeroy, J. W., & Male, D. H. (1992). Steady-state suspension of snow. Journal of Hydrology (Amster-
dam), 136(1-4), 275–301. doi:10.1016/0022-1694(92)90015-N
Puri, S., Chickos, J. S., & Welsh, W. J. (2002). Three-Dimensional Quantitative Structure−Property
Relationship (3D-QSPR) Models for Prediction of Thermodynamic Properties of Polychlorinated Bi-
phenyls (PCBs): Enthalpy of Sublimation. Journal of Chemical Information and Computer Sciences,
42(1), 109–116. doi:10.1021/ci010081y PMID:11855974
Roberson, L. B., Kowalik, J., Tolbert, L. M., Kloc, C., Zeis, R., Chi, X., & Wilkins, C. et al. (2005).
Richard Fleming, and Charles Wilkins, Pentacene Disproportionation during Sublimation for Field-Effect
Transistors. Journal of the American Chemical Society, 127(9), 3069–3075. doi:10.1021/ja044586r
PMID:15740146
Robertson, G. R. (1932). Sublimation. Journal of Chemical Education, 9(10), 1713. doi:10.1021/
ed009p1713
Ruan, C., Ouyang, J., & Liu, S. (2012). Multi-scale modeling and simulation of crystallization during
cooling in short fiber reinforced composites. International Journal of Heat and Mass Transfer, 55(7-8),
1911–1921. doi:10.1016/j.ijheatmasstransfer.2011.11.046
Sahin, A. Z., & Dincer, I. (2000). Analytical modeling of transient phase-change problems. International
Journal of Energy Research, 24(12), 1029–1039. doi:10.1002/1099-114X(20001010)24:12<1029::AID-
ER640>3.0.CO;2-J
Satoh, H., Uchida, K., Koizumi, A., . . .. (2010). Effective cleaning of Si by the dry ice blasting for future
dry process technology. IEEE Int. Symposium on Semiconductor Manufacturing Conference Proceed-
ings, Article no. 5750253.
Scarchilli, C., Frezzotti, M., Grigioni, P., De Silvestri, L., Agnoletto, L., & Dolci, S. (2010). Extraordinary
blowing snow transport events in East Antrctica. Climate Dynamics, 34(7-8), 1195–1206. doi:10.1007/
s00382-009-0601-0

147

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Schach, M. O., Oyzrzun, B., Schramm, H., Schneider, R., & Repke, J. U. (2011). Feasibility study of
CO2 capture by anti-sublimation. Energy Procedia, 4, 1403–1410. doi:10.1016/j.egypro.2011.02.005
Schinzer, S., & Kinzel, W. (1998). Modeling sublimation by computer simulation: Morphology-dependent
effective energies. Surface Science, 401(1), 96–104. doi:10.1016/S0039-6028(97)00916-3
Schlichthaerle, P., & Werther, J. (2000). Influence of the particle size and superficial gas velocity on
the sublimation of pure substances in fluidized beds of different sizes. Drying Technology, 18(10),
2217–2237. doi:10.1080/07373930008917839
Shalev, O., & Shtein, M. (2013). Effect of crystal density on sublimation properties of molecular organic
semiconductors. Organic Electronics, 14(1), 94–99. doi:10.1016/j.orgel.2012.09.033
Sherman, R. (2007). Carbon dioxide snow cleaning. Particulate Science and Technology, 25(1), 37–57.
doi:10.1080/02726350601146424
Siddiqi, M. A., Siddiqui, R. A., & Atakan, B. (2009). Thermal Stability, Sublimation Pressures, and
Diffusion Coefficients of Anthracene, Pyrene, and Some Metal β-Diketonates. Journal of Chemical &
Engineering Data, 54(10), 2795–2802. doi:10.1021/je9001653
Skorov, Y. V., Marov, M. Y., & Korolev, A. E. (2002). Mass transfer in the near-surface layer of a cometary
nucleus: A gas-kinetic approach. Solar System Research, 36(2), 99–109. doi:10.1023/A:1015254614701
Smilauer, P., & Vvedensky, D. D. (1995). Coarsening and slope evolution during unstable spitaxial
growth. Physical Review B: Condensed Matter and Materials Physics, 52(19), 14263–14272. doi:10.1103/
PhysRevB.52.14263 PMID:9980648
Smolik, J., & Vitovec, J. (1983). Transient heat and mass transfer in the two-phase system: Subliming
solid-vapor-gas mixture. International Journal of Heat and Mass Transfer, 26(7), 975–980. doi:10.1016/
S0017-9310(83)80122-7
Somorjai, G. A. (1968). Mechanism of sublimation. Science, 162(3855), 755–760. doi:10.1126/sci-
ence.162.3855.755 PMID:17794808
Somorjai, G. A., & Lester, J. E. (1967). Evaporation mechanism of solids. Progress in Solid State Chem-
istry, 4, 1–42. doi:10.1016/0079-6786(67)90004-0
Song, C. F., Kitamura, Y., Li, S.-H., & Ogasawara, K. (2012). Design of a cryogenic CO2 capture system
based on Stirling coolers. International Journal of Greenhouse Gas Control, 7, 107–114. doi:10.1016/j.
ijggc.2012.01.004
Sovershenny, V. D., & Tirsky, G. A. (1961). Sublimation of a solid near a critical point in flat and axi-
symmetrical gas flows. International Journal of Heat and Mass Transfer, 4, 119–126. doi:10.1016/0017-
9310(61)90067-9
Spalding, D. B., & Christie, F. A. (1964). Solid carbon-dioxide sublimation at an axisymmetric stagnation
point. International Journal of Heat and Mass Transfer, 8(4), 511–514. doi:10.1016/0017-9310(65)90041-4
Spur, G., Uhlmann, E., & Elbing, F. (1999). Dry-ice blasting for cleaning: Process, optimization and
application. Wear, 233, 402–411. doi:10.1016/S0043-1648(99)00204-5

148

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Stewart, W. E., & Prober, R. (1962). Heat transfer and diffusion in wedge flows with rapid mass-transfer.
International Journal of Heat and Mass Transfer, 5(12), 1149–1163. doi:10.1016/0017-9310(62)90191-6
Stull, D. R., & Sinke, G. C. (1956). Thermodynamic properties of the elements in their standard states.
Washington: American Chemical Society.
Subramaniam, S. (2013). Lagrangian-Eulerian methods for multiphase flows. Progress in Energy and
Combustion Science, 39(2-3), 215–245. doi:10.1016/j.pecs.2012.10.003
Sumarokov, V. V., Stachowiak, P., & Jeżowski, A. (2007). The peculiarities of heat transfer in CO2 and
N2O solids at low temperatures. Low Temperature Physics, 33(6), 595. doi:10.1063/1.2755197
Tang, S. J., Kodambaka, S., Swiech, W., Petrov, I., Flynn, C. P., & Chiang, T. C. (2006). Sublimation
of Atomic Layers from a Chromium Surface. Physical Review Letters, 96(12), 126106. doi:10.1103/
PhysRevLett.96.126106 PMID:16605934
Tao, W. Q., & He, Y. L. (2009). Recent advances in multiscale simulation of heat transfer and fluid flow
problems. Progress in Computational Fluid Dynamics, 9(3/4/5), 150–157. doi:10.1504/PCFD.2009.024813
Tatarenko, S., Daudin, B., & Brun, D. (1995). Zn and Te desorption from the (100) ZnTe surface. Ap-
plied Physics Letters, 66(14), 1773. doi:10.1063/1.113363
Thomas, N., Portyankina, G., Hansen, C. J., & Pommerol, A. (2011). Sub-surface CO2 gas flow in Mars’
polar regions: Gas transport under constant production rate conditions. Geophysical Research Letters,
38(8), L08203. doi:10.1029/2011GL046797
Tricot, S., Semmar, N., Lebbah, L., & Boulmer-Leborgne, C. (2010). ZnO sublimation using a polyener-
getic pulsed electron beam source: Numerical simulation and validation. Journal of Physics. D, Applied
Physics, 43(6), 065301. doi:10.1088/0022-3727/43/6/065301
Tsuji, Y. (2007). Multi-scale modeling of dense phase gas-particle flow. Chemical Engineering Science,
62(13), 3410–3418. doi:10.1016/j.ces.2006.12.090
Tuinier, M. J., & Annaland, M. (2010). Cryogenic CO2 capture using dynamically operated packed beds.
Chemical Engineering Science, 65(1), 114–119. doi:10.1016/j.ces.2009.01.055
Tuinier, M. J., Annaland, M. van S., & Kramer, G. J. (2011). A novel process for cryogenic CO2 capture
using dynamically operated packed beds-An experimental and numerical study. International Journal
of Greenhouse Gas Control, 5(4), 694–701. doi:10.1016/j.ijggc.2010.11.011
van der Hoef, M. A., Ye, M., van Sint Annaland, M., Andrews, A. T., Sundaresan, S., & Kuipers, J. A.
M. (2006). Multiscale modeling of gas-fluidized beds. Advances in Chemical Engineering, 31, 65–149.
doi:10.1016/S0065-2377(06)31002-2
Varela, G. G. A., Castellano, N. E., & Avila, E. E. (2009). Formation and sublimation of ice structures
over cylindrical collectors. International Journal of Heat and Mass Transfer, 52(21-22), 5167–5172.
doi:10.1016/j.ijheatmasstransfer.2009.05.002

149

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Veiga, da W.R., & Meyer, J.P. (2002). Heat transfer coefficient of a snow bag. International Journal of
Refrigeration, 25, 1043-1046.
Velazquez-martin, B., Gracia-Lopez, C., & Tordera-Tordera, C. (2006). A solid carbon dioxide (dry ice)
cooling system for the mechanized aerial release of sterile male Ceratitiscapitata. Transactions of the
ASABE, 49, 335–340. doi:10.13031/2013.20401
Vitovec, J. (1975). Heat transfer between a heated surface and a fluidized bed in sublimation. Chemical
Engineering Journal, 10(2), 235–239. doi:10.1016/0300-9467(75)80042-6
Volkov, A. N. (2012). Asymptotic thermal analysis of a rotating and sublimating sphere illuminated by
a light source. International Journal of Heat and Mass Transfer, 55(5-6), 1441–1456. doi:10.1016/j.
ijheatmasstransfer.2011.09.038
Walter, R. I. (1968). An apparatus for large-scale sublimation. Journal of Chemical Education, 45(8),
538. doi:10.1021/ed045p538
Wang, Q., Yang, Q., Wu, X., Luo, Z., & Cen, K. (2011). Research progress of particle rotation charac-
teristics in multi-phase flows. CIESC Journal, 62, 2381–2390.
Wang, W., Lu, B., Shi, Z., & Li, J. (2010). A review of multiscale CFD for gas-solid CFB modeling.
International Journal of Multiphase Flow, 36(2), 109–118. doi:10.1016/j.ijmultiphaseflow.2009.01.008
Weida, M. J., Sperhac, J. M., & Nesbitt, D. J. (1996). Sublimation dynamics of CO2 thin films: A high
resolution diode laser study of quantum state resolved sticking coefficients. Journal of Physical Chem-
istry, 105(2), 749–766. doi:10.1063/1.472814
Wu, B., Briens, L., & Zhu, J. X. (2006). Multi-scale flow behavior in gas-solids two-phase flow systems.
Chemical Engineering Journal, 117(3), 187–195. doi:10.1016/j.cej.2005.11.010
Wu, B., & Zhang, H. (2005). Isotropic and anisotropic growth models for the sublimation vapor
transport process. Modelling and Simulation in Materials Science and Engineering, 13(6), 861–873.
doi:10.1088/0965-0393/13/6/006
Xu, J., & Li, Y. (2007). Boundary conditions at the solid-liquid surface over the multiscale channel size
from nanometer to micron. International Journal of Heat and Mass Transfer, 50(13-14), 2571–2581.
doi:10.1016/j.ijheatmasstransfer.2006.11.031
Yamaguchi, H., Niu, X. D., Sekimoto, K., & Neksa, P. (2011). Investigation of dry ice blockage in an
ultra-low temperature cascade refrigeration system using CO2 as a working fluid. International Journal
of Refrigeration, 34(2), 466–475. doi:10.1016/j.ijrefrig.2010.11.001
Yamaguchi, H., & Zhang, X. R. (2009). A novel CO2 refrigeration system achieved by CO2 solid–gas
two-phase fluid and its basic study on system performance. International Journal of Refrigeration, 32(7),
1683–1693. doi:10.1016/j.ijrefrig.2009.05.003
Yamaguchi, H., Zhang, X. R., & Fujima, K. (2008). Basic study on new cryogenic refrigeration using
CO2 solid–gas two phase flow. International Journal of Refrigeration, 31(3), 404–410. doi:10.1016/j.
ijrefrig.2007.08.001

150

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Yim, J. W. L., Xiang, B., & Wu, J. (2009). Sublimation of GeTe Nanowires and Evidence of Its Size
Effect Studied by in Situ TEM. Journal of the American Chemical Society, 131(40), 14526–14530.
doi:10.1021/ja905808d PMID:19807191
Zaichik, L. I., & Pershukov, V. A. (1996). Problems of modeling gas-particle turbulent flows with
combustion and phase transitions [review]. Fluid Dynamics, 31(5), 635–646. doi:10.1007/BF02078213
Zhang, X. R., & Yamaguchi, H. (2011). An experimental study on heat transfer of CO2 solid-gas two
phase flow with dry ice sublimation. International Journal of Thermal Sciences, 50(11), 228–2234.
doi:10.1016/j.ijthermalsci.2011.05.019
Zhou, L., Xuan, Y., & Li, Q. (2010). Multiscale simulation of flow and heat transfer of nanofluid with
lattice Boltzmann method. International Journal of Multiphase Flow, 36(5), 364–374. doi:10.1016/j.
ijmultiphaseflow.2010.01.005
Zhu, R. S., Wang, J. H., & Lin, M. C. (2007). Sublimation of Ammonium Salts: A Mechanism Re-
vealed by a First-Principles Study of the NH4Cl System. The Journal of Physical Chemistry C, 111(37),
13831–13838. doi:10.1021/jp073448w

KEY ITEMS AND DEFINITIONS

Carbon Capture and Sequestration (CCS): Carbon capture and storage (CCS) (or carbon capture
and sequestration) is the process of capturing waste carbon dioxide (CO2) from large point sources, such
as fossil fuel power plants, transporting it to a storage site, and depositing it where it will not enter the
atmosphere, normally an underground geological formation.
Cold Energy: Cold Energy comes from the materials or environment that contains ‘cold’, or low
temperature. Many valuable industrial byproducts can be produced using cold energy of LNG. Also,
the ice remains frozen by way of its own cold energy, and it can be used to chill guests’ food throughout
the summer.
Cryogenics: Cryogenics is the study of the production and behaviour of materials at very low tem-
peratures (below −150 °C, −238 °F or 123 K).
Evaporation: Evaporation is a type of vaporization of a liquid that occurs from the surface of a liquid
into a gaseous phase that is not saturated with the evaporating substance. Evaporation that occurs directly
from the solid phase below the melting point, as commonly observed with ice at or below freezing or
moth crystals (napthalene or paradichlorobenzine), is called sublimation.
Multi-Scale Analysis: Multi-Scale Analysis means the solution or the investigation of a specific
problem that coves several time, spatial or other scales. Usually in physics study, the time scaling and
spatial scaling are the major challenges of a problem, such as the sublimation problem, the particles may
changes from micro to nano scale during the process.
Solid-Gas Two-Phase Flow: Solid-Gas Two-Phase Flow is one type of multi-phase flows that in-
cludes only solid phase and gas phase. Usually the situation is particle dispersed in a gaseous phase and
moves inside some kind of geometry.

151

New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Sublimation Dynamics: Sublimation Dynamics is the study of basic process, mass, momentum and
Energy transportation topics that involved in the sublimation process. The dynamic behaviors of sub-
limation surface, the movement of molecules and the basic dynamic testing of the sublimation system
parameters are also the main branches of the center topic.
Thermal Equilibrium: Thermal equilibrium is an axiomatic concept of classical thermodynamics.
It is an internal state of a single thermodynamic system, or a relation between several thermodynamic
systems connected by permeable walls. In thermodynamic equilibrium there are no net macroscopic
flows of matter or of energy, either within a system or between systems.

152
New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

APPENDIX: NOMENCLATURE

A = area (m2)
CCS = Carbon Capture and Sequestration
COPLPC = coefficient of performance, the ratio of cooling capacity in LPC to compressor work in LPC
COPSL = coefficient of performance, the ratio of cooling capacity in LPC to all the compressor works
d = diameter (m); gap width (m)
E* = activation energy (J)
FET = freeze-etch technique
H = height (m)
J = vaporization/sublimation rate (mol/(cm2·s))
k = rate constant for vaporization/sublimation (s-1); Boltzmann constant (J/K)
L = length (m)
LES = Large Eddy Simulation
Nu = Nusselt number
P = pressure (MPa)
Pr = Prandtl number
Q = heat flux input (W/m2)
R = universal gas constant (J/(mol·K)); Radius (m)
t = temperature (K)
T = temperature (°C)
u = velocity (m/s)
VR = volume ratio
Wh = compressor work in the high-pressure cycle (kW)
Wl = compressor work in the low-pressure cycle (kW)
x = horizontal length (m)
z = vertical length (m)
Δhlcool = cooling capacity in the low-pressure cycle (kJ/kg)
Δhlc1, Δhlc2 = heat exchanging quantities in the condenser of hot water, and the condenser of cool water,
in the low-pressure cycle, respectively (kJ/kg)
Δhlc1, Δhlc2 = heat exchanging quantities in the gas cooler of hot water, and the gas cooler of cool water,
in the high-pressure system, respectively (kJ/kg)

Greek Symbols

α = ratio coefficient
δ = boundary thickness (m)
ξ = non-dimensional vertical direction
η = non-dimensional radial direction
θ = angle (degree)
φ = angle (rad)

153
New Mechanisms for Cryogenic Solid-Gas Sublimation Refrigeration

Subscripts

0 = reference value
cr = critical point
eq = equilibrium state
exp = experimental value
i = interface value
l = liquid phase
max = max value
p = particle
s = static state
ther = theoretical value
tr = triple point
v = vaporization; vapor phase
w = wall value

154
155

Chapter 5
Ejection Refrigeration Cycles
Dariusz Józef Butrymowicz Jerzy Gagan
Białystok University of Technology, Poland Białystok University of Technology, Poland

Kamil Leszek Śmierciew Jarosław Karwacki


Białystok University of Technology, Poland Institute of Fluid-Flow Machinery of PASc,
Poland

ABSTRACT
The chapter presents the development of ejector refrigeration technology that strongly reduces the green-
house gases emission by using natural refrigerants and also dramatically reduces the need for the electric
power. This is accomplished by using free or inexpensive heat – either solar or waste heat, as the main
source of energy instead of electricity. Nowadays, the thermal driven refrigeration system, especially
with low-temperature heat source became more and more popular. The operation of the ejection cycle
using low-temperature heat source can be considered as very attractive and the ejection cycles becomes
truly competitive in comparison with the absorption refrigeration systems.

INTRODUCTION

An ejection refrigeration systems are a novel approach to the vapour compression cycle for air-conditioning
in buildings or even in mobile applications. The specific innovation in the ejection air-conditioning
system that:

• Utilizes natural refrigerants and therefore operates without any ozone depletion effects;
• Utilizes the solar or waste heat energy as a main source of energy;
• Eliminates a mechanical compressor, which is a main user of electricity.

While similar systems were proposed in the past (but mostly with CFC or HCFC refrigerants), none
of them has ever emerged as a commercial product, mainly due to difficulties in matching the ejector
geometry to the specific thermodynamic cycle. By eliminating the mechanical compressor from a cycle,
the ejection system has the potential of providing the same amount of refrigeration (cooling capacity)
with using only a fraction of electrical energy (for a pump only) as compared to conventional vapour

DOI: 10.4018/978-1-4666-8398-3.ch005

Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

Ejection Refrigeration Cycles

compression systems. This is due to a thermodynamic principle that compressing the liquid requires
substantially less work than compressing vapour. Less electricity results in a considerable reduction in
emission of carbon dioxide and other greenhouse gases. Additionally, by using the natural refrigerants
we are completely eliminating ozone layer depletion effect.
The schematic diagram of the ejection system is shown in Figure 1.
Momentum transfer from motive vapour (primary fluid) to vapour leaving the evaporator is the ba-
sis of functioning an ejector. The motive vapour passing through the nozzle undergoes expanding and
achieves large velocities. Expanded and the sped up driving vapour sucks vapour leaving the evaporator,
thanks to the momentum exchange between primary and secondary streams. Momentum transfer of mo-
tive vapour flowing from the generator is the driving force of ejection process. Therefore heat supplied
to the vapour generator is motive energy source for the discussed system.
The current direction in research in the area of refrigeration concentrates on decreasing and/or elimination
of adverse environmental effects. Refrigeration systems are the source of two types of pollution: ozone-
depletion from chlorofluorocarbon refrigerants and greenhouse gas emission from electricity production.
This creates the urgent need to replace the conventional refrigerants with environment-friendly working
fluids as well as applying renewable and non-polluting energies to run these systems. Approximately
15% of the world electrical consumption is used for refrigeration and air-conditioning applications, and
additionally, the demand for air-conditioning is proportional to solar radiation. Therefore the utiliza-
tion of solar energy is a logical way to meet the increasing demand for cooling and consequently, much
research has been conducted on this subject in recent years but it was concentrated on the absorption
cycle. Unfortunately, current absorption systems have low efficiency and require high temperatures to
regenerate the refrigerant, usually not achievable with presently used flat panel solar collectors.
There are limited available literature where experimental or numerical results of ejection cooling
cycles are presented. From the practical point of view COP is the crucial parameter describing the ejec-
tion cooling device. The basic feature of the ejection refrigeration systems is relatively low value of
COP in comparison with classical vapour compression systems. COP strongly depends on the working
fluid and operating parameters, i.e. condensation pressure, evaporation pressure, and vapour generation
pressure. Selection of the working fluid for the refrigeration or air-conditioning system is the crucial

Figure 1. Schematic diagram of ejector solar air-conditioning system

156

Ejection Refrigeration Cycles

problem because of the strong influence of the thermodynamic fluid properties on the system efficiency.
Moreover, the working fluid should fulfil the environment criteria such as zero ODP (Ozone Depletion
Potential) and as low as possible GWP (Greenhouse Warming Potential). Therefore the natural fluids
are thought as the best option.

VAPOUR JET APPARATUS

Ejector as the element of the cooling device is known from years, however appearing new refrigerant
influenced the growth of the application of ejection refrigeration cycles. Generally ejector is device to
transportation and compression of gases, vapours, liquid, mixtures and suspensions. In general two kind
of ejectors may be applied in refrigeration systems: gas (or vapour) one phase ejectors and two-phase
ejectors where liquid is the motive fluid and gas (vapour) is the secondary fluid. Concerning the ejection
refrigeration systems one phase vapour ejectors are considered. The classification of vapour jet apparatus
proposed by Sokolov & Zinger (1989) is presented in Table 1. In refrigeration systems in most cases the
vapour ejectors are applied with compression ratio πc > 2.5.
Schematic diagrams of vapour ejectors are presented in Figure 2. It is seen that in the case of the
conical-cylindrical mixing chamber the static pressure distribution along the ejector is different. The
typical example of the ejector of the geometry presented in Figure 2b is a steam ejector, e.g. jet water

Table 1. Classification of jet apparatus

Type Compression Ratio πc


Jet injectors < 1.2
Jet compressors 1.2 ÷ 2.5
Ejectors >2.5

Figure 2. Vapour ejectors and static pressure distributions: a) with cylindrical mixing chamber; b) with
conical-cylindrical mixing chamber

157

Ejection Refrigeration Cycles

chiller or jet vacuum pump to remove inert gases from steam condenser. In above applications compres-
sion ratio may πc = 4÷7 may be achieved. The characteristic feature of this type of ejectors is very high
expansion ratio of the primary fluid which can even reach the level of πe ≈ 500.
In most of ejection refrigeration systems ejectors of the cylindrical mixing chambers (Figure 2a and
Figure 3) may be applied. For example, for the isobutane refrigeration system dedicated for air-conditioning
the expansion ratio of the primary fluid is approximately πe ≈ 5 with the compression ratio πc = 4÷7.
The physical basis for the operation of the jet apparatus is momentum transfer between primary and
secondary vapour flows. In most cases the supercritical ejectors are applied, i.e. primary fluid flow is
supercritical at the motive nozzle outlet. Therefore de Laval nozzles are used for these apparatus. Vapour
parameters at the inlet to the motive nozzle are treated as stagnation parameters due to almost negligible
vapour velocity at motive vapour pipe. As an effect of expansion pressure decreases and velocity increases
reaching supercritical level wo at the nozzle outlet. The static pressure at the nozzle outlet po follows from
the geometry of the divergent part of the nozzle as well as velocity at the outlet wo. Secondary vapour
static pressure decreases following Bernoulli equation due to increasing velocity. It is assumed that static
pressure at the suction line is stagnation pressure pe0. For the fixed geometry of the motive nozzle there
exists the only one pressure pe = po for which outflow occurs with no shock waves inside or outside of
the nozzle. However, in most cases the operating parameters cannot be precisely fixed at required level
so the oblique waves or expansion focused waves are generated in the form of shock waves train of the
diamond shape. As a result of intensive momentum transfer due to velocity difference between primary
and secondary fluids occurs acceleration of the secondary fluid along with decrease of the static pres-
sure. Also deceleration of the motive vapour occurs. Supercritical flow transition to subcritical occurs
through a shock wave with increase of the static pressure and decrease of velocity. Vapour compression
in shock wave is the main mechanism of the compression that occurs in the ejector. Further compression
and vapour deceleration at the diffuser occurs as a result of the cross section area increasing and plays
minor role for the operation of the ejector.
The capacity of the ejector may be thought as both compression of the secondary fluid as well as
entrainment ratio. Therefore operation of the ejector is analysed as a relationship between compression
and entrainment ratios.

Figure 3. Schematic diagram of vapour ejector of the cylindrical mixing chamber

158

Ejection Refrigeration Cycles

Figure 4 presents regimes of the operation of the vapour ejector. This is relation of the entrainment
ratio on the backpressure which in the case of refrigeration systems is the condensation pressure pc for
fixed motive vapour pressure pg and suction pressure pe. It is seen in the Figure 4 that if the backpres-
sure is lower than the critical one pcr then the ejector operates under so called on-design conditions. In
this case both streams: primary and secondary ones are chocked so the ejector is double-chocked. This
means that the ejector operates under the maximum possible entrainment ratio. Therefore the change
of the backpressure does not influence the entrainment ratio if only pc ≤ pcr. Under these conditions
increase of the backpressure causes changes of the shock wave location towards the inlet to the mixing
chamber. If the backpressure exceeds the critical level pc > pcr then the secondary fluid flow is subcritical
and it means that entrainment ratio decreases. At certain level of the backpressure pco the ejector stops
the suction of the secondary vapour, i.e. U = 0. This is the limiting operating condition for the ejector.
Therefore under condition of pcr < pc < pco the operation of the ejector is called as off-design operation.
The difference pco - pcr depends on the geometry of the ejector as well as thermodynamic properties of
the working fluid. Further increase of the backpressure causes reverse flow of the vapour, i.e. malfunc-
tion of the ejector. It is worth to note that the maximum possible compression of the ejector is produced
under off-design conditions.

REFRIGERATION EJECTION CYCLE

The schematic diagram of the ejection refrigeration system is presented in Figure 5. The ejection cycle
consists of driving cycle and refrigeration cycle. In the ideal system there are no losses so the driving
vapour (primary fluid) at state 1 expand isentropically to state 2. During this process static pressure of
vapour decreasing while the velocity is increasing. Driving nozzles are designed this way, to get the
possibly largest velocity of fluid on the outlet from nozzle. Vapour with low pressure and large velocity

Figure 4. Regimes of operation of vapour ejectors

159

Ejection Refrigeration Cycles

Figure 5. Schematic diagram of ejection system

(state 2) sucks in vapour (secondary fluid) at state 7. Mixing of streams follows on certain section of
length of the ejector. Mixing is the result of momentum transfer between the streams. The effect of this
process is decreasing of velocity of primary fluid and increasing of velocity of secondary fluid. The
velocity drop causes the simultaneous static pressure rise. In spite of the velocity drop in most cases the
velocity of the mixture is supercritical after total mixing (state 3). In certain section of ejector the flow
changes from supercritical to subcritical. As an effect the shock wave will occur and static pressure rapidly
increases. In most cases the shock wave can occur at the inlet to the diffuser or at the outlet section of
the mixing chamber. The further compression follows in the diffuser as the consequence of enlargement
of the cross-section surface area. The mixture leaving ejector (state 4) has pressure corresponding to the
condensation pressure. After condensation liquid is split into two flow paths. The first one flows through
the throttling valve were it expands to the evaporation pressure (state 6), and the other one flows through
the pomp to the vapour generator achieves state 8. During the vaporisation in the generator, refrigerant
changes its state parameters from 8 to 1.
Estimation of the entrainment ratio is one of the most important task. The entrainment ratio is de-
fined as ratio of the secondary fluid to the primary fluid. Formulation of the model of the ejection cycle
requires to define the following operating parameters:

• Evaporation pressure pe; condensation pressure pc; motive pressure (generator pressure) pg;
• Temperature of the primary fluid at the outlet of the vapour generator Tg;
• Superheat of the secondary fluid at the outlet of the evaporator ∆Te;
• Subcooling of liquid refrigerant at the outlet of the condenser ∆Tl;
• Overall efficiency of the ejector.

Above mentioned assumptions allow unambiguously define the essential parameters of state (see
Figure 6) using the following relations:

160

Ejection Refrigeration Cycles

Figure 6. Defining the essential characteristic points of the ejection cycle

Point 1: T1 = Tg , p1 = pg , h1 = f (T1, p1 ), s1 = f (T1, p1 ) ;


Point 2: s1 = s2, p2 = pe ,T2 = f (s2, p2 ), h2 = f (s2, p2 ) ;
Point 5: p5 = pc ,T5 = f (p5, x = 0), h5 = f (T5, p5 ), s5 = f (T5, p5 ) ;
Point 6: h6 = h5, p6 = pe , x 6 = f (h6, pe ), s6 = f (x 6, p6 ) ;
Point 7: p7 = pe ,T7 = f (pe , x = 1), h7 = f (T7 , p7 ), s7 = f (T7 , p7 ) ;
Point 8: s 8 = s5, p8 = p1,T8 = f (s 8 , p8 ), h8 = f (s 8 , p8 ) .

The most difficult problem is to estimate the parameters of fluid inside ejector, i.e. inside the mixing
chamber and the diffuser. This is caused by simultaneous process of mixing and compression of both
streams. For determining of the parameters of fluid inside an ejector a specific model of the mixing
process must be assumed and then the ejection cycle will be adequate to the chosen mixing model.

Constant Pressure Mixing Model

The cycle of constant pressure mixing is presented in Figure 7. This is very classic approach. The model
assumed that the process of mixing and compression of streams, namely change parameters from state
2 to 4 proceeds independently. Basic assumptions for this type of model are:

161

Ejection Refrigeration Cycles

Figure 7. Ejector refrigeration cycle of constant pressure mixing

• Flow inside the ejector is steady state;


• Working fluid is treated as an ideal gas with constant specific heats;
• Velocities at motive and suction inlets are negligible;
• Compression and expansion processes are isentropic and mixing and compression losses are taken
into account by means of ejector efficiency determined experimentally;
• Mixing process occurs at constant pressure;
• Ejector is adiabatic;
• The shock wave occurs in the mixing chamber.

Motive vapour leaving the primary nozzle (state 2) sucks the secondary vapour (state 7). Both streams
are mixed at constant pressure in the mixing chamber (state 3) and compressed in the diffuser form state
3 to state 4. Energy conservation equation for the motive nozzle can be written as follows:

1 1
h1 + w12 = h2s + w22s . (1)
2 2

According to above mentioned assumption the inlet velocity of primary fluid w1 is negligible with
respect to outlet velocity w2 from the motive nozzle, thus the outlet velocity:

162

Ejection Refrigeration Cycles

w2s = 2(h1 − h2s ) . (2)

The efficiency of the driving nozzle:

h1 − h2
ηn = . (3)
h1 − h2s

Therefore the actual velocity at the nozzle outlet is:

w2 = 2ηn (h1 − h2s ) , (4)

where:

h2s = f (s2s , p2 ), s2s = s1, p2 = pe .

From eq. (1) and (4):

h2 = h1 − ηn (h1 − h2s ). (5)

The remaining parameters at state 2 are given by relations:

T2 = f (h2, p2 ), p2 = pe , s2 = f (T2, p2 ) .

The efficiency of the supercritical motive nozzle in most cases achieves η = 0.85÷0.90 (Paliwoda,
1971; Sun, 1997; Yu, Chen, Ren, & Li, 2006). Mass entrainment ratio U is defined as:

U = m e / m g . (6)

The momentum conservation equation is:

m g w2 + m ew 7 = (m g + m e )w 3 . (7)

Assuming that the inlet velocity of the secondary fluid w7 is negligible in relation to the velocity of
primary fluid leaving the motive nozzle w2, then the velocity of mixture w3 is given by:

w2
w3 = . (8)
1 +U

163

Ejection Refrigeration Cycles

The efficiency of mixing chamber ηm takes into account loss during mixing process, i.e. changes of
the profile of velocity, density, temperature, and pressure due to momentum transfer. This efficiency
may be defined as:

ηm = w 32 / w 32 , (9)

where w 32 is average velocity of the mixture in mixing chamber after mixing process. In most cases the
efficiency of the mixing chamber can achieve ηm = 0.85 ÷ 0.95 . The energy conservation equation for
mixing chamber has a form:

 w 2   w 2   w 2 
m g h2 + 2  + m e h7 + 7  = (m g + m e ) h3 + 3  . (10)
 2   2   2 

Enthalpy of mixture is equal to:

h2 + w 22 +Uh7 w 32
h3 = − , (11)
1 +U 2

then

T3 = f (h3, p3 ), p3 = pe , s 3 = f (T3, p3 ) .

Assuming isentropic compression in the diffuser, enthalpy of point 4 can be found as a function

h4s = f (s 4 = s 3, pc ) h4s = f (s 4 = s 3, pk ), kJ / kg  .

The diffuser efficiency is given by:

h 4s − h 3
ηd = , (12)
h4 − h3

which most often does not exceed ηd = 0.75, but there are a few publications where efficiency of dif-
fuser reaching ηd = 0.85 (e.g. Sun, 1996; Sun, Eames, & Aphornatana, 1996). Enthalpy of point 4 can
be found on the basis of energy conservation equation for the ejector:

m g h1 + m eh7 = (m g + m e )h4 . (13)

On the basis of eqs. (8), (9), (11), and (12) the entrainment ratio is:

164

Ejection Refrigeration Cycles

h1 − h2s
U = ηn ηm ηd −1 . (14)
h 4s − h 3

Capacity of the separate devices of the ejector cycle may be then calculated. For the evaporator:

Qe = m e (h7 − h6 ) , (15)

vapour generator:

Q g = m g (h1 − h8 ) , (16)

condenser:

Qc = (m g + m e ) (h4 − h5 ) , (17)

and circulating pump:

m g (pg − pc )
Pp = , (18)
ρ ' ηp

where: ρ ′ = ρ5 is the average density of liquid flowing through the pump; ηp is the overall efficiency of
the pump.
Coefficient of performance of the entire ejection cycle is defined as a ratio of cooling capacity and
energy necessary to generate of motive vapour including electric energy driving the pump:

Q
COP =  e . (19)
Qg + Pp

Ejection Cycle with Adiabatic Mixing

In the adiabatic mixing cycle simultaneous mixing and compression process is assumed that occurs in
the mixing chamber. Both of these processes occur adiabatically. This assumption corresponds to real
operation of the vapour ejector. However, this type of the cycle may not be thought as a classic approach
since it requires more complicated calculation procedure. The calculation approach of this type was
proposed by Paliwoda (1971). The actual and ideal operation of the ejector assuming adiabatic mixing
is presented in Figure 8.
In an ideal case motive vapour expands in the nozzle along the path 1-2s. The difference of enthalpy
∆h1 represents the specific work for this process. In the actual process of the motive vapour expansion

165

Ejection Refrigeration Cycles

Figure 8. The interpretation of isentropic and actual processes in ejector in h-s and p-h diagrams

the increase of entropy is present and the process runs along polytropic line 1-2, and representing this
process difference of enthalpy is ϕn 2∆h1 , where ϕn is the coefficient of velocity of the motive nozzle.
Expanded motive vapour (state 2) sucks in the stream of vapour of the state 7. Then mixing and com-
pression of streams to state 3 follows in the mixing chamber, and subsequently compression in the dif-
fuser to state 4 occurs. Compression process from pressure pe to pc causes the isentropic increase of
enthalpy ∆h2.
The isobars and isentrops in the p-h diagram are almost parallel in the pressures range pe ÷ pc, then
it is possible to assume that the isentropic increase of enthalpy is equal when s=const crosses by the
point 2s, 2, 3 or 7, see Figure 9. Therefore the real process proceeds along the lines 2 → 3 → 4 and
7 → 3 → 4 . The actual enthalpy rise is equal then:

1
∆h2r = ∆h2 . (20)
ϕd 2

This model defines a hypothetic ideal ejector of no any energy losses during the mixing process.
Moreover this model assumes that there are no any changes of the specific enthalpy during the mixing
process of both streams. Taking into above Paliwoda (1971) proposed to apply the following equation
describing the mixing process:

w22 w2 w2
m g + m e 7 = (m g + m e ) 3 . (21)
2 2 2

166

Ejection Refrigeration Cycles

Figure 9. Graphic interpretation of entrainment ratio in an ideal and real ejector

Assuming that w7 = 0 and expressing velocities w1 i w4 as the isentropic difference of enthalpy (w2
= 2∆h) the theoretical entrainment ratio can be found:

∆h1
Ut = −1 = χ −1 . (22)
∆h2

It should be noticed that the momentum exchange is the main process occurring in the ejector, therefore
the correct value of the entrainment ratio can be obtain from the equation of the momentum conserva-
tion, and only these values will be taken into account in the given analysis. Equation of momentum
conservation for mixing chamber is:

m g w2 + m ew 7 = (m g + m e )w 3 , (23)

167

Ejection Refrigeration Cycles

hence:

w1 ∆h1
U = −1 = −1 = χ −1 . (24)
w4 ∆h2

More exact result we will get in practical conditions introducing coefficient taking into account the
losses in the flow, defined as the product of relations of velocity in separate parts of ejector:

K = ϕnϕmϕd , (25)

where: φn – velocity coefficient of the motive nozzle, φm - velocity coefficient of the mixing chamber,
φd – velocity coefficient of the diffuser. Then:

U = K χ − 1. (26)

Equation 26 is the same as Equation 14 taking into account Equation 25 and relation: φ2 = η. The
entrainment ratio for an ideal ejector that is required by the cycle can be obtain from equation of energy
conservation:

m g h1 + m eh7 = (m g + m e )h4 , (27)

after transformations

h1 − h4
U = . (28)
h 4 − h7

As it was mentioned above, lines of p = const in the pressures range pe ÷ pc are almost parallel.
Therefore it may be assumed that:

∆h2 ≈ hw − h2 ≈ h4s − hH ≈ hM − h2s ,


(29)
h1 − hM = ∆h1 − ∆h2 .

Because as the consequence of losses the increase of the enthalpy and entropy that occur in a real
ejector the condition of refrigerant at the outlet of the ejector are defined by point 4 and not 4s. Therefore
the entrainment ratio in this case is:

h1 − h4 ∆h1 line 1 − N
U = =K −1 = . (30)
h4 − h2 ∆h2 line N − 7

168

Ejection Refrigeration Cycles

As it was shown by Śmierciew, Butrymowicz, & Karwacki (2008a) it may be applied correction that
accounts the non-parallel paths of the isobars in the h-s diagram, therefore:

∆h1
U =K −1. (31)
C ∆h2

Graphic interpretation of correction coefficient C is presented in Figure 10. Point A0 represents satura-
tion condition for given evaporation temperature te, and corresponding pressure pe(te). Given condensation
temperature tc determines condensation pressure pc(tc). Intersection of constant entropy line sA0 = const
crossing by the point A0, with isobar pc = const determines point B0. The enthalpy difference between
points B0 and A0, ∆h0 = hB0 - hA0, gives the basis enthalpy difference.

Figure 10. Graphic interpretation of corrective coefficient C

169

Ejection Refrigeration Cycles

Non-parallel of isobars on the graph h-s causes, that the superheating of secondary vapour up to the
temperature tA1 with pe = const gives enthalpy difference increasing ∆h1 = hB1 - hA1 >∆h0. With increasing
of vapour superheating, the isobars scatter increases and the differences of enthalpy rise ∆h0 <∆h1<∆h2.
In wet vapour region quality of two-phase flow x should be used. In this region with lower x the smaller
is the difference of enthalpy ∆hi = hBi - hAi in the comparison with the basis enthalpy difference ∆h0.
Putting the coefficient C reducing influence of changing of enthalpy difference caused by the isobars
scatter in the h-s graph. This coefficient is defined as:

∆h
C = , (32)
∆h0

where: ∆h0 = hB0 - hA0, ∆h = hBi - hAi, i = 1,2,... .


In wet vapour region coefficient C < 1, in superheat vapour region C > 1.
The relationship between the correction coefficient C and dimensionless vapour superheating ∆Te/
Te and the two-phase flow quality x, for four various refrigerants are presented in Figure 11 and Figure
12. The following correlation was proposed by Śmierciew et al. (2008a) for the correction coefficient:

1.143∆Te / Te + 1.00 super heated vapour region


C =  (33)
1.00 − 1.252(1 − x ) wet vapour region.

Figure 11. Dependence of correction coefficient C as the function of dimensionless vapour superheating
∆Te/Te and quality x; tc= 30 °C, te = -5 °C

170

Ejection Refrigeration Cycles

Figure 12. Dependence of correction coefficient C as the function of the dimensionless vapour super-
heating ∆Te/Te and the quality x for water vapour; tc= 30 °C, te = 5 °C

Comparison of calculated and estimated on the basis of eq. (33) values of correction coefficient C
is shown in Figure 13.
For entrainment ratio U calculated from Equation 31 fundamental parameter of fluid at the outlet of
the diffuser can be estimated transforming Equation 28. Capacity of several devices and coefficient of
performance COP can be calculated from Equation 15 to Equation 19.

Figure 13. Corrective coefficient C in comparison with actual valuesin superheated and wet vapour region

171

Ejection Refrigeration Cycles

Own Approach of Cycle Analysis

The idea of the own approach is to split the entire cycle into the cycle performance and ejector per-
formance (Śmierciew et al., 2008b) as schematically shown in Figure 14. This will allow to separate
investigation of the very difficult flow of refrigerant in the ejector with the simple equations describing
the operation of the whole of system.
Applying existing models based on the balance equations requires the driving pressure to be as-
sumed. However, in refrigeration systems condensation pressure and evaporation pressure are normally
the only fixed parameters. The first one results from the thermal capacity of condenser, the temperature
and the mass flow rate of cooling fluid, and also heat transfer conditions. Similarly, evaporation pres-
sure results from the cooling capacity of evaporator, but mainly required chamber temperature and heat
transfer conditions. However, it is difficult to unambiguously predict the driving pressure for the ejector
of given geometry.
Preparing two characteristics as dependences of dimensionless compression ratio Π from the entrain-
ment ratio U is the aim of this proposed method of analysis of the ejection cycle. The knowledge of
performance curves of the system and ejector will allow to find the operating point of the system. This
way is analogous to the analysis of the cooperation of pump with the liquid feeding system. Compres-
sion ratio is given by Equation 34:

pc − pe
Π= . (34)
pg − pe

The work of the installation is describes by energy conservation equation only. Because pg and U are
unknown parameters in ejection systems, there is a need to seek other relation that makes possible to
predict the driving vapour pressure.

Figure 14. Schematic diagram of ejection device

172

Ejection Refrigeration Cycles

For every ejector it could be prepared the individual performance curve. This may be based on the
experimental investigations or modelling. The prediction of the performance of the ejector is very com-
plicated because of processes and phenomena occurring in the ejector such as: mixing of the streams,
mass, momentum and energy transfer, eddies, separation of the stream from boundaries, shock wave
and other phenomena. Because of all these phenomena the model of ejector operation is very complex.
However, results received from experiments carried over on similar ejector and working in similar con-
ditions, can be applied to simplify the problem.
Performance of the system results from the energy balance and does not contain any information about
the ejector, therefore it treats the ejector as the black box. The energy conservation for the system is:

Qc = Qe + Q g + Pp . (35)

Because of very small value of motive power of pump Pp in comparison with remaining values in
Equation 35 to further calculations it can be dropped out. Expressing the individual thermal capacity of
devices using Equation 15 to Equation 17 and making suitable transformations it is obtained:

(m h
g 1
+ m eh7 ) = (m g + m e )h4 . (36)

In refrigeration system quantity of transferred heat has to be known, therefore has to be known a
cooling capacity of evaporator. From Equation 15, mass flow rate of refrigerant flowing through the
evaporator can be calculated. Enthalpy h7 is a function of pressure and temperature of evaporation h7 =
f(pe,te) and te = ts (pe ) + ∆Te . Enthalpy h4 can be calculated from equation 37:

h1(pg ) +Uh7
h4 (pg ,U ) = . (37)
1 +U

In the proposed method the state of refrigerant at the outlet from the ejector has to be known. This
is the reason for making following assumption: mixing takes place at constant pressure and processes
of expansion and compression are isentropic. If the pressure and superheating of the driving fluid are
known then temperature of fluid is: tg = ts (pg ) + ∆Tg , and then enthalpy h1 can be obtain: h1 = f(pg,
tg). Assuming initially the entrainment ratio U, then enthalpy at outlet of the ejector can be found from
Equation 37. Energy balance equation between inlet and outlet of nozzle can be written as:

1
h1(pg ) = h2 + w22 . (38)
2

Energy equation for mixing chamber is as follows:

1  1   1 
h2 + w22 +Uh7 = (1 +U ) h3 + w 32  = h1(pg ) +Uh7 = (1 +U ) h3 + w 32  . (39)
2  2   2 

173

Ejection Refrigeration Cycles

Assuming zero velocity of vapour after mixing process (state 3), the mixture velocity in the mixing
chamber before compression is defined:

w2
w3 = . (40)
1 +U

Combining the Equations 39 and 40 leads to:

 
 1 w2 
2

h1(pg ) +Uh7 = (1 +U ) h3 + . (41)
 2 (1 +U )2 
 

Expressing the outlet velocity from the driving nozzle by using the difference of enthalpy given by
Equation 4, and making a few transformations:

h1(pg ) − h2 h1(pg ) + Uh7


h3 + = , (42)
(1 + U )2
(1 + U )

thus, enthalpy h3 is a function of pg and entrainment ratio U as follows:

h1(pg ) +Uh7 h1(pg ) − h2


h3 (pg ,U ) = − . (43)
(1 +U ) (1 + U )2

Using Equation 37 it can be written:

h3 (pg ,U ) = h4 (pg ,U ) − ∆h2 (pg ,U ) , (44)

where:

h1(pg ) − h2
∆h2 (pg ,U ) = (45)
(1 +U )2

is the kinetic energy changed at the diffuser resulting in static pressure raise. Results from this is the h3,
which is not total enthalpy but the static one. According to above considerations, the total enthalpy is:

h3 ' (pg ,U ) = h3 (pg ,U ) + h2 = h4 . (46)

Calculating the entropy at this state, pressure at the outlet from the ejector can be defined as
p4 = f (s 3=4 , h4 ). If obtained in this way pressure p4 differs from defined pc, then the entrainment ratio

174

Ejection Refrigeration Cycles

U should be recalculated. Therefore the procedure should be repeated with different U value so long,
until we recognise that the difference between p4 and pc is not important. In this way one point of the
ejector performance Π = f(Ui) is obtained. The above procedure should be repeated for various driving
vapour pressures which allows to prepare the entire performance curve of the system.
The model describing the work of isentropic ejector is presented below. For given geometry, con-
stant values of pe, ∆Te, pc, ∆Tg and variable value of driving vapour pressure pg, the specific parameters
of state of refrigerant inside the ejector were obtained. The driving pressure was assumed initially. The
enthalpy and density of driving vapour was calculated as functions: h1 = f(pg, tg) and ρ1 = f(pg, tg). As the
consequence of expanding in the nozzle vapour achieves pressure p2=p7=pe. The remaining parameters
of expanded vapour can be found as: h2 = f(pg, s1=2), t2 = f(pg, s2), ρ2 = f(p2, t2). Constant value of specific
heat is assumed. Critical velocity in the nozzle throat is given by:

κ
wcr = 2
κ −1 g 1
(
p ρ −1 1 − β (κ−1)/κ . ) (47)

Mass flow rate of motive vapour can be calculated from equation:

2  κ
2 κ +1 
κ
m g = pg At β − β κ  , (48)
κ − 1 Tg R  

where At is the cross-section area of throat of nozzle, κ = cp/cv, β is the critical pressure ratio.
Because two streams of different temperatures enter separately to the mixing chamber their average
temperature was estimated as:

t2 +Ut7
tma (U ) = , (49)
1 +U

and temperature of mixture at the outlet of the cylindrical part of the ejector is given by isentropic relation:

(κ−1)/κ
p 
tm (pm ,U ) = tma (U )  m  . (50)
 p2 

In above equation pm is pressure of mixture. This is unknown parameter and which depends on en-
trainment ratio U. Using similar isentropic relation density of mixture ρm(pm) and the outlet temperature
of mixture t4 = tc can be calculated. Velocity of mixture can be calculated from mass balance equation,
which can be written as:

m g +Um g = Am ρm (pm )wm , (51)

thus

175

Ejection Refrigeration Cycles

m g (1 +U )
wm (pm ,U ) = . (52)
Am ρm (pm )

Velocity at the outlet from ejector is given by:

m g (1 +U )
w 4 (pm ,U ) = , (53)
Ad ρ4

where Ad is area of cross-section of diffuser, and ρ4 = f(p4, t4) is density of mixture at the outlet of dif-
fuser. Velocity achieved by the secondary vapour at the inlet to the mixing chamber is given by following
equation:

m gU
w 7 (U ) = , (54)
A7ρ7

where A7 = Am-A2, and ρ7 = f(pe, te)≈ ρ2 .


Because of values of pm and U are unknown therefore it is required to find the auxiliary relation in
order to find solution. This relation can be momentum balance equation prepared for mixing chamber
and diffuser separately, see Figure 15. For mixing chamber the following equation can be developed:

Figure 15. Momentum transfer at the mixing chamber (a) and the diffuser (b)

176

Ejection Refrigeration Cycles

Am
(p − pm ) − (1 +U )wm + w2 +Uw 7 = 0 , (55)
m g 2

and for diffuser:

Am pm + Ax px − Ad p4 = (m g + m e ) (w 4 − wm ) , (56)

where Axpx is the average force acting on the inner surface of the diffuser. This force in the simplest form
and can be written as:

1
Ax px = (A − Am )(pm + p4 ) ,
2 d
(57)

thus,

1  
A p + 1 (A − A )(p + p ) − A p  − (1 +U )w + (1 +U )w = 0 . (58)
m g  m m
2 d m m 4 d 4  4 m
 

Point of intersection of curves described Equations 55 and 58 is the ejector solution. This solution
defines pressure of mixture pm and entrainment ratio U. Obviously these parameters are related with
themselves because rise of entrainment ratio U with constant mass flow rate of driving vapour m g ,
causes mass flow rate of the secondary fluid increase, and this causes that the velocity w7 grows up and
the difference between velocities w2 and w7 decreasing because w2 = const. As a consequence the ve-
locities of the mixture increasing, and therefore the pressure falls at the end of the mixing chamber, so
above mentioned equations should been consider together.
It should be noticed that the momentum balance equations are prepared in the simplest possible
forms. Friction force has not been considered here. Moreover, the above equations do not contain any
information about shock wave. This wave is one of the causes of pressure raise in the ejector. Regard-
less of shock wave in the presented approach it is assumed that the compression ratio is the result of
momentum exchange only.
The following data was taken to prepare the exemplary performance calculations:

• R-123 is the working fluid;


• Ejector geometry: nozzle throat diameter dn = 2 mm, ejectors throat diameter (constant cross-
section) dm = 10 mm, diffuser outlet diameter dd = 25 mm;
• Driving fluid parameter: pressure range pg = 600÷770 kPa;
• Superheating ∆Tg = 4K;
• Evaporator pressure pe = 80 kPa, superheating ∆Te = 9K;
• Condensation pressure pc = 133 kPa;
• Heat capacity ratio κ = 1.15.

177

Ejection Refrigeration Cycles

Intersection the system performance curve and the ejector performance curve is the operation point,
see Figure 16. The obtained compression ratio will be approximately Π≈0.079. Therefore, to get on the
outlet from ejector pressure pc=133 kPa, with evaporation pressure pe =80 kPa, generator pressure from
Equation 34 should be pg= 751 kPa.
On the basis of numerical modelling as well as experimentation the performance of the ejector could
be applied to the presented approach which makes possible usage of this approach for optimization of
the cycle and ejector geometry.

INVESTIGATION OF EJECTION REFRIGERATION SYSTEMS

Ejection Solar Air Conditioning

Individual residential air conditioning consumes over 11 percent of electric energy used by buildings.
For buildings with central air conditioning it comprises a substantially larger share, even one-third of
electric energy. Moreover, air-conditioning creates two sources of environmental pollution: 1) the direct
emission of greenhouse gases for working fluids belonging to HFC groups, and 2) the emission of the

Figure 16. Exemplary case of the ejection system operation point

178

Ejection Refrigeration Cycles

greenhouse gases connected with the generation of electricity. Both sources are contributing significantly
to the global warming effect. Additionally, with energy cost rising constantly, industry is looking to
reduce electricity expenses as a means of lowering their fixed costs in order to stay competitive.
This chapter presents the development of air-conditioning technology that strongly reduces the
greenhouse gases emission by using natural refrigerants and also dramatically reduces the need for
the electric power. This is accomplished by using free or inexpensive heat – either solar or waste heat,
as the main source of energy instead of electricity. The described system is a modification of a well-
known vapour compression cycle. Instead of pressurizing the refrigerant by a mechanical compressor,
an ejector compresses the vapour flowing from the evaporator and discharge it to the condenser. The
motive vapour is generated in the generator which is heated by low-temperature hest source. Therefore,
the main difference between this cycle and the conventional refrigeration cycle, besides elimination of
a compressor, is that it requires three heat sources at different temperatures rather than two, namely at
the generator level, which is the temperature of the solar or waste heat source, at a condensing level,
which is the ambient temperature (actually this is a heat sink) and the evaporator temperature required
for cooling effect.
Presented in this chapter experimental and numerical investigation of ejection air conditioning system
were executed within the national and international projects. Theoretical and experimental investigation
of the solar air-conditioning operating with natural fluids were presented by Śmierciew (2011). The
details of this investigation were also presented in Final Report of Project by Butrymowicz, Śmierciew,
& Karwacki (2010).
The experimental rig for investigation of the vapour ejection refrigeration system is presented in
Figure 17 and Figure 18. The main elements of the rig are listed in the figure caption. The geometry of
the test ejector is presented in Figure 19. The testing stand was equipped with the temperature sensors
and pressure transducers installed in the critical locations and other locations of interest. The test rig was
equipped with two additional loops: the first for the thermal load and the second for condenser cooling.
These systems allow for adjusting refrigerant flow rates as well as for changing of thermal parameters
in the wide range. The condenser cooling system was equipped with an automatically controlled fan
cooler. The thermal load system was equipped with automatically controlled electrical heater. Both sys-
tems are fully instrumented with transducers for measuring temperatures, pressures and flow rates with
high accuracy. The design of the entire test rig allows to maintain all test parameters within a specific
range. The rig is equipped with control valves enabling the adjustment of the operating parameters of
the motive vapour at the inlet to the motive nozzle of the ejector. The measuring system consists of a
NI SCXI module system.
Investigations were carried out with isobutane (Śmierciew, Gagan, Butrymowicz, & Karwacki,
2014). This fluid was applied in the investigation because it is a natural fluid and also its gives very high
COP for air-conditioning in comparison with most HFC and natural fluids (Butrymowicz, Śmierciew,
Karwacki & Gagan, 2014).
Generally, the performance of the ejector was changed by means of adjusting the condensation pres-
sure by changing the massflow rate of the coolant in condenser cooling loop. In present investigations
saturation pressure in the generator, superheating and temperature of the motive primary vapour were
set at four levels and kept constant. Also the evaporating pressure was kept constant.
The basic parameters describing the ejection cycle were calculated: mass entrainment ratio, Equa-
tion 6 and compression ratio given by Equation 34. Additionally, the momentum loss coefficient K was
calculated using Equations 25 and 26 proposed by Paliwoda (1971).

179

Ejection Refrigeration Cycles

Figure 17. Photo of the stand dedicated to investigation of the vapour ejector refrigeration system

The tested ejector was equipped with the mixing chamber of the diameter Dm = 6 mm, driving nozzle
of the throat diameter Dn = 3.5 mm. Average operating parameters are listed in Table 2.
Experimental results in form of performance curve Π = f(U) is presented in Figure 20 and momentum
loss coefficient is shown in Figure 21. The operation parameters are given in Table 2.
Performance line in co-ordinate compression ratio Π vs. mass entrainment ratio U is presented in
Figure 20. It is clear that for the case of tg4 = 63.5 °C ejector operates both at the on-design and off-design
conditions: the vertical part of the performance line covers the on-design conditions while the sloped
part covers the off-design operating conditions. On the other hand for tg1 = 55 °C and tg2 = 58 °C it is
clear that there is no on-design conditions.

Table 2. The operating parameters of the tested ejection system

Series Unit 1 2 3 4
pressure of motive vapour pg MPa 0.77 0.83 0.90 0.94
superheating of motive vapour ΔTg K 8.0 8.0 8.0 8.0
saturation temperature of motive vapour tgs °C 55.0 58.0 61.5 63.5
suction pressure pe MPa 0.20 0.20 0.20 0.20
superheating of suction vapour ΔTe K 6.5 6.5 6.5 6.5
saturation temperature of evaporation tes °C 7.0 7.0 7.0 7.0

180

Ejection Refrigeration Cycles

Figure 18. Schematic of the testing stand: 1 – tested ejector; 2 – internal heat exchanger (not used in
the reported investigations); 3 – condenser; 4 – refrigerant pump; 5, 6, 12,14 – mass flow meters; 7 –
vapour generator; 8 – evaporator; 9 –dry cooler; 10 – circulation pump in the condenser cooling loop;
11 – control valve; 13 - glycol pump; 15 auxiliary electric heater, 16 - electric heater in evaporator heat
load loop, 17 – circulation pump in evaporator heat load loop; 18 - solar panels

Figure 19. The schematic of tested ejector

181

Ejection Refrigeration Cycles

Figure 20. Experimental results for compression ratio Π versus mass entrainment ratio U for tested ejector

Figure 21. Momentum loss coefficient K versus mass entrainment ratio U

In Figure 21, the loss coefficient K obtained from Equation 26 is shown. It is seen that for all tested
cases loss coefficient K is higher than 0.75. For two analysed cases i.e. tg3 = 61 °C and tg4 = 63 °C it
is clearly seen that below tc* corresponding to the critical pressure pc* coefficient K slightly increasing,
however, for other tested cases i.e. tg1 = 55 °C and tg2 = 58 °C this relation is not observed. This could be
results of small amount of experimental data obtained in on-design operating regime. For condensation
temperature higher than tc* coefficient K decrease to approximately 0.75 for all tested cases. Informa-
tion about value of loss coefficient K could be very helpful and useful in engineering calculation and
designing of solar ejector air-conditioning system. In that case application of sofisticated and complex
model for calculation of entrainment ratio and ejector performance is not necessary.

182

Ejection Refrigeration Cycles

Using definition proposed by Dvořák & Vit (2005) the ejector efficiency was calculated. The results
are presented in Figure 22.

κ−1

η=
(
m e 1 − pe pc
κ
) (59)
m g κ−1

(
pg pc κ
) −1

It can be seen from Figure 22 that for two cases when ejector operates at on-design condition (i.e.
tg3 = 61.5 °C and tg4 = 63.5) the efficiency of the ejector slightly increasing with increasing of the
condensation temperature. According to Equation 59, this efficiency increase is related with increas-
ing of the condensation pressure. At off-design conditions the efficiency decreases with increasing of
the condensation temperature which is a result of the decreasing of the mass entrainment ratio at these
operation conditions. It can be also suggested basing on the results presented in Figure 20 that for off-
design operating conditions the ejector performance is independent of the motive vapour temperature.
Classic approach based on Keenan (1950) model with constant area mixing flow for air and later
developed by Munday & Bagster (1977) and Huang, Chang, Wang, & Petrenko (1999) is most popular
in theoretical analysis of the ejector. Based on the previous experience of author and author’s research
team the model was improved by additional equations for shock waves and for friction loss in the mixing
chamber and the diffuser. The shock wave is calculated on the basis of Fanno and Rayleigh equations.
Model could be used for chocked suction conditions which is necessary in gas cooling ejector. This type
of ejector’s operation results maximum value of cooling capacity and biggest COP of the system. For this
case the mass entrainment is fixed for given geometry of the motive nozzle, so compression produced
by the ejector is calculated (i.e. the discharge pressure). Thermodynamic properties of isobutane was
applied from Helmholtz free energy equation on the basis of Buecker & Wagner (2006). For given ejec-
tor geometry and given inlet parameters for primary and secondary fluid the mass entrainment ratio and
outlet parameters can be found. Then, the parameters of the vapour at the characteristic cross section of

Figure 22. Efficiency of the ejector vs. condensation temperature for various driving temperatures

183

Ejection Refrigeration Cycles

the ejector also can be found. The location of these characteristic cross sections are presented in Figure
23. Generally the model requires to solution of 6 equations given below: Equation 60, Equations 62 to
65, and Equation 67. The subscripts for equation are listed in Nomenclature.
Momentum balance equation for mixing chamber is given by:

Am (pC − pX ) − Rm = m g w pC + m ewsC − (m g + m e ) w X . (60)

The reaction force Rm for friction in the mixing chamber:

1
Rm = K ρ w2 A . (61)
2 m X X bm

Energy balance equation for the mixing chamber can be written as:

 2 
w pC   w 2   w 2 
  
m g hpC +  + m e hsC + sC  = m hX + X  . (62)
 2   2   2 

The pressure rise and change of specific volume is calculated using Fanno line and Rayleigh line,
Equations 63 and 64 respectively:

2 2
1  m  1  m 
hX + v   = hY + vY2   ,
2
(63)
2  Am 
X
2  Am 

2 2
 m   m 
pX vX   = pY vY   . (64)
 Am   Am 

Figure 23. Location of characteristic cross section inside ejector

184

Ejection Refrigeration Cycles

Momentum balance equation for diffuser is given by:

1
 Y = 0.
 D + mw
Am pY + (AD − Am )(pY + pD ) − AD pD − Rd − mw (65)
2

Analogically, the component Rd denotes the reaction force for friction in diffuser, and has a form:

1
Rd = Kd ρDwD2 Abd . (66)
2

Energy balance equation for entire ejector:

 w 2 
m g hA + m ehO = m hD + D  . (67)
 2 

Presented above equations are valid only in on-design operating conditions. This feature can be
considered as a limitation of the model, however the on-design operation regime is most promising and
desirable operation mode of the ejector in air-conditioning. Therefore the model can be successfully
used for evaluation of the ejectors geometry.
The development of the IT branch makes the numerical simulation of gas ejector with CFD techniques
becomes more and more popular. Commercial CFD code was used for modelling of isobutane ejector
operating in superheated vapour region. The main goals of numerical calculation are: verification and
validation of the experimental results; validation of the 0D models; application of the CFD into design
processes; numerical analysis of the flow in order to better understand the physical process occurring
inside ejector.
Numerical results presented in this chapter deals with calculations done for boundary conditions
proper for air-conditioning. Two-dimensional axi-symmetric model was applied for simulations. The
thermodynamic properties of working fluid are implemented from NIST database (Lemmon, McLinden,
& Huber, 2010) using equation of state given by Buecker & Wagner (2006).
Exemplary comparison of numerical results of pressure, velocity and temperature obtained from CFD
simulations and 0D calculation in characteristic cross section are collected in Table 3 and presented in
graphic form in Figure 24, Figure 25 and Figure 26, respectively.
On the basis of the calculation results it could be concluded that the lumped parameter model predicts
of the main operation parameters with reasonably good agreement in respect to pressure distribution,
temperature distribution as well as velocity distribution in the ejector. The highest disagreement between
0D and CFD simulation are found in parameters in cross section Y. In authors opinion it could be results
of assumption of the model, in 0D model the shock wave is perpendicular which leads to discontinuity,
in CFD model the shock waves in clearly not perpendicular.
The results of simulation in comparison with experimental results of measured pressure at ejector
wall are presented in Figure 27 and in Table 4. It may be concluded that reasonably very good agreement
between CFD theoretical results and experimental data has been observed.

185

Ejection Refrigeration Cycles

Table 3. Numerical results of pressure, velocity and temperature obtained from CFD and 0D model

CFD 0D Model CFD 0D Model CFD 0D Model


Pressure p [MPa] Temperature t [°C] Velocity w [m/s]
A 1.100 1.100 90.0 90.0 5.68 0.0
T 0.620 0.627 73.8 71.2 201.68 200.21
B 0.189 0.166 35.8 32.9 390.27 401.97
C 0.169 0.127 29.5 30.5 413.83 441.45
X 0.200 0.149 37.4 39 382.32 355.37
Y 0.391 0.429 57.8 69.2 286.21 131.90
D 0.449 0.441 67.8 67.9 15.89 11.7

Figure 24. Comparison of static pressure distribution calculated from lumped parameter model (0D
model) and CFD model

Systematic investigation of the isobutane ejection system, both experimental and numerical using 0D
model and CFD techniques were basics to estimation of the required operations parameters and geometry
of the prototype ejectors for solar air-conditioning device (Butrymowicz, Trela, Karwacki, Ochrymiuk,
& Śmierciew, 2008; Butrymowicz, Śmierciew, & Karwacki, 2010; Butrymowicz et al., 2010).
The main operation parameters of the ejection system are evaporation and condensation tempera-
tures. It was assumed that the evaporation temperature te = 10 °C, the condensation temperature tc = 30
°C and that the secondary vapour is at saturated state. The corresponding saturation pressures for these
temperatures are: pe = 0.22 MPa and pc = 0.40 MPa. The compression of liquid produced by available

186

Ejection Refrigeration Cycles

Figure 25. Comparison of average velocity distribution calculated from lumped parameter model (0D
model) and CFD model

Figure 26. Comparison of static temperature distribution calculated from lumped parameter model (0D
model) and CFD model

187

Ejection Refrigeration Cycles

Figure 27. Comparison of static pressure distribution along the ejector wall

Table 4. Comparison of CFD exemplary results with experimental results

pg tg pe te pc tc
[MPa] [°C] [MPa] [°C] [MPa] [°C] Primary Flow m g Secondary Flow m e
[kg/h] [kg/h]

EXP CFD EXP CFD


78.9 0.265 20 0.340 53 35.2 35.5 15.3 18.7
1.068

mechanical liquid pump was estimated as Δp = 0.70 MPa so the generator pressure pg = 1.10 MPa. For
these conditions the specific enthalpy of vaporization hfg = 269 kJ/kg and saturation temperature tgs =
70.5 oC. In solar based systems the motive temperature of the heat source from solar collector may be
estimated at the level 90 ÷ 100 oC so the superheating of the motive vapour may be estimated as ΔTg
= 20÷30 K. On the basis of the above formulated conditions the following motive specific heat load is
necessary 43÷65 kJ/kg. Assuming the basic available heat load for solar collectors as 20 kW it is easy to
predict required motive mass flow rate for the system 180 kg/h. For such operating conditions required
throat diameter of the motive nozzle is Dn = 7 mm. It is suggested to limit the capacity of the system due
to possible loss of the thermal output of the solar collectors and chose the motive nozzle throat diameter
Dn = 4.0 mm. On the basis of CFD as well as 1D numerical approach the mixing chamber diameter was
predicted Dm = 5.94 mm.
Commercial CFD code was used for modelling and analysing the isobutane ejector operating in
superheated vapour region. The boundary conditions used for simulation are presented in Table 5.
Performance line of ejector operated in on-design and off-design regime with isobutane is presented
in Figure 28. Pressure distribution along ejector axis for various back-pressure for both regime are pre-
sented in Figure 29 and Figure 30.
It is important to note that there is a significant gap of the knowledge of modelling of the ejector
for isobutene as well as experimental validation of the operation of the system. Therefore the results
presented above may be thought as first complex approach for investigation of the isobutene ejection
system which is necessary to develop the rational design of the system for solar based air-conditioning.

188

Ejection Refrigeration Cycles

Table 5. Boundary conditions and simulation results

Boundary Conditions Results


pg tg pc tc(pc)
[MPa] [°C] [MPa] [°C] m g m e
[kg/s] [kg/s]

11 90 0.35 29 0.0285 0.00730


pe te 0.40 29.6 0.0285 0.00730
[MPa] [°C]
0.45 33.8 0.0285 0.00730

0.22 10 0.50 37.7 0.0285 0.00539

0.53 39.9 0.0285 0.00074

0.54 40.6 0.0285 0.00000

Figure 28. Performance line Π = f(U) for solar ejector operation with isobutane for air-conditioning
conditions

Working Fluids for Ejection Systems

Selection of the working fluid for the solar air-conditioning system is the crucial problem because of the
strong influence of the thermodynamic fluid properties on the system efficiency. Moreover, the working
fluid should fulfil the environment criteria such as zero ODP (Ozone Depletion Potential) and as low
GWP (Greenhouse Warming Potential) as possible. Therefore the natural fluids are thought as the best
option. The energy efficiency of the system for the following fluids have been done: isobutane, ammonia,
propane, methanol, water. The following conditions have been assumed in the analysis:

189

Ejection Refrigeration Cycles

Figure 29. Pressure distribution along the ejector in “double-chocked region”

Figure 30. Pressure distribution along the ejector in “single-chocked region”

190

Ejection Refrigeration Cycles

• Superheating of the motive vapour: for isobutane ΔTg = 5 K, for water ΔTg = 60 K, and for the
other fluids ΔTg = 50 K;
• Secondary vapour leaves the evaporator as dry saturated vapour;
• All loses at the ejector are represented by the overall efficiency η.

Table 6 contains selected operating parameters of the ejection system for temperature of the motive
vapour tg = 130 oC. The specific volume vapour to liquid ratio vv/vl was calculated at saturation condi-
tions for condensation (tc, pc). The results of the calculation are presented in Figure 31 and Figure 32.
It is evident therefore that isobutene is the best option as a working fluid for ejection system in ap-
plication to air-conditioning since it offers the highest COP. It is important to note that this conclusion
is valid for the other liquid subcooling and vapour superheating in the system.

Ejection Cycle Improvement by Means of Internal Heat Transfer

Since the ejection refrigeration devices generally are low efficient systems it is a clear need for further
development of simple and low cost method for improving of the COP for such systems. The internal
heat transfer with warm internal heat exchanger – located at the ejector discharge line/ liquid line is the
most simple approach. The plate heat exchanger was applied in the tested system. Schematic diagram of
ejection refrigeration system with regenerative heat exchanger is presented in Figure 33, and the diagram
of thermodynamic cycle of this system is presented in Figure 34.
In general, the COP coefficient of ejection refrigeration system is defined as ratio of heat capacity
of evaporator and heat put into the system which is capacity of generator and the work rate of the pump:

Q
COP =  e . (68)
Qg + Pp

According to Figure 34, capacity of evaporator is given by

Qe = m e ⋅ (h7 − h6 ) , (69)

Table 6. Comparison of the operations parameters for ejection system for various working fluids: tg =
130 oC, tc = 35 oC, te = 10 oC, η = 0.70, ΔTg = 5 Κ for isobutane, ΔTg = 60 Κ for water, ΔTg = 50 Κ for
ammonia, propane and methanol; U is mass entrainment ratio

Working h1 h2s h3 h5=h6 h7 pc vl vv vv/vl U COP


Fluid
[kJ/kg] [kJ/kg] [kJ/kg] [kJ/kg] [kJ/kg] [kPa] [m3/kg] [m3/kg] -- -- --
Isobutane 709.5 603.8 567.8 271.3 276.1 468 1.9E-3 8.3E-2 49 0.54 0.37
Ammonia 1813.0 1547.4 1615.3 495.0 489.6 1351.0 1.7E-3 9.6E-2 56.2 0.37 0.32
Propane 776.3 681.0 585.7 278.8 282.7 1217.9 2.1E-3 3.8E-2 17.9 0.32 0.20
Methanol 1228.1 967.7 1048.2 -92.8 -92.6 28.0 1.3E-3 2.8E+0 2150.2 0.38 0.33
Water 2742.9 2283.6 2519.2 125.7 125.8 5.6 1.0E-3 2.5E+1 25057 0.29 0.26

191

Ejection Refrigeration Cycles

Figure 31. COP for various working fluid for given evaporating temperature te = 10 oC as a function of
motive vapour temperature

Figure 32. COP for various working fluid for given temperature of motive vapour tg = 130 oC as a func-
tion of evaporating temperature

192

Ejection Refrigeration Cycles

Figure 33. Schematic diagram of ejection refrigeration system with internal heat exchanger

Figure 34. Diagram of the augmented ejection cycle

193

Ejection Refrigeration Cycles

and capacity of generator is defined as

Q g = m g ⋅ (h10 − h1 ) . (70)

The mechanical work rate of the pump is given by:

P = m ⋅ (h10 − h5 ) . (71)

Power consumption by pump may be assumed as negligible small in comparison with the thermal
capacity of generator (which is motive source of the system). The internal heat exchanger is applied to
preheat liquid feeding the generator by means of desuperheating of vapour discharged from the ejec-
tor (Butrymowicz, Śmierciew, & Karwacki, 2014). Therefore the thermal capacity of the generator is
smaller in this case:

Q g ,x = m g ⋅ (h8 − h1 ) . (72)

In effect Q g ,x < Q g , and the coefficient of performance

Q Q
COPx =  e >  e . (73)
Qg ,x Qg

The suction pressure (evaporation pressure) was treated as the variable parameter. Pressure and
temperature of the motive vapour were kept constant. The back pressure at the ejector outlet also was
constant. Superheating of secondary vapour was also kept constant. First run (No. 0) was recorded with
suction pressure pe = 0.181 MPa when system operating without internal heat exchange and then pe =
0.189 MPa with internal heat exchange.
The stand was equipped with a single-phase ejector of 0 mm diameter and 40 mm length of mixing
chamber. The diameter of de Laval nozzle throat was 2.0 mm. The suction pressure (evaporation pres-
sure) was treated as the variable parameter. Pressure and temperature of the motive vapour were kept
constant. The back pressure at the ejector outlet also was constant. Superheating of secondary vapour
was also kept constant. First run (No. 0) was recorded with suction pressure pe = 0.181 MPa when
system operating without internal heat exchange and then pe = 0.189 MPa with internal heat exchange.
Table 7 contains the operating parameters. The difference of temperature of vapour phase at inlet and
outlet of heat exchanger is given by:

∆Tvap = T9 −T4 , (74)

and the difference of temperature of liquid phase is calculated according to eq.

∆Tliq = T8 −T10 . (75)

194

Ejection Refrigeration Cycles

Table 7. Average pressure and temperature measurements during experiments

Parameter Operating without Heat Exchanger Operating with Heat


Exchanger
Pressure of motive vapour MPa 0.911 0.916
Superheating of motive vapour K 9.6 9.4
Temperature of motive vapour o
C 71.6 71.7
Suction pressure MPa 0.154 ÷ 0.181 0.153 ÷ 0.189
Superheating of suction vapour K 1.2 0.9
Temperature of suction vapour o
C -0.4 ÷ 2 -0.8 ÷ 5.3
Backpressure MPa 0.220 0.220

The results of measured differences of temperature are presented in Figure 35. Heat transfer coef-
ficient was calculated as follows:

Q x ,liq m g ⋅ (h8 − h10 )


K= = , (76)
∆Tlog ⋅ Ax ∆Tlog ⋅ Ax

where ∆Tlog is logarithmic difference of temperature and Ax is surface of heat transfer area. The results
of calculation of heat transfer coefficient are presented in Figure 36.

Figure 35. Temperature difference calculated from eq.(74) and (75) for both phases

195

Ejection Refrigeration Cycles

Figure 36. Overall heat transfer coefficient for tested internal heat exchanger

The NTU parameter is defined as:

K ⋅ Ax
NTU = , (77)
C min

where:

 
C min = min (m ⋅ cp ) , (m g ⋅ cp )  . (78)
 vap liq 

The results of calculation of NTU are presented in Figure 37.

Figure 37. NTU for tested internal heat exchanger

196

Ejection Refrigeration Cycles

The heat-exchanger effectiveness, ε , is defined as the ratio of actual rate of heat transfer from the
hot (vapour) to cold (liquid) fluid to the maximum possible rate of heat transfer. The maximum possible
heat transfer is given by:

Q max = C min (Tin ,vap −Tin ,liq ) = C min (T4 −T10 ) , (79)

where Tin ,vap and Tin ,liq are the inlet temperatures of the hot (vapour) and cold (liquid) fluids. Thus, the
actual rate of heat transfer

Q = ε ⋅ C min (Tin ,vap − Tin ,liq ) . (80)

Effectiveness of internal heat exchanger and its influence of COP increase in ejector solar air-con-
ditioning system is presented in Figure 38 and Figure 39, respectively.
The application of the internal heat exchange in tested ejection refrigeration system caused increase
of COP up to 20%. The increase of COP scarcely depends on the evaporation temperature. Since the
experimental investigation was carried out for isobutane as the working fluid, the ejector operated with
the superheated vapour. It is a clear need for further investigations of internal heat transfer in the systems
operating with other fluids of different thermodynamic properties, e.g. methanol since the ejector will
partly operate in two-phase region which may lead to significant change of the internal heat transfer
efficiency.

Figure 38. Effectiveness of tested internal heat exchanger

197

Ejection Refrigeration Cycles

Figure 39. Comparison of COP of the ejection system operating with and without internal heat exchange
as a function of evaporation temperature

Table 8. Exemplary COP results for operation with and without internal heat exchanger

tevap [oC] COP COP with


Heat Exchanger
0 0.324 0.378
1 0.400 0.452
2 0.459 0.534
3 0.527 0.608
4 0.556 0.666

EXAMPLES OF NEW APPLICATION OF EJECTION REFRIGERATION SYSTEMS

Gas ejectors become more and more often used device in refrigeration technology. This chapter presents
an exemplary applications of the gas ejector which were done by the author research team in frame of
the national and international projects. Author of this book was the head of those projects. Main goal of
those project was design and manufacture the prototype ejector for dedicated working fluid and nominal
parameters. The design process is preceded by the numerical analysis for various operating condition
and geometry modification in order to find the highest ejector performance. Numerical results are verify
by experimental investigation.

198

Ejection Refrigeration Cycles

The ejector geometry and testing stand used in experiments as well as results of numerical and ex-
perimental investigation of solar air-conditioning were presented earlier (Śmierciew, 2011; Butrymowicz
et al., 2010). In following figures the photographs or sketches and exemplary results of designed ejector
for different application are shown.
In Figure 40 the schematic diagram of the ejector for refrigeration system dedicated to cooling of
electronic devices is presented by Butrymowicz, et al. (2011). Follow, in Figure 41 the photograph of
prototype mini-ejector is shown. The discretized geometry of the tested ejector is shown in Figure 42.

Figure 40. Schematic of the prototype mini-ejector dedicated to cooling of electronic devices

Figure 41. Photograph of the prototype mini-ejector

199

Ejection Refrigeration Cycles

Figure 42. Part of the calculation geometry of the analyzed micro-ejector in 3D

Results of numerical and experimental investigation, e.g. the performance line Π = f(U) are shown
in Figure 43, while the velocity field is shown in Figure 44. Mini-ejector operated with isobutane.
Other application of gas ejector is ejection air-conditioning for cars (Butrymowicz, Śmierciew, &
Karwacki, 2011). Photo of the designed and manufactured prototype ejector was showed in Figure 45.
Ejector was designed for operating with refrigerant R134a in the supercritical state. Theoretical analysis
with 0D model and CFD techniques were used to design and manufacture the prototype ejector.

Figure 43. Mass entrainment ratio vs. compression ratio; comparison of CFD calculations with mea-
surements

200

Ejection Refrigeration Cycles

Figure 44. Velocity field inside the ejector operating with isobutane (2D plane)

Figure 45. Prototype ejector for mobile ejection air-conditioning

Flow filed in range of Mach number for both regime on- and off-design was presented in Figure 46
and Figure 47, respectively.
The ejector operating with R143a will be also use in refrigeration system in cogeneration with Or-
ganic Rankin Cycle (see Figure 48). The project is still in progress, therefore only concept of prototype
ejector is shown in Figure 49.

CONCLUSION

The recent activities of the own numerical and experimental investigation of the ejection refrigeration
systems in aplication to air-conditioning system driven by low temperature heat source were presented
in this chapter. The basics principles of the operation of the ejection refrigeration system were presented.
Also, the most commonly numerical modelling approach including CFD were presented. The influence
of the operating parameters such as motive vapour temperature and condensation temperature based on
the experimental data were presented and discussed. The experimental investigations confirmed that the

201

Ejection Refrigeration Cycles

Figure 46. Mach number field in the ejector operating in the on-design conditions

Figure 47. Mach number field in the ejector operating in the off-design conditions

Figure 48. Schematic of cogeneration with ejection cooling device

ejection cycle operating with natural fluids, e.g. isobutane can effectively be driven by low temperature
heat source, e.g. lower that tg = 80 °C. Under this range of motive temperature heat sources the ejection
cycles can be considered as truly competitive in comparison with absorption refrigeration systems. It
was shown that there is a maximum value of the mass entrainment ratio as a function of the motive va-
pour temperature. It was shown that for higher value of the motive vapour the ejector operates at higher

202

Ejection Refrigeration Cycles

Figure 49. Concept of prototype ejector geometry for refrigeration device for ORC system

values of the condensation temperature. Also it was shown that the motive vapour temperature does
not influence on the ejector efficiency at the on-design operating conditions. The obtained value of the
refrigeration system efficiency COP = 0.15 agrees very well with the results of the theoretical analysis.
A simple method for improvement of the COP of the system by means of application of the internal
heat exchanger were presented, discussed and experimentally confirmed. Several exemplary potential
applications of the ejector were presented.

REFERENCES

Buecker, D., & Wagner, W. (2006). Reference Equations of State for the Thermodynamic Properties of
Fluid Phase n-Butane and Isobutane. Journal of Physical and Chemical Reference Data, 35(2), 929–1019.
doi:10.1063/1.1901687
Butrymowicz, D., Karwacki, J., Śmierciew, K., Przybyliński, T., Lackowski, M., & Ochrymiuk, T. …
Maxfield, B. (2011). Design of ejector refrigeration systems for cooling of electronic equipment. Un-
published Final Report, Magnetic Development, Madison, CO, USA.
Butrymowicz, D., Śmierciew, K., & Karwacki, J. (2010). Investigation of ejection air-conditioning systems
operating with natural working fluid – Unpublished Final Report of the Project N512 013 31/1179 (in
Polish), The Szewalski Institute of Fluid-Flow Machinery of Polish Academy of Sciences, Gdańsk, Poland.
Butrymowicz, D., Śmierciew, K., & Karwacki, J. (2010). Solar, Ejector-Based VCC Air-Conditioner
Utilizing Natural Refrigerants. Unpublished Final Report of supported by the National Science Founda-
tion and by the US Department of Energy, Magnetic Development, Madison, CO, USA.
Butrymowicz, D., Śmierciew, K., & Karwacki, J. (2011). Motive nozzles with diffusers, Unpublished
Final Report, Technische Universität Berlin, Germany.

203

Ejection Refrigeration Cycles

Butrymowicz, D., Śmierciew, K., & Karwacki, J. (2014). Investigation of internal heat transfer in
ejection refrigeration systems. International Journal of Refrigeration, 40, 131–139. doi:10.1016/j.
ijrefrig.2013.11.016
Butrymowicz, D., Śmierciew, K., Karwacki, J., & Gagan, J. (2014). Experimental investigations of
low-temperature driven ejection refrigeration cycle operating with isobutane. International Journal of
Refrigeration, 39, 196–209. doi:10.1016/j.ijrefrig.2013.10.008
Butrymowicz, D., Trela, M., Karwacki, J., Ochrymiuk, T., & Śmierciew, K. (2008). Investigation and
modeling of ejector for air-conditioning systems. Archives of Thermodynamics, 29(4), 27–40.
Dvořák, V., & Vit, T. (2005). Experimental And Numerical Study Of Constant Area Mixing. Paper pre-
sented at 16th International Symposium on Transport Phenomena, Prague.
Huang, B. J., Chang, J. M., Wang, C. P., & Petrenko, V. A. (1999). A 1-D analysis of ejector performance.
International Journal of Refrigeration, 22(5), 354–364. doi:10.1016/S0140-7007(99)00004-3
Keenan, J. H., Neuman, E. P., & Lustwerk, F. (1950). An Investigation of Ejector Design, Analysis and
Experiment. Journal of Applied Mechanics. Trans ASME, 72, 299–309.
Lemmon, E. W., McLinden, M. O., & Huber, M. L. (2010). NIST Standard Reference Database 23:
Reference Fluid Thermodynamic and Transport Properties-REFPROP, Version 9.0, National Institute
of Standards and Technology, Standard Reference Data Program, Gaithersburg.
Munday, J. T., & Bagster, D. F. (1977). A New Ejector Theory Applied to Steam Jet Refrigeration.
Industrial & Engineering Chemistry Process Design and Development, 16(4), 442–449. doi:10.1021/
i260064a003
Paliwoda, A. (1971). Ejector cooling devices (in Polish). Warsaw: WNT.
Śmierciew, K. (2011). Theoretical and experimental investigation of driving ejector in ejection air-
conditioning devices (in Polish) (Unpublished doctoral dissertation), The Szewalski Institute of Fluid-
Flow Machinery of Polish Academy of Sciences, Gdańsk, Poland.
Śmierciew, K., Butrymowicz, D., & Karwacki, J. (2008a). Analysis of thermodynamic cycles of ejection
systems, part II [in Polish]. Technika Chłodnicza i Klimatyzacyjna, 152, 408–412.
Śmierciew, K., Butrymowicz, D., & Karwacki, J. (2008b). Analysis of thermodynamic cycles of ejection
systems, part III [in Polish]. Technika Chłodnicza i Klimatyzacyjna, 154, 502–508.
Śmierciew, K., Gagan, J., Butrymowicz, D., & Karwacki, J. (2014). Experimental investigations of
solar driven ejector air-conditioning system. Energy and Building, 80, 260–267. doi:10.1016/j.en-
build.2014.05.033
Sokolov, E. I., & Zinger, N. M. (1989). Jet apparatuses. Moscow: Energoatomizdat. (in Russian)
Sun, D. W. (1996). Variable geometry ejectors and their applications in ejector refrigeration systems.
Energy, 21(10), 919–928. doi:10.1016/0360-5442(96)00038-2

204

Ejection Refrigeration Cycles

Sun, D. W. (1997). Solar powered combined ejector-vapour compression cycle for air conditioning and
refrigeration. Energy Conversion and Management, 38(5), 479–491. doi:10.1016/S0196-8904(96)00063-5
Sun, D. W., Eames, I. E., & Aphornatana, S. (1996). Evaluation of a novel combined ejector – absorption
refrigeration cycle – I: Computer simulation. International Journal of Refrigeration, 19(3), 172–180.
doi:10.1016/0140-7007(96)00010-2
Yu, J., Chen, H., Ren, Y., & Li, Y. (2006). A new ejector refrigeration system with an additional jet
pump. Applied Thermal Engineering, 26(2-3), 312–319. doi:10.1016/j.applthermaleng.2005.04.018

KEY TERMS AND DEFINITIONS

Coefficient of Performance: Ratio of the cooling capacity of the system to the motive power.
Compression Ratio: Difference between discharge pressure, usually equal condensation pressure
and suction pressure, usually equal evaporation pressure divided by difference of the motive pressure
and suction pressure.
Ejection Refrigeration Systems: Refrigeration systems applied with supersonic gas ejector instead
mechanical compressor.
Low Temperature Heat Source: Heat source with temperature lower than 80 °C, e.g. waste heat
used as motive energy for ejector refrigeration systems.
Mass Entrainment Ratio: Defined as the amount of motive fluid required to entrain and compress
a given amount of suction fluid.
Natural Working Fluid: Naturally occurring substances (refrigerants), i.e: ammonia, hydrocarbons
(propane, butane), carbon dioxide, water.
Off-Design: Operating regime of the ejector at which only primary fluid is chocked (supersonic
speed) and the secondary one is unchecked (subsonic speed).
On-Design: Operating regime of the ejector at which primary and secondary are chocked, velocities
of both fluids are sonic or supersonic.
Primary and Secondary Fluid: Motive fluid and fluid sucked in by ejector, respectively.

205
Ejection Refrigeration Cycles

APPENDIX: NOMENCLATURE

A = area, m2
C = coefficient
COP = coefficient of performance
D = diameter, m
h = specific enthalpy, J/kg
k = heat transfer coefficient, W/(m2⋅K)
K = momentum loss coefficient,
L = length, m
m = mass flow rate, kg/s
p = pressure, Pa
P = driving power, W
Q = heat flux, W
R = reacting force, N
s = specific entropy, J/(kg⋅K)
t = temperature, °C
T = temperature, K
U = mass entrainment ratio,
w = velocity, m/s

Greek Symbols

β = critical pressure ratio


Δ = difference
= effectiveness of heat exchanger
φ = velocity coefficient
η = efficiency
κ = coefficient of isentropic expansion
Π = compression ratio of ejector, Π = (pd − pi ) (pwi − pi )
p = pump, primary flow
s = isentropic process, secondary flow
t = throat
v = vapour
x = internal heat exchanger

206
207

Chapter 6
Current and Future Trends of
Refrigerants Development
M. V. Duarte P. D. Silva
University of Beira Interior, Portugal University of Beira Interior, Portugal

L. C. Pires P. D. Gaspar
University of Beira Interior, Portugal University of Beira Interior, Portugal

ABSTRACT
In this chapter is addressed the thematic of refrigerants: its historical evolution; properties; legislation
applied in the area and future trends. The first refrigerant being marketed on a large scale was ethyl
ether (R610), in 1834. Since then, the evolution of the utilized refrigerants was stimulated, initially due
to constructive issues in the refrigeration system and later to environmental issues. This evolution may
be divided into four generations: 1st use of any fluid that worked; 2nd safety and durability of the equip-
ment; 3rd ozone layer protection and 4th increase of global warming concerns. During the process of
evolution many refrigerants were tested to understanding of their properties. Currently, environmental
concerns are taken as guide in the search for new refrigerants. The most promising refrigerants to be
used in future are the HFEs, HFOs and HFCs with low-GWP, natural refrigerants and blends between
(HCs/HFCs and HFCs/HFOs) refrigerants.

INTRODUCTION

Refrigerants are all the fluids that have a high capacity to “absorb” large amounts of thermal energy
during the evaporation process at low temperature. This characteristic makes it ideal for use in refrigera-
tion systems that operate a refrigeration vapour compression cycle, the cycle applied in air condition-
ers, heat pumps and cooling systems. Refrigerants can be classified as primary or secondary. Primary
refrigerants are those that circulate inside the cooling system, being compressed in the compressor and
expanded in the expansion device, i.e. those whose width of biphasic zone measured in (p, h) diagram
is important. Some examples of primary refrigerants are the halogenated compounds, hydrocarbons,
inorganic compounds, azeotropic and zeotropic mixtures. Secondary refrigerants are usually liquids used

DOI: 10.4018/978-1-4666-8398-3.ch006

Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

Current and Future Trends of Refrigerants Development

as heat carriers at low temperature from primary refrigerant to other fluid. This type of refrigerants is
only used in refrigeration vapour compression systems that operate by indirect expansion. Are examples
of secondary refrigerants, the brines and antifreezes.
Exceptionally, absorption refrigeration systems use two substances (refrigerant and absorbent). In
these systems it is intended to take advantage of the chemical affinity existing between these two fluids,
in order to its combination and dissociation allows the desired cooling effect.
The ethyl ether (R610) was the first fluid known with excellent thermodynamic properties that makes
it a good refrigerant (i.e. the evaporative effect at low temperature). This characteristic was perceptible
when a small amount is spilled on the hand. The almost instantaneous evaporative effect causes a cool-
ing sensation in the skin, while a portion of substance is evaporated “stealing” thermal energy at hand.
For this reason, the ethyl ether (R610) was used in the first refrigerating machine patented to operate
a refrigeration vapour compression cycle, machine built in 1834 by Jacob Perkins. This was the first
refrigerant marketed in a large scale (Dinçer & Kanoglu, 2010).
Since then, there has been an impressive evolution of the used refrigerants, stimulated initially due
to constructive issues related with the refrigeration system and later as environmental issues.
This evolution may be divided into four generations (Calm, 2008):

• 1st marked by the use of any fluid that worked – the natural refrigerants were the firsts to be used
because they abound in nature. But the use of some natural refrigerants was very dangerous as
they are highly toxic and flammable, having occurred several accidents, including people’s death;
• 2nd marked by the safety and durability of the equipment – in attempting to find more stable re-
frigerants whose use was safer for equipment and users, were discovered the CFCs compounds
in 1928. These compounds revolutionized the chemical industry, having been used in several ap-
plications beyond refrigeration;
• 3rd marked by the ozone layer protection – for being too stable the CFCs compounds are very
harmful to environment, with great participation in the destruction process of ozone layer and
increase of global warming;
• 4th marked by the increase of global warming – currently, the trend is towards the disuse of CFCs
and HCFCs compounds and bet on HFCs with a low-GWP, HFOs, HFEs and the return to natural
refrigerants.

In this chapter are discussed the thematic related with the refrigerants and its historical evolution,
taking into account the main events that marked this evolution. The main characteristics and properties
that characterize the refrigerants in order to allow its classification according to the respective family
also are presented. Finally are discussed the different replacement possibilities for the currently most
used refrigerants and those that are indicated as long-term future possibilities, the different substitution
methods and collection and recycling of refrigerants.

FRAMEWORK

Refrigerants are currently used in air conditioning systems, commercial refrigeration systems, domestic
refrigerators and freezers, mobile refrigeration systems, cryogenic and low temperature freezers and
industrial refrigeration systems.

208

Current and Future Trends of Refrigerants Development

Air Conditioning Systems

The unitary air conditioning systems (including heat pumps) have an average lifespan of 15 years. China
is the largest manufacturer of air condition equipment, holding about half of world production of these
systems. The electrical power consumption of these devices ranges from 1 - 135 kW and the refrigerant
charge between 0.5 - 90 kg. Estimative indicate that air conditioners were responsible for the emission
of 91 million metric tonnes of carbon dioxide equivalent (MMtCO2-eq.), equivalent to 8% of global
consumption of HFCs refrigerants in 2010. Unitary air conditioning systems are responsible for 11% of
refrigerants consumption in the refrigeration/air conditioning sector. It is expected that this percentage
will rise to 38% in next year’s due to transition from HCFCs to HFCs refrigerants with low-GWP value.
This growth is driven by the increasing use of air conditioning systems in developing countries (U. S.
EPA, 2010a).
The most used refrigerant in unitary air conditioning systems is R22 (HCFC), used in about 60% of
worldwide systems in 2010. Since 2000, in developed countries began to be used the refrigerants R410A
and R407C (both HFCs). In 2010, these refrigerants were used in approximately 39% of air conditioning
systems worldwide. In the same year, unitary air conditioning systems are already being marketed using
R290 (HC), representing about 1% of worldwide systems. In addition to propane (R290), are also being
studied for application in air conditioning systems the CO2 (R744), HFCs (R32) and HFO (R1234yf).
In several service buildings (i.e. offices and hospitals), chillers are used for distribution of cold water
in the buildings. The cool water is used to cooling the air in the individual rooms. Previously these sys-
tems operate with CFCs, HCFCs and HFCs refrigerants, but recently, ammonia based and hydrocarbon
based refrigeration systems started to be used. Some small systems based on HFCs refrigerants (with
charge below 10 kg) are still being commercialized (Pedersen, 2012).

Commercial Refrigeration Systems

Commercial refrigeration systems include all refrigerated equipment existing in supermarkets, convenience
stores, restaurants and other food service establishments (including vending machines). These systems
can be characterized according to its level of operating temperature: chilled food in the range of 1 to 14
ºC and frozen food in the range of -12 ºC to -20 ºC (Devotta et al., 2005), influencing this characteristic
the choice of ideal refrigerant to used.
In 2006, there were about 530,000 supermarkets worldwide, containing about 546,000 metric tonnes
of refrigerant. From this amount 55% are HCFCs, 30% CFCs and 15% HFCs. Estimative indicate that
commercial refrigeration systems are responsible for the emission of 346 million metric tonnes of carbon
dioxide equivalent (MMtCO2-eq.), equivalent to 32% of global consumption of HFCs refrigerants in
2010. From this value, the developing countries are responsible for an estimated 131 million metric tonnes
of carbon dioxide equivalent (MMtCO2-eq.) or 38% of the global HFCs consumption in this end-use.
Commercial refrigeration systems have a lifespan range of 15 - 20 years and an approximate refriger-
ant charge between 0.15 - 20 kg. Several designs of refrigeration systems and its components have been
studied in order to reduce: the amount of refrigerant required for systems; the probability of leakage and
the risk of use toxic or flammable refrigerants (U. S. EPA, 2010b).
Today the most used refrigerants in commercial refrigeration systems are R22 (HCFC), R134a,
R404A, R407C and R507A (HFCs). In 2010, the R22 (HCFC) was the most widely used in multiplex

209

Current and Future Trends of Refrigerants Development

rack systems. As alternative refrigerants to be used in this type of refrigeration systems are highlighted
the hydrocarbon compounds (R290 and R600a), propylene (R1270), ammonia (R717), CO2 (R744) and
other alternatives as mixtures between the HFCs and HFOs compounds.
In 2007 a group of international companies launched the program called “Refrigerants Naturally”.
With this program, companies intent to promote the utilization of alternative HFCs-free refrigeration
technologies in the food and drink, food service and retail sectors. This program stimulated the use of
R290, R744 and R600a refrigerants. Unilever Company has introduced freezers for ice cream conserva-
tion using R290 refrigerant, about 2000 units in Europe with a refrigerant charge of 90 grams. These
devices present an increased efficiency of 9% and do not require a greater maintenance when compared
with traditional devices using R404A refrigerant. To the same application, Embraco and Nestlé have
developed freezers using R744 (replacing the R404A freezers). Currently, several units using the R744
are being tested in Europe (Melo, 2011).
Transcritical CO2 systems technology is currently the most used in the construction of new refrigera-
tion units in supermarkets exceeding 10 kg of refrigerant. It was estimated that in 2011, the total amount
of transcritical CO2 systems existing in supermarkets was about 2000 units (Pedersen, 2012).

Domestic Refrigerators and Freezers

The expected life time for domestic refrigerators and freezers varies between 15 - 20 years. In 2009, it
was estimated the existence of 1.5 - 1.8 billion of this equipment operating worldwide. Every year are
produced approximately 100 million units of new refrigerators and freezers systems typically containing
between 0.05 - 0.25 kg of refrigerant. The amount of refrigerant charge in this type of equipment has
been reduced due to the introduction of technological innovations in the system and its components.
Domestic refrigeration systems were responsible for the consumption of about 2% of global HFCs
refrigerants consumed in 2010. In developing countries, the global consumption of HFCs refrigerants was
approximately 12% of the total amount consumed by the refrigeration/air conditioning sector in 2010.
R12 was the first to be used on a large scale in domestic refrigerators and freezers. Due to CFC
compounds phase-out, the refrigerant R134a was selected as a substitute, although the hydrocarbon
compounds (HCs) would start to be used in Europe and Japan. A large part of refrigerators/freezers
still use the R134a refrigerant. In China, 75% of new refrigerators/freezers use isobutane (R600a) as
refrigerant (U. S. EPA, 2010c).
R600a dominates the market of domestic refrigerators in Europe and Asia, but is not used in USA
due its high flammability. In Europe, 95% of domestic refrigerators used operate with R600a. Since
1993 more than 200 million of R600a refrigerators and freezers have been commercialized and until the
moment no accident (in normal operation) has been reported. Refrigerators using R600a are more silent
and efficient than R134a refrigerators (Melo, 2011).

Mobile Refrigeration Systems

As in other applications, also in Mobile Air Conditioning - MAC (i.e. the refrigeration systems installed
in cars, trains, aircrafts, ships and containers) the refrigerant R134a (HFC) come replaced the R12 (CFC)
due to impositions agreed in the Montreal Protocol. R134a refrigerant has been employed in other types

210

Current and Future Trends of Refrigerants Development

of refrigerating systems, but was in the automotive air conditioning systems that became famous (Min-
jares, 2011). Normally a typical refrigerant charge of R134a is 0.8 kg for an existing car (i.e. refill), 0.6
kg for new cars, 1.5 kg for a truck and 5 – 12 kg for a bus. The leakage rate is about 10 - 20% of the total
charge per year (Pedersen, 2012).
With the European MAC Directive publication in 2006, occurred the abolition of refrigerants with a
GWP value > 150 (low-GWP). As the refrigerant R134a presents a high-GWP value, this was abolished,
being in progress its phase-out (until January 1, 2017). The R134a refrigerant is already being replaced
by R152a (HFC) that presents a GWP = 120. The function of R152a refrigerant is to ensure the transi-
tion to a refrigerant that presents a residual GWP value, at least this is the desire of the European Union.
R1234yf was chosen by the automotive industry in 2009 to replace the currently used refrigerant
R134a. After this decision a manufacturer (Daimler) was questioned the safety of R1234yf use. This
situation leads the KBA (German Motor Authority) to proceed with its own testing. The automotive in-
dustry has developed the risk assessment and fault tree analysis of the use of R1234yf in MAC systems,
in cooperation with SAE International, which concludes that this refrigerant was safe to use in MAC
systems (EU, 2014a).
A compilation of information developed through SAE Cooperative Research Programs, chemical sup-
pliers, Original Equipment Manufacturers (OEMs), and other sources, EPA (US Environment Protection
Agency) has approved through its SNAP (Significant New Alternatives Policy) program three low-GWP
refrigerants candidates to be used in air conditioners in new light-duty vehicles. Are also presented the
respective use conditions concerning to security features for each alternative: R152a; R744 and R1234yf
(Sciance, 2013). Presently, automobile and refrigerants manufacturers together with the European Union
try to reach an agreement to which is the refrigerant to be used in this application, in future.
General Motors account already with the following vehicle models on the market using the R1234yf:
in North America (Cadillac XTS in May 2012 and Chevrolet Spark Battery Electric Vehicle in June
2013) and in Europe (Chevrolet Malibu in 2012, Chevrolet Trax and Opel Mokka in January 2013). GM
already has more than 100,000 vehicles on the roads globally, using as refrigerant R1234yf. In Europe,
there are several OEMs (as Hyundai, Subaru, Ford and BMW) that produce components for air condi-
tioning systems of passenger vehicles using R1234yf (Sciance, 2013).
Refrigerated transport growth significantly since the 50s. According to last RTOC report that evalu-
ates in 4 million the terrestrial fleet of refrigerated transport equipment in service worldwide in 2010.
The worldwide fleet of marine reefer containers is evaluated to 850,000 units in 2013. In European
Union-27, the overall refrigerated terrestrial transport equipment is estimated on some 1.1 million equip-
ment in 2012 by Cemafroid. The refrigerated transport sector is dominated by one technology based on
vapour-compression systems using HFCs as refrigerant, namely R404a and R134a. The phase down of
HFCs with high-GWP in European Union (i.e. FGas regulation revision in 2014), in the next years will
directly impact refrigerated transport equipment technology. The transport refrigeration units must change
tremendously during the next years. Between the alternatives, CO2 is considered as a possible long term
alternative in the coming 5 to 10 years. New HFCs and blends of HFCs may be a transitional solution
for short term. The utilization of HFOs is still considered a possibility. Hydrocarbons are certainly too
flammable (Michineau et al., 2014).
In Iceland and Norway, about 20 – 30% of fishing vessels use ammonia as refrigerant. New large
fishing vessels build in Norway have CO2/ammonia cascade systems. To air condition in trains and
aircrafts a cold-air refrigeration system has been developed, using air as refrigerant (Pedersen, 2012).

211

Current and Future Trends of Refrigerants Development

Cryogenic and Low Temperature Freezers

Cryogenic systems and low temperature freezers are a particular area of refrigeration widely used in
laboratories. This equipment normally consists on a cascade system. The R507, or R23 or R508 refriger-
ants can be used in the stage of high temperature (approximately -50 ºC) and hydrocarbons to the stage
of low temperature – ethane R170 (approximately -80 to -90 ºC) and ethane R1150 (approximately
-100 to -120 ºC). As R23, R507 and R508 present very high-GWP values, these have been replaced by
propylene R1270 (Pedersen, 2012).

Industrial Refrigeration Systems

Industrial refrigeration systems are normally very large systems, used in process of refrigeration and
cold storage on food industry, chemical/biochemical industry (oil and gas industries) and ice-making
industry. This type of systems is built on site (Pedersen, 2012).
The capacity of units may vary from 25 kW to 30 MW and with a refrigerant charge size between 20
to 60,000 kg. These types of refrigeration system are normally located in industrial areas with limited
public access, and ammonia is the most widely used refrigerant. The current big trend is to use indirect
refrigeration in order to reduce the refrigerant charge (in 75 to 85%) and avoid ammonia in working areas
(due to its high toxicity). The use of industrial refrigeration plants with CO2 operating a refrigeration
cycle at low temperature in cascade systems is becoming common, in which the ammonia is used in
the high temperature stage and CO2 in the low temperature stage (Pedersen, 2012; Devotta et al., 2005).

Refrigerants Used in the European Union

In UE-27, CO2-eq emissions have decreased from 170 metric tonnes in 1990 to 147 metric tonnes in 2010,
a decrease of 13.5% in 20 years (Figure 1), despite the use of refrigerants having more than doubled in
the same period from 200,000 metric tonnes in 1990 to about 510,000 metric tonnes in 2010, as can be
seen in Figure 2, meaning an average market growth of 8% per annum. This increase may be due to the
accelerated phase-out of CFCs and HCFCs refrigerants under the EU ODS Regulations and introduction
of the EU F-Gas Regulation in 2006 (Clodic & Barrault, 2011).
Although the growing refrigerants utilization in EU, CO2-eq emissions significantly decrease under
the predicted scenario, from 147 metric tonnes in 2010 to 124 metric tonnes in 2030, as can be seen in
Figure 3, meaning a reduction of 15%. According to predicted scenario (i.e. F-Gas Scenario), the total
amount of used refrigerants in EU will grow about 43% between 2010 and 2030, reaching the 900,000
metric tonnes, as can be seen in Figure 4.
Currently several scenarios to the evolution of refrigerants utilization and their respective emis-
sions are proposed, having been presented the least favourable in Figures 3 and 4. In Figures 3 and 4 is
also notable the success of the European Community Directives through reduction of use of CFCs and
HCFCs refrigerants and increasing use of HFCs refrigerants. It is anticipated that HFCs refrigerants will
dominate in terms of use over the next decade.

212

Current and Future Trends of Refrigerants Development

Figure 1. Total CO2-eq. emissions per family of refrigerants in the EU from 1990 to 2010
(Clodic & Barrault, 2011)

Figure 2. Total amount (metric tonnes) of used refrigerants per family in EU from 1990 to 2010
(Clodic & Barrault, 2011)

213

Current and Future Trends of Refrigerants Development

Figure 3. Total CO2-eq. emissions per family of refrigerants in the EU from 1990 to 2030
(Clodic & Barrault, 2011)

CHARACTERIZATION OF REFRIGERANTS

In this section are presented the main characteristics and properties of refrigerants that allow its clas-
sification, according to the respective family.

General Characteristics of a Refrigerant

During the selection process of a refrigerant to use in a refrigeration system or heat pump, its general
characteristics must be considered so it can be applied the most adequate refrigerant to installation and
be able to meet the desired cooling or heating conditions.
The ideal refrigerant must present the following general characteristics (Dinçer & Kanoglu, 2010):

• Environmentally friendly (does not contribute to destruction of ozone layer and increase of green-
house effect);
• Low evaporation temperature;
• Low flow rate per unit of system capacity;
• Pressure of evaporation higher than ambient pressure;

214

Current and Future Trends of Refrigerants Development

Figure 4. Total amount (metric tonnes) of used refrigerants per family in EU from 1990 to 2030
(Clodic & Barrault, 2011)

• High enthalpy of evaporation;


• Low flammability and toxicity;
• Do not be explosive or corrosive;
• Do not be reactive, but compatible with the compressor lubricating oils;
• Chemically stable;
• Suitable thermal and physical properties (i.e. thermal conductivity, viscosity …);
• High commercial availability;
• Easily detectable on leakage;
• Low cost;
• Recyclable.

The saturation properties of refrigerants must also be considered. To achieve a reasonable rate of
thermal energy transfer, the temperature difference between the refrigerant inside the heat exchanger
and the external ambient should be about 10 ºC (i.e. a Δ ≈ 10 ºC). The lower pressure obtained in the

215

Current and Future Trends of Refrigerants Development

refrigeration cycle occurs on evaporator, this pressure must be higher than atmospheric pressure to
prevent any air entry into the refrigeration system. Also the temperature and pressure of refrigerant
on condenser must depend on the thermodynamic and physical properties of the ambient to which the
thermal energy is rejected (i.e. should be considered the amount of thermal energy that ambient can
receive) (Dinçer & Kanoglu, 2010).

Properties of Refrigerants

The characterization of refrigerants is only possible through the study of their properties. The proper-
ties that allow the characterization of a refrigerant are related to safety (toxicity and flammability),
environmental (ODP, GWP, TEWI, LCCP, LCA and residence time in atmosphere), compatibility with
materials (e.g., metals, insulators and lubricants) and performance (efficiency, properties of thermal
energy transfer and critical point). Other important properties are the physical properties such as sound
speed, glide and work pressure (McLinden, 2011; McQuay, 2002).
Are now studied the properties of refrigerants described in the previous paragraph.

Toxicity and Flammability

Toxicity and flammability are the two parameters used by ASHRAE (American Society of Heating,
Refrigerating and Air Conditioning Engineers) to evaluate the level of security resulting of a refrigerant
using. To facilitate this classification, the ASHRAE Standard 34 (2010) has adopted a standard matrix
that relates both parameters, as can be seen in Figure 5.
The toxicity of a substance is defined as its concentration in an environment to cause injury in the
human body (ASHRAE, 2010). A substance that has a high stability in the atmosphere (i.e. remains in
the environment with the same characteristics over a long period of time) is capable to penetrate more
easily into the human body through the respiratory tract. This substance can easily reach to several organs
of the human body with the same characteristics when are emitted, concentrating himself on several

Figure 5. Classification matrix of refrigerants regarding the safety of their use


(ASHRAE, 2010)

216

Current and Future Trends of Refrigerants Development

organs of the human body, thus causing more profound damages. On the other hand, a substance that
is more reactive (i.e. the opposite of stable) interact violently when in contact with the human body,
causing damage in its surface, namely, irritation on skin, respiratory systems and eyes (McQuay, 2002).
Acute toxicity is the term used to indicate human exposure to high concentrations of a substance
for short periods of time. For example: the exposure to a leakage of refrigerant inside a closed space.
Chronic toxicity refers the repetitive human exposure to low concentrations of a substance for ex-
tended periods of time. For example: the exposure experienced by a maintenance and installer technician.
Threshold Limit Value (TLV) is the maximum concentration value of a substance present in an envi-
ronment to prevent the occurrence of problems in the human health, after an exposure to chronic toxicity.
TLV may vary from person to person, according to their habits (smoking, alcohol or other drugs), health
status, age, medication, among others (ASHRAE, 2010).
According to the toxicity level presented by refrigerants, the ASHRAE Standard 34 (2010) classifies
them into two groups, class A and B, as can be seen in Figure 5.
The flammability of a substance is defined as the ease with which it ignites and burn. Lower Flam-
mability Limit (LFL) represents the minimum concentration value of a substance in air at the STP con-
ditions, in a perfect homogeneous mixture, in which exist the possibility of the substance ignite, being
expressed in [g/cm3]. Flame spread velocity shows the speed at which the combustion reaction occurs,
being expressed in [m/s]. Heat Of Combustion (HOC) represents the thermal energy released during the
combustion of a substance. This index can be expressed in the following units [kJ/kg] (ASHRAE, 2010).
According to the level of flammability presented by refrigerants, the ASHRAE Standard 34 (2010)
classifies them into four groups, class 1, 2, 2L, and 3, as illustrated in Figure 5.
From study of the matrix present in Figure 5 it can be concluded that:

Class A: For refrigerants belonging to this class, there are no reported cases of intoxication in humans
for concentrations less than or equal to 400 ppm by volume. This determination was based on the
TVL of the respective refrigerant;
Class B: For refrigerants belonging to this class, were found cases of intoxication in humans exposed
to concentrations below 400 ppm by volume. This determination was based on the TVL of the
respective refrigerant;
Class 1: For refrigerants belonging to this class, the flame propagation on STP conditions is not observed;
Class 2: The refrigerants belonging to this class have a low LFL value (i.e. low flammability), above
0.17 [g/cm3], a HOC value below 18.996 [kJ/kg] and a low velocity of flame spread, below 10
[cm/s], for STP conditions;
Class 2L: The refrigerants belonging to this class have a low LFL value (i.e. low flammability), above
0.17 [g/cm3], a HOC value higher to 18.996 [kJ/kg] and a high velocity of flame spread, above 10
[cm/s], for STP conditions;
Class 3: The refrigerants belonging to this class are very flammable. These has a high LFL value, below
0.17 [g/cm3], a HOC value higher to 18.996 [kJ/kg] and a high velocity of flame spread, above 10
[cm/s] for STP conditions.

It is possible to predict some characteristics of refrigerants, including flammability, toxicity and the
residence time in atmosphere, through the study of the chemical composition of substances, as can be
seen in Figure 6. In table 2 is presented the flammability for several families of refrigerants.
From the analysis of Figure 6 it may be concluded that:

217

Current and Future Trends of Refrigerants Development

Figure 6. Triangle expressing the relation between chemical constitution, flammability, toxicity and the
residence time in atmosphere of organic refrigerants
(Calm & Didion, 1998)

• Refrigerants rich in hydrogen (H) are highly flammable and have a very low residence time in the
atmosphere (consequence of its high reactivity). Example: the hydrocarbon compounds.
• Refrigerants rich in chlorine (Cl) and bromine (Br) are highly toxic. Example: ammonia.
• Refrigerants rich in fluorine (F) and chlorine (Cl) present a high residence time in atmosphere
(consequence of its low reactivity). Examples: the family of CFCs compounds (halogenated
compounds).

Efficiency

The higher the efficiency of a refrigeration system, the less the amount of electrical energy required to
promote a similar cooling capacity. The efficiency of a refrigeration machine, thermal efficiency to first
law or COP (Coefficient of Performance) may be defined as the ratio between the energy objective of
the cycle and the amount of energy introduced into the cycle to reach the energy aim.
Carnot cycle represents the theoretical cycle completely reversible for a refrigeration system whose
efficiency may be expressed only depending on temperatures of the heat sources (i.e.,
Tf Tq
COPMF (CARNOT ) = and COPBC (CARNOT ) = ). Note that properties of refrigerants are not
Tq −Tf Tq −Tf

218

Current and Future Trends of Refrigerants Development

included in this equation. So efficiency can be considered a characteristic of the refrigeration system,
not a property of refrigerant. There are other parameters inherent to the refrigeration system, thus not
depending on the property of refrigerants, which directly influence the efficiency of these systems. Some
of these parameters are the efficiency of electric motor, compressor efficiency (at full load or partial
load), design and construction of heat exchangers and operating conditions. Others properties of refrig-
erant that may affect the efficiency of the refrigeration system are the transfer properties (compress-
ibility), thermal energy transfer properties and acoustic velocity, among others (McQuay, 2002). In table
2 is presented the system efficiency for several families of refrigerants.

Thermal Energy Transfer Properties

The efficiency of a refrigeration cycle is maximum, only if the refrigerant presents a high capacity to
transfer thermal energy. The use of a refrigerant that presents good properties of heat transfer, property
that is reflected in its heat transfer coefficient, allows the construction of more compact refrigeration
systems, especially as regards the heat exchangers. Thus, the decrease in size of heat exchangers will
result in a price decrease of the refrigeration system. There are several factors that influence the heat
transferred by a refrigerant. Some of these factors are related with characteristics of the refrigeration
system components. For example: the layout and design of piping and other elements, and flow rate
(Reynolds number). Other properties of refrigerants that affect the overall heat transfer ability of the
refrigeration system are dynamic viscosity (μ), specific heat (cp) and thermal conductivity (k). These
 c 

properties are used to calculate the Prandtl number Pr = µ × p  , a factor widely used on the sizing
 k 
and design of heat exchangers.
The objective is to use as refrigerant, a substance capable to “carry” large amounts of energy (i.e.
presents a higher specific heat) and easily transfers energy (i.e. have a higher thermal conductivity). It is
also desirable that substance presents a low viscosity, thus allowing the increase of turbulence on flow and
reduces the mechanical power produced by the compressor to promote its movement (McQuay, 2002).

ODP (Ozone Depletion Potential)

The ODP value is the index that characterizes the participation of a refrigerant molecule in the process
of ozone layer destruction. The value of this index is calculated through the comparison with the contri-
bution to the process of ozone layer destruction of the reference molecule, the molecule of R11 or R12,
both presenting a ODP = 1. This value is considered the maximum potential of ozone layer destruction
(Benhadid-Dib & Benzaoui, 2012).
With the signature of the Montreal Protocol in 1989, it was agreed between the signatory countries
the abolition of all refrigerants that present a value of ODP > 0. These were gradually replaced by others
that do not promote the ozone layer depletion (Calm, 2008). Currently the production and commercial-
ization of refrigerants that present a value of ODP > 0 is prohibited in the European Union, as well as
the commercialization of refrigeration systems prepared to use this type of refrigerants (Benhadid-Dib
& Benzaoui, 2012). It is possible to predict the contribution of refrigerants to the process of ozone layer
destruction through the study of its chemical composition, as can be seen in Figure 7.

219

Current and Future Trends of Refrigerants Development

Analysing the Figure 7 it can be concluded that the ODP value of a refrigerant increases with the
concentration of chlorine (Cl) or bromine (Br) in chemical constitution of substance.
Halogenated compounds (CFCs), rich in chlorine and fluorine atoms are very stable in nature. When
they reach the Earth’s stratosphere, these compounds react with ozone molecules due to interaction with
the incident sunlight, dissociating them according to the chemical equation

(CFC +UV solar radiation + 2O 3 )


→ 3O2 + Cl * – named photodissociation.

The result of this reaction is chlorine free radicals that present a high residence time in atmosphere.
During his lifetime, free radicals of chlorine can destroy thousands of ozone molecules, until his com-
bination with other substance beyond the ozone molecules.
The ozone layer destruction leads to increased entry of solar ultraviolet radiation on Earth atmosphere,
which represents a high risk to public health, including the increase incidence of skin cancer and burns
caused by prolonged exposure to solar radiation (Dinçer & Kanoglu, 2010). In tables 2 and 8 is presented
the ODP for several families of refrigerants.

GWP (Global Warming Potential)

GWP value is an index that characterizes the participation of a refrigerant molecule in the greenhouse
increase. The value of this index is calculated through the comparison with the contribution to the
greenhouse effect of the reference molecule, a molecule of CO2, presenting a GWP = 1.

Figure 7. Triangle expressing the relation between chemical constitution and the ODP value of organic
refrigerants
(Calm & Didion, 1998)

220

Current and Future Trends of Refrigerants Development

Figure 8. Triangle expressing the relation between chemical constitution and the GWP value of organic
refrigerants
(Calm & Didion, 1998)

This comparison can be made according to heating effect caused by a molecule of refrigerant during
different time periods (20, 100 or 500 years), being the most common used the 100 years (GWP100).
With the application of the Kyoto Protocol in 2005, it was agreed between the signatory countries,
the abolishing of all refrigerants that present a high-GWP value. These were gradually replaced by others
that do not promote the increased greenhouse effect (Benhadid-Dib & Benzaoui, 2012).
Recently several countries have published legislation in order to reach the targets agreed in the Kyoto
Protocol. Are signatories of the Kyoto Protocol 175 countries, except the United States of America and
others (UN, 2014). It is also possible to predict the contribution of refrigerants to the process of global
warming through the study of its chemical composition, as can be seen in Figure 8.
Analysing the Figure 8 it can be concluded that GWP value of a refrigerant increases with the con-
centration of fluorine (F) atoms present in chemical constitution of substance. GWP value of a refrig-
erant increases with the ability of its molecules in absorb solar radiation reflected by Earth (infrared
radiation) and its residence time in the atmosphere. The increase of greenhouse effect may result in: an
intermediate warming of the atmosphere (estimated as 3 to 5 ºC by 2050); a rise in the level of oceans
(estimated as 20 cm by 2050) and climatic effects (increases in drought, rain, snow, warming and cool-
ing) (Dinçer & Kanoglu, 2010).
Molecules with higher GWP values are those that present chemical bonds of carbon-fluorine (C-F).
Currently, about 95% of refrigerants of the HFCs family used worldwide present a GWP value be-
tween 700 and 4000 (Kohler, 2012). TEAP (Technology & Economic Assessment Panel), technical and

221

Current and Future Trends of Refrigerants Development

research unit created by UNEP (United Nations Environment Programme - Ozone Secretariat), proposes
the classification of refrigerants according its GWP value could be seen in Table 1. In tables 2 and 8 is
presented the GWP for several families of refrigerants.

TEWI (Total Equivalent Warming Impact)

TEWI value expresses the performance ratio between the refrigeration cycle and its overall impact on
the environment resulting from use of a refrigerant. Applying this parameter is possible to determine
the total amount of GHG emissions related to use of a refrigerant through the accounting of direct emis-
sions (leakages) and indirect emissions (GHG emissions resulting from electricity production that is
consumed by the refrigeration system - parameter related with the refrigeration system efficiency), as
can be seen in Figure 9.
The value of indirect GHG emissions (registered during the refrigerant lifespan) represents the largest
share of GHG emissions related to a refrigerant use. Thus, TEWI index depends on how the electricity
power is generated. Replacing a refrigerant that presents a high-GWP value by other with a lower GWP
value, counting only with the value of direct GHG emissions is not the best solution. If the substitute
refrigerant has worst thermophysical and thermal energy transfer properties, the replacing will result in
a decreased performance of the refrigeration cycle. Thus the refrigeration systems need to consume a
larger amount of electric energy to achieve an equal cooling or heating capacity.
In summary, the replacement of a refrigerant needs to be properly planned or will lead to increased
costs of operating the system and increase of GHG emissions. Through this parameter is possible to
evaluate more comprehensively the feasibility of replacing a refrigerant.
As weak points of this parameter, not include the GHG emissions related with production, handling
and transportation of refrigerants.
The value of this parameter is obtained by applying the following equation (Benhadid-Dib & Ben-
zaoui, 2012):

( )
TEWI = GWP L × n + m 1 − C  + (n × E × β ) (1)

where:

Table 1. Classification table of refrigerants according its GWP value

GWP100 Classification
GWP > 30 Ultra low-GWP
GWP > 100 Very low-GWP
GWP > 300 Low-GWP
GWP > 1000 Moderate-GWP
GWP > 3000 High-GWP
GWP > 10 000 Very High-GWP
GWP > 30 000 Ultra High-GWP
(Kohler, 2012)

222

Current and Future Trends of Refrigerants Development

Figure 9. Simplified scheme to the analysis of TEWI concept


(Kohler, 2012; Hwang, 2013a)

GWP – Global Warming Potential;


L – Annual amount of the refrigerant emissions (leakage) [kg];
n – Lifetime of the refrigeration system [years];
m – Refrigerant charge [kg];
C – Recovery factor/refrigerant recycling [between 0 and 1];
E – Annual energy consumption of the refrigeration system [kWh];
 kg 
β – Carbon dioxide emissions  .
 kWh 
 

223

Current and Future Trends of Refrigerants Development

LCCP (Life Cycle Climate Performance)

The LCCP is the natural broadening of the TEWI index, which focused on refrigerant during their manu-
facture and usable life. All the direct and indirect contribution to Global Warming is taken into account
from its manufacturing to their disposal (include: recovery and final treatment), passing through their
use (operation), see Figure 10.
To evaluate the global impact of a refrigerant is necessary to define its production method, how is
used and wasted (Riva et al., 2006).
The value of LCCP can be obtained by adding the GWP values resulting from emissions during the
refrigerant manufacturing process on the Equation 1 of TEWI, namely the GWP resulting from indirect
emissions - RIE (energy consumption during manufacturing) and GWP resulting from direct emissions
- FE (fugitive emissions during the refrigerant production), as can be seen on Equation 2.

( )
TEWI = (GWP + EIR + FE )× L × n + m 1 − C  + (n × E × β ) (2)

where:

FE – Fugitive emissions;
RIE – Refrigerant Indirect Emissions during the manufacturing process.

There are specific software applied to each area of refrigerants use to calculate the value of LCCP.
This software has in its database, pre-defined data on CO2-eq emissions resulting from the production
and use of a refrigerant according to the climate conditions existing in the location where the refriger-
ant is applied. An example of this software is the GREEN-MAC, applied by the automotive industry
(Hwang, 2013b).

LCA (Life Cycle Assessment – LCCO2)

LCA also known as life cycle analysis, ecobalance or cradle-to-grave analysis is the valuation of the en-
vironmental impacts of a given product or service caused by its existence. This parameter joins together

Figure 10. Simplified scheme to the analysis of LCCP concept


(Hwang, 2013a)

224

Current and Future Trends of Refrigerants Development

with the Global Warming Impact, also the eutrophisation potential, the acidification potential and the
ozone depletion potential. The advantage that LCA present in relation to LCCP is in its standardiza-
tion: LCA is ruled under ISO 14000 Standards. Thus, the LCA is most accurate in calculating the GHG
emissions (Rebitzer et al., 2004).
The “Carbon Footprint” is considered an output from the LCA and is considered one the most ob-
jective result of a LCA study. To calculate the “Carbon Footprint” parameter is used the LCA method
(POST, 2006).
Applying the concept of “Carbon Footprint” to a refrigerant allows the calculation of the total amount
of CO2 and other GHG, emitted over the full life cycle of a refrigerant, as can be seen in Figure 11. The
LCA final result is expressed in grams of CO2-eq per kilowatt hour of generation (gCO2-eq/kWh).
LCA is a calculation method of CO2-eq emissions more detailed and expansive than TEWI and
LCCP. When applied, LCA method allows the calculation of total direct and indirect emissions resulting
from the whole lifecycle of refrigerant and the devices that supports it (i.e. the entire components of a
refrigeration system), namely during (Rebitzer et al., 2004):

• Design and development of technologies;


• Row material acquisition and transformation;
• Production of system devices components and refrigerant;

Figure 11. Simplified scheme of the environmental impact of a refrigerant


(Hwang, 2013a; 2013b)

225

Current and Future Trends of Refrigerants Development

• Process of assembly;
• Operation (includes servicing);
• Recovery, recycling or destruction of refrigerant and system devices.

LCA method also includes the CO2-eq emissions resulting from energy consumption, transportation and
solid, liquid and gaseous waist produced between all steps of a refrigerant and system devices life cycle.

Compatibility with Materials

Before the commercialization of a new refrigerant it is important to understand how it reacts when
in contact with the materials constituents of the refrigeration system (copper, steel, brass, lubricants,
gaskets, seals, among others). All the compatibilities between the materials and refrigerant should be
studied in order to understand if will cause the degradation of thermophysical properties and decrease
of thermal energy transfer by the refrigerant. The damage caused on the components of the refrigeration
system due to non-compatibilities with the refrigerant will allow the refrigerant leakage to the ambient.
The use of inert refrigerants, substances that present a high stability in the ambient, it would be the ideal
option in terms of materials compatibility. Thus, the compatibility with materials is presented as one of
the greatest benefits resulting from utilization of CFCs refrigerants. Now is analysed the compatibility
between the refrigerants and some of the most sensitive materials that constitute the refrigeration systems
(McQuay, 2002):

• Electric Motor: The hermetic compressors are the most used compressors. In these types of com-
pressors the electric motor finds it permanently exposed to refrigerant. This is important because
it allows the direct refrigeration of windings of the electric motor by the refrigerant in circulation.
From interaction between the refrigerant and insulation may result the absorption of refrigerant
by the isolation and extraction of polymer from insulation by the refrigerant. This may degrade
the insulation and promote the occurrence of faults in the electric motor. The polymer extracted
by the refrigerant may deposited on any part of the refrigeration system and obstruct the refriger-
ant flow. The resin used for the insulation of the electric motor windings may also be affected if
is exposed to refrigerant at high temperatures. Thus, during a process of refrigerant replacement,
the selection of the new refrigerant is very important, because is necessary protect the compressor
existing on the facility. The new refrigerant must be compatible with the origin materials of the
refrigeration system;
• Elastomers and Plastics: There are several materials that can be made from elastomers, namely,
gaskets and rubbers (i.e. sealing materials). Refrigerants and lubricants can cause the extraction of
the filler material and damage the properties of sealing materials, through their expansion or con-
traction. The properties of plastic materials may also be affected by the exposure to refrigerants.
Its degradation decreases with the increasing concentration of fluorine atoms in the composition
of the refrigerant;
• Metals: The reactivity between refrigerants and metals increases depending on the conditions of
the system. For example, the increase of refrigerant temperature or pressure.

226

Current and Future Trends of Refrigerants Development

Lubricants

All the mechanical components that produce movement suffer abrasion, thus requiring lubrication.
In the refrigeration systems case, some of the components that require lubrication are the compressor
and various valves scattered throughout the system. The utilization of a good lubricant should protect
these equipment against wear caused by the contact between the moving parts and increase the com-
pressor sealing. The lubricant should also be chemically compatible with the refrigerant and the other
materials present in the refrigeration system to not damage them. It is expected that the lubricant and
refrigerant are chemically compatible, allowing the respective homogeneous mixture. The lubricant
should be sufficient miscible in the refrigerant to minimize the negative effects on heat transfer and
to ensure the oil return to compressor. The proper operation and longevity of the refrigeration system
directly depend on the interaction between the lubricant and refrigerant. Lubricating oils are classified
according to their origin/composition. There are the following types of lubricants: mineral oils (MO);
alkylbenzenes (AB); polyol esters (POE); polyalkylene glycol (PAG); modified polyalkylene glycols
and polyvinyl ethers (EVP). Mineral oil and POE lubricants are the two types of lubricants more used
in air conditioning systems. Mineral oils are natural (petroleum products) whereas the POE lubricants
are synthetic (alcohols). Mineral lubricants are not compatible with POE oils, although it can happen
mixing these lubricants after replacing for a new refrigerant. Synthetic lubricants are more hygroscopic
(i.e. have a greater ability to absorb moisture) than natural lubricants. So the synthetic lubricants should
never be exposed to atmospheric moisture because its combination will cause a decrease in the refrig-
eration cycle performance. Refrigeration systems must be properly sealed from outside ambient to not
allow the penetration of external moisture. If exists the penetration of moisture into the refrigeration
system, this may chemically combine with the carbon and form carbonic acid, which will contribute to
the degradation of the refrigeration system materials. In table 2 are presented the lubricants compatible
with the several families of refrigerants.

Table 2. Summary containing some properties of refrigerants by family of refrigerants

Fluid Synthetic Natural Refrigerants


Properties
HFCs HFOs HFEs HCs CO2 Ammonia
GWP High Low Low Low Low Low
Flammability No Can be ignited No Yes No Can be
ignited
Toxicity No No No No Only at high Yes
concentrations
Price of Moderate High Low Low Low Low
refrigerant
Price of system Low Medium to high Low Low to medium Medium Medium to
high
Theoretical Good Medium Good Good Medium Good
system
efficiency
Traditional Alky benzenes (AB) Poly alkaline Mineral oils Mineral oils (MO) Mineral oils (MO)
lubricants Polyol ester (POE) glycol (PAG) (MO) Alky benzene (AB) Hydro treated mineral oil
Poly vinyl ether Polyol ester Polyol ester Poly alpha olefin
Poly alkaline glycol (POE) (POE) (PAO)
(PAG) Polyol ester (POE)

227

Current and Future Trends of Refrigerants Development

Critical Point

The critical point of a refrigerant is found using its (p, h) or (t, s) diagrams. This is the point where the
refrigerant properties in liquid and vapour state meet and become indistinguishable. At the critical point
the temperature, density and composition of refrigerant are identical in the liquid/gaseous phase. If the
refrigeration cycle operates above the critical point, it is difficult to separate the gas and liquid phases
of the refrigerant. Operating a refrigeration cycle below (but very close) to critical point conditions
brings some benefits, namely high volumetric capacity. On the other hand, there is a decrease in system
efficiency, caused by the reduction of cooling capacity.

Glide

The term Glide was introduced in the refrigerants market as response to the emerging use of blends of
refrigerants (zeotropic mixtures).
Blends of refrigerants consist on a mixture of several refrigerants in the same or different proportions.
Thus, the resulting refrigerant adopts the behaviour of all its components.
Glide is defined as the temperature difference between the beginning and end phase change of a
refrigerant in the evaporator and condenser (see Figure 12). This happens because the refrigerants that
constitute the mixture have different evaporation temperatures and the blend component that presents a
lower evaporation temperature is the first to evaporate. At the moment when the first component begins
to evaporate, there is a change in composition of the resulting mixture, and thus the establishment of a
different average evaporation point of the mixture. The change in composition of the mixture is known as
fractionation. The change on average evaporation temperature of the refrigerant mixture is called Glide.

Acoustic Velocity

The acoustic velocity is a parameter considered during the designing process of a refrigeration system.
This parameter defines the maximum speed at which the refrigerant may circulate in the system. If the
refrigerant circulates with a velocity higher or equal to supersonic velocity, shock waves will be produced
and damage the equipment of the refrigeration system. Normally the refrigerant (gas) at the entrance of
centrifugal compressors acquires a supersonic speed (Mach = 1). The acoustic velocity of a refrigerant
increases with the temperature increasing and decreases with the pressure increasing.

Physical Properties

Through the study of the physical properties of refrigerants is possible to know its potentialities.
The evaporation and freezing temperatures are two practical examples of the physical properties of
a refrigerant. The evaporation temperature of a refrigerant indicates the temperature levels at which the
refrigeration system can operate. The freezing temperature of the coolant must be lower than minimum
temperature attained in the refrigeration cycle in order to prevent the refrigerant solidification in the system.
Table 2 is a summary containing some properties of refrigerants previously studied by family of
refrigerants.

228

Current and Future Trends of Refrigerants Development

Figure 12. Simplified scheme of the glide concept

After the analysis of Table 2, can be concluded:

• Natural refrigerants are abundant in nature so have a low cost;


• CO2 presents the most theoretical low efficiency, because operates according to a transcritical
cycle. Its refrigeration systems present a medium cost due to need of use special materials capable
to containing the refrigerant or and compress it to a very high pressure;
• Ammonia refrigeration systems present a higher cost due to incompatibility with the materials
which constitute them, in particular metals (ammonia is corrosive when mixed with moisture);
• Due to its high flammability (HCs) and high toxicity (ammonia) the amount of refrigerant used on
refrigeration systems should be as smaller as possible to reduce the risk of explosion (HCs) and
health problems in humans (ammonia);
• HFOs can be flammable under certain environmental conditions, thus the refrigeration systems for
this family of refrigerants need some improvements in comparison to traditional systems, which
makes them more expensive. How they are synthetic substances, present a high cost. This family
of refrigerants is compatible with the commonly used lubricants;

229

Current and Future Trends of Refrigerants Development

• HFEs have a low applicability in refrigeration industry, being widely used as cleaning solvers for
electronic equipments. There are few studies with HFEs operating as refrigerants. This family of
refrigerants is compatible with the commonly used lubricants, cheaper than HFCs and present a
theoretical efficiency near to CFCs compounds more used.

REFRIGERANTS AND ITS HISTORICAL EVOLUTION

As already been mentioned, the first patented refrigeration machine with the capacity to operate continu-
ously was built by Jacob Perkins in 1834 and use ethyl ether (R610) as refrigerant. Ethyl ether (R610)
was the first fluid known with good thermodynamic properties (i.e. evaporative effect). These properties
were discovered by William Cullen in 1755, when obtained a little amount ice through the evaporation
of ether (Melo, 2011).
Ether was also been used in many refrigeration machines built in early nineteenth century. Thus
ether became the first refrigerant to be commercialized on a large scale. For being the first refrigerant
to be used, ether was not the most suitable for this purpose, being its utilization dangerous due to highly
flammability (Dinçer & Kanoglu, 2010).
In the following years, new refrigerants were studied, namely, ammonia (R717), carbon dioxide (R744),
the monocloroetano (R160), isobutane (R600a), dichloromethane (R30), sulfur dioxide (R764) and air
(R729). Between all these fluids, only three became commercial. The ammonia and sulfur dioxide have
been widely used in small refrigeration units and the carbon dioxide preferably used in cold rooms of
ships. All of these fluids have thermochemical interesting properties, except the components (SO2 CH3
Cl, NH3 and C2H5Cl) toxic to humans, (CH3Cl, C2H5Cl and NH3) highly flammable and (CO2) which
can only be used at very high pressures (Benhadid-Dib & Benzaoui, 2012). The intention was find any
fluid that worked.
In early twentieth century, domestic refrigerators consisted on ice boxes. The using of ice boxes had
the following disadvantages: need for a daily ice supply, water drainage from thawing and variable rate
of cooling. Until then, the ice used in ice boxes was extracted from cold places in nature, for example,
from Hudson River in the USA and from Lousã Mountain in Portugal.
In 1928 the Company Frigidaire asked a group of scientists led by Thomas Midgley to discover
a non-toxic and non-flammable refrigerant. As a result of this investigation were introduced the first
halogenated hydrocarbons compounds in refrigeration industry, the chlorofluorocarbons (CFCs), in the
beginning of 30’s decade. This event leads to abandonment of natural substances and beginning of the
security and durability generation. The production of CFCs compounds has revolutionized not only the
refrigeration industry, because they have been also used in many other applications, including cellulosic
foams, aerosols and cleaning solvents. Between all the CFCs compounds known, the most widely used
as refrigerants were the dichlorodifluoromethane (R12) and trichlorofluoromethane or Freon (R11).
These fluids were initially produced and marketed by DuPont and Nemours companies. Obtained from
petroleum distillation (i.e. are petroleum products), CFCs are composed by atoms of carbon, chlorine
and fluorine.
For about 70 years the CFCs compounds dominated the domestic and slight commercial refrigeration.
This was the time required to realize that CFCs compounds do not present the characteristics which were
initially determined, namely, the safety of their use and his non-aggressiveness to the ambient. Several

230

Current and Future Trends of Refrigerants Development

accidents, including the death of people through inhalation of these gases in confined spaces had been
registered. The high stability of these compounds allows their migration into the Earth stratosphere and
its combination with the ozone molecules, contributes substantially to the destruction of ozone layer.
This effect was discovered by Sherwood Rowland and Molina Mario, in 1974, when associated the ozone
layer depletion with the release of CFCs compounds to environment. This theory was confirmed in 1985
with the discovery of a hole in the ozone layer over Antarctica (Dinçer & Kanoglu, 2010).
Thus it has become necessary to ban the CFCs compounds and look for new refrigerant replacements.
In 1987, 25 countries together with the EC signed the Montreal Protocol, whose purpose would be to
protect the integrity of the ozone layer through the implementation of measures to control the worldwide
emissions, production and consumption of substances belonging to the CFCs family. The signing of this
document is still considered a major milestone in the history of refrigeration, marking the beginning
of environmental concern with the study of impact resulting from use of refrigerants (UNEP, 2000). It
begins the ozone layer protection generation.
The measures and deadlines set by the Montreal Protocol have been intensified in the following
years to its subscription, during the conferences of London (1990), Nairobi (1991), Copenhagen (1992),
Bangkok (1993), Vienna (1995), Montreal (1997) and Beijing (1999). One year after the signature of
the Montreal Protocol was created the Intergovernmental Panel on Climate Change (IPCC), a scientific
intergovernmental organ under the competence of the United Nations. Its goal is the production of reports
that support the United Nations Framework Convention on Climate Change (UNFCCC), which is the
key international treaty to reduce global warming and cope with the consequences of climate change.
The first IPPC Assessment Report was published in 1990 (IPCC, 2007a).
In Germany, was published the Regulation CFC/Halon Prohibition Ordinance (FCKW-Halon-Verbots-
Verordnung) in 1991. This document outlined a time scale for the CFCs compounds phase-out, according
the implications of Montreal Protocol. Later, in 2006, the Regulation CFC/Halon Prohibition Ordinance
was replaced by the Regulation Chemical Ozone Layer V (ChemOzonschichtV) (Refripro, 2014). The
main aims of the Regulation Chemical Ozone Layer V was the prevention of the emission of ozone
depleting substances (CFCs and HCFCs) into the atmosphere, through the regular inspection activities
to monitoring and maintain in good conditions the refrigeration and air conditioning plants with regard
to leakages and recovery of refrigerants.
ChemOzonschichtV serves as complement to CE Regulation nº 2037/2000. This last one emerged
from adaptation of the Montreal Protocol, created to regulate the emission of substances that lead to
destruction of ozone layer in the European space (EU, 2000). The signature of the Montreal Protocol
promoted the gradual replacement of CFCs compounds by hydrochlorofluorocarbons (HCFCs) and
hydrofluorocarbons (HFCs), less harmful compounds.
The utilization of CFCs compounds was finally abandoned in 1995, in the European Union with the
interruption of its production. HCFCs compounds have a lower stability on environment, which makes
them less harmful to the environment than CFCs compounds. These substances are considered refriger-
ants of transition, because it is predicted that its utilization be temporary until the development of new
substances more environmentally friendly.
The monoclorodifluormetano (R22), the dichlorotrifluoroethane (R123), the monoclorotetrafluoretano
(R124), the dicloroflouretano (R141b) and certain blends (R401A and R408A and R409A) were some
of the most commonly used refrigerants of HCFCs family. Several countries have published their own
laws to accelerate the phase-out of halogenated compounds (CFCs and HCFCs). For example, in 1994

231

Current and Future Trends of Refrigerants Development

was published the Regulation EC 3093/94, document that specifies the phase-out periods to HCFCs
compound in the European Community, see Table 3. In USA the phase-out period of R22 (HCFCs) ends
on 2020 (Copeland & Emerson, 2008).
Later in 2000 was published a rectification to EC 3093/94, the CE Regulation 2037/2000. The main
measures of Regulation 2037/2000 were: continuing the gradual recovery of the CFCs and HCFCs
compounds, performed during the maintenance process of refrigeration and air conditioning equipment
already installed, allowing the recycling, regeneration or destruction of these compounds; and the creation
of mechanisms that promote the regular inspection and maintenance of refrigeration and air conditioning
equipment in order to prevent refrigerant leakage.
The molecules of hydrofluorocarbons (HFCs) are known as the replacement refrigerants. These com-
pounds do not have chlorine atoms in its constitution, which prevents their participation in the process of
ozone layer depletion. However, while not participating in the process of ozone layer destruction, HFCs
compounds contribute significantly to the greenhouse effect increase (i.e. global warming).
The trifluoromethane (R23), difluoroethane (R32), pentafluoroethane (R125), tetrafluoroethane
(R134a), trifluoroethane (R143a), difluoroethane (R152a) and other blends (R404A, R407C R410A,
R422A, R422D, R427A and R507A), were some of the most commonly used refrigerants of HFCs family.
The signature of the Kyoto Protocol is also known as one of the most important factors in the history
refrigerants. This Protocol arises as a consequence of a series of conferences whose theme was the cli-
mate changes and its main objective was the reduction of GHG emissions by 5% on the period between
2008 and 2012, with reference to GHG emissions of 1990. To accomplish this objective each signa-
tory country was established its own goals. For example, the European Union has committed initially
to reduce the GHG emissions by 8%. The polluter’s agents are, in this case, carbon dioxide, methane,
nitrous oxide and fluorinated gases (UN, 1998). With the signature of the Kyoto Protocol arises the
global warming generation.
Rectified and signed by 175 countries, the Kyoto Protocol was published in 2005. All countries that
have signed this Protocol have committed to take strict measures to reduce the emissions of GHG, con-
sidered as anthropogenic cause of global warming and responsible for global climate changes (UN, 1998).
According to the Kyoto Protocol, the reduction of GHG can be achieved through the application of
three mechanisms: the Clean Development Mechanism (CDM), the emissions trading, and Activities
Implemented Jointly (AIJ) (UN, 1998).
To achieve the rectified targets imposed by the Kyoto Protocol, the European Union committed itself
to complete three objectives until 2020: reduce the GHG emissions by 20%, reduce the power consump-
tion by 20%; and increase the amount of energy produced through renewable sources for 20% of total

Table 3. Phase-out period of HCFCs in the European Union, according the Regulation EC 3093/94

Regulation EC 3039/94 Deadline


The level of HCFCs marketed or used by producers and importers does not exceed the 1989 level. 1/1/1995
35% reduction in the level of HCFCs marketed based on calculated in 1995. 1/1/2004
60% reduction in the level of HCFCs marketed based on calculated in 1995. 1/1/2007
95% reduction in the level of HCFCs marketed based on calculated in 1995. 1/1/2010
Prohibition of placing on the market of virgin HCFCs. 1/1/2015
(Benhadid-Dib & Benzaoui, 2012)

232

Current and Future Trends of Refrigerants Development

primary energy consumed. Posteriorly have been defined new goals by the European Union, long-term
goals, to be achieved until 2050: reduce the GHG emissions by 60%; reduce the power consumption
by 40%; and increase the amount of energy produced from renewable sources to 33% of total primary
energy consumed (Morais, 2011).
In January 2006 was published the CE Regulation 842/2006, concerning to the utilization of certain
fluorinated gases that contribute to greenhouse effect, the F-Gases. This Regulation was created with
the intent to reduce the GHG emissions in the European Union, through the abolition of some refriger-
ants of the HFCs family, partially halogenated compounds. In this context the most important measures
were: the contention (prevention, minimization and immediate repair of refrigerant leakage); recovery of
F-Gases for recycling, reclamation or destruction; reporting by manufacturers, importers and exporters
of refrigerant quantities greater than one tonne per year; refrigerant container and refrigeration system
properly labelled with the designation of the refrigerant used and the respective main classifications
(toxicity, flammability, GWP, ODP, among others); and limiting the use of certain refrigerants, under
certain operating conditions (EU, 2006a).
It was also published on May 17, 2006 the MAC Directive 2006/40/EC regarding to emissions from
air conditioning systems in motor vehicles. This directive restricts the use of refrigerants with a GWP100
higher than 150 in automotive air conditioning systems for all new automobiles and trucks, manufactured
from 1st January 2017 (EU, 2006b). It should be referred that the automotive industry is responsible for
50% of worldwide sales of refrigerant R134a (Melo, 2011).
On April 16, 2014 was published the Regulation EU 517/2014 on F-Gases and repealing the Regulation
EC 842/2006. This Regulation was created based on the 4th Assessment Report of the Intergovernmental
Panel on Climate Change of the UNFCCC, in which it was resolved that F-Gases emissions should be
reduced by 72% to 73% by 2030 and by 70% to 78% by 2050, values compared to 1990 levels (EU, 2014b).
The publication of Regulation EU 517/2014 shows the continuing efforts of the European Union
in promoting the use of refrigerants with a very low-GWP. This specifies the phase-out periods in the
European Union to HFCs compounds that present a high and medium GWP, see Table 4.
Also, according an appropriate transitional period (not defined yet), the use of refrigerants with a
very high-GWP of 2500 or more, subject to maintenance processes, with a charge size of 40 tonnes of
CO2-eq or more should be banned.

Table 4. Phase-out period of HFCs in the European Union, according the Regulation EU 517/2014

Regulation EU 517/2014 Deadline


Domestic refrigerators and freezers that contain HFCs with GWP of 150 or more. 1/1/2015
Refrigerators and freezers for commercial use that contain HFCs with GWP of 2500 or more. 1/1/2020
(hermetically sealed equipment)
that contain HFCs with GWP of 150 or more. 1/1/2022
Stationary refrigeration equipment that contains HFCs with GWP of 2500 or more. Except equipment intended for 1/1/2020
application designed to cool products to temperature below - 50 ºC.
Multipack centralised refrigeration systems for commercial use with a rated capacity of 40 kW or more, or F-Gases with 1/1/2022
GWP of 150 or more, except in the primary refrigerant circuit of cascade systems where are used F-Gases with a GWP
of less than 1500 may be used.
Hermetically sealed equipment that contains HFCs with GWP of 150 or more. 1/1/2020
Single split air conditioning systems containing less than 3 kg of F-Gas with GWP of 750 or more. 1/1/2025
(EU, 2014b)

233

Current and Future Trends of Refrigerants Development

On March 7, 2014 was published a Scientific review of the research regarding the safety aspects of
R1234yf use in MAC systems. This study not found evidence of serious risks related to the R1234yf
use in MAC systems under normal and foreseeable conditions. The publication of this study serves as
an argument to persuade the vehicle manufacturers of safety resulting from the HFOs refrigerants use
(EU, 2014c).
Currently, the demand for alternative refrigerants that are less harmful to the environment is one of
the research areas with most featured in the refrigeration industry. Thus, the aim is to find refrigerants
which do not contribute to the destruction of ozone layer or to increase greenhouse effect. In the absence
of an ideal alternative refrigerant, pure or blend, it is necessary to define priorities on the research for
a refrigerant that can be more environmentally friendly, has a good energy performance, be compatible
with the currently used refrigeration systems, safe and has a low cost. To try to overcome this problem
some future alternatives are available in the literature, namely:

The Study of Pure Natural Refrigerants (Ammonia,


Hydrocarbons, and Carbon Dioxide)

According to the theoretical analysis made with the refrigerants (R134a, R404A, R407C, R410A, R717,
R290, R600a and R1270) in chillers unites used to air conditioning, ammonia (R717) presents the highest
efficiency ratio (COP) for the different temperature ranges (-10/+35 ºC and +5/+45 ºC), preceded by
isobutene (R600a) with the second highest COP. Although having lower theoretical efficiency, R410A
has a higher volumetric efficiency, which allows the construction of more compact, cheap and competi-
tive systems (Pedersen, 2012).
The use of CO2 (R744) in MAC systems was also been studied: due to technical and safety issues
remaining unresolved by the automotive OEM industry, despite years of development and large invest-
ments, this technology cannot currently compete with systems using HFCs and HFOs compounds in
terms of cost/benefit (DuPont, 2008).
Cascade cycles are currently the commonly used technology to achieve safe and economical operating
costs for very low temperature conditions in the refrigerant industry. CO2/Ammonia or CO2/N2O systems
are considered as viable alternatives for the replacements of the typical fluids. In this study CO2, or R41
or dimethyl ether (DME) was used in the lower cycle and R1234yf, or R1234ze or ammonia was used
in upper cycle. A blend called XP10 (R1234yf with R134a) was used in the upper cycle. The results for
these fluids are compared against the traditional CO2/ammonia cycle. CO2 cycle has the lowest COP.
DME cycles in lower circuit shows the best performance (DME is highly flammable, thereby necessitating
the system of different designs in comparison to traditional systems which make the installations more
expansive). In the upper cycles the difference between R1234yf, R1234ze, ammonia or XP10 appears
to be small. Therefore the choice of the upper circuit refrigerant is most significant and ammonia does
not offer a significant advantage, due to its toxicity (Naicker, 2014).
The low critical point conditions (T = 31.1 ºC) of CO2 refrigerant when compared with more con-
ventional refrigerants can make CO2 refrigeration systems inefficient. Above this point, there is no phase
change of the refrigerant, meaning in a lower capacity of refrigerant to reject heat, being the transcritical
cycle less efficient. In climates where ambient temperatures surpass 20 – 23 ºC, transcritical operation
is often unavoidable and CO2 systems do not present the same success as in colder climes. It becomes
necessary another additional medium (i.e. no air) to maintain the subcritical cycle operating when

234

Current and Future Trends of Refrigerants Development

ambient temperatures are above the transcritical threshold of CO2. A cooling medium can be provided
by coupling the high pressure side of the CO2 refrigeration system to the ground with an aqueous loop.
Since ground temperatures are typically lower than ambient temperatures during summer, subcritical
cycle can be realised more often and vast improvements to coefficient of performance can be achieved.
According to data collected at two supermarkets, the first utilizing ground coupling and second using a
typical air cooled refrigeration cycle. The results show that geo-coupling lowers condensing temperatures
provides a reduction rounding 25% of energy consumption. This improves opens up opportunities for
CO2 using in warmer climes (Leiper et al., 2014).
Using ammonia as refrigerant has great benefits in large refrigeration units, but can be relatively
expansive for small to medium sized systems. However, a new concept using ammonia has now been
developed for cold store and freezer plant which draws together existing technologies and integrates these
within an efficient, compact, standalone packaged unit complete with air cooled condenser and housing.
The packaged ammonia plant is present as a viable alternative to an HFC/HCFC based system for
small to medium sized storage or freezer plant. Traditional site installed ammonia pumped circulation
system would be more costly and may introduce hazards to human being health and safety due to the
higher ammonia charge. With the package ammonia plants the disadvantages commonly associated to
ammonia plants for this size range can be overcome and an efficient, low refrigerant charge (i.e. more
safety), environmentally responsible solution can be offer with an attractive total cost of ownership
(Tomlinson, 2014).

The Study of the HFOs Refrigerants and HFOs/HFCs Mixtures

By presenting a comparable performance and lower GWP to R134a, the refrigerant R1234fy is currently
considered the most promising substitute to R134a, especially in automobile air conditioning systems.
Otherwise, the R1234fy presents a lower performance than R410A witch difficult its application in
residential air conditioners. Thus, is studied the use of mixtures between R32 and R1234yf at two mass
fractions (80/20 and 50/50 in mass %), in order to test their use in residential air conditioners. The re-
frigerant mixture R1234fy+R32 with 20% of R32 mass fraction present a less heat transfer coefficient
than pure R1234fy and the mixture with 50% of R32 mass fraction has a greater heat transfer coefficient
than pure R1234fy at large heat and mass fluxes. The heat transfer coefficient of mixture at two con-
centrations are 20 – 50% lower than that of pure R32. The effects of mass diffusion on the heat transfer
of a mixture are significant. Large mass and heat fluxes can improve the heat transfer coefficient of the
mixture (Li et al., 2012).
After several studies to various climate and use conditions, the Japanese Automotive Manufacturers
Association (JAMA) determined that R1234yf has the potential to reduce the total LCCP, or contribution
to climate change, by 20 – 30% compared to R134a or R744 systems. DuPont estimate that R1234yf
adoption in all new cars by the year 2017 has de potential to save more than 2,200 million litters (590
million gallons) of fuel annually due to emissions resulting from leakages – the equivalent of take ap-
proximately 1.5 million cars off road each year. In terms of safety of R1234yf use, this present a low
toxicity and despite being a bit flammable is safe for use in automotive air conditioning. R1234yf can
provide a low cost transition due its compatibility with proven current R134a based automotive air con-
ditioning technology used today (DuPont, 2008).

235

Current and Future Trends of Refrigerants Development

Was also tested the use of R1234ze as a blend component as an alternative to R134a and R410A. In
Table 5 are shown some properties of the refrigerants studied (Low & Cooper, 2014). Are also included
the properties of some of the most widely used HFCs refrigerants worldwide to enable the comparison
between these fluids and new blends in study.
The mixtures R444A and R445A have a similar performance to R134a. These two mixtures have
been investigated with the purpose of being applied in MAC systems. R444A presents a similar flam-
mability to R1234yf while R445A in non-flammable at ambient temperature, becoming flammable at
high temperatures. A theoretical cycle calculation comparison was made for conditions typical of those

Table 5. Some properties of the refrigerants studied

Alternative Fluids to R134a and R1234yf

R134a R444A R445A R1234yf

Molecular mass 102.03 96.7 103.1 114.04

GWP100 1430 92 130 4

Critical Temperature (ºC) 101.1 103.2 98 94.7

Critical pressure (MPa) 4.059 4.278 3.973 3.382

Components Single R32 R744 Single


R125a R134a
R1234ze R1234ze
Atmospheric boiling point/range (ºC) -26.4 -34.3 to -50.3 to -29.4
-24.3 -23.4
Lower Flammable limit at 60 ºC None 6.9% 8% 5.4%

Latent heat of vaporisation at atmospheric pressure (kJ/kg) 217 232 237 180

Fluids for Stationary Sector Applications


Fluid AC5X LTR4X HPR1D R407A R410A R32
Molecular mass 100.87 85.09 62.98 90.11 72.58 52.02
GWP100 620 1295 407 2107 2088 675
Critical Temperature (ºC) 102.2 85.8 80.9 82.3 71.4 78.1
Critical pressure (MPa) 4.24 4.74 4.515 5.657 4.902 5.782
Components R32 R32 R744 R32 R32 Single
R134a R125a R32 R125 R125
R123ze R134a R1234ze R134a
R1234ze
Atmospheric boiling point/range (ºC) -31.3 to -45.3 to -57.3 to -45 to -51.4 -51.7
-26.1 -36.8 -43.6 -38.6
Lower Flammable limit at 60 ºC None None 13% None None 14%
Latent heat of vaporisation at atmospheric pressure (kJ/kg) 222 250 332 236 273 382
(Low & Cooper, 2014)

236

Current and Future Trends of Refrigerants Development

in air conditioning application (evaporator and condenser refrigerant inlet temperature - 5/55 ºC). Us-
ing R134a as reference, the others refrigerants have presented a COP relative to reference of: 101.6% to
R444A; 98.5% to R445A; 100.3% to AC5X and 94.5% to R1234yf.
Applying the metric LCCP witch is evaluated using the car industry-developed GREEN-MAC soft-
ware is possible to predict the ambient impact of a car air conditioning system filled with R134a, R444A,
R445A or R1234yf, in a range of climates. The performance of R444A and R445A are comparable to
R-1234yf although their direct GWP values are somewhat higher, because their energy efficiency can
be better in some operation conditions. Acceptable performance was achieved through the use of R444A
and R445A in systems manufactured for R134a or R1234yf. Have also been evaluated other fluids based
on R1234ze for use in stationary sector applications, as can be seen in Table 5.
The AC5X is a non-flammable drop-in alternative to R134a, LTR4X is a non-flammable alternative
for R32, R407 or R404A DX applications, and HPR1D is an alternative to R410A for DX applications,
this blend presents a flammable range similar to R32. The AC5X and HPR1D have been evaluated for
commercial and scroll compressors used in supermarket refrigeration.
The HPR1D blend has teste as drop-in substitute in a light commercial air conditioner/heat pump unit
designed for R410A. The test results shown that HPR1D presents an improved performance comparted
to R410A, approaching performance of R32, but operating at lower compression discharges pressures
and temperature (Low & Cooper, 2014).

The Study of HFCs Refrigerants with Low-GWP

Several authors have been studying the characteristics of HFCs family of refrigerants with low-GWP to
be used as replacements to current used refrigerants.
In the next study were analysed the operation parameters of a refrigeration cycle at same evaporator
and condenser refrigerant inlet temperature (4/40 ºC) using different refrigerants (R22 - HCFC, R134a
- HFC, R410A - HFC and R407C - HFC), as can be seen in Table 6 (Jain et al., 2011).
For studied conditions, the R407C presents the highest refrigerant efficiency (%) with the highest
COP and lower compression work from the HFCs family in studied. R410A refrigerant has the lowest
refrigerant efficiency (%) and COP.

Table 6. Table with several parameters for a refrigeration cycle operating at same evaporator and con-
denser refrigerant inlet temperature with the R22, R134a, R410A and R407C refrigerants

Parameters HCFC HFCs


R22 R134a R410A R407C
COP 2.352 2.31 2.06 2.329
Compressor Work (kW) 28.352 38.921 32.337 28.628
Refrigerant efficiency % 47.39 46.64 41.85 64.93
Mass flow rate (kg/s) 0.475 0.535 0.499 0.511
Condenser pressure (MPa) 1.938 1.314 3.091 2.21
Evaporation pressure (MPa) 0.432 0.25 0.696 0.492
Heat transfer rate (kW) 94.36 98.12 94.36 94.36
(Jain et al., 2011)

237

Current and Future Trends of Refrigerants Development

In other work was study the performance of R152a and R32 to replace R134a in domestic refrigerators.
In this study was tested a refrigerator designed and developed to work with R134a, and its performance
using R152a and R32 was evaluated and compared with its performance when R134a was used. The
design temperature and pull-down time set by ISO for small refrigerator were achieved earlier using
refrigerant R152a and R134a than using R32. The highest COP was obtained using R152a. The average
COP obtained using R152a is 4.7% higher than that of R134a, while average COP of R32 is 8.5% lower
than of R134a. Discharge pressure of 152a is practically equal to R134a with average percentage reduc-
tion of 0.8%. The discharge pressure of R32 was highest with average value of 8.1 and 7.2% higher than
R152a and R134a, respectively. R152a offers lower energy consumption. The compressor consumes 4%
and 3.2% less energy when 152a was used than R134a and R32, respectively. Summarizing, the system
presents a better performance with R152a than with both R134a and R32. Thus, R152a can be used as
replacement refrigerant for R134a in domestic refrigerators (Bolaji, 2010).

The Study of HCs Refrigerants and HCs/HFCs Mixtures

The HCs, HCs mixtures and HFCs/HCs mixtures appear as alternatives to be used in domestic refrigera-
tion. A theoretical analysis with some HCs/HFC and HCs mixtures are made to the conditions obtained
using the R12 and R22 at same evaporator and condenser refrigerant inlet temperature (-10/50 ºC). In
Table 7 are presented the performance of different refrigerant in study (R600a/R290, R600/R290, R1270/
R290 and R152a/R290).
For the conditions studied, the pure refrigerants (R12, R22 and R134a) have a slightly increased COP
of all HCs and HCs/HFCs mixtures of refrigerants. Propane (R290) systems have a heating capacity,
when used as a heat pump, which is about 8% lower than R22, while the COP is 5 – 7% higher. It was
also shown that the refrigeration charge of a propane system is reduced by about 50%, when compared
with R22, for the same geometry. Other mixtures of refrigerants which are also identified as promising
are (R600/R290/R134a) and (R290/R600/R600a). Based on the results regarding the performance, HC
mixtures and R152a are found as the better substitutes for R12 and R152a in domestic refrigeration
sector. R29, R1270, R290/R152a, R744 and HCs/HFCs mixtures are considered as the best long-term
alternatives for R22 in air conditioning and heat pump applications (Paul et al., 2013).

The Study of HFEs Refrigerants

Currently HFE compounds are used as cleaning solvents, blowing agents, refrigerants and dry etching
agents in semiconductor manufacturing. There are few HFEs to still possess some environmental hazards

Table 7. Table with the performance of different HCs refrigerants and HCs/HFCs mixtures

Fluid Property Single Mixtures


R12 R22 R134a 40% 40% 50% 50% 80% 20%
R600a R600 R600a R600 R1270 R152a
60% 60% 50% 50% 20% 80%
R290 R290 R290 R290 R290 R290
COP 3.233 3.180 3.097 2.907 2.826 2.921 2.886 2.997 2.820
(Paul et al., 2013)

238

Current and Future Trends of Refrigerants Development

(GWP, flammability and toxicity). Some HFEs such as HFE-7500 is considered as replacer to HCFCs,
HFCs and chlorinated solvents in applications in the cleaning of electronic components, and heat transfer
systems due to their relatively low-GWP, very low toxicity and non-flammability (Tsai, 2005).
A theoretical comparison between the R11 and R12 (CFCs); R123 (HCFC); R134a (HFC) and
R347mcc, R245fa and R245mc (HFEs) was performed and the results are presented in Table 8. The
conditions were simulated for a heat pump system operating to a temperature of 30/80 ºC (evaporator/
condenser) (Sekiya & Misaki, 2000).
For the conditions studied, the R245fa (HFE with the best energy performance) present a COP: 7.4%
lower than R11; 13% higher than R12; 4.8% lower than R123; 22.3% higher than R134a and a heating
capacity: 26.2% higher than R11; 189% lower than R12; 46.3% lower than R123; 196% higher than R134a.
In the Figure 13 is shown a scheme with the historical evolution of used refrigerants, where is syn-
thesized the information contained in this section.

CLASSIFICATION OF REFRIGERANTS

In this section will be presented the classification of primary refrigerants, according to the five main
groups (families):

• Halocarbons;
• Hydrocarbons (HCs);
• Inorganic compounds;
• Azeotropic mixtures;
• Zeotropic mixtures or non-azeotropic.

Halocarbons

Halocarbon compounds contain in their chemical constitution atoms of chlorine, fluorine or bromine.
Typically designated as chlorofluorocarbons (CFCs), these compounds were widely used in the past as
refrigerants, solvents and blowing agents for foams. The most commonly used refrigerants of the CFCs
family were the R11 and R12 (Benhadid-Dib & Benzaoui, 2012).
After signing the Montreal Protocol in 1987, the utilization of these compounds rapidly decreased
during the respective phase-out period (between 1987 and 2000) in European Union, because of its nega-

Table 8. Table with the performance of some HFEs refrigerants

Fluid Property CFCs HCFC HFC HFEs


R11 R12 R123 R134a R347mcc R245fa R245mc
COP 6.08 5.01 5.93 4.63 5.59 5.66 5.27
Heating capacity (kJ/m3) 1145 4176 988 4277 873 1445 1723
(Sekiya & Misaki, 2000)

239

Current and Future Trends of Refrigerants Development

Figure 13. Historical evolution of used refrigerants and the legislation applied on this field
(Duarte, 2013)

tive impact on the ambient resulting from destruction of the ozone layer. The lubrication of refrigeration
systems that use refrigerants of CFCs family is performed using mineral oils. The increase of chlorine
concentration in the chemical composition of the refrigerants, allows them present better properties of
lubrication. Halogenated refrigerants are very stable when exposed to metals (McQuay, 2002). CFCs
compounds are odorless, non-toxic and heavier than atmospheric air, which makes them dangerous when
handled inside indoor ambient. The products resulting from combustion of CFCs compounds are quite
toxic if they are inhaled (Dinçer & Kanoglu, 2010). In Table 9 it is possible to see the characteristics pre-
sented by the CFCs compounds and others, allowing the comparison between several types of refrigerants.
The perfluorocarbon compounds (PFC’s) are compounds belonging to the CFCs family. These are
constituted by atoms of carbon and fluorine. Analyzing the characteristics of the PFC’s compounds
presented in Table 9, it can be concluded that they have a high residence time in the atmosphere and a
strong participation in the greenhouse effect increase, but not contributing to the ozone layer depletion.
Attending to its high-GWP value, PFC’s compounds will be banned in the European Union by the ac-
tion of Regulation EU 517/2014.

240

Current and Future Trends of Refrigerants Development

Table 9. Some of the features (ODP, GWP and life cycle) and phase-out period for some of the families
of refrigerants

Refrigerants Classification ODP GWP100 Life Cycle in the Phase-Out


Atmosphere
(Years)
R11 1 3 800 45 By the year 2000
R12 1 8 100 100 (ODP > 0)
Bromomethane 0.6 1 300 0.7 By the year 2015
HCFC 0.05 400 to 1800 1 to 20 (in EU)
HFC 0 400 to 11700 1 to 300 By the year 2017, depending on the
application (in EU)
PFC 0 6500 to 9200 10 000 to 50 000
HC 0 2 to 20 Days Not defined
(ODP = 0)
HFE 0 6 to 500 1 to 40
HFO 0 4 to 6 7
Ammonia 0 0 Days
Carbon dioxide 0 1 50-200
(Duarte, 2013)

Hydrocarbons

Hydrocarbons (HCs) are compounds mainly constituted by chemical bounds between atoms of carbon
and hydrogen. The methane (R50), ethane (R170), propane (R290) cyclopropane, n-butane (R600),
cyclopentane and isobutane (R600a) are some of the compounds that belong to the HCs family.
This type of compounds present the following characteristics: highly flammable and explosives; low
toxicity; and environmentally friendly because it does not are involved in the process of ozone layer de-
struction and have a very low participation in the greenhouse effect increase (Dinçer & Kanoglu, 2010).
Because they are high flammable and explosive, it becomes necessary to reduce the refrigerant charge
in systems to minimize the risks of its utilization (Pedersen, 2012). No problems of compatibility are
found between these compounds and all materials traditionally used in refrigeration systems (Bolaji &
Huan, 2013).
HCs emerged as the most advantageous solution to be applied in small packaged air-cooled equipment,
namely, propane (R290) and isobutene (R600a) in domestic refrigerators and appliances. The energy
efficiency of hydrocarbon chillers in medium - large range (50 – 400 kW) is about 10% better than in
HFCs systems and its price about 20% higher compared to HFCs systems. The typical payback time of
this equipment is 1 to 2 years (Pedersen, 2012). For example, the Aarhus University Hospital, Skejby in
Denmark is cooled and heated by HC chillers and heat pumps, and McDonald’s chain in Denmark has
installed cascade system using HCs and CO2 in their restaurants (Kohler, 2012).
There are several families of HCs compounds, namely: hidrobromofluorocarbonos (HBFC’s); hy-
drochlorofluorocarbons (HCFCs), hydrofluorocarbons (HFCs); bromomethane (CH3Br); and methane
chloroform (CH3CCl3). The compounds belonging to these families are synthetic derived from fossil
fuels. In this work will be discussed only the most important families, the hydrochlorofluorocarbons
(HCFCs), hydrofluorocarbons (HFCs), the hydrofluoroether (HFEs) and hidrofluoroleofinas (HFOs)
(Dinçer & Kanoglu, 2010):

241

Current and Future Trends of Refrigerants Development

1. HCFCs Compounds: Are constituted by hydrogen, chlorine, fluorine and carbon atoms. These
were created in order to replace the CFCs compounds. By having chlorine atoms in its chemical
composition these compounds participate in the ozone layer destruction process, but in much lower
proportions when compared with the CFCs compounds. The HCFCs compounds have a very low
ODP value, as can be seen in Table 9. This type of refrigerants can use mineral (MO) or synthetic
(POE) oils. In some cases, additives are used to improve the lubricant performance (McQuay,
2002);
2. HFCs Compounds: Are constituted by hydrogen, fluorine and carbon atoms. These are consid-
ered as the true class of substitutes for CFCs compounds, not having in its constitution chlorine
or bromine atoms, resulting in its non-participation in the process of the ozone layer destruction.
However, there are several HFCs compounds that have a high-GWP value, as can be seen in Table
9. The HFCs refrigerants generally require the use of a synthetic lubricant, such as a POE or PAG
lubricant (esters). Are not compatible with MO lubricants (Melo, 2011). Some metals, especially
metals in which chemical constitution exist atoms of zinc show increased corrosion with ester based
oils and refrigerants of this family (Bolaji & Huan, 2013). Because of their safety, efficiency and
cost, this type of compounds must remain an option in many regions and applications: HFCs and
blends of HFOs and HFCs can provide highly efficient non-flammable solutions for developing
regions; in large commercial air-conditioning with significant charge amounts and large refrigeration
applications in densely populated areas to prevent toxic emissions, or explosion or inflammation of
the systems (Kohler, 2012). HFCs refrigerants are widely used on ships today, but they are being
replaced by ammonia-based refrigeration systems and, in large fishing vessels, by CO2/ammonia
cascade systems. This substitution was necessary, because there is a large leakage of refrigerant
from the refrigeration systems existing on ships, due to rough physical actions of sea (Pedersen,
2012). The automobile sector is responsible for 50% of global sales of R134a (Melo, 2011);
3. HFEs Compounds: Are constituted by hydrogen, fluorine, carbon and oxygen atoms. As this
compounds does not contain chlorine, iodine or bromine, does not contribute to ozone depletion.
Due to chemical arrangement of their atoms, these compounds have a very low-GWP. As they
have fluorine atoms in its constitution, they are chemically stable, presenting a low reactivity (low
flammability and toxicity – being more safety). Otherwise, its OH radicals were designed to react
easily in the troposphere because of the inclusion of H atoms in its chemical structure. Widely
used in industrial applications as cleaning solvents and anesthetic agents, HFEs are considerate as a
promisor choice to replace CFCs, HCFCs, HFCs and PFCs compounds on refrigeration industry (as
refrigerants and blowing agents), once some of these compounds have physical properties similar
to refrigerants commonly used. These compounds are alcohol derivatives, synthetic, presenting the
next properties: colorless, odorless, tasteless and low viscosity. Depending on the molecular weight,
these compounds can be gaseous or liquid at ambient conditions (Sekiya & Misaki, 2000). They
dissolve slightly into the mineral oil and polyester typo oil was been tested successfully. From the
experience of its utilization as cleaning solvents, these compounds are perfectly compatible with
the commonly used metals, plastics and elastomers, according with 3M Novec Company (3M,
2003). The lack of studies using these fluids as refrigerants is its biggest weakness;
4. HFOs Compounds: Are constituted by hydrogen, fluorine and carbon atoms. These compounds
do not participate in the process of the ozone layer destruction and have a very low-GWP value,
as can be seen in Table 9. The most promising applications of these compounds are as refriger-
ants (MT and LT refrigeration systems) in AMC systems, aerosol propellant and insulation foams

242

Current and Future Trends of Refrigerants Development

(Dinçer & Kanoglu, 2010). As challenges to use of HFOs compounds, the production capacity of
R1234yf, the persistence of safety questions and new competitive technologies, may be presented
as barriers to the proliferation of these compounds (Sciance, 2013). For refrigerants of this fam-
ily (and blends of these with other refrigerants) is acceptable to use PAG or POE lubricants. The
reactivity between HFCs compounds and PAG lubricants varies depending on oil formulation and
moisture content (Low & Cooper, 2014).

Inorganic Compounds

Currently, inorganic compounds are used as refrigerants in several types of refrigeration systems, such
as, commercial, industrial and mobile refrigeration systems with big cooling powers to cooling or freez-
ing. Examples of inorganic compounds used as refrigerants are the ammonia (R717), water (R718), air
(R729), carbon dioxide (R744) and sulfur dioxide (SO2). Among all these compounds, ammonia stands
out as the more utilized in industrial applications, because it presents interesting thermophysical and
heat transfer properties.
Air and CO2 are the most widely used refrigerants to operate a supercritical refrigeration cycle. This
happens because these two refrigerants can reject thermal energy at supercritical temperatures (McQuay,
2002).
Within the whole range of inorganic refrigerants the ammonia, carbon dioxide and air are those that
present the most interesting characteristics for being used in refrigeration systems (Dinçer & Kanoglu,
2010):

1. Ammonia: Has excellent thermodynamic characteristics when operates in refrigeration systems


whose the evaporating temperatures are between -35 and +2 ° C. As disadvantages, ammonia
is highly toxic and flammable. Despite the security problems that ammonia presents, their ther-
mophysical and heat transfer properties are so good that it became, in some countries, the most
requested refrigerant to operate in industrial applications. When combined with water, ammonia
can be corrosive to cooper, zinc and copper and zinc alloys. Ammonia presents a little miscibility
with lubricant used in refrigeration systems. As an advantage, it becomes easier separate the two
components and reduce the risk of fire. On the other hand, as a disadvantage it is needed to introduce
a separate mechanism to promote the return of oil to the compressor. Due to its properties (toxicity
and flammability) several countries has restricted the use of ammonia as refrigerant, especially in
commercial and residential applications (Paul et al., 2013). For example, ammonia heat pumps are
used to heat a district in Norway, and the Hotel Scandinavia and the connected Aarhus Congress
Centre in Aarhus was installed ammonia chillers in 1996 (Kohler, 2012);
2. Carbon Dioxide: Is not toxic (at concentrations below 5% by volume in mixture with air), not
flammable, but thermodynamically inefficient. Has a good chemical compatibility with the mate-
rial constituents of a refrigeration system and with a large number of lubricant oils. Its use involves
high pressures (a large energy consumption for its mechanical compression), normally operating
in transcritical cycles, necessitating of the utilization of specific compressors (to achieve the high
pressures of the transcritical refrigeration cycle). An advantage of its utilization is that it has an
ODP = 0 and GWP = 1 (i.e. non harmful to the ambient, compared with all other used refriger-

243

Current and Future Trends of Refrigerants Development

ants). The use of CO2 as refrigerant was proposed in 1850 by Alexander Twining. The peak of its
use took place between 20s and 30s and was completely abandoned in 1960. In 1993, G. Lorentzen
reintroduced the CO2 on refrigeration by improving the transcritical thermodynamic cycle (Melo,
2011). Currently, CO2 used in cascade refrigeration systems (R717/R744), dry-ice production and
food freezing applications. For example, in Denmark several centralized refrigeration systems using
CO2 as refrigerant were built and tested. Today this technology is standard and hundreds of systems
are installed presenting good performance, energy efficiency and economy (Pedersen, 2012);
3. Air: Was used for many years in cold-air refrigeration systems to cool passenger cabins in airplanes.
When air is compressed a simple joule process is used to remove heat from air to the surrounding
ambient. Afterwards, the air is expanded in a turbine, where it becomes cold. In Germany, a cold-air
refrigeration system has been developed for trains (Pedersen, 2012). The coefficient of performance
(COP) associated to the utilization of air as refrigerant is low due to its low specific weight, thus
resulting in the transfer of small thermal energy quantities. Thus, air used as refrigerant is not
energy efficient, but the lightness of the system components is the main advantage of its use. This
makes the cold-air refrigeration systems interesting to be used in mobile transport applications. In
some factories air is used as refrigerant to promote a quick freezing of the food.

Mixtures or Blends

Mixtures or blends are constituted by multiple substances. Currently, the future options of refrigerants
are slim, if researchers only focus in pure natural refrigerants. Thus, the use of mixtures of refrigerants
increases the possibilities of choice as well, the chances of replacement in older installations by “new”
refrigerants. As advantages, this type of refrigerants takes advantage of existing technology, allowing the
creation and optimization of “new” refrigerants (McQuay, 2002). There are two types of mixed refriger-
ants, the azeotropic mixtures and zeotropic mixtures (Dinçer & Kanoglu, 2010):

1. Azeotropic Mixtures: Consist of two or more substances with different chemical and physical
properties, but behaving as a single substance. In azeotropic mixture is not possible to separate the
two components using the distillation process, because these substances present the same evapora-
tion temperature to the same ambient conditions.
2. Zeotropic Mixtures or Non-Azeotropic: Are constituted by multiple substances with different
volatilities (i.e. different evaporation temperatures). When they are used as refrigerants, change
their composition during the processes of evaporation and condensation. The application of these
compounds as refrigerants was proposed in the early twentieth century. Recently, a great interest
has been shown by the zeotropic mixtures, especially for operation as the refrigerant in heat pumps
due its adaptive chemical composition during heating and cooling process, providing conditions
that allow the development of new heat exchangers designs.

In Table 10 is presented a synthesis of the information relative to several refrigerants, namely the main
applications and properties collected in date cheats of companies that commercialize them, including the
DuPont, National Refrigerants, Linde, Forane, Honeywell, and the Report of UNEP (2010).

244

Current and Future Trends of Refrigerants Development

Table 10. Physical, safety and environmental features of some refrigerants. The features were collected
from data cheats of refrigerants commercialized by the companies

Family Refrigerant Mixture ODP GWP100 Life TBOILING TCRÍTICAL PCRÍTICAL Security Applications
[% Cycle [°C] [°C] [MPa] [ASHRAE]
Constituents] [Years]
CFC R11 Single 1.00 4000 45 23.7 197.78 4.41 A1 Industrial refrigeration
units that use centrifugal
compressors, operating at
LP, LT and VLT.
R12 Single 1.00 2400 100 -29.8 112.22 4.12 A1 Small refrigeration
systems that utilizes
reciprocating compressors
at MT.
R13 Single 1.00 14400 640 -81.5 28.89 3.87 A1 Used in VLT applications
in medicine, laboratory
analysis, electronics and
equipment testing.
R502 R22 (48.8) 0.311 4600 - -45.3 81.5 4.07 A1 Used in LT applications in
R115 (51.2) commercial refrigeration
systems.
HCFC R22 Single 0.055 1700 11.9 -40.8 96.11 4.98 A1 Air conditioning systems
and heat pumps used in
homes. Refrigeration
systems that operate at
MT.
R123 Single 0.02 77 1.3 27.8 183.89 3.67 B1 Systems operating LP
compressors (substitute
for R11).
R124 Single 0.07 620 5.9 -12 122.2 3.614 A1 Cooling systems that use
centrifugal compressors.
R401A R22 (53) 0.037 1200 - -32.9 105 4.61 A1 Refrigeration systems that
R152a (13) operate at low and MT.
R124a (28) Applied in supermarket
freezers, equipped with
reciprocating compressors
(replacement for R12).
R408A R125 (7) 0.026 3152 - -44.6 83.17 4.283 A1 Refrigeration systems that
R134a (46) operate at MT and LT.
R22 (47) Applied in supermarket
freezers (substitute for
R502).
R409A R22 (53) 0.005 1585 - -34.4 109.28 4.605 A1 Systems operating at
R124 (13) MT and LT. Applied
R142b (28) in supermarket
freezers, equipped with
reciprocating compressors
(replacement for R12).

continued on following page

REPLACEMENT OF REFRIGERANTS

In this section are presented the existing approaches to the process of refrigerants replacement. It follows
the exposure of the alternative refrigerants that present better characteristics (to short and long terms)
to replace the CFCs and HCFCs compounds, according the characteristics of refrigeration system and
the respective practical application.

245

Current and Future Trends of Refrigerants Development

Table 10. Continued

Family Refrigerant Mixture ODP GWP100 Life TBOILING TCRÍTICAL PCRÍTICAL Security Applications
[% Cycle [°C] [°C] [MPa] [ASHRAE]
Constituents] [Years]
HFC R23 Single 0 5700 222 -82 25.8 4.82 A1 Systems operating at VLT
(substitute for R13 and
R503).
R32 Single 0 650 5.2 -51.7 78.10 5.782 A2L Refrigeration systems
that operate at VLT.
Air conditioning and
industrial applications.
R125 Single 0 3400 28.2 48.1 66.015 3.629 A1 Air conditioning,
industrial refrigeration
and cooling units of water.
R134a Single 0 1300 13.4 -26.1 101.1 4.06 A1 Automotive and
household air
conditioning and heat
pumps (replacement for
R12). Also used as a
blowing agent of foams.
R143a Single 0 3800 8.2 -47.2 72.707 3.761 A2 Air conditioning,
industrial refrigeration
and cooling units of water.
Refrigeration systems that
operate at MT.
R152a Single 0 120 1.5 -24 113.26 4.52 A2 Automotive and housing
air conditioning systems
(replacement for R134a).
Domestic and commercial
freezers.
R404A R125 (44) 0,04 3300 - -46.2 72.1 3.74 A1 Refrigeration systems that
R134a (4) operate at MT and LT.
R143a (52)
R407C R32 (23) 0 1610 - -43.6 86.2 4.62 A1 Systems with reciprocal
R125 (25) compressors in air
R134a (52) conditioners and heat
pumps to MT (substitute
R22). Freezes systems
that operate at MT and
LT.
R410A R32 (50) 0 1890 - -51.4 72.22 4.76 A1 Air conditioning and
R125 (50) heat pump systems
(replacement for R22).
Refrigeration systems that
operate at MT.
R422D R125 (65.1) 0 2700 - -43.2 79.6 3.9 A1 Used in LT and MT
R134a (15) applications in chillers,
R600a domestic, commercial and
industrial refrigeration
systems.
R437A R125 (19.5) 0 1684 - -32.9 96 4.096 A1 Air conditioning systems
R134a (78.5) (including automotive).
R600 (1.4) MT freezers, processing
R601 (0.6) and storage of foods
(replacement for R12).
R507A R125 (50) 0 3800 - -46.7 70.9 3.79 A1 Refrigeration systems that
R134a (50) operate at MT and LT.
PFC R508B R23 (46) 0 10350 - -87.6 13.7 3.935 A1 Refrigeration systems
R116 (54) that operate at VLT (ex:
refrigeration systems used
in medical applications).

continued on following page

246

Current and Future Trends of Refrigerants Development

Table 10. Continued

Family Refrigerant Mixture ODP GWP100 Life TBOILING TCRÍTICAL PCRÍTICAL Security Applications
[% Cycle [°C] [°C] [MPa] [ASHRAE]
Constituents] [Years]
HFE R143a Single 0 656 5.7 -23.75 104.7 3.63 A1 Refrigerant.
R245mc Single 0 697 4.7 6 133.7 2.89 - Refrigerant and blowing
agent.
R347mcc Single 0 575 4.8 4.8 164.5 2.48 - Refrigerant.
R356mff Single 0 39 0.3 63.75 203.15 2.78 - Refrigerant.
R7000 Single 0 450 4.7 34.05 164.6 2.48 - Refrigerant and cleaning
solvent.
HFO R1234yf Single 0 4 0.029 -29.5 94.7 3.382 A2L Automotive and housing
air conditioning systems
to (replacement for
R134a). Domestic and
commercial freezers.
R1234ze Single 0 6 0.045 -19 109.36 3.635 A2L Blowing agent of foams.
HC R170 Single 0 20 0.041 -88.6 32.22 4.9 A3 Used in VLT refrigeration
(ethane) systems.
R290 Single 0 20 0.21 -42.1 98.8 4.25 A3 Air conditioning systems
(propane) and heat pumps for
domestic utilization
(replacement for R22).
Used to MT.
R600a Single 0 4 0.016 -11.7 152.22 3.80 A3 Air conditioning,
(isobutane) commercial refrigeration,
process chilling or heat
extraction.
R1150 Single 0 4 0.004 -103.8 9.44 5.12 A3 Used in VLT refrigeration
(ethylene) applications, such as those
found in LNG liquefaction
process.
R1270 Single 0 2 0.001 -47.6 91.67 4.62 A3 Commercial refrigeration,
(propylene) chill cabinets, cold storage
and food processing,
industrial refrigeration,
small air conditioning
systems and large air
conditioning and chiller
systems. Refrigeration
systems that operate at
VLT.
Inorganic R717 Single 0 0 < 0.02 -33.3 132.78 11.42 B2L Industrial units, freezing,
(ammonia) meat conservation …
Widely used as first
refrigerant.
R718 Single 0 0 - 100 373.94 22.06 A1 Absorption systems
(water) (water - lithium
bromide). Gas jet ejector
refrigeration systems
and air conditioning.
Widely used as secondary
refrigerant.
R744 Single 0 1 > 50 - 31.11 7.38 A1 Industrial refrigeration
(CO2) units. Widely used in
cascade systems with
direct expansion.
Cryogenic R729 Single 0 0 - -194.2 - - A1 Absorption and gas jet
(Air) ejector refrigeration
systems.

(Duarte, 2013; UNEP, 2010; IPCC, 2007b)

247

Current and Future Trends of Refrigerants Development

Different Approaches to the Replacement Process

The refrigerants replacement process can be performed through two distinct approaches: retrofit type and
drop-in type replacement’s (Melo, 2011). A replacement process of retrofit type requires a modification
(i.e. upgrade) of some components of the refrigeration system. The retrofit term is used in engineering
to describe the process of an equipment modernization. Normally the process of substitution by a blend
happens according to retrofit process type, because is usual the lubricant substitution, thus as some equip-
ment of the refrigeration system. This is due to incompatibility issues between the new refrigerant and
the refrigeration systems equipment. On the other hand, a replacement process of drop-in type allows
the direct substitution of refrigerants, without exists any incompatibility issues between the lubricant
and each refrigerant (substitute and substituted) or any other components of the refrigeration system
due to incompatibility of materials.

Most Common Refrigerant Replacements

The refrigerant replacement should only happen in two circumstances. The first, to comply the tight
legislation applied to use of refrigerants, if the refrigerant present in the refrigeration system is dis-
missed or obsolete. The second, to increase the efficiency of the refrigeration cycle replacement with
a new refrigerant that presents better thermophysical and thermal energy transfer properties, or also,
if the new refrigerant fits best to the refrigeration system and its operation conditions. In this case, the
replacement reflects the necessity for companies in reduce the operation costs related, in this case, with
the cold production. Moreover, the acquisition of refrigerants more “recent” or less commercial, that
exhibit improved properties, has the disadvantage of the high initial investment.
According to DuPont (American Company), the acquisition of refrigerants of low quality causes
damage on compressors. The deposition of salts in the pipes and other components will decrease the
system efficiency and increase the periods of maintenance and refrigerant leakages (i.e. increase costs).
This happens, usually due to excess impurities and existence of moisture in refrigerants. In Table 11 is
possible to see the alternatives that present better characteristics (to short and long terms) for replace-
ment the more used CFCs, HCFCs and HFCs with high-GWP compounds. To facilitate the analysis,
the refrigerants are grouped according to the type of refrigeration system (refrigeration, air conditioning
and freezing) and applications (residential, commercial, industrial and automotive).
The CFCs and HCFCs refrigerants most used and whose replacement is required due to non-com-
pliance with the current legislation are the R12 (CFC), R409A (HCFC), R502 (CFC), R11 (CFC) and
R22 (HCFC). Currently, the CFCs refrigerants are discontinued, having its phase-out period occurred
to 2000. Besides, the HCFCs refrigerants are currently in the phase-out period, to 2015 (target imposed
by the Regulation EC 3039/94). According to the Regulation EU 517/2014 the HFCs with a GWP value
higher than 150, must be discontinued until 2017 (depending on the application). This is the reason
because the replacement options to long-term are present.
As options of replacement to short-term, the HFCs with high-GWP refrigerants were chosen. This
family of refrigerants presents characteristics of compatibility of materials identical to the refrigerants
replaced, allowing an approach to substitution of drop-in type, do not being necessary the changing
of the system components or lubricating (with the exception of R134a refrigerant). The refrigerants
chosen to the short-term replacement, result of the orientation obtained through the guidelines to the

248

Current and Future Trends of Refrigerants Development

Table 11. Table presenting the alternative refrigerants (to short and long term) that presents the best
features to replace the CFCs, HCFCs and HFCs with high-GWP compounds, taking into account the
characteristics of the refrigerating systems and its respective practical application

System Applications Replaced Replacement Replacement Options to Long-Term


Refrigerants Options to Short- (HFCs, HFOs e Naturals)
(CFCs e HCFCs) Term Retrofit
(HFCs)
Drop-In
Refrigeration Housing • R12 • R134a* • R152a
(domestic) • R409A • R437A • R600a
• R290
• R1234yf
• Blends of HCs/low-GWP HFCs
• Blend of HFOs/low-GWP HFCs
Commercial • R502 • R407C • R32
• R22 • R422D • R152a
     R410A • R600a
• R290
• Cascade systems (R717/R744)
• Blend of HFOs/low-GWP HFCs
• Blends of HCs/low-GWP HFCs
Industrial • R11 • R123 • R123
• R717
• R744
• Cascade systems (R717/R744)
Air conditioning Residential and • R22 • R407C • R32
commercial • R422D • R152a
• R410A • R290
• R600a
• R1234yf
• Blend of HFOs/low-GWP HFCs
• Blends of HCs/low-GWP HFCs
Automotive • R12 • R134a* • R152a
• R437A • R1234yf
• R744
Freezing Industrial • R11 • R123 • R123
• R717
• R744
• Cascade systems (R717/R744)
(Duarte, 2013)
* In the case of R134a, the replacement process needs the change of lubricant and expansion device, being this substitution a Retrofit,
according with the data provided by the DuPont company.

replacement of refrigerants proposed by the DuPont Company. It is intended with the analysis of the
short-term replacement, a quick response to the replacement of refrigerants in the refrigeration systems
(air conditioning, refrigeration and freezing), present in companies, public and private buildings, shop-
ping centers and automobiles, so that the current legislation that regulates its utilization can be followed.
It can be seen in Table 11, that in the group of options to short and long-term replacement, the R123
refrigerant is underlined. This happens because it belongs to the HCFCs family and not the HFCs family.
This happens because the R123 refrigerant as a low value of GWP and ODP, see Table 10, and excellent
thermodynamic properties when operate at very low temperatures. Thus, this refrigerant remains pointed

249

Current and Future Trends of Refrigerants Development

by investigators as a refrigerant with high potential to long-term when applied in industrial refrigera-
tion systems. In cases where the replacement is performed by the R134a refrigerant, the replacement
process requires the exchange of lubricant and an expansion device (i.e. a replacement of the retrofit
type), as related in the orientation documents provided by the company DuPont (2011). In systems that
operate with the R12 refrigerant is recommended to use lubricants of types AB (alkylbenzenes) and OM
(mineral oils) and in systems that operate with R143a refrigerant is recommended to use the lubricants
of type POE (polyol esters).
As long-term replacement options, were chosen the HFCs with low-GWP, HFOs, natural refriger-
ants and mixtures (HCs/HFCs and HFOs/HFCs) refrigerants. These refrigerants have a lower level of
pollution (ODP and GWP), and for this reason are considered the future refrigerants. The choice of
these refrigerants, such as long-term replacement options is based on experimental studies performed
by several authors referred in this chapter. The refrigerants to be used as long-term replacement options
are only possibilities of use.
The necessity to perform profound modifications on the installation/components of the refrigeration
system is the disadvantage to perform a replacement by a refrigerant belonging to the long-term group,
on an existing installation (i.e. a replacement of retrofit type). In this type of replacement, the charac-
teristics of the refrigerant substitutes are not compatible with the installation features, may requiring the
implementation of a new system, capable to operate the required conditions set by the new refrigerant.
For example, it is not possible to use the CO2 as a substitute refrigerant on systems which operates the
refrigerant R11. Buying a new system or installation is too expensive.
The use of a refrigerant substitute may still result in a loss of performance of the refrigeration system,
caused by improper adjustment of refrigeration systems that are not properly qualified to receive these
new refrigerants. The increased reactivity of the used refrigerants as replacement choice to a long-term
with the materials that compose the refrigeration system may result in a faster degradation of these
components and increase of refrigerant leakage.
From observation of Table 11, it can be concluded that:

• In domestic refrigeration applications the refrigerants R12 and R409A are replaced. As replace-
ment options to short-term are proposed the refrigerants R134a and R437A. As replacement op-
tions to long-term are proposed the refrigerants R152a, R600a, R290, R1234yf and blends of
(HCs/low-GWP HFCs and HFOs/low-GWP HFCs);
• In commercial refrigeration applications the refrigerants R502 and R22 are replaced. As replace-
ment options to short-term are proposed the refrigerants R422D, R407C and R410S. As replace-
ment options to long-term are proposed the refrigerants R32, R152a, R600a, R290, cascade sys-
tems (R717/R744) and blends of (HCs/low-GWP HFCs and HFOs/low-GWP HFCs);
• In industrial refrigeration applications the refrigerant R11 is replaced. As replacement options to
short-term are proposed the refrigerants R123. As replacement options to long-term are proposed
the refrigerants R123, R717, cascade systems (R717/R744) and R744;
• In residential and commercial air conditioning applications the refrigerant R22 is replaced. As
replacement options to short-term are proposed the refrigerants R407C, R422D and R410A. As
replacement options to long-term are proposed the refrigerants R32, R600a, R290, R1234yf and
blends of (HCs/low-GWP HFCs and HFOs/low-GWP HFCs);

250

Current and Future Trends of Refrigerants Development

• In automotive air conditioning applications the refrigerant R12 is replaced. As replacement op-
tions to short-term are proposed the refrigerants R134a and R437A. As replacement options to
long-term are proposed the refrigerants R152a, R744, R1234yf and blends of HFOs/low-GWP
HFCs;
• In industrial freezing applications the refrigerant R11 is replaced. As replacement options to
short-term are proposed the refrigerants R123. As replacement options to long-term are proposed
the refrigerants R123, R744, R717 and cascade systems (R717/R744).

REFRIGERANT RECOVERY, RECYCLING, RECLAMATION, AND DISPOSAL

The creation of the concepts of refrigerant recovery, recycling, reclamation and disposal, was one of the
Montreal Protocol guidelines. This was the first document referencing the introduction of these concepts
on the politics of environment preservation. These concepts were defined with the objective to promote
the creation of incentives that allows the reduction of halogenated compounds (F-Gases) into the atmo-
sphere, and thus, mitigate the resulting consequences by the emission of these compounds. However,
only after a few years of the signature of the Montreal Protocol, the concern of world governments to
adopt legislation to regulate this issue arises.
The Heat Pump Centre has a project which the objective focuses on data collection and analysis related
to the recovery, recycling and reclamation R/R/R of refrigerants, to facilitate the planning of strategies
that encourage the practice these practices (Bouma, 2003). The overall objective of this program is the
documentation of practices and policies implemented for refrigerant recovery and recycling in major world
markets of refrigerants use, namely, Asia, Europe and North America. This would allow the quantification
of the impact of reducing emissions of refrigerants and comparing the effectiveness of different policy
approaches. Some of the planned actions in course of this program are the development of standards and
regulations with the competent authorities (i.e. in this case the U.S. Environment Protection Agency),
promoting the creation of mandatory procedures for refrigerant R/R/R and the assessment of economic
costs/benefits of leakage reduction, recovery, recycling and reutilization of refrigerants.
Currently, already exists legislation published in different countries that encourages the practice of
refrigerants R/R/R and application of good practice rules, starting by the companies that selling refrig-
erants. For example, in Brazil, the National Plan for Elimination of CFCs refrigerants - PNC plans to
install 114 decentralized units for the recovery and recycling of CFCs, HCFCs and HFCs refrigerants.
All centrals will be fitted with recycling equipment and will operate on all the Brazilian nation territory,
as stated in the Decree Nº 462 of December 22 of 2009 (LB, 2009).
In Portugal, Regulation EC 842/2006 presents the guide lines to R/R/R of certain fluorinated green-
house gases (F-Gases). This Regulation obliges operators of a number of installations to follow a restricted
plane for the proper recovery of F-Gases by certified personnel, to ensure the adequate R/R/R, disposal
or destruction of these gases. According the Regulation EC 842/2006 (EPEE, 2014):

• Recovery means removal, collection and storage of gas or fluids from equipment into a closed
vessel, appropriately identified with the designation of the collected fluid;
• Recycling means reuse of a recovered gas or fluid in the same system after having performed a
basic cleaning process;

251

Current and Future Trends of Refrigerants Development

• Reclamation means cleaning or reprocessing of gas or fluid to meet official specifications and to
be used like virgin material;
• Disposal or destruction means transformation or decomposition of gas or fluid in one or more
substances which are not fluorinated greenhouse gases anymore.

The minimum requirements and conditions for mutual recognition for training and certification of
personal and companies involved in installation, maintenance or servicing of refrigeration, air condition-
ing and heat pump equipment are also defined in the Regulation EC 842/2006 (EPEE, 2014).
To be an authorized installer (i.e. certified) is necessary to complete successfully a theoretical and
practical examination by the competent authorities. The installer must also be responsible to recover
from systems the F-Gases in process of decommissioning in cases of maintenance work require the
removal of the refrigerant charge. The companies involved in installation, maintenance, servicing, leak-
age checking and recovery should be required to be certified to perform these activities and to obtain a
certificate, a company should present to competent authorities the number of certified staff members
per category, and an indication of the expected volume of activity and a document with the necessary
tools and equipment available to the personnel engaged in activities subject to certification.
Also according the Regulation EC 842/2006 in installations containing 3 kg of F-Gases or more is
required the presence of an operator responsible for the installation to keep accurate the records of all
activities performed on the system. The producers, importers and exporters of F-Gases must report their
transactions annually to the European Commission and the national authorities if the transitioned amount
of refrigerant is greater than one tonne per year. In this context, the company that performs the operation
of substitution is responsible for collection and storage of the refrigerant. As there are no facilities in
Portugal to promote the regeneration of refrigerants, these are sent by the companies who execute the
replacement to other countries were exist plants in which the refrigerants can be recycled or destroyed.

CONCLUSION

In this chapter of the general characteristics and properties of refrigerants were presented. As charac-
teristics, the ideal refrigerant should be harmless to the environment, presents good heat transfer and
thermophysical properties, be safe (low toxicity, low flammability and non-explosive), highly applicable,
available, with low cost and recyclable.
Historical evolution of refrigerants was boosted by environmental concerns and constructive features
of the refrigeration systems. Thus, the historical evolution of refrigerants can be divided into four genera-
tions: 1ª any refrigerant that work; 2ª security and durability; 3ª ozone layer protection; 4ª global warming.
The signature of the Montreal Protocol in 1987 and Kyoto Protocol in 2005 are known as the main
events that marked the historic evolution of refrigerants, actions that lead to global rise of environmental
concerns.
Refrigerants composed by chemical bounds of (C-Cl) have the ability to absorb light radiation in
the ultraviolet range and have a long residence time in the atmosphere, thus intervening in the process
of destruction of the ozone layer.
Refrigerants composed by chemical bounds of (C-F) have the ability to absorb light radiation in
the infrared range and have a long residence time in the atmosphere, thus intervening in the process of
destruction of the ozone layer. Because they block the infrared radiation emitted by the Earth to space
and converting it into heat.

252

Current and Future Trends of Refrigerants Development

These two effects are responsible for climate change and consequent destruction of ecosystems. Thus
it became necessary to take measures to reverse this trend, so emerged the Protocols above mentioned.
There is currently a wide variety of data compiled from studies in which low-GWP refrigerants
features are compared with the traditional refrigerants (i.e. the refrigerants commonly used currently).
The use of new refrigerants may require modified refrigeration cycles and expensive system components.
The determination of indexes as the TEWI, LCCP and LCA to refrigerants before promote a replace-
ment process is presented as an asset.
In refrigeration systems of the 2nd generation the indirect effect of GHGs emissions accounted for
80% of TEWI value. In 3rd generation, this value increased to approximately 98%. This value depends
strongly on the energy matrix of the country under review, i.e. if electricity is or not produced from
renewable or fossil fuels. From the 3rd generation of refrigerant, the direct impact of GHGs emissions
related to a refrigerant on TEWI become marginal. Currently, the energy consumption by refrigeration
systems becomes much more relevant. Thus, must be sought the improvement of thermal efficiency and
heat transfer of refrigerants.
The European Union stands out as the group that more effort has demonstrated in searching and
applying alternative refrigerants more environmentally friendly. This is visible through the amount of
Directives and Regulations that have published in the last two decades on this field.
From the literature, the refrigerants with a more promising future: in terms of industrial and com-
mercial applications are the natural refrigerants (ammonia and CO2) and some blends of HFCs/HCs;
in terms of automotive applications are the HFOs refrigerants and blends of HFCs/HFOs; in terms of
domestic applications are the HCs and HFOs refrigerants, and blends of HFCs/HFO and HFCs/HCs.
During the next years the evolution on the field of refrigerants will be large, namely with respect to
the control and use of HCs, ammonia and CO2 as a refrigerant (technologies and equipment which en-
able a reduction in price, size and charge of the systems) and HFOs family (as the future of automotive
air conditioning).
Finally it was introduced the concepts of refrigerant recovery, recycling, reclamation R/R/R, having
these concepts as goal the reduction of refrigerants emissions into the atmosphere, thus promoting the
environmental sustainability of the Planet.

REFERENCES

3MTM NovecTM Engineered Fluids Company. (2003). The science of precision & electronics cleaning,
8/03. Retrieved June 10, 2014, from http://multimedia.3m.com/mws/media/229339O/novec-fluids-the-
science-of-precision-and-electronics-cleaning.pdf?fn=bro_nvcclng.pdf
American National Standards Institute & American Society of Heating. Refrigerating and Air-Condi-
tioning Engineers. (2010). ANSI/ASHRAE Standard 34-2010: Designation and Safety Classification
of Refrigerants. Atlanta, GA: ASHRAE SSPC.
Benhadid-Dib, S., & Benzaoui, A. (2012). Refrigerants and their environmental impact hydro chloro-
fluorocarbon substitution of HCFC and HFC hydro fluorocarbon. Search for an adequate refrigerant.
Energy Procedia, 18, 807–816. doi:10.1016/j.egypro.2012.05.096

253

Current and Future Trends of Refrigerants Development

Bolaji, B., & Huan, Z. (2013). Ozone depletion and global warming: Case for the use of natural refrig-
erant – a review. Renewable & Sustainable Energy Reviews, 18, 49–54. doi:10.1016/j.rser.2012.10.008
Bolaji, B. O. (2010). Experimental study of R152a and R32 to replace R134a in a domestic refrigerator.
Energy, 35(9), 3793–3798. doi:10.1016/j.energy.2010.05.031
Bouma, J. (2003). Refrigerant conservation programs in international perspective. In Proceedings of the
Xth European Conference - Technological innovations in refrigeration, air conditioning and the food
industry. Milano, Italy.
Brazilian Government. (2009). PORTARIA Nº 462, de 22 de Dezembro de 2009. Retrieved June 25, 2014,
from http://www.mma.gov.br/estruturas/ozonio/_arquivos/portaria_462_unidades_de_reciclagem_130.
pdf
Calm, J. M. (2008). The next generation of refrigerants – Historical review, considerations, and outlook.
International Journal of Refrigeration, 31(7), 1123–1133. doi:10.1016/j.ijrefrig.2008.01.013
Calm, J. M., & Didion, D. A. (1998). Trade-offs in refrigeration selections: Past, present, and future.
International Journal of Refrigeration, 21(4), 308–321. doi:10.1016/S0140-7007(97)00089-3
Clodic, D., & Barrault, S. (2011). 1990 to 2010 Refrigerant inventories for Europe. Previsions on banks
and emissions from 2006 to 2030 for the European Union. Brussels: Armines/ERIE.
Copeland & Emerson. (2008). Refrigerants for Residential and Commercial Air Conditioning Applications,
10/08. Retrieved June 10, 2014, from http://www.emersonclimate.com/asia/en-ap/whitepapers/6051-
r410a-whtpaper_9.pdf
Devotta, S., & Sicars, S. et al.. (2005). Safeguarding the ozone layer and the Global Climate System.
Cambridge, England: Cambridge University Press.
Dinçer, I., & Kanoglu, M. (2010). Refrigeration systems and applications (2nd ed.). United Kingdom:
John Wiley & Sons Ltd. doi:10.1002/9780470661093
Duarte, M. V. (2013). Estudo da bomba de calor – Fluidos de trabalho e eficiência energética. (Master
dissertation). Retrived from Ubi Thesis. (http://hdl.handle.net/10400.6/2472)
DuPont Refrigerants. (2008). Next Generation Refrigerants for Automotive Air Conditioning. Retrieved
June 10, 2014, from http://www2.dupont.com/Refrigerants/en_US/assets/downloads/SmartAutoAC/
next_gen_fact_sheet.pdf
DuPont Refrigerants. (2011). Guia de fluidos refrigerantes alternativos aos HCFCs e CFCs. Retrieved
June 10, 2014 from http://www2.dupont.com/Refrigerants/pt_BR/assets/downloads/Tabela_Suva.pdf
EU. (2006a). Regulation (EC) Nº 842/2006 of the European Parliament and of the Council on certain
fluorinated greenhouse gases. Official Journal of the European Communities. Brussels: Official Journal
of the European Communities.
EU. (2006b). Regulation (EC) Nº 2006/40/EC of the European Parliament and of the Council relating
to emissions from air-conditioning systems in motor vehicles and amending Council Directive 70/156/
EEC. Brussels: Official Journal of the European Communities.

254

Current and Future Trends of Refrigerants Development

EU. (2014a). MEMO/14/168 – MAC Directive: no evidence to support the safety concerns. Retrieved
June 10, 2014, from http://europa.eu/rapid/press-release_MEMO-14-168_en.htm
EU. (2014b). Regulation (EU) Nº 517/2014 of the European Parliament and of the Council on of fluo-
rinated greenhouse gases. Brussels: Official Journal of the European Communities.
EU. (2014c). Safety aspects of the use of refrigerant R1234yf - scientific review. Retrieved June 28, 2014,
from http://ec.europa.eu/enterprise/sectors/automotive/environment/macs/index_en.htm#h2-1
European Parliament – EU. (2000). Regulation (EC) Nº 2037/2000 of the European Parliament and
of the Council on substances that deplete the ozone layer. Brussels: Official Journal of the European
Communities.
European Partnership for Energy and the Environment – EPEE. (2014). The F-gas Regulation. Retrieved
June 28, 2014, from http://www.epeeglobal.org/refrigerants/f-gas-regulation/
Hwang, Y. (2013a). IIR Working Party on life cycle climate performance evaluation. In Proceedings of
the 4th International Conference on Thermophysical Properties and Transfer Process of Refrigerants.
Delft, Netherlands: International Institute of Refrigeration.
Hwang, Y. (2013b). Review of life cycle climate performance analysis and IIR Working Party. In Pro-
ceedings of the Conference on Advancing Ozone and Climate Protection Technologies: Next Steps.
Bangkok, Thailand: UNEP.
Intergovernmental Panel on Climate Change – IPCC. (2007). History. Retrieved May 30, 2014, from
https://www.ipcc.ch/organization/organization_history.shtml
IPCC. (2007). The F-gas Regulation. Retrieved June 28, 2014, from http://www.epeeglobal.org/
refrigerants/f-gas-regulation/
Jain, V., Kachhwaha, S. S., & Mishra, R. S. (2011). Comparative performance study of vapour compres-
sion refrigeration system with R22/R134a/R410A/R407C/M20. International Journal of Energy and
Environment, 55(2), 297–310.
Kohler, J. (2012). Safety considerations for adopting zero-ODP and low-GWP refrigerants. In Proceed-
ings of the 2nd Ozone2 Climate Technology Road Show and Industry Roundtable. Beijing, China: UNEP.
Leiper, A., Skelton, J., Rivers, N., & Zaynulin, D. (2014). Preventing transcritical operation of CO2
refrigeration systems with ground coupling. In Proceedings of the 3rd IIR International Conference on
Sustainability and the Cold Chain. London, England: The Institute of Refrigeration.
Li, M., Dang, C., & Hihara, E. (2012). Flow boiling heat transfer of HFO1234yf and R32 refrigerant
mixtures in a smooth horizontal tube: Part I. Experimental investigation. International Journal of Heat
and Mass Transfer, 55(13-14), 3437–3446. doi:10.1016/j.ijheatmasstransfer.2012.03.002
Low, R., & Cooper, M. (2014). New refrigerant mixtures with low GWP. In Proceedings of the 3rd
IIR International Conference on Sustainability and the Cold Chain. London, England: The Institute of
Refrigeration.

255

Current and Future Trends of Refrigerants Development

McLinden, M. O. (2011). Property data for low-GWP refrigerants: what do we know and what don’t
we know? In Proceedings of the 6th Seminar – Removing Barriers for low-GWP refrigerants. Boulder,
USA: ASHRAE.
McQuay. Air Conditioning (2002). Refrigerants – Applications guide, AG 31-007. Retrieved June 28,
2014, from http://lit.daikinapplied.com/bizlit/literature/lit_ch_wc/AppGuide/AG31-007_1.pdf
Melo, C. (2011). Uso de refrigerantes alternativos em refrigeração doméstica e equipamentos compactos
de refrigeração comercial. In A. Lage, E. & Amorim, F. (Ed.), Uso de fluidos alternativos em sistemas
de refrigeração e ar condicionado (45-56). Brasília, Brasil: Brazilian Ministry of Environment.
Michineau, T., Cavalier, G., & Devin, E. (2014). F-Gases in refrigerated transport. In Proceedings of the
3rd IIR International Conference on Sustainability and the Cold Chain. London, England: The Institute
of Refrigeration.
Minjares, R. (2011 Refrigerants for light-duty passenger vehicle air conditioning systems: Technical
assessment of alternatives to HFC-134a. In Proceedings of the 7th International Council on Clean
Transportation Conference. Mexico: ICCT.
Morais, J. C. (2011). Energia Offshore. Retrieved June 10, 2014, from http://www.fem.pt/lac/
Apresentacoes/29Nov4/edp.pdf
Naicker, P. K. (2014). A study of some next generation candidates for cascade refrigeration cycles. In
Proceedings of the 3rd IIR International Conference on Sustainability and the Cold Chain. London,
England: The Institute of Refrigeration.
Paul, S., Starkar, A., & Mandak, B. K. (2013). Environmental impacts of halogenated refrigerants and
their alternatives: Recent developments. International Journal of Emerging Technology and Advanced
Engineering, 3(3), 400–409.
Pedersen, H. (2012). Low GWP Alternatives to HFCs in Refrigeration. Copenhagen, Denmark: Danish
Technological Institute Press.
Rebitzer, G., Ekvall, T., Frischknecht, R., Hunkeler, D., Norris, G., Rydberg, T., & Pennington, D. et al.
(2004). Life cycle assessment: Part 1: Framework, goal and scope definition, inventory analysis, and ap-
plications. Environment International, 30(5), 701–720. doi:10.1016/j.envint.2003.11.005 PMID:15051246
Refripro (2014). Legislation: Chemikalien-Ozonschicht-Verordnung. Retrieved May 30, 2014, from
http://www.refripro.eu/en/03-know-how/03-legislation.html
Riva, M., Flohr, F., & Meurer, C. (2006). LCCP vs. Eco-Efficiency. In Proceedings of the International
Refrigeration and Air Conditioning Conference. Lafayette, USA: Purdue University.
Sciance, F. (2013). The Transition from HFC-134a to a Low-GWP Refrigerant in Mobile Air Conditioners.
In the Mobile Source Technical Review Subcommittee Meeting. Washington, DC: U. S. Environmental
Protection Agency (EPA).
Sekiya, A., & Misaki, S. (2000). The potential of hydrofluoroethers to replace CFCs, HCFCs and PFCs.
Journal of Fluorine Chemistry, 101(2), 215–221. doi:10.1016/S0022-1139(99)00162-1

256

Current and Future Trends of Refrigerants Development

Tomlinson, R. (2014). Ultra low charge direct ammonia cold store and freezer plant. In Proceedings
of the 3rd IIR International Conference on Sustainability and the Cold Chain. London, England: The
Institute of Refrigeration.
Tsai, W.-T. (2005). Environmental risk assessment of hydrofluoroethers (HFEs). Journal of Hazardous
Materials, 119(1-3), 69–78. doi:10.1016/j.jhazmat.2004.12.018 PMID:15752850
U. S. EPA. (2010a). Transitioning to low-GWP: Alternatives in unitary air conditioning. Retrieved June
10, 2014, from http://www.epa.gov/ozone/downloads/EPA_HFC_UAC.pdf
U. S. EPA. (2010b). Transitioning to low-GWP: Alternatives in commercial refrigeration. Retrieved June
10, 2014, from http://www.epa.gov/ozone/downloads/EPA_HFC_ComRef.pdf
U. S. EPA. (2010c). Transitioning to low-GWP: Alternatives in domestic refrigeration. Retrieved June
10, 2014, from http://www.epa.gov/ozone/downloads/EPA_HFC_DomRef.pdf
UN. (2014). Status of ratification of the Kyoto Protocol. Retrieved June 28, 2014, from https://ozone.
unep.org/Assessment_Panels/TEAP/Reports/RTOC/RTOC-Assessment-report-2010.pdf
UNEP. (2000). Montreal Protocol on substances that deplete the ozone layer as either adjusted and/or
amended in London 1990, Copenhagen 1992, Vienna 1995, Montreal 1997 and Beijing 1999. Ozone
Secretariat United Nations Environment Programme. Retrieved June 10, 2014, from http://www.unep.
ch/ozone/Montreal-Protocol/ Montreal-Protocol2000.shtml
UNEP. (2010). 2010 Report of the refrigeration, air conditioning and heat pumps technical options
committee. Retrieved June 10, 2014, from http://www.unep.ch/ozone/Montreal-Protocol/ Montreal-
Protocol2000.shtml
United Kingdom Parliamentary Office Of Science and Technology. (2006). Carbon Footprint of Electric-
ity generation. Retrieved June 10, 2014, from http://www.parliament.uk/documents/post/postpn268.pdf
United Nations - UN. (1998). Kyoto protocol to the United Nations Framework Convention on Climate
Change. Retrieved June 28, 2014, from http://www.unfccc.int/resource/docs/convkp/kpeng.pdf

KEY TERMS AND DEFINITIONS

Coefficient of Performance: Expresses the ratio between the energetic objective of the refrigeration
cycle and the quantity of energy provided to the system to comply the energetic objective of the cycle.
Drop-In Replacement: The refrigerant replacement does not require modifications (upgrade) in the
refrigeration system components. The replacement could be direct, because does not exist incompat-
ibilities between the new refrigerant and the installation materials.
Flammability of a Refrigerant: Ease with which a refrigerant comes into combustion.
Global Warming Potential: Index that characterizes the participation of one molecule of refrigerant
to the increase of global warming. This index is determined through the comparison with the molecule
of CO2 (R744) =1, the molecule of reference.
Residence Time in the Atmosphere: Residence time of a refrigerant in the atmosphere.

257

Current and Future Trends of Refrigerants Development

Lubricant: It is a substance, such as grease or oil, which reduces friction when applied as a surface
coating to moving parts. In other words, lubricants reduce the mechanical wear between moving parts.
Mixtures: Two or more substances which have been combined such that each substance retains its
own chemical identity.
Ozone Depletion Potential: Index that characterizes the participation of one molecule of refriger-
ant in the process of ozone layer destruction. This index is determined through the comparison with the
molecule of R12 (CFC) =1, the maximum potential value of ozone layer destruction.
Refrigerant: Are all the fluids with a high capacity to “absorb” large amounts of thermal energy
during the evaporation process at low temperature, characteristic that makes it ideal for use in refrigera-
tion systems operating by a refrigeration vapour compression cycle.
Retrofit Replacement: The refrigerant replacement require modifications (upgrade) of some com-
ponents of the refrigeration system.
Toxicity of a Refrigerant: Capability of a refrigerant to poison or do damage on the health of a person.
Vapour Compression Cycle: Thermodynamic cycle widely used in refrigeration systems to transfer
heat from an environment to another. A fluid is used (refrigerant) in a closed cycle, and submitted to
processes of compression, cooling with condensation, expansion and heating with evaporation.

258
Current and Future Trends of Refrigerants Development

APPENDIX: ABBREVIATION LIST

ASHRAE = American Society of Heating, Refrigerating and Air Conditioning Engineers;


EC = European Community;
COP = Coefficient of Performance;
EU = European Union;
GHG = Greenhouse Gas;
GWP = Global Warming Potential;
HOC = Heat Of Combustion;
IPCC = Intergovernmental Panel on Climate Change;
JAMA = Japanese Automotive Manufacturers Association
LCA = Life Cycle Assessment;
LCCP = Life Cycle Climate Performance;
LFL = Lower Flammability Limit;
LP = Low pressure;
LT = Low temperature;
MAC = Mobile Air Conditioning;
MT = Medium temperature;
ODP = Ozone Depletion Potential;
OEM = Original Equipment Manufacturer;
TEAP = Technology & Economic Assessment Panel;
TEWI = Total Equivalent Warming Impact;
TLV = Threshold Limit Value;
UNEP = United Nations Environment Programme;
VLT = Very low temperature;
EPA = US Environmental Protection Agency.

259
260

Chapter 7
Retrofitting of R404a
Commercial Refrigeration
Systems with R410a and
R407f HFCs Refrigerants
Mauro Gamberi Alessandro Graziani
University of Padua, Italy University of Padua, Italy

Marco Bortolini Riccardo Manzini


University of Bologna, Italy University of Bologna, Italy

ABSTRACT
This chapter presents an experimental analysis about the retrofitting of two commercial stationary
refrigeration systems marketed by an Italian leading company of the sector. Such systems operate both
at medium temperature (MT) and low temperature (LT) and they are originally designed to work with
the high global warming potential (GWP) hydrofluorocarbon (HFC) R404a fluid (GWP = 3922). The
purpose is to investigate the performances of HFCs R410a (GWP = 2088) and R407f (GWP = 1825)
chosen as effective alternatives to HFC R404a, due to their compatibility, non-flammability and market
availability. Furthermore, such fluids meet the EU restrictions in force in the next future for high GWP
HFCs. The experimental analysis compares the performances, in terms of COP and cooling capacity,
of R404a and the two identified alternatives under different operating conditions, i.e. chamber and con-
denser inlet air temperatures. In case of comparable performances, significant environmental benefits
are introduced by the adoption of R407f and R410a in the MT and LT refrigeration systems.

INTRODUCTION

According to the fifth assessment report (AR5) of the International Panel on Climate Change (IPCC),
the global emissions of greenhouse gases (GHGs) are at unprecedented level despite the rising atten-
tion to policies limiting the climate change. As in Figure 1, the total GHG emissions reach 49(±4.5)
GtCO2eq/year in 2010, while the average annual GHG emission growth from 2000 to 2010 is of about

DOI: 10.4018/978-1-4666-8398-3.ch007

Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

Retrofitting of R404a Commercial Refrigeration Systems

Figure 1. Global annual anthropogenic GHG emissions by group of gases 1970-2010


(IPCC, 2014)

1.0 GtCO2eq/year (+2.2%), compared to the 0.4 GtCO2eq/year (+1.3%) measured from 1970 to 2000,
showing an exponential trend. However, the climate change scenarios assessed in the AR5 of the IPCC
show that to have a likely chance of limiting the increase in global mean temperature to 2°C, and thus
prevent undesirable climate effects, means lowering GHG emissions by 40 to 70 percent compared with
2010 by mid-century, and to near-zero by the end of this century (IPCC, 2014).
In accordance with the international agreements on the global warming and climate change, e.g.
Kyoto Protocol in the 1997, the European Union (EU) establishes and continuously updates a roadmap
to progressive reduce GHG emissions in different sectoral areas. The growing concern on climate
change reported on the forth assessment (AR4) of the IPCC (IPCC, 2007) accelerates this process and
the European Directive (29/2009/EC), known as ‘’20/20/20 climate and energy package”, sets ambitious
targets by 2020:

• 20% reduction in EU GHG emissions from 1990 levels;


• 20% increase of the share of EU energy consumption produced from renewable sources;
• 20% improvement of the EU energy efficiency.

261

Retrofitting of R404a Commercial Refrigeration Systems

Refrigeration and air conditioning systems contribute to GHG emissions due to the energy consump-
tion, which is identified as indirect emission, and the adoption of working fluids with significant GWP,
which represents a direct emission due to the risk of accidental gas leakage into the atmosphere (Bovea,
Cabello, & Querol, 2007). Following the Montreal Protocol (1987), fluorinated gases (F-Gases), mostly
the hydrofluorocarbons (HFCs), replace the chlorofluorocarbons (CFC) and hydrochlorofluorocarbons
(HCFC) ozone-depleting refrigerants. They have no ozone depletion potential because chlorine is elimi-
nated from their chemical structure, low toxicity, non-flammability and thermally stability. However,
HFCs are a family of powerful GHG, listed into the basket of seven GHGs (Kyoto Protocol, 1997), with
a warming effect on the atmosphere up to 23’000 times greater than that of CO2 and responsible of the
2% of EU overall GHG emissions.
The EU defines measures to control their emissions (842/2006/EC), while the EU Directive (40/2006/
EC) sets significant limits, i.e. 100yr GWP<150, to the emissions from air-conditioning systems in motor
vehicles. In the other sectors, after several discussed proposals (EU Legislative Resolution, 2014), the
EU Regulation No 517 of the European Parliament and of the council of 16 April 2014 on F-gases for
the most common refrigeration and air-conditioning systems defines a program to reduce F-Gas emis-
sions and reach IPCC targets. Particularly, non-CO2 emissions, including F-gases, should be reduced by
72% to 73% by 2030 and by 70-78% by 2050, compared to the 1990 levels. To achieve such purpose, a
two-third reduction in F-gas emissions by 2030 is mandatory. A phase-down program to decrease F-gas
sold in the EU and to ban F-gases in some of new equipment is set. Such a program is cost-effective
because proven and tested alternatives are available in many sectors (517/2014/EU).
Table 1 shows the introduced HFCs market prohibition program, which plans to stop the production
of new refrigeration systems working with HFCs with GWP values overcoming a specific limit, depend-
ing on the type of refrigeration system and use, from 2015 to 2025.
Finally, Figure 2 presents a timeline of the previously discussed EU Directives/Regulations and
international agreements progression on CFCs, HCFCs and HFCs refrigerants (Calm, 2008; Benhadid-
Dib & Benzaoui, 2012).
According to such a phase-out program, two strategies are outlined. The former deals with the devel-
opment of new equipment working with natural refrigerants, hydrocarbons (HCs) and new generation
fluids. The latter deals with the retrofitting of existing plants using lower GWP refrigerants.

Table 1. HFCs phase-out program of the European Union

New Equipments
Domestic refrigerators and freezers that contain HFCs with GWP of 150 or more 1 January 2015
Refrigerators and freezers for commercial use that contains HFCs with GWP of 2500 or more 1 January 2020
(Hermetically sealed equipment)
that contain HFCs with GWP of 150 or more 1 January 2022
Stationary refrigeration equipment, that contains, or whose functioning relies upon, HFCs with GWP of 2500 1 January 2020
or more
Movable air conditioning that contain HFCs with GWP of 150 or more 1 January 2020
Single split air-conditioning systems containing less than 3kg of fluorinated greenhouse gases with GWP of 1 January 2025
750 or more
Maintenance Operations
With an appropriate transitional period, the use of refrigerants with a very high global warming potential of 2500 or more to service or
maintain refrigeration equipment with a charge size of 40 tonnes of CO2 equivalent or more should be banned.

262

Retrofitting of R404a Commercial Refrigeration Systems

Figure 2. International protocol and EU Regulation progression on CFC, HCFC, and HFC refrigerants

This chapter presents an experimental analysis about the retrofitting of two commercial stationary
refrigeration systems, proposed by a leading company in the refrigeration sector. The plants are originally
designed for the HFC R404a (AR4 100yr GWP = 3922) fluid. The purpose is to investigate the feasi-
bility and effectiveness of the switch to the HFCs R410a (AR4 100yr GWP = 2088) and R407f (AR4
100yr GWP = 1825) fluids. R410a and R407f fluids show a good potential in terms of compatibility
with the considered plants, they are non-flammable, their market availability and cost are competitive
and they match with the future EU market restrictions (see Table 1). The aim of the experimental study
is to verify the thermodynamic performances of R407f and R410a respect to the R404a fluid. In case
of comparable or better performances, the adoption of R410a and R407f for the existing plants, either
new or in use, leads to significant environmental impact reduction.

BACKGROUND

Kyoto Protocol (1997) and Montreal Protocol (2000) reflect the world concern about the GHG effect
increase due to anthropogenic activities. GHG emissions in the field of refrigeration and air conditioning
are the consequence of two main aspects: the former is the direct effect due to the GHG emissions of the
used refrigerants, generally belonging to the HFC fluid family. HFCs are listed into the basket of seven
GHGs published in the Kyoto Protocol. The latter is the indirect effect caused by the energy consump-
tion of cooling facilities (Bovea et al., 2007). To limit the environmental impact of cooling plants, three
strategies, or even a mix of them, are identified:

• The mitigation of the direct effect by using new refrigerants which are harmless for the green-
house effect;
• The mitigation of the direct effect by the retrofitting of existing plants, either new or in use, using
lower GWP refrigerants;
• The improvement of the efficiency of the refrigeration systems and the reduction of the indirect
effect by either modifying or implementing systems with greater energy efficiency.

Considering the first strategy, several scientific contributions show the potential technical and eco-
nomic feasibility of the adoption of natural refrigerants, hydrocarbons (HCs) and new generation fluids.

263

Retrofitting of R404a Commercial Refrigeration Systems

• Calm (2008) reviews the progression of refrigerants, from early use to present, and then addresses
future directions and candidates concluding that governing selection criteria for new generation
adds low GWP to old requirements for suitability, safety and material compatibility. The new gen-
eration must also offer high efficiency or the low GWP refrigerant may increase the environmental
impact due to indirect GHG emissions;
• Mohanraj, Jayaraj, & Muraleedharan (2009) review the suitable environment friendly alternatives
to halogenated refrigerants highlighting the importance of use of natural refrigerants and HC/HFC
mixtures in the next refrigeration systems;
• Palm (2008) review the adoption of HCs as refrigerants in small heat pump and refrigeration sys-
tems showing that using HCs it is possible to obtain performances equal or higher than those of
similar HFC systems;
• Bolaji & Huan (2013) reviews the use of natural refrigerants defining them the most suitable long
time alternatives. Among natural refrigerants, propane (R290) is one of the most interesting solu-
tions to guarantee energetic performance with a low direct environmental impact. Safety concerns
related to its flammability are limiting its introduction in some fields of refrigeration, therefore the
research is focuses on new solutions to reduce refrigerants charge and costs (Mastrullo, Mauro,
Menna, & Vanoli, 2014; Hwang, Jin, & Radermacher, 2007);
• McLinden, Kazakov, Steven Brown, & Domanski (2014) analyze the possibilities and tradeoffs
for low-GWP refrigerants, including natural refrigerants and hydrofluoroolefins (HFO) conclud-
ing that only few refrigerants are promising solutions;
• Finally, Duvedi & Achenie (1997) proposes a mathematical programming approach to design new
environmentally-friendly refrigerant mixtures.

The second strategy, i.e. the retrofitting of existing plants with lower GWP refrigerants, reduces the
direct GHG emissions while, the indirect impact is a function of the thermodynamic performances. For
this reason, many contributions carry out experimental studies to compare the performances of refrig-
eration systems with the original refrigerant and the identified lower GWP alternatives. As examples,
experimental studies evaluate the replacement of ozone depleting CFC-12 and HCFC-22 due to the
ongoing and future global phase-out, with HFCs, HCs, HFC/HC mixtures (Yang & Wu, 2013), while
other experimental studies are focused on the retrofitting of HFCs with lower GWP HFCs, HCs and
hydrofluoroolefins/ hydrofluoroolefin-based mixtures. A summary of such retrofitting operations is in
Table 2. Among them, some experimental analysis show that an optimized retrofitting of the existing
refrigeration systems may improve or have comparable efficiency and reduce both direct and indirect
GHG emissions without significant investments for the system re-design.
Starting from such a background, this chapter presents an experimental analysis about the retrofitting
of two commercial stationary refrigeration systems. Such systems are commercial solutions proposed by
a leading company in the refrigeration sector and they are originally designed for the HFC R404a (AR4
100yr GWP = 3922). The HFC R404a is the global industry standard HFC refrigerant for commercial
refrigeration in both small and large systems and the literature proposes few contributions about his re-
placement in existing plants. Mastrullo et al. (2014) proposes the replacement of R404a with the natural
refrigerant Propane (R290) in a light commercial freezer. After the optimization of the performances
with respect to the refrigerant charge and the regenerative capillary tube lenght, a remarkable increase
of the performance is obtained. However, safety concerns about R290 flammability are still present.
Minor (2012) proposes the adoption of the new low GWP refrigerant HFO1234yf (AR4 100 year GWP

264

Retrofitting of R404a Commercial Refrigeration Systems

Table 2. Scientific contributions about the retrofitting of existing plants

Retrofitting Refrigerants
Original Scientific Contribution HFCs HCs HFC / HC HFOs /
Refrigerant Mixtures HFO -Based
Mixtures
HCFC-22 (Spatz & Yana Motta, 2004) ✓ ✓
(Jabaraj, Narendran, Lal, & Renganarayanan, 2007) ✓
(Chen, 2008) ✓
(La Rocca & Panno, 2011) ✓
(Aprea, Maiorino, & Mastrullo, 2011) ✓
(Bukola Olalekan Bolaji, 2011) ✓
(Llopis, Torrella, Cabello, & Sánchez, 2012) ✓
CFC-12 (S.Joseph Sekhar, Lal, & Renganarayanan, 2004) ✓
(S.J. Sekhar & Lal, 2005) ✓
(Jerald & Senthilkumaran, 2014) ✓
HFC (Xuan & Chen, 2005) ✓
(Wongwises, Kamboon, & Orachon, 2006) ✓
(Hwang et al., 2007) ✓
(Mohanraj, Jayaraj, Muraleedharan, & Chandrasekar, ✓
2009)
(Rasti, Hatamipour, Aghamiri, & Tavakoli, 2012) ✓
(Jarall, 2012) ✓
(Minor, 2012) ✓
(Han et al., 2012) ✓
(Xu, Hwang, & Radermacher, 2013) ✓
(Navarro-Esbrí, Mendoza-Miranda, Mota-Babiloni, ✓
Barragán-Cervera, & Belman-Flores, 2013)
(Yu & Teng, 2014) ✓
(Mastrullo et al., 2014) ✓

= 4), which is a suitable component in mixtures able to match R404a performance while reducing GWP.
Two approaches are proposed. The first is to develop the lowest GWP R404a possible alternative (GWP
~1400) while maintaining non-flammability. The second approach is to drive to the lowest GWP (e.g.
~250), but with mild flammability. Both the developed mixtures show thermodynamic performances
comparable to R404a. Finally, Han et al. (2012) present a new ternary non-azeotropic mixture of HFC-
161/125/143a (AR4 100 year GWP = 3300), as a low GWP alternative to R404a fluid. Experimental
evidences show that the new mixture obtains better performances and thermodynamic conditions. The
purpose of this experimental study is to investigate the switch to the HFCs R410a (AR4 100yr GWP =
2088) and R407f (AR4 100yr GWP = 1825). The selection of HFCs R410a and R407f is the results of
a preliminary screening of the refrigerants available in the market, which shows an optimal trade-off
between the compatibility with considered refrigeration systems, the non-flammability, the availability
and the future market restrictions imposed by the EU. Indeed, the described retrofitting operations are
in accordance with the aforementioned 517/2014/EU Regulation on stationary refrigeration systems (see

265

Retrofitting of R404a Commercial Refrigeration Systems

Table 1) and suitable for maintenance operations in systems designed for HFC R404a and non-flammable
refrigerants. The aim of the experimental study is to verify whether the R407f and R410a refrigerants
show thermodynamic performances comparable to those of the R404a, so that the adoption of the two
identified alternatives reduces the direct GHG emissions without increasing the indirect GHG emissions
due to the energy consumption.

RETROFITTING OF REFRIGERATION SYSTEMS

The retrofitting of a refrigeration system represents the replacement of the original refrigerant, generally
with a high impact on the environment, with a lower environmental impact fluid. No significant plant
re-design and component changes are required. In addition, the adoption of the new refrigerant has to
ensure that the existing plant operates until the end of its economic life.
Thus, the purpose of a retrofitting operation is the environmental impact reduction (both the direct
and indirect effects) of a cooling plant during its lifetime. The direct effect depends on the GWP of the
refrigerant, the amount of refrigerant leakage, the equipment run time, etc., while the indirect emissions
depend on the energy consumption during operation and the CO2 emissions per unit of generated energy.
This Section presents and discusses the results of the developed experimental study on the retrofit-
ting of two existing refrigeration systems. The Section also includes a preliminary overview of the main
indices used to estimate the environmental impact.

Environmental Impact of Refrigeration Systems

Common evaluation indices for the global warming impact are the GWP, the total equivalent warming
impact (TEWI), the life cycle global warming impact (LCWI), the life cycle climate performance (LCCP)
and carbon footprint. GWP is used as a general evaluation index for global warming impact while TEWI,
LCWI, LCCP and carbon footprint are evaluation indexes specially developed for refrigeration systems.

Global Warming Potential (GWP)

The GWP is used to measure the GHG effect of a gas based on its radiative properties over a given time
frame, considering CO2 as common benchmark. GWP is 1. The GWP is measured in mass of equivalent
carbon dioxide (CO2e). The GWP is a function of its infrared radiation absorptivity, its stability in the
atmosphere and the selected time frame. Thus, the same gas can have different GWPs for different time
frames. 100 years is the normally used standard time frame (Wu, Hu, & Mo, 2013).
In the EU Regulation (842/2006/EC) GWP is defined as “the climatic warming potential of a green-
house gas relative to that of carbon dioxide and is calculated in terms of the 100-year warming potential
of one kilogram of a gas relative to one kilogram of CO2”.

Total Equivalent Warming Impact (TEWI)

The TEWI index considers both direct and indirect emissions related to a refrigeration system. The
former represents the quantity of refrigerant leaked from the equipment during its use phase and main-
tenance, the latter represents the amount of GHG gases released because of the energy consumed by

266

Retrofitting of R404a Commercial Refrigeration Systems

the cooling plant. The TEWI index is calculated as the sum of direct and indirect emissions related to a
refrigeration unit, limited to its use phase, only. The TEWI is a suitable term of comparison between dif-
ferent cooling plants or different refrigerant retrofitting options. Several TEWI definitions are proposed
in the literature. Davies and Caretta (2004) consider only refrigerant leakage as direct GHG emissions
while Wu et al. (2013) evaluate the contributions of the residual amount of refrigerant and its recovery
rate at the end of system lifetime. Formally:

TEWI = Direct GHG emissions + Indirect GHG emissions (1)

Direct GHG emissions = l ⋅ n ⋅ GWP + R ⋅ (1 − a ) ⋅ GWP (2)

Indirect GHG emissions = n ⋅ E ⋅ ε (3)

where l is the refrigerant leakage per year [kg/year]; n is cooling plant lifetime [year]; R is the residual
amount of refrigerant at the end of cooling system lifetime [kg]; a is the refrigerant recovery rate; E
is the energy consumption per year [kWh/year]; ε is the amount of CO2 emissions for 1 kWh energy
generation [kgCO2/kWh].

Life Cycle Global Warming Impact and Life Cycle Climate Performance

Life cycle global warming impact (LCWI) extends the concept of TEWI on the basis of the assumption
that indirect emissions resulting from manufacturing, delivery and recycling of refrigerants are a consid-
erable addition to the environmental burden and must be included in the environmental impact assess-
ment (Papasavva & Moomaw, 1997). Life cycle climate performance (LCCP) is comparable to LCWI
and extends the concept of TEWI including the consideration of the warming impact associated to the
energy consumed to manufacture both the refrigerant and the raw materials used for the manufacturing
of the refrigerant and the direct warming impact of any fugitive GHGs emitted during the refrigerant
manufacture (Dieckmann & Little, 1999).
LCCP is defined as:

LCCP = l ⋅ n + R ⋅ (1 − a ) ⋅ (GWP + Er + Fr ) + n ⋅ E ⋅ ε (4)


 

where Er is the global warming impact due to the manufacturing process of refrigerant, including energy
consumed and raw materials (embedded energy) and Fr is the direct warming impact of any fugitive
GHG emitted during the refrigerant manufacture (“fugitive” emissions). Estimated values of Er and Fr
for common refrigerants are deeply investigated by the literature (Wu et al., 2013).

267

Retrofitting of R404a Commercial Refrigeration Systems

Carbon Footprint

The carbon footprint describes the CO2 emissions of a given product or process through its whole life.
Such a basic concept is detailed by the recent literature. Wiedmann & Minx (2007) define the carbon
footprint as a measure of the total amount of carbon dioxide emissions directly and indirectly caused
by an activity over the product lifetime. The direct emissions are the GHG effect caused by the leakage
of the various GHGs in each process, including the emissions of byproducts with GHG impact in the
production of the refrigerants and refrigeration systems, the leakage of the refrigerants during assembly
and transport and the leakage of the refrigerants during use, repair and recycling. The indirect emissions
are the CO2 equivalent emissions due to the energy consumption in each process including production,
transport, use and recycling of the refrigerants and refrigeration systems (Wu et al., 2013). Figure 3
shows an example of the system boundaries of a carbon footprint assessment (CFA) for a refrigeration
system. The refrigeration unit and refrigerant are included in the analysis providing a cradle to the grave
assessment.

Experimental Analysis

The experimental analysis is focused on the performance comparison of different fluids combined to two
stationary refrigeration systems produced by a leading company in the refrigeration and air conditioning
sector. This subsection presents:

• The description of the reference refrigeration cycle;


• The overview of the considered refrigerants;
• The description of the laboratory test equipment and procedure;
• The multi-scenario analysis carried out during the experimental study.

Figure 3. System boundaries of a CFA for a refrigeration system

268

Retrofitting of R404a Commercial Refrigeration Systems

Refrigeration Systems

Single Stage Vapor Compression Cycle


Both stationary refrigeration systems are based on the so-called single stage vapor compression refrigera-
tion cycle. The vapor compression cycle uses a circulating refrigerant as the medium that absorbs and
removes heat from the volume to be cooled and subsequently transfers that heat elsewhere. Such systems
have four basic functional units: a compressor, a condenser, a thermal expansion valve or a capillary
tube and an evaporator. There are four refrigeration processes in an ideal single-stage vapor compression
cycle, as shown in the diagram of Figure 4 and p-h diagram shown in Figure 5.
Isentropic compression process (1 - 2). The circulating refrigerant enters the compressor in the ther-
modynamic state known as saturated vapor (1) and it is compressed isentropically from point 1 to 2,
increasing pressure and temperature. The hot, compressed vapor is in the thermodynamic state known
as superheated vapor (2). The work input to the compressor w12 , in kJ/kg, is:

w12 = (h1 − h2 ) (5)

The greater the difference in temperature and pressure between the condensing pressure, pcon, and
the evaporating pressure pev, the higher the work input to the compressor is.
Isothermal condensation process (2 - 3). Hot gaseous refrigerant discharged from the compressor is
condensed in the condenser into liquid so that the latent heat for condensation is transferred to the con-
denser water or the ambient air. At the condenser outlet, the condensed liquid refrigerant is in a thermo-
dynamic state known as saturated liquid (3). The heat rejection during condensation, −q 23 , in kJ/kg, is:

−q 23 = (h3 − h2 ) (6)

Figure 4. Schematic diagram of a single-stage vapor compression cycle

269

Retrofitting of R404a Commercial Refrigeration Systems

Figure 5. p-h diagram of a single-stage vapor compression cycle

1. Throttling process (3 - 4). Liquid refrigerant flows through a throttling device (e.g. an expansion
valve, a capillary tube or orifices) and its pressure is reduced to the evaporating pressure. A portion
of the liquid flashed into vapor (adiabatic flash evaporation) and enters the evaporator. This is the
only irreversible process of the described ideal cycle, represented as a dotted line in Figure 5. For
a throttling process, assuming that the heat gain from the surrounding is negligible:

h3 = h4 (7)

The auto-refrigeration effect of the adiabatic flash evaporation and the pressure drop generate a de-
crease of the temperature of the liquid and vapor refrigerant mixture to a value that it is colder than the
temperature of the enclosed volume to be refrigerated.
Isothermal evaporation process (4 - 1). The cold mixture is routed through the coil or tubes in the
evaporator. The circulating refrigerant absorbs and removes heat from the low temperature room which
is subsequently rejected to the condenser and transferred elsewhere by the water or air used in the con-
denser. The refrigerant evaporates completely in the evaporator and the latent heat of evaporation is
absorbed by the refrigerant producing the refrigeration effect q14 , in kJ/kg:

q14 = (h1 − h4 ) (8)

270

Retrofitting of R404a Commercial Refrigeration Systems

An ideal single stage vapor compression refrigeration cycle on a ph diagram is divided into two
pressure regions called high pressure, pcon, and low pressure, pev. The saturated vapor at the evaporator
outlet is routed back into the compressor to complete the refrigeration cycle.
The COP is a dimensionless index used to indicate the performances of a thermodynamic cycle or
thermal system. The magnitude of COP is generally greater than 1. The COP is the ratio between the
useful effect and the compressor work input.

COP = q14 / w12 considering the refrigeration effect (refrigeration system) (9)

COP = q 23 / w12 considering the condensation effect (heat pump) (10)

The condensed liquid is often subcooled to a temperature lower than the saturated temperature cor-
responding to the condensation pressure pcon, as in Figure 6.
Subcooling increases the refrigeration effect to:

q14′ = (h1 − h4 ' ) > (h1 − h4 ) (11)

The sub cooling is achieved by any of the following methods:

Figure 6. p-h diagram of a single-stage vapor compression cycle with subcooling

271

Retrofitting of R404a Commercial Refrigeration Systems

• By passing the liquid refrigerant from the condenser through a heat exchanger through which the
cold vapor at suction from the evaporator is allowed to flow to the reversed direction. This process
subcools the liquid but superheats the vapor. The COP is not improved, while the refrigeration
effect increases.
• By making use of enough quantity of cooling water so that the liquid refrigerant is further cooled
below the temperature of saturation. In some cases, a separate subcooler is also made for this pur-
pose. In this case, COP is improved.

Vapor superheating (See Figure 7) further prevents the liquid refrigerant to flood back into the com-
pressor and causing damages. The degree of superheating mainly depends on the types of refrigerant
feed, the construction of the suction line and the type of compressor. The state point of the vapor refrig-
erant after superheating of an ideal system must be at the evaporating pressure with a specific degree of
superheat (Wang & Lavan, 1999).

Commercial Stationary Refrigeration Systems


The experimental analysis considers two stationary refrigeration systems. They are versatile monoblock
units to be installed in small and medium size cold rooms, either in straddle position or through-the-wall
with an auxiliary thru-wall kit. The refrigeration systems are designed to operate in the medium tempera-
ture (MT) range, i.e. [-5; +10] °C, and low temperature (LT) range, i.e. [-25; -15] °C. In the following,
MT and LT indicate both the refrigeration systems and their working temperatures. Figure 8 and Figure
9 present general views of the systems and the process diagram, respectively. The key technical data of
the systems are in Table 3.

Figure 7. p-h diagram of a single-stage vapor compression cycle with vapor superheating

272

Retrofitting of R404a Commercial Refrigeration Systems

Figure 8. Views of the MT and LT systems

Figure 9. Process diagram of the adopted experimental system

273

Retrofitting of R404a Commercial Refrigeration Systems

Table 3. Technical data of the MT and LT commercial refrigeration systems

Parameter MT LT
Working temperature range [-5; +10]°C [-25; -15]°C
Refrigerant R404a
Compressor type Hermetic
Compressor nominal power 0.56kW 1.3kW
Nominal current consumption 4.90A 5.10A
Max current consumption 8.30A 10.9A
Expansion type Capillary tube
Capillary tube diameter 1.83mm 1.49mm
Capillary tube length 3400mm 2500mm
Condenser type Micro-channels
Condenser flow rate 600m3/h
Evaporator type Finned heat exchanger
Evaporator flow rate 600m3/h
De-frost system Hot fluid (bypass)

The key system features are shortly commented in the following bulleted list.

• Both MT and LT systems use hermetically sealed reciprocating compressor, a widely adopted
technology for the refrigeration and air conditioning applications. In hermetically sealed compres-
sor, the compressor and the motor are enclosed to a welded steel case and they are connected by
a common shaft. This structure makes the whole compressor and the motor a single compact and
portable unit that can be handled easily;
• The fluid evaporation is through a finned heat exchanger, while the fluid condensation through a
micro-channel heat exchanger. Micro-channel condensers, whose channels has a hydraulic diam-
eter below 1 mm, increase the efficiency and reduces the energy consumption;
• The fluid expansion is through a capillary tube. The capillary tube is one of the most commonly
used throttling devices in the refrigeration and the air conditioning systems. It is a copper tube of
small internal diameter and long length reduced by coiling it. The internal diameter of the capil-
lary tube used for the refrigeration and air conditioning applications generally ranges from 0.5 to
2.5 mm. When the refrigerant leaves the condenser and enters the capillary tube its pressure drops
down suddenly. The pressure drop of the refrigerant depends on its diameter and its length. The
smaller the diameter and the higher the length, the higher the pressure drop of the refrigerant is.
The capillary tube is a non-adjustable device. The control of the refrigerant flow is not possible
as commonly done with automatic throttling valves. According to this, the flow of the refrigerant
through the capillary changes together with the boundary conditions change. For instance, if the
condenser pressure increases due to the high atmospheric temperature and the evaporator pres-
sure decreases due to the lower refrigeration load the flow of the refrigerant through the capillary
changes. The capillary tube is designed for a given ambient condition range. However, if properly
selected, it works reasonably well over a given wide range of conditions;

274

Retrofitting of R404a Commercial Refrigeration Systems

• The subcooling of the condenser outlet flow and the superheating of the evaporator outlet flow are
provided by transferring heat to the evaporator outlet flow inside the capillary tube. Subcooling of
the condenser outlet flow increases the refrigeration effect;
• A bypass tube to de-frost the evaporator coils is, finally, provided to avoid reduction of the evapo-
rator heat exchange efficiency. The hot vapor leaving the compressor is bypassed through the
evaporator coils for a period of time melting down the existent frost. The resulting water drains
through a duct and it flows out.

Refrigerants

Required Properties
A refrigerant is a primary working fluid used to produce a cooling effect in a refrigeration system. All
refrigerants extract heat from a low temperature and low pressure environment during evaporation and
eliminate such a heat to the high temperature and pressure environment during condensation. Refrigerants
should not cause ozone depletion and low GWPs are required. Additional considerations for refrigerant
selection are in the following:

1. Safety aspects, including toxicity and flammability. ANSI/ASHRAE Standard 34-1992 classifies
the toxicity of refrigerant as Class A and Class B. Class A refrigerants are of low toxicity. Class B
refrigerants are of higher toxicity. ANSI/ASHRAE Standard 34-1982 classifies the flammability
of refrigerants as Class 1, no flame propagation; Class 2, lower flammability; and Class 3, higher
flammability. The safety classification of refrigerants is based on the combination of toxicity and
flammability: A1, A2, A2L, A3, B1, B2, B2L and B3 groups are provided (See Figure 10);
2. Effectiveness of the refrigeration cycle. The energy consumed per ton of produced refrigeration is
an index of this assessment;
3. Oil miscibility. Refrigerants should be miscible with mineral lubricant oil because of a mixture of
refrigerant and oil helps to lubricate pistons, discharge valves, bearings and other moving parts of
a compressor;
4. Compressor displacement. Compressor displacement per ton of produced refrigeration directly
affects the size of the positive displacement compressor and therefore it compactness;
5. Desired properties:
a. Evaporating pressure should be higher than atmospheric pressure so that non condensable
gases will not leak into the system;
b. Low condensing pressure for lighter construction of compressor, condenser, piping etc;
c. High thermal conductivity and therefore high heat transfer coefficient in the evaporator and
condenser;
d. Dielectric constant should be compatible with air when the refrigerant is in direct contact
with the motor windings in the hermetic compressor;
e. An inert refrigerant that does not chemically react with material will avoid corrosion, erosion
and damages to the system component;
f. Refrigerant leakage has to be easily detected (Wang & Lavan, 1999).

275

Retrofitting of R404a Commercial Refrigeration Systems

Figure 10. ASHRAE refrigerant safety classification

Refrigerant Selection
Both MT and LT systems are originally designed to work with R404a HFC fluid. R404a is a global
industry standard HFC refrigerant for commercial refrigeration in both small and large systems. Due
to the high global warming potential of R404a (AR4 100 year GWP = 3922), the interest is growing to
identify low GWP alternatives which match with the R404a performances. The components of a specific
air conditioning or refrigeration equipment are engineered around the properties of the refrigerant used.
When the replacement of that refrigerant becomes necessary for technical, regulatory or economic rea-
sons, the new refrigerant should have as many similar properties as the original one. In this way hardware
changes, controls adjustments, or other time consuming operations such as oil changes are reduced.
Table 4 shows a list of common refrigerants available in the market. Their composition, the ozone
depleting potential (ODP), the GWP and the ASHRAE safety group are presented. In addition, Figure
11 shows the unitary cost of some of these refrigerants obtained through a market survey.
Among these alternatives, the choice of HFCs R410a and R407f comes from a preliminary screen-
ing that highlights their optimal trade-off between the compatibility with the considered refrigeration
systems (capillary tube length and diameter and the amount of refrigerant mass charged), evaporation
and condensation refrigerant pressures in the operating temperature ranges, the non-flammability, the
availability in the market, the costs and the future market restrictions imposed by the EU. The described
retrofitting operations are in accordance with the aforementioned 517/2014/EU Regulation on station-
ary refrigeration systems (see Table 1) and suitable for maintenance operations in systems designed for
HFC R404a and non-flammable refrigerants. Particularly, R404a and R407f fluids operate with exactly
the same system components (drop-in replacement), while R410a requires the substitution of the sole
hermetic compressor. Table 5 summarizes the environmental impact, the ASHRAE safety group and
the chemical composition of the three considered fluids.

276

Retrofitting of R404a Commercial Refrigeration Systems

Table 4. R404a and potential alternatives

Standard 34 Refrigerant Molecular Formula ODP GWP Safety Group


ASHRAE Name ASHRAE
R134a 1,1,1,2-tetrafluoroethane C2H2F4 0 1300 A1
R404a HFC - Blend R125/R143a/R134a 0 3700 A1
(44/52/4)
R407f HFC - Blend R32/R125/R134a 0 1530 A1
(30/30/40)
R410a HFC - Blend R32/R125 (50/50) 0 1730 A1
R152a Difluoroethane-1,1 C2H4F2 0 140 A2
R161 Ethylfluoride CH3 CH2F 0 12 A2
R32 Difluoromethane CH2F2 0 675 A2
R717 Ammonia NH3 0 <1 B2
R744 Carbon Dioxide CO2 0 1 A1
R290 Propane C3H8 0 3 A3
R600a Isobuthane CH(CH3)3 0 3 A3
R1270 Propylene C3H6 0 3 A3
R1234yf HFO blend C3H2F4 0 4 A2
R1234ze HFO blend C3F4H2 0 6 -

Table 5. R404a, R410a and R407f characteristics

Working Fluid Ozone Global ASHRAE Composition


Depletion Warming Safety Group
R143a R125 R32 R134a
Potential Potential
R404a 0 3922 A1 52% 44% - 4%
R410a 0 2088 A1 - 50% 50% -
R407f 0 1825 A1 - 30% 30% 40%

R410a and R407f fluids have a considerable higher GWP compared to the new generation fluids,
natural gases and HCs but it is significantly lower than R404a. As a consequence, significant environ-
mental benefits are introduced by their adoption in the case of comparable efficiencies. The technical
performances of such fluids, when coupled to the two described commercial systems, are field checked
and compared during the experimental analysis.

Thermostatic Chamber and Control System

The experimental analysis is focused on the assessment of the COP and the cooling capacity of HFCs
R407f and R410a identified as alternatives to the HFC R404a (benchmark). The laboratory system used
for the experimental tests is now described before presenting the key outcomes
The experimental tests are developed using a walk-in thermostatic chamber as the controlled tem-
perature environment. The thermostatic chamber, shown in Figure 12, is entirely made of stainless steel.

277

Retrofitting of R404a Commercial Refrigeration Systems

Figure 11. Unitary cost of refrigerants

The external dimensions are 180x180x220(h)cm while the internal dimensions are 160x160x200cm,
with an internal volume of 5.12m3. The insulation is provided by a 100mm layer of rigid expanded
polyurethane with a reference thermal conductivity of k = 0.022 W/mK. The chamber integrates the
refrigeration systems, equipped with several sensors to monitor the temperature inside and outside the
chamber, as well as the thermodynamic conditions of the working fluid, i.e. temperatures and pressures.
Table 6 lists the sensors currently integrated in the climatic chamber, their range of calibration and
accuracy and the monitored parameters.
The temperature measurements are made by means of PT100 termoresistances with ± 0.3°C accu-
racy, while pressure monitoring is by using piezoresistive transducers with ± 0.025bar accuracy. Finally,
current is measured by a Hall effect current transducer with ± 0.15A accuracy. The acquisition of the
sensor signals and the data processing are provided by a customized and easy-use real-time interface
developed in LabviewTM integrated development environment (IDE). The front panel of the monitoring
and control tool is in Figure 13 and it includes the output of each sensor (4-20mA output signals), the
calculation of the parameters to evaluate the system performances and the storage data settings. The
control system also allows adjusting the condenser inlet air temperature and the thermostatic chamber

278

Retrofitting of R404a Commercial Refrigeration Systems

Figure 12. Thermostatic chamber view

Table 6. List of sensors equipped in the thermostatic chamber

Notation Measurement Calibration Range and Accuracy


Tchamb Chamber temperature [-43; +47] ± 0.3°C
Tair_ext Ambient temperature [-43; +47] ± 0.3°C
T1 Compressor inlet temp. [-43; +47] ± 0.3°C
T2 Compressor outlet temp. [0; +159.5] ± 0.3°C
T3 Condenser inlet temp. [0; +86] ± 0.3°C
T4 Evaporator inlet temp. [-43; +47] ± 0.3°C
T5 Evaporator outlet temp. [-43; +47] ± 0.3°C
Tjet Cooling jet temperature [-43; +47] ± 0.3°C
P1 Compressor inlet press. [0; +10] ± 0.025bar
P2 Compressor outlet press. [0; +40] ± 0.1bar
P3 Condenser press. [0; +40] ± 0.1bar
P4-5 Evaporator pressure [0; +10] ± 0.025bar
(2-Way valve)
Tc_in Air condenser inlet temp. [0; 101.5] ± 0.3°C
Tc_out Air condenser outlet temp. [0; 101.5] ± 0.3°C
Ic Compressor current input [0; +10]A
Ir Thermal load current input [0; +10]A

279

Retrofitting of R404a Commercial Refrigeration Systems

Figure 13. Automatic real-time monitoring and control tool

internal temperature through the regulation of a series of resistors placed in the condenser inlet duct and
inside the thermostatic chamber (see Figure 14). The external air temperature ranges from the current
ambient temperature to 50-55°C, matching with all the possible ambient conditions.

Test Procedure

The cooling capacity, Qf, and the COP are the most frequently adopted parameters to measure the ef-
ficiency and the range of application of a refrigeration system. The former represents the ability of a
cooling system to remove heat from a low temperature ambient, while the latter is defined as the ratio

Figure 14. Process diagram of the adopted refrigeration cycle and sensor positions

280

Retrofitting of R404a Commercial Refrigeration Systems

between the cooling capacity and the supplied electric power. Both these parameters are significantly
affected by the temperature conditions inside and outside the thermostatic chamber. In the experimental
tests, Qf is calculated in stationary conditions, i.e. maintaining a constant temperature inside the ther-
mostatic chamber through the aforementioned resistors, as shown in the schematic of Figure 15. The
refrigerant circulating through the evaporator absorbs the heat that enters through the walls and the heat
generated by the resistors placed inside the chamber. The resistors are feedback regulated to keep the
desired constant temperature setpoint.
The following Equation 12 describes such heat balance:

k ⋅ S ⋅ (Text −Tchamb )
Q f = Qr + Qw = I r ⋅V + (12)
s

where Qr is the thermal load coming from the resistors and Qw is the power dissipated through the cham-
ber walls. Qr is calculated as the product between the resistor absorbed current, Ir, and the grid voltage,
V, while Qw depends on the thermal conductivity of expanded polyurethane, k, the thickness of the wall
insulation, s, the thermostatic chamber exchange surface, S, and the gap between the outdoor and indoor
temperatures, Text −Tchamb .
According to the technical data sheet, the thermostatic chamber features are the following:

Figure 15. Heat exchanges in the thermostatic chamber during the experimental tests

281

Retrofitting of R404a Commercial Refrigeration Systems

Thermal conductivity of expanded polyurethane k = 0.0221 W/mK


Thickness of the expanded polyurethane insulation s = 0.1 m
Thermostatic chamber exchange surface S = 22.32 m2
k ⋅S
As a consequence, the factor is equal to 4.93 W/°C. However, during the experimental study
s
setup, the thermostatic chamber features are experimentally validated in order to take into account heat
k ⋅S
dispersions through the glass window and the manufacturing peculiarity. The factor is field evalu-
s
ated at a constant chamber temperature of 50°C (See Figure 16).
In stationary conditions, the heat dispersions through the walls, Qw, equal the thermal load, Qr, inside
k ⋅S
the chamber so that is from Equations 13 and 14.
s
k ⋅ S ⋅ (Tchamb −Text )
Qr = Qw = with Tchamb > Text (13)
s

k ⋅S
Figure 16. Heat exchange in the thermostatic chamber during evaluation
s

282

Retrofitting of R404a Commercial Refrigeration Systems

k ⋅S Qr I r ⋅V
= = (14)
s (Tchamb −Text ) (50 −Text )

k ⋅S
The analysis shows = 8.2 W/°C, 60% higher than the theoretical values previously calculated.
s
Finally, the COP is evaluated as the ratio between the cooling capacity and the electric power ab-
sorbed by the compressor and the auxiliaries, calculated by the product between the measured current,
Ic, and the grid voltage, V. Formally:

Qf
COP = (15)
Pc

Pc = I c ⋅V (16)

Multiscenario Analysis

Several parameters affect the behaviour of a refrigeration system. Only a multi-scenario analysis allows a
comprehensive assessment. The performances of the refrigeration systems, coupled with the three work-
ing fluids under different operating conditions, are investigated. The condenser inlet air temperature and
the chamber temperature are two of the free variables set to simulate typical system working conditions.
The condenser inlet air temperature and the chamber temperature affect the condenser and the evaporator
temperature. Such values have a considerable effect on the efficiency and the thermodynamic conditions
of a refrigeration system (Reddy, Panwar, & Kaushik, 2012).
The condenser inlet air temperature, Tc_in, ranges between 25°C and 45°C, step 10°C, while the chamber
temperature Tchamb, are -5, 0, 5°C for the MT system and -25, -20, -10°C for the LT system. Including the
three different fluids, 54 scenarios occur (see Table 7) and 30 of them are investigated.
For the sake of brevity and without loss of relevant information the 35°C and 45°C Tc_in scenarios are
tested with Tchamb = 0ºC and Tchamb = -20ºC only, for the MT and LT systems, respectively.
The next subsection discusses the outcomes coming from the experimental analysis to draw general
conclusions about the fit of the different fluids with both the systems and the working operative conditions.

Table 7. Investigated scenarios during the experimental study

System MT LT
Refrigerant R404a R407f R410a R404a R407f R410a
Tchamb [°C] -5 0 5 -5 0 5 -5 0 5 -25 -20 -15 -25 -20 -15 -25 -20 -15
Tc_in [°C] 25 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
35 ✓ ✓ ✓ ✓ ✓ ✓
45 ✓ ✓ ✓ ✓ ✓ ✓

283

Retrofitting of R404a Commercial Refrigeration Systems

Results and Discussions

Figure 17 compares the three working fluids for the MT refrigeration system. The condenser inlet air
temperature is set to 25°C. Both Qf and the COP are shown.
R404a fluid allows a lower cooling capacity and COP respect to both R407f and R410a fluids for all
such scenarios. In addition, R407f and R410a fluids present opposite behaviours, the former provides
a higher cooling capacity, the latter presents a higher efficiency (COP). Particularly, for Tchamb = 0 °C,
COP of R404a, R407f and R410 are, respectively, 1.61, 1.73, 1.98, while the cooling capacities are
1395, 1520 and 1422. Such results are confirmed by varying the condenser inlet air temperature and
therefore the condensation temperatures. Figure 18 shows the cooling capacity and the COP when the
temperature inside the chamber is set to 0°C.
Refrigerant R407f is able to generate higher cooling capacity, between 1215W and 1520W, while
R410a has a higher COP, between 1.2 and 2.0, for all scenarios. Finally, for the higher condenser inlet
air temperature (45°C), the performances of the refrigerant R404a overcome the R407f and R410a fluids
for the generated cooling capacity and COP, respectively. A further outcome of interest refers to the
comparison of HFC refrigerants when coupled with the LT system. Concerning the R410a fluid, the low
chamber setpoint causes a significant increase of the compressor outlet temperature, fully compromising

Figure 17. MT cooling capacity and COP, Tchamb = -5, 0, 5°C, Tc_in = 25°C

284

Retrofitting of R404a Commercial Refrigeration Systems

Figure 18. MT cooling capacity and COP, Tchamb = 0°C, Tc_in = 25, 35, 45°C

the compressor operations. As a consequence, R410a is not suitable for R404a replacement in the LT
system. Concerning the other fluids, Figure 19 shows the cooling capacity and COP of such refrigerants
with Tchamb of -15°C, -20°C, -25°C and Tc_in of 25°C.
The experimental tests demonstrate comparable performances of R407f and R404a fluids. The
cooling capacity ranges between 1100W for Tchamb of -25°C to 1400W for Tchamb of -15°C, while the cor-
respondent COP is between 1 and 1.2. Finally, R404a and R407f performances are investigated vary-
ing the condenser inlet air temperature. At Tc_in of 35°C and 45°C relevant performance reductions are
experienced (see Figure 20). R407f has about 10% average higher cooling capacity than R404a, while
the COP values are comparable.
A summary framework of the cooling capacity and COP is in the Table 8 and Table 9.
During the experimental analysis temperatures and pressures in the circuit key points are monitored
to control the behaviour of different refrigerants in the refrigeration cycle. For the sake of brevity, the
values of the pressures and the temperatures for MT and LT systems during the tests with a condenser
inlet temperature of 25°C are shown in Table 10 and Table 11, only.
R407f shows higher temperature at the outlet of compressor, from 104°C to 112°C, than R404a and
R410a, which show similar values from 90°C to 98°C. Furthermore, R410a has higher evaporation and

285

Retrofitting of R404a Commercial Refrigeration Systems

Figure 19. LT cooling capacity and COP, Tchamb = -25, -20, -15°C, Tc_in = 25°C

condensation pressures than the other two alternatives. Considering 0°C as the thermostatic chamber
temperature setpoint, the R410a evaporating pressure is in the range [5; 6]bar, while the evaporating
pressure range for R407f and R404a is [3; 3.5]bar.
Considering LT system, R407f shows higher temperature after vapor compression than R404a as for
MT. The evaporation pressures are comparable, while R407f has a slightly higher condensing pressure
than R404a.

FUTURE RESEARCH DIRECTIONS

The experimental analysis discussed in this Chapter demonstrates the effectiveness of the replacement of
the R404a refrigerant with R407f and R410a fluid for the two commercial refrigeration systems. R407f
and R410a have lower GWP than R404a with comparable or even better efficiencies. As a consequence,
the direct GHG emissions, due to the refrigerant GWP, decrease without the increase of GHG indirect
emissions, due to the energy consumption. However, an exhaustive and complete evaluation of the envi-
ronmental impact of different refrigerants in cooling systems suggests adopting indices able to estimate
the amount of GHG emissions due to the manufacturing and the end-life processes. Future researches

286

Retrofitting of R404a Commercial Refrigeration Systems

Figure 20. LT cooling capacity and COP, Tchamb = -20°C, Tc_in = 25, 35, 45°C

deal with the calculation of GHG emissions of the two commercial refrigeration systems with the most
suitable of the previously described indices, i.e. life cycle global warming Impact, life cycle climate
performance and carbon footprint.

CONCLUSION

The growing concern on the climate change issues and the high global warming potential (GWP) fluid
on-going dismissal program suggest two strategies for the refrigeration sector. The development of new
equipment using natural refrigerants, hydrocarbons, new generation fluids and the retrofitting of exist-
ing systems with lower GWP refrigerants. This chapter follows such a second strategy and presents an
experimental analysis about the retrofitting of two commercial stationary refrigeration systems. Both
systems are originally designed to work with the hydrofluorocarbon (HFC) R404a fluid (GWP=3922).
The chosen fluid alternatives are the HFCs R410a (GWP=2088) and R407f (GWP=1825) refrigerants,
presenting good compatibility with the existing plants, non-flammability, availability in the market and

287
288
Table 8. Cooling capacity of the investigated scenarios

System MT LT
Refrigerant R404a R407f R410a R404a R407f
Thermostatic chamber -5 0 5 -5 0 5 -5 0 5 -25 -20 -15 -25 -20 -15
temperature
Tchamb [°C]
Condenser inlet 25 1279 1395 1433 1370 1520 1610 1313 1422 1531 1123 1145 1437 1211 1222 1465
air temperature
35 1282 1368 1326 924 1001
Tc_in [°C]
45 1098 1215 1030 668 866

Table 9. COP of the investigated scenarios

System MT LT
Refrigerant R404a R407f R410a R404a R407f
Thermostatic chamber temperature Tchamb [°C] -5 0 5 -5 0 5 -5 0 5 -25 -20 -15 -25 -20 -15
Condenser inlet air temperature 25 1.60 1.61 1.62 1.67 1.73 1.74 1.90 1.98 2.04 0.95 1.02 1.20 1.01 1.05 1.18
Tc_in [°C]
35 1.40 1.35 1.67 0.73 0.75
45 1.10 1.05 1.16 0.47 0.56
Retrofitting of R404a Commercial Refrigeration Systems


Retrofitting of R404a Commercial Refrigeration Systems

Table 10. MT 25°C temperature and pressure values

System MT
Refrigerant R404a R407f R410a
Thermostatic chamber -5 0 5 -5 0 5 -5 0 5
temperature
Tchamb [°C]
T1 -1.3 4.2 9.6 -5.1 4.1 9.1 -12.2 -4.7 2.8
T2 92.1 95.3 97.5 104.1 108.2 111.4 90.8 93.5 97.7
T3 43.3 45.2 46.5 41.3 41.5 44.6 42.2 43.6 45.6
T4 -16.1 -13.3 -11.8 -25.5 -17.1 -14.1 -20.2 -17.4 -14.9
T5 -8.5 -2.0 3.2 -14.2 -4.1 3.1 -14.9 -11.8 -5.1
P1 2.7 3.1 3.2 2.8 3.1 3.5 4.7 5.2 5.8
P2 19.1 20.0 20.5 19.6 19.9 21.2 25.6 26.5 27.8
P3 18.8 19.8 20 19.4 19.8 21.0 24.9 25.9 27.2
P4 2.9 3.4 3.6 3.1 3.4 3.8 5.4 6.0 6.5

Table 11. LT 25°C temperature and pressure values

System LT
Refrigerant R404a R407f
Thermostatic chamber -25 -20 -15 -25 -20 -15
temperature
Tchamb [°C]
T1 -22.8 -20.2 -12.3 -20.2 -18.0 -16.0
T2 105.0 107.1 109.8 133.3 136.2 138.8
T3 42.3 43.6 45.0 40.3 43.7 45.7
T4 -32.4 -30.1 -21.4 -32.2 -28.5 -25.5
T5 -31.1 -27.2 -17.4 -31.9 -28.2 -25.3
P1 1.2 1.3 1.6 1.1 1.4 1.6
P2 19.9 20.6 21.7 20.4 22.1 23.1
P3 19.9 20.6 21.7 20.4 22.1 23.1
P4 1.2 1.4 1.6 1.1 1.4 1.6

match with the restrictions on HFCs imposed by the European Union (EU) in the next future. The ex-
perimental analysis compares the system performances, in terms of coefficient of performance (COP)
and cooling capacity, of the HFC R404a fluid (benchmark) and the two identified alternatives under
different operating conditions, i.e. varying the chamber and the condenser inlet air temperatures.
Results highlight that for the medium temperature (MT) system [-5; +10] °C the R407f and R410a
fluids allow to obtain higher COP and cooling capacity than R404a. Among them, R407f has higher
cooling capacity but presents a lower COP respect to R410a. The average performances among the
investigated scenarios are (1.47; 1297W), (1.51; 1417W), (1.75; 1324W) for R404a, R407f, R410a,

289

Retrofitting of R404a Commercial Refrigeration Systems

respectively. Such results indicate the R410a is a potential alternative for the MT system even if R407f
also represents a suitable solution. Concerning the low temperature (LT) system [-25÷-15] °C, R410a
is not usable because of the compressor overheating. On the contrary, R407f presents comparable COP
values respect to R404a, while it increases of about 10% the cooling capacity.
Concluding, the presented experimental study shows the feasibility of the replacement of R404a
with the R410a and R407a refrigerants in two commercial refrigeration systems without or with limited
component replacements. The adoption of the identified alternatives in different commercial refrigeration
systems working with the R404a fluid in the medium and low temperature ranges is potentially feasible
but it requires further analysis and experimental investigations.

ACKNOWLEDGMENT

The authors are grateful to Zanotti S.p.A. (Italy) and “Fondazione Organismo di Ricerca - GTechnology”
foundation for the technical and financial support to the research project.

REFERENCES

Aprea, C., Maiorino, A., & Mastrullo, R. (2011). Change in energy performance as a result of a R422D
retrofit: An experimental analysis for a vapor compression refrigeration plant for a walk-in cooler. Ap-
plied Energy, 88(12), 4742–4748. doi:10.1016/j.apenergy.2011.06.049
Benhadid-Dib, S., & Benzaoui, A. (2012). Refrigerants and their Environmental Impact Substitution of
Hydro Chlorofluorocarbon HCFC and HFC Hydro Fluorocarbon. Search for an Adequate Refrigerant.
Energy Procedia, 18, 807–816. doi:10.1016/j.egypro.2012.05.096
Bolaji, B. O. (2011). Performance investigation of ozone-friendly R404A and R507 refrigerants as al-
ternatives to R22 in a window air-conditioner. Energy and Building, 43(11), 3139–3143. doi:10.1016/j.
enbuild.2011.08.011
Bolaji, B. O., & Huan, Z. (2013). Ozone depletion and global warming: Case for the use of natural refrig-
erant – a review. Renewable & Sustainable Energy Reviews, 18, 49–54. doi:10.1016/j.rser.2012.10.008
Bovea, M. D., Cabello, R., & Querol, D. (2007). Comparative Life Cycle Assessment of Commonly
Used Refrigerants in Commercial Refrigeration Systems. Green House Gas Emissions, 12(5), 299–307.
Calm, J. M. (2008). The next generation of refrigerants – Historical review, considerations, and outlook.
International Journal of Refrigeration, 31(7), 1123–1133. doi:10.1016/j.ijrefrig.2008.01.013
Chen, W. (2008). A comparative study on the performance and environmental characteristics of R410A
and R22 residential air conditioners. Applied Thermal Engineering, 28(1), 1–7. doi:10.1016/j.applthe-
rmaleng.2007.07.018
Davies, T. W., & Caretta, O. (2004). A low carbon, low TEWI refrigeration system design. Applied
Thermal Engineering, 24(8-9), 1119–1128. doi:10.1016/j.applthermaleng.2003.12.026

290

Retrofitting of R404a Commercial Refrigeration Systems

Dieckmann, J., & Little, A. D. (1999). Global Comparative Analysis of HFC and Alternative Technologies
for Refrigeration, Air Conditioning, Foam, Solvent, Aerosol Propellant, and Fire Protection Applications.
Retrieved from https://unfccc.int/2860.php
Duvedi, A., & Achenie, L. E. K. (1997). On the design of environmentally benign refrigerant mixtures:
A mathematical programming approach. Computers & Chemical Engineering, 21(8), 915–923. http://
linkinghub.elsevier.com/retrieve/pii/S0098135496003109 doi:10.1016/S0098-1354(96)00310-9
European Commission. (2006a). Directive 2006/40/EC of the European Parliament and of the Council
of 17 May 2006 relating to emissions from air conditioning systems in motor vehicles and amending
Council Directive 70/156/EEC. Retrieved from http://ec.europa.eu/index_en.htm
European Commission. (2006b). Regulation (EC) 842/2006 of the European Parliament and of the Coun-
cil of 17/05/2006 on certain fluorinated gases, (166). Retrieved from http://ec.europa.eu/index_en.htm
European Commission. (2009). Directive 2009/29/EC of the European Parliament and of the council of
23 April 2009 amending Directive 2003/87/EC so as to improve and extend the greenhouse gas emission
allowance trading scheme of the Community. Retrieved from http://ec.europa.eu/index_en.htm
European Parliament and Council. (2014a). European Parliament legislative resolution of 12 March
2014 on the proposal for a regulation of the European Parliament and of the Council on fluorinated
greenhouse (Vol. 0305, pp. 1–60). Retrieved from http://ec.europa.eu/index_en.htm
European Parliament and Council. (2014b). Regulation (EU) No 517/2014 of the European Parliament
and of the council of 16 April 2014 on fluorinated greenhouse gases and repealing Regulation (EC) No
842/2006. Official Journal of the European Union, 195–230.
Han, X., Qiu, Y., Xu, Y., Zhao, M., Wang, Q., & Chen, G. (2012). Cycle performance studies on a new
HFC-161/125/143a mixture as an alternative refrigerant to R404A. Journal of Zhejiang University SCI-
ENCE A, 13(2), 132–139. doi:10.1631/jzus.A1100154
Hwang, Y., Jin, D.-H., & Radermacher, R. (2007). Comparison of R-290 and two HFC blends for
walk-in refrigeration systems. International Journal of Refrigeration, 30(4), 633–641. doi:10.1016/j.
ijrefrig.2006.10.007
International Panel on Climate Change (IPCC). (2007). European Parliament legislative resolution
of 12 March 2014 on the proposal for a regulation of the European Parliament and of the Council on
fluorinated greenhouse gases. Retrieved from https://www.ipcc.ch/
International Panel on Climate Change (IPCC). (2014). Summary for policymakers of the 5th Assessment
Report (AR5) (pp. 1–33). Retrieved from https://www.ipcc.ch/
Jabaraj, D. B., Narendran, A., Lal, D. M., & Renganarayanan, S. (2007). Evolving an optimal composi-
tion of HFC407C/HC290/HC600a mixture as an alternative to HCFC22 in window air conditioners.
International Journal of Thermal Sciences, 46(3), 276–283. doi:10.1016/j.ijthermalsci.2006.05.004
Jarall, S. (2012). Study of refrigeration system with HFO-1234yf as a working fluid. International
Journal of Refrigeration, 35(6), 1668–1677. doi:10.1016/j.ijrefrig.2012.03.007

291

Retrofitting of R404a Commercial Refrigeration Systems

Jerald, L., & Senthilkumaran, D. (2014). Investigations on the Performance of Vapour Compression
System Retrofitted With Zeotropic Refrigerant R404a. American Journal of Environmental Sciences,
10(1), 35–43. doi:10.3844/ajessp.2014.35.43
La Rocca, V., & Panno, G. (2011). Experimental performance evaluation of a vapour compression re-
frigerating plant when replacing R22 with alternative refrigerants. Applied Energy, 88(8), 2809–2815.
doi:10.1016/j.apenergy.2011.01.051
Llopis, R., Torrella, E., Cabello, R., & Sánchez, D. (2012). HCFC-22 replacement with drop-in and
retrofit HFC refrigerants in a two-stage refrigeration plant for low temperature. International Journal
of Refrigeration, 35(4), 810–816. doi:10.1016/j.ijrefrig.2012.01.001
Mastrullo, R., Mauro, A. W., Menna, L., & Vanoli, G. P. (2014). Replacement of R404A with propane in
a light commercial vertical freezer: A parametric study of performances for different system architectures.
Energy Conversion and Management, 82, 54–60. doi:10.1016/j.enconman.2014.02.069
McLinden, M. O., Kazakov, A. F., Steven Brown, J., & Domanski, P. (2014). A thermodynamic analysis
of refrigerants: Possibilities and tradeoffs for Low-GWP refrigerants. International Journal of Refrigera-
tion, 38, 80–92. doi:10.1016/j.ijrefrig.2013.09.032
Minor, B. (2012). Low GWP R-404A Alternatives for Commercial Refrigeration. ASHRAE Transac-
tions, 666–672.
Mohanraj, M., Jayaraj, S., & Muraleedharan, C. (2009). Environment friendly alternatives to halogenated
refrigerants—A review. International Journal of Greenhouse Gas Control, 3(1), 108–119. doi:10.1016/j.
ijggc.2008.07.003
Mohanraj, M., Jayaraj, S., Muraleedharan, C., & Chandrasekar, P. (2009). Experimental investigation
of R290/R600a mixture as an alternative to R134a in a domestic refrigerator. International Journal of
Thermal Sciences, 48(5), 1036–1042. doi:10.1016/j.ijthermalsci.2008.08.001
Navarro-Esbrí, J., Mendoza-Miranda, J. M., Mota-Babiloni, A., Barragán-Cervera, A., & Belman-Flores,
J. M. (2013). Experimental analysis of R1234yf as a drop-in replacement for R134a in a vapor compres-
sion system. International Journal of Refrigeration, 36(3), 870–880. doi:10.1016/j.ijrefrig.2012.12.014
Palm, B. (2008). Hydrocarbons as refrigerants in small heat pump and refrigeration systems – A review.
International Journal of Refrigeration, 31(4), 552–563. doi:10.1016/j.ijrefrig.2007.11.016
Papasavva, S., & Moomaw, W. (1997). Life-Cycle Global Warming Impact of CFCs and for Refrigera-
tion. Refrigeration. J. Ind. Ecol, 1(4).
Rasti, M., Hatamipour, M. S., Aghamiri, S. F., & Tavakoli, M. (2012). Enhancement of domestic refrigera-
tor’s energy efficiency index using a hydrocarbon mixture refrigerant. Measurement, 45(7), 1807–1813.
doi:10.1016/j.measurement.2012.04.002
Sekhar, S. J., & Lal, D. M. (2005). HFC134a/HC600a/HC290 mixture a retrofit for CFC12 systems.
International Journal of Refrigeration, 28(5), 735–743. doi:10.1016/j.ijrefrig.2004.12.005

292

Retrofitting of R404a Commercial Refrigeration Systems

Sekhar, S. J., Lal, D. M., & Renganarayanan, S. (2004). Improved energy efficiency for CFC domestic
refrigerators retrofitted with ozone-friendly HFC134a/HC refrigerant mixture. International Journal of
Thermal Sciences, 43(3), 307–314. doi:10.1016/j.ijthermalsci.2003.08.002
Siva Reddy, V., Panwar, N. L., & Kaushik, S. C. (2012). Exergetic analysis of a vapour compression
refrigeration system with R134a, R143a, R152a, R404A, R407C, R410A, R502 and R507A. Clean
Technologies and Environmental Policy, 14(1), 47–53. doi:10.1007/s10098-011-0374-0
Spatz, M. W., & Yana Motta, S. F. (2004). An evaluation of options for replacing HCFC-22 in medium
temperature refrigeration systems. International Journal of Refrigeration, 27(5), 475–483. doi:10.1016/j.
ijrefrig.2004.02.009
United Nations. (UN). (1987). Protocol on Substances that Deplete the Ozone Layer (Vol. 1522). Re-
trieved from http://montreal-protocol.org/new_site/en/index-paris.php
United Nations. (UN). (1997). Kyoto Protocol to the United Nations Framework Convention on Climate
Change. Retrieved from http://www.un.org/en/
United Nations. (UN). (2000). The Montreal Protocol on Substances that Deplete the Ozone Layer.
Retrieved from http://montreal-protocol.org/new_site/en/index-paris.php
Wang, S. K., & Lavan, Z. (1999). “Air-Conditioning and Refrigeration” Mechanical Engineering Hand-
book. (CRC Press LCC, Ed.) (Ed. Frank.).
Wiedmann, T., & Minx, J. (2007). A definition of “carbon footprint.” ISAUK Research Report 07-01.
Retrieved from http://info.ornl.gov/sites/rams09/k_fernandez/Documents/carbon_footprinting.pdf
Wongwises, S., Kamboon, A., & Orachon, B. (2006). Experimental investigation of hydrocarbon mixtures
to replace HFC-134a in an automotive air conditioning system. Energy Conversion and Management,
47(11-12), 1644–1659. doi:10.1016/j.enconman.2005.04.013
Wu, X., Hu, S., & Mo, S. (2013). Carbon footprint model for evaluating the global warming impact
of food transport refrigeration systems. Journal of Cleaner Production, 54, 115–124. doi:10.1016/j.
jclepro.2013.04.045
Xu, X., Hwang, Y., & Radermacher, R. (2013). Performance comparison of R410A and R32 in vapor
injection cycles. International Journal of Refrigeration, 36(3), 892–903. doi:10.1016/j.ijrefrig.2012.12.010
Xuan, Y., & Chen, G. (2005). Experimental study on HFC-161 mixture as an alternative refrigerant to
R502. International Journal of Refrigeration, 28(3), 436–441. doi:10.1016/j.ijrefrig.2004.04.003
Yang, Z., & Wu