Practical Application of Supercritical

Fluid Chromatography for

Pharmaceutical Research and
Development Hicks M.
Visit to download the full and correct content document:
https://ebookmass.com/product/practical-application-of-supercritical-fluid-chromatogra
phy-for-pharmaceutical-research-and-development-hicks-m/
SEPARATION SCIENCE AND TECHNOLOGY
VOLUME 14

A reference series edited by

SATINDER AHUJA
 PRACTICAL APPLICATION OF
SUPERCRITICAL FLUID
CHROMATOGRAPHY FOR
PHARMACEUTICAL RESEARCH
AND DEVELOPMENT

VOLUME 14
Edited by

MICHAEL HICKS
Analytical Research & Development, Merck Research Labs, Rahway, NJ, United States

PAUL FERGUSON
New Modalities & Parenteral Development, Pharmaceutical Technology & Development, Operations,
AstraZeneca, Macclesfield, United Kingdom
Academic Press is an imprint of Elsevier
125 London Wall, London EC2Y 5AS, United Kingdom
525 B Street, Suite 1650, San Diego, CA 92101, United States
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom

Copyright © 2022 Elsevier Inc. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or
mechanical, including photocopying, recording, or any information storage and retrieval system, without
permission in writing from the publisher. Details on how to seek permission, further information about the
Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance
Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher (other
than as may be noted herein).

Notices
Knowledge and best practice in this field are constantly changing. As new research and experience broaden our
understanding, changes in research methods, professional practices, or medical treatment may become necessary.

Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using
any information, methods, compounds, or experiments described herein. In using such information or methods
they should be mindful of their own safety and the safety of others, including parties for whom they have a
professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability
for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from
any use or operation of any methods, products, instructions, or ideas contained in the material herein.
ISBN: 978-0-323-88487-7
ISSN: 1877-1718

For information on all Academic Press publications

visit our website at https://www.elsevier.com/books-and-journals

Publisher: Susan Dennis

Acquisitions Editor: Charlotte Rowley
Editorial Project Manager: Zsereena Rose Mampusti
Production Project Manager: Sruthi Satheesh
Cover Designer: Vicky Pearson Esser
Typeset by STRAIVE, India
 Dedication

For the two constants in my life, Isobel and Emily, without who’s patience and support this book
would not have been possible
Paul Ferguson, November 2022

For my family, Thomas, Brendan, Georgina, and Kathy, for inspiring me and for my colleagues for
sharing this journey with me
Michael Hicks, November 2022
 Contents

Contributors xi Other operating parameters: Temperature, pressure

and flow rate 64
Preface xiii Predicting retention and selectivity 65
Summary 66
1. Evolution of packed column SFC References 66
as a greener analytical tool for
pharmaceutical analysis 4. Measurements of drugs and metabolites
Susan Olesik and Raffeal Bennett in biological matrices using SFC
and SFE-SFC-MS
Discovery of supercritical fluids and its relevance to Brady W. Drennan, A. Paige Wicker, Blair K. Berger,
analytical separation science 1 and Kevin A. Schug
Sustainable aspects of subcritical and supercritical
chromatographic separation methods 9 Introduction to drug and metabolite analysis
Analytical scale subcritical SFC 10 in biological matrices using SFC and
Analytical enhanced-fluidity liquid SFE-SFC-MS 73
chromatography 12 On-line SFE-SFC-MS method development for
Preparative and analytical scale instrument biological matrices 75
improvements 13 Drug metabolism and pharmacokinetics (DMPK)
Current utility and recent advancements 20 monitoring in discovery and development 88
Future directions 22 Discovery and dereplication of natural products 93
References 22 Conclusions 94
References 94
2. Application space for SFC in 5. Synthetic chemistry screening for robust
pharmaceutical drug discovery
analysis and purification from discovery
and development
through to development
Gioacchino Luca Losacco, Amandine Dispas, Jean-Luc Veuthey,
and Davy Guillarme Tim Underwood, Sean Hindley, Andy Knaggs, and Craig White

Introduction 29 Introduction 101

Discussion 33 Screening strategies 103
Considerations on SFC as an analytical tool in drug Practicality 119
discovery and development 38 Aspects for consideration 123
Conclusions and perspectives 42 Future refinements 128
References 43 References 129

6. Application of preparative SFC in the

3. Selection of SFC stationary pharmaceutical industry
and mobile phases € en,
Jennifer Kingston, Hanna Leek, Astrid Buica, Kristina Ohl
Caroline West and Eric Lesellier Katie Proctor, Joanna Raubo, Matthew Sanders, and Linda Thunberg

Introduction 49 Introduction to the use of preparative SFC 133

Current stationary phase chemistries for SFC 49 Preparative SFC instrumentation and
Flexible mobile phase composition 59 infrastructure 134

vii
viii Contents

Features and controls during preparative SFC 138 Role of water in modern SFC 261
Temperature control 143 Enhanced-fluidity liquid chromatography 269
Detection systems 143 Applications of SFC to biomolecules 279
Fraction collection 144 Concluding remarks 290
Method development in preparative SFC 145 References 290
Preparative SFC applications: Case studies within
AstraZeneca research laboratories 148 10. Different detectors used with SFC
SFC as a sustainable chromatographic technique 159 G. John Langley, Sergio Cancho-Gonzalez,
Conclusions 163 and Julie M. Herniman
References 163
Introduction to detectors used with
7. Method development approaches modern SFC 299
for small-molecule analytes Generic detectors used with SFC 300
Coupling SFC to mass spectrometric
Syame Khater, Paul Ferguson,
and Alexandre Grand-Guillaume-Perrenoud detectors 313
Conclusions 321
Introduction 167 Acknowledgment 321
Method development “prework” 170 References 321
Screening tool utilization to identify optimal SFC
parameters 174 11. SFC in GMP testing and quality
Method validation 205 control of medicinal drug products
Continuous method performance verification 210
Adrian Clarke, Paul Ferguson, and Michael Hicks
Summary 211
Acknowledgments 212 Introduction 325
References 212 Current use of SFC in pharmaceutical
development 327
8. Application of SFC for the Transfer of methods to manufacturing QC
characterization of formulated facilities 332
drug products Future requirements toward regulatory acceptance of
Paul Ferguson, Rebecca Cross, and Gesa Schad SFC methods 344
Conclusions 348
Introduction 221 Acknowledgments 348
Drug formulations 222 References 348
Sample preparation procedures 225
Water and organic solvents as sample diluents in 12. Best practices and instrumental
SFC for APIs and solid-oral dosage forms 230 troubleshooting for successful
Alternative approaches to solubilize analytes in SFC
SFC methods
compatible solvents 237
Fiona Bell, Paul Ferguson, Rebecca Poulten, Emily Roddy,
Characterization of polymer excipients 241
and William Farrell
Conclusions 246
Acknowledgments 247 Introduction 353
Appendix: Constituents of formulated drugs discussed System configuration 354
in chapter 247 Best practice for system setup 356
References 252 System performance checks 357
Cylinder issues 361
9. Expanding the boundaries of SFC: Instrument troubleshooting and errors 364
Analysis of biomolecules Eluent effects 369
Martin Beres Chromatographic troubleshooting 371
Optimization of detector sensitivity 374
Historical problems analyzing polar molecules via Conclusions 374
SFC 257 References 375
 Contents ix
13. The state-of-the-art and future Method development, performance and prediction
perspectives for SFC aspects 386
Method scaling 390
Paul Ferguson and Michael Hicks
Future directions and applications of SFC 392
Introduction 377 Sustainable instrument design 394
Reflection on previous chapters 378 Conclusions 397
Theoretical performance of SFC 382 Acknowledgments 398
Influence of new column particle technologies and References 398
instrument design on SFC performance 383
Considerations for future instrument application and List of abbreviations 403
design 384 Index 409
 Contributors
Fiona Bell Global Chemical Development,
Pharmaceutical Technology & Development,
Operations, AstraZeneca, Macclesfield,
United Kingdom
Raffeal Bennett Merck & Co., Inc., MRL, Ana-
lytical Research & Development, Boston, MA,
United States
Martin Beres Separation & Analysis Technology
Team, Bristol Myers Squibb, Princeton, NJ,
United States
Blair K. Berger Department of Chemistry & Bio-
chemistry, The University of Texas at Arling-
ton, Arlington, TX, United States
Astrid Buica Early Chemical Development,
Pharmaceutical Sciences, BioPharmaceuticals
Research & Development, AstraZeneca,
Gothenburg, Sweden
Sergio Cancho-Gonzalez Faculty of Engineer-
ing & Physical Sciences, School of Chemistry,
University of Southampton, Southampton,
United Kingdom
Adrian Clarke Chemical & Analytical Develop-
ment, Novartis Pharma AG, Technical R&D,
Basel, Switzerland
Rebecca Cross Global Chemical Development,
Pharmaceutical Technology & Development,
Operations, AstraZeneca, Macclesfield,
United Kingdom
Amandine Dispas University of Liege (ULiege),
CIRM, Vibra-Sant
e Hub, Laboratory of Pharma-
ceutical Analytical Chemistry; University of
Liege (ULiege), CIRM, Mas-Sant
e Hub, Labora-
tory for the Analysis of Medicines, Liege,
Belgium
Brady W. Drennan Department of Chemistry &
Biochemistry, The University of Texas at
Arlington, Arlington, TX, United States
xi
William Farrell Oncology Medicinal Chemistry,
Pfizer, Inc., Worldwide Medicinal Chemistry,
La Jolla Laboratories, San Diego, CA,
United States
Paul Ferguson New Modalities & Parenteral
Development, Pharmaceutical Technology &
Development, Operations, AstraZeneca, Mac-
clesfield, United Kingdom
Alexandre Grand-Guillaume-Perrenoud Chem-
ical & Analytical Development, Novartis
Pharma AG, Technical R&D, Basel, Switzerland
Davy Guillarme School of Pharmaceutical Sci-
ences; Institute of Pharmaceutical Sciences of
Western Switzerland, University of Geneva,
CMU, Geneva, Switzerland
Julie M. Herniman Faculty of Engineering &
Physical Sciences, School of Chemistry,
University of Southampton, Southampton,
United Kingdom
Michael Hicks Analytical Research & Develop-
ment, Merck Research Labs, Rahway, NJ,
United States
Sean Hindley GlaxoSmithKline Medicine Re-
search Centre, Stevenage, Hertfordshire,
United Kingdom
Syame Khater Technologie Servier, Orl
eans,
France
Jennifer Kingston Oncology Chemistry, Oncol-
ogy Research & Early Development,
AstraZeneca, Cambridge, United Kingdom
Andy Knaggs GlaxoSmithKline Medicine Re-
search Centre, Stevenage, Hertfordshire,
United Kingdom
G. John Langley Faculty of Engineering &
Physical Sciences, School of Chemistry,
University of Southampton, Southampton,
United Kingdom
xii
Hanna Leek Early Chemical Development,
Pharmaceutical Sciences, BioPharmaceuticals
Research & Development, AstraZeneca,
Gothenburg, Sweden
Eric Lesellier University of Orleans, CNRS,
Institute of Organic & Analytical Chemistry
(ICOA), UMR, Orl
eans, France
Gioacchino Luca Losacco School of Pharmaceu-
tical Sciences; Institute of Pharmaceutical Sci-
ences of Western Switzerland, University of
Geneva, CMU, Geneva, Switzerland; Analytical
Research and Development, MRL, Merck & Co,
Inc., Rahway, NJ, United States
€ en Early Chemical Development,
Kristina Ohl

Pharmaceutical Sciences, BioPharmaceuticals
Research & Development, AstraZeneca,
Gothenburg, Sweden
Susan Olesik Department of Chemistry, The
Ohio State University, Columbus, OH,
United States
Rebecca Poulten Early Chemical Development,
Pharmaceutical Sciences, Research & CMU, Geneva, Switzerland
Development, AstraZeneca, Macclesfield,
United Kingdom
Katie Proctor Oncology Chemistry, Oncology
Research & Early Development, AstraZeneca,
Cambridge, United Kingdom
Joanna Raubo Oncology Chemistry, Oncology
Research & Early Development, AstraZeneca,
Cambridge, United Kingdom
Contributors
Emily Roddy Early Chemical Development,
Pharmaceutical Sciences, Research &
Development, AstraZeneca, Macclesfield,
United Kingdom
Matthew Sanders Oncology Chemistry, Oncol-
ogy Research & Early Development,
AstraZeneca, Cambridge, United Kingdom
Gesa Schad Shimadzu Europa GmbH,
Duisburg, Germany
Kevin A. Schug Department of Chemistry &
Biochemistry, The University of Texas at
Arlington, Arlington, TX, United States
Linda Thunberg Early Chemical Development,
Pharmaceutical Sciences, BioPharmaceuticals
Research & Development, AstraZeneca,
Gothenburg, Sweden
Tim Underwood GlaxoSmithKline Medicine
Research Centre, Stevenage, Hertfordshire,
United Kingdom
Jean-Luc Veuthey School of Pharmaceutical Sci-
ences; Institute of Pharmaceutical Sciences of
Western Switzerland, University of Geneva,
Caroline West University of Orleans, CNRS,
Institute of Organic & Analytical Chemistry
(ICOA), UMR, Orl
eans, France
Craig White Exscientia AI Ltd, Abingdon,
Oxfordshire, United Kingdom
A. Paige Wicker Department of Chemistry &
Biochemistry, The University of Texas at
Arlington, Arlington, TX, United States
 Preface
Supercritical fluids
The term “supercritical” refers to the state
of a substance above a certain temperature
(termed its critical temperature, Tc) and pres-
sure (critical pressure, Pc). The state transitions
are often described through a phase diagram
plotting state boundaries (solid, liquid, gas,
supercritical regions) as a function of temper-
ature and pressure. In the supercritical re-
gion, the fluid exhibits properties that are
intermediate between that of a liquid and a
gas. In particular, supercritical fluids possess
liquid-like densities, gas-like viscosities, and
diffusivities intermediate to that of a liquid
and a gas.
A modified phase diagram for carbon
dioxide (CO2) is shown in Fig. P1. Above
modest values of 31°C and 74 bar (7.3 MPa),
the fluid exists in a supercritical state. Near
the critical point, small changes in pressure
or temperature can result in large changes
in density. This fact can be conveniently
exploited in SFC, to modify properties of
supercritical fluids to elute analytes from a
chromatographic column by modification
of the temperature, pressure, or both. The
properties like density, viscosity, and diffu-
sivity of supercritical fluids are intermediate
between the gas and a liquid. As there is no
distinct gas or liquid phase and supercritical
fluids have very low or no surface tension,
this allows the use of high flow rates for fast
separations, or longer columns for higher
efficiency/resolution separations and short
reequilibration times.
xiii
Although the physicochemical properties
of CO2 are important to its selection as a
supercritical fluid chromatography (SFC)
mobile phase, the elution of many analytes
will only occur with the inclusion of a
cosolvent modifier, typically methanol.
Although the phase diagram of CO2 is com-
monly shown (see Chapter 1, Fig. 1.1),
Fig. P1 shows the impact on critical point
with the addition of methanol as a modifier.
Even the addition of small volumes of meth-
anol can shift the properties away from su-
percritical conditions into a “subcritical”
region below the critical temperature and
pressure of the mixture. However, for all
practical chromatographic purposes, it is just
a matter of semantics if the CO2-solvent mod-
ifier is sub- or supercritical as it maintains its
monophasic behavior and advantageous
chromatographic properties, e.g., low viscos-
ity and high diffusivity.
Historical perspectives
Supercritical fluid behavior was first
reported in 1822 by a French engineer and
physicist Charles Cagniard de La Tour
through his “cannon barrel” experiments [1]
at the Academie des Sciences in Paris. He
listened to discontinuities in the sound of a
rolling flint ball in a sealed gun barrel filled
with a variety of solvents (water, ethanol,
diethyl ether, and petroleum spirit) at differ-
ent temperatures and he noted that the
splashing sound at the liquid–air interface
xiv Preface

Pressure
(MPa) 0% 5% 10% 20%
15

l)
/m
14

5g
Subcritical fluid Supercritical fluid

0 .7
13
r = 0.75-1.1 g/ml r = 0.2-0.9 g/ml

e(
lin
12 10.7% MeOH

ic
cn
74∞C

py
11 13 MPa

iso
8.8% MeOH
10 60∞C
11 MPa
9
3.5% MeOH
8 40∞C
8.3 MPa
7 Liquid
100% CO2
31∞C
6 7.3 MPa Gas

10 20 30 40 50 60 70
Temperature (°C)
FIG. P1 Phase diagram for carbon dioxide (blue area) or carbon dioxide/methanol mixtures (green area). The red
lines denote a constant density (pertinent to the elution characteristics of the system) at different percentages of meth-
anol in the mobile phase. Reproduced with permission of Elsevier from E. Lesellier, C. West, The many faces of packed column
supercritical fluid chromatography—a critical review, J. Chromatogr. A 1382 (2015) 2–46.

disappeared at temperatures well above the
liquid’s boiling points. He had unknowingly
discovered their transition points to a
supercritical phase. In a related experiment,
he heated a sealed glass vial of ethanol noting
it expanded to twice its original volume, and
the liquid appeared to vanish. On cooling, a
cloud of mist appeared that corresponded to
the solvent’s critical temperature [2]. In a sub-
sequent paper, he reported the critical temper-
ature of a series of liquids when the interface
tension vanished—visualized through the
disappearance of the liquid’s meniscus [3].
More than 45 years later, Andrews authored
the first systematic study of gas–liquid criti-
cal point of carbon dioxide (generated from
carbonic acid) and nitrous oxide [4]. The
first application of “supercriticality” was
reported in 1879 by Hannay and Hogarth
who described the solubilization of cobalt
chloride in supercritical ethanol [5]. These
fundamental experiments set in place the
cornerstones for further pivotal works in
the 20th century (see Chapter 1).
In 1957, James Lovelock, one of the pio-
neers of gas chromatography (GC), first pro-
posed the use of supercritical fluids for
chromatographic mobile phases for the anal-
ysis of nonvolatile compounds (“critical state
chromatography”—[6,7]). In 1962, Ernst
Klesper and coworkers presented the first
study on the use of an inorganic gas as a chro-
matographic mobile phase [8]. They used
chlorofluorocarbons (CHClF2 and CCl2F2)
above their critical points (pressures of
800–2300 psi and 115°C) with a 30-in. packed
column containing Carbowax 20 M (polyeth-
ylene glycol) on a 180–250 μm Chromosorb W
 Preface
diatomaceous earth support to separate
colored nickel–porphyrin complexes. He
termed this “high pressure gas chromato-
graphy.” The elution strength was found to
be proportional to the system pressure,
which was controlled through restriction
capillaries and mobile phase flow rate [6].
As the work was undertaken on modified
gas-chromatographic instrumentation (GC),
the technique was coined “high pressure
GC”. This is considered the first practical
demonstration of what was later to become
known as SFC.
The term SFC was first proposed by Sie
and coworkers at Shell Research Laboratories
(Amsterdam, The Netherlands) in a series of
papers in the late 1960s. In these experiments,
they investigated the effects of pressure and
temperature in open-tubular (squalene and
glycerol coated, and later packed column)
gas chromatography with carbon dioxide as
the carrier gas [9–11]. They described how in-
creasing the carbon dioxide pressure altered
the density of the gas and reduced analyte
retention. At a similar timepoint ca. 1968,
J. Calvin Giddings group proposed the term
“dense gas chromatography” [12] and soon
after reported the influence of high pressures
on analyte retention in GC with carbon diox-
ide and ammonia as mobile phase/carrier
gas [13].
As is often the case in early research, the late
1960s through to the mid-1980s saw the rise of
certain claims that ultimately hindered the
pace of development of the technique, notably
the assertion that the solvent strength of car-
bon dioxide was similar to isopropyl alcohol
[14], publications highlighting the deficiencies
of packed columns on chromatographic per-
formance [15], and the belief that modifiers
did not increase mobile phase solvent strength
[16]. Similarly, there were limitations in the
available instrumentation, which meant that
the full benefits of supercritical mobile phases
could not be exploited [17].
xv
All the aforementioned experiments,
among others, generated a curiosity that
paved the way for numerous advances in
the field. At the heart of these develop-
ments were improvements in understanding
the impact of organic solvent on the mobile
phase elution strength, the role of mobile
phase additives, and the significant role
stationary phase interactions play on analyte
retention and selectivity. Alongside these
theoretical developments, improvements in
instrument design saw the transition from
early modification of gas chromatographic
instruments, to modified HPLC instrumen-
tation and then the concerted development
of instrumentation dedicated to the task.
At each juncture, improved understanding
of CO2 fluid metering allowed for more
accurate delivery of volumes of the com-
pressible gas mobile phase and cosolvent
modifiers. These advances were supported
by developments in thermostated high-
pressure UV flow cells for improved detec-
tion sensitivity, improved pump check
valves materials, and advances in back
pressure system regulation.
SFC in the pharmaceutical industry
Having been available for many decades,
SFC is a chromatographic technique that
many analytical chemists are aware of, but
may have not yet used. It is often considered
a niche technique with only limited applica-
tion for preparative and chiral chromatogra-
phy. SFC has a perceived complexity over
more established techniques such as GC
and liquid chromatography (LC), and these
perceptions have prevented its application
as a standard lab technique. However, if an
analyst has experience in liquid chromatog-
raphy, then much of the knowledge and in-
strumental understanding is relevant and
transferable to SFC, and application to SFC
xvi
is not as high a barrier to overcome as one
might believe.
SFC has several interesting aspects as a
separation technique that are pertinent to
the analysis of both small and large phar-
maceutical molecules as well as related
materials such as certain formulation excipi-
ents. In many cases, SFC can provide a
better solution than LC for the characteriza-
tion of pharmaceutically relevant molecules.
Fig. P2 shows the number of publications
reporting the use of SFC for pharmaceutical
analysis year on year from 1985. There is a
clear exponential upward trend reflecting
the growing interest and importance of the
technique in the industry. In fact, application
of the technique is expanding so rapidly in
pharmaceutical science that we found it chal-
lenging to propose possible future applica-
tions areas (see Chapter 13).
Some of the aspects that make SFC attrac-
tive for pharmaceutical analysis include the
following:
• Typically faster analysis times and
gradient reequilibration than (U)HPLC
FIG. P2 CAS Scifinder search using keywords “SFC” + “pharmaceutical” for the period 1985–2021.
Preface
due to the low viscosity of the mobile
phase allowing high flow rates to be used;
• Effective with a wide range of organic
solvents commonly used in synthetic
chemistry reactions, and is particularly
useful for analysis of water labile
molecules;
• Virtually no stationary phase limitations
in terms of selection choice, using carbon
dioxide–based mobile phases (see
Chapter 3);
• Compatible with many sample diluents
ranging from fully organic to fully
aqueous samples (depending on mobile
phase composition, see Chapter 8);
• As a result of the low mobile phase
viscosity, highly efficient columns with
sub-2 μm stationary phases can be
successfully employed to achieve higher
efficiency and resolution than analogous
LC separations (see Chapter 7);
• Is becoming the industry technique-of-
choice for chiral and preparative separations
due to the technique using less solvent
and instrumental energy than the analogous
LC separations (see Chapters 5 and 6);
 Preface
• Produces less solvent waste than LC and
uses a “recyclable” gas as the primary
mobile phase component. In addition,
alcohols are often used as a cosolvent that
further improves its “green” credentials.
Scope and relevance for this text
These are exciting times for SFC. There has
been a proliferation in new understanding,
workflows, and instrumental developments
in recent years, and the pace is not slowing.
As analytical chemists working in the phar-
maceutical industry, we wanted to outline
the opportunities to utilize SFC throughout
the complete development cycle of a drug
molecule, from drug discovery through to
development and commercial release of the
formulated medicine, in a manufacturing en-
vironment. We believe this text fills a crucial
gap in the market, one that focuses on fully
utilizing the technique in a pharmaceutical
setting. We envisage this book will appeal
not only to analytical scientists in the phar-
maceutical industry but also to analytical sci-
entists in other industries. We also hope that
the book will serve as a useful reference for
academics offering in-depth and practical
insight and also provide a platform to help
teach the next generation of separation scien-
tists about this environmentally sustainable
technique.
In our final comments, we would like to
acknowledge the contributions of our col-
leagues and collaborators. In developing this
book, we wanted to partner with scientists
who have extensive theoretical knowledge
and practical experience with SFC. This text
was therefore compiled by practitioners of
the technique and includes understanding
gathered over many years (often decades)
of experience. We are indebted to them for
their contribution, insight, and dedication
in helping us compile this text. We hope
xvii
you enjoy the book and find it helpful in
expanding your knowledge of SFC.
Michael Hicks
Rahway, NJ, United States
Paul Ferguson
Macclesfield, United Kingdom
November 2022
References
[1] C. Cagniard de Latour, Expose de quelques
resultats obtenu par l’action combinee de la chaleur
et de la compression sur certains liquides, tels que
l’eau, l’alcool, l’ether sulfurique et l’essence de
petrole rectifiee, Ann. Chim. Phys. 21 (1822)
127–132.
[2] B. Berche, M. Henkel, R. Kenna, Critical phenom-
ena: 150 years since Cagniard de Latour, J. Phys.
Stud. 13 (2009) 3001–3005.
[3] C. Cagniard de Latour, Nouvelle note de
M. Cagniard de Latour, sur les effets qu’on obtient
par l’application simultanee de la chaleur et de la
compression à certains liquids, Ann. Chim. Phys.
22 (1823) 410–415.
[4] T. Andrews, On the continuity of the gaseous and
liquid states of matter, Philos. Trans. R. Soc. Lond.
159 (1869) 575–590.
[5] J.B. Hannay, J. Hogarth, VI. On the solubility of
solids in gases, Proc. R. Soc. Lond. 29 (1879)
324–326.
[6] T.A. Berger, The past, present, and future (?) of an-
alytical supercritical fluid chromatography—a 2018
perspective, Chromatogr. Today (2018) 4–8.
[7] L.T. Taylor, Past, current, and future direc-
tions in supercritical fluid chromatography, LCGC
31 (2013). https://www.chromatographyonline.
com/view/past-current-and-future-directions-
supercritical-fluid-chromatography. (Accessed
on April 3, 2022).
[8] E. Klesper, A.H. Corwin, D.A. Turner, High
pressure gas chromatography above critical
temperatures, J. Organomet. Chem. 27 (1962)
700–706.
[9] S.T. Sie, G.W.A. Rijnders, High-pressure gas chro-
matography and chromatography with supercriti-
cal fluids. III. Fluid-liquid chromatography, Sep.
Sci. 2 (1967) 729–753.
[10] S.T. Sie, G.W.A. Rijnders, High-pressure gas
chromatography and chromatography with super-
critical fluids. II. Permeability and efficiency of
packed columns with high-pressure gases as mobile
xviii
fluids under conditions of incipient turbulence, Sep.
Sci. 2 (1967) 699–727.
[11] S.T. Sie, G.W.A. Rijnders, High-pressure gas chro-
matography and chromatography with supercriti-
cal fluids. IV. Fluid-solid chromatography, Sep.
Sci. 2 (1967) 755–777.
[12] L. McLaren, M.N. Myers, J.C. Giddings, Dense-
gas chromatography of nonvolatile substances
of high molecular weight, Science 159 (1968)
197–199.
[13] J.C. Giddings, M.N. Myers, L.M. McLaren, R.A.
Keller, High pressure gas chromatography of non-
volatile species, Science 162 (1968) 67–73.
Preface
[14] J.C. Giddings, M.N. Myers, J.W. King, Dense gas
chromatography of pressures to 2000 atmospheres,
J. Chromatogr. Sci. 7 (1969) 276–283.
[15] M. Novotny, W. Bertsch, A. Zlatkis, Temperature
and pressure effects in supercritical-fluid chroma-
tography, J. Chromatogr. 61 (1971) 17–28.
[16] B.W. Wright, R.D. Smith, Investigation of polar
modifiers in carbon dioxide mobile phases for
capillary supercritical fluid chromatography,
J. Chromatogr. 355 (1986) 367–373.
[17] T.A. Berger, Separation of polar solutes by packed
column supercritical fluid chromatography,
J. Chromatogr. A 785 (1997) 3–33.
 C H A P T E R

1
Evolution of packed column SFC
as a greener analytical tool
for pharmaceutical analysis
Susan Olesika,∗ and Raffeal Bennettb
a
Department of Chemistry, The Ohio State University, Columbus, OH, United States bMerck &
Co., Inc., MRL, Analytical Research & Development, Boston, MA, United States
∗Corresponding author: E-mail: olesik@chemistry.ohio-state.edu

Discovery of supercritical fluids and its relevance to

analytical separation science
The International Union on Pure and Applied Chemistry [1] and the American Society of
Testing and Materials [2] both describe the region on a phase diagram (Fig. 1) above the crit-
ical temperature and pressure as a “supercritical fluid.” The critical point conditions (temper-
ature and pressure) are specific to each substance. At, or above, the critical point, the interface
between the gas and liquid phase disappears and a homogenous fluid results. Supercritical
fluids were discovered by Cagniard de Latour where he noted the disappearance of the me-
niscus at the critical point for methyl alcohol and ether [3]. Later, Hannay and Hogarth illus-
trated that ionic compounds, such as CoCl2, KBr, KI, and FeCl3, dissolve in supercritical
fluids, such as CS2 and methyl alcohol [4]. They stated the “the liquid condition of the fluids
has very little to do with their solvent power, but only indicates molecular closeness.”
Another important point described by Hannay and Hogarth is that moving from the fluid
condition above the critical pressure and temperature to a liquid (down or to the left on
the phase diagram relative to the supercritical region) does not cause a phase transition or
a meniscus to form, illustrating a continuum of homogenous conditions. The resultant high
fluidity conditions between a supercritical fluid and conventional liquid conditions are
increasingly used in SFC pharmaceutical analyses.

Separation Science and Technology, Volume 14 1 Copyright © 2022 Elsevier Inc. All rights reserved.
https://doi.org/10.1016/B978-0-323-88487-7.00006-1
2 1. Evolution of packed column SFC as a greener analytical tool

100

90
Compressible Supercritical
80 liquid Phase
31.1 °C 73.8 bar

70 Critical Point
Pressure (bar)

60

50

40 Solid Liquid Phase

Phase
30
Triple Point
20 P = 5.18 bar,
T = 56.6 °C
Sublima on Point Gaseous Phase
10 P = 1.0 bar,
T = -78.5 °C
0
-100 -75 -50 -25 0 25 50 75 100
Temperature (°C)
FIG. 1 Phase diagram for carbon dioxide.

The progression of analytical supercritical fluid chromatography

The physical attributes of sub-/supercritical fluid led to the first supercritical chromato-
graphic separation that involved the resolution of thermally labile porphyrins that could
not be separated by gas chromatography [5]. Chlorofluorocarbons were used as the mobile
phase with a diatomaceous earth packing material. Shortly thereafter, Giddings, illustrated
the possibility of dissolving, separating, and detecting compounds with molecular weights
as high as 400,000 amu, including carotenoids, cortisol steroids, sterols, nucleosides, amino
acids, carbohydrates, with either supercritical NH3 or CO2 as the mobile phase at tempera-
tures of 140°C and 40°C, respectively, which was above their critical temperatures, and
Porosil B, Chromosorb-W, or Ucon packing material in short 1–1.5 m long columns with a
flame ionization detector [6,7].
In the 1970s, instrumentation for high-performance liquid chromatography (HPLC)
became available and quickly resulted in a flurry of applications in many fields of science.
With the advancement of HPLC, further development of packed column supercritical fluid
chromatography (SFC) did not keep pace with HPLC until the 1980s. In the early 1980s, a
renewed interest in packed column SFC was shown. The first commercial SFC was developed
by Hewlett-Packard through the leadership of Dennis Gere. The first SFC instrument had
independent control of flow, mobile phase composition, and outlet pressure and used a com-
bination of gas and liquid chromatography components. Using a 10 cm
4.6 mm column
packed with 3 μm ODS (octadecyl silica or C18) particles, oligomeric ubiquinones extracted
from Legionella pneumophilia cell walls were separated and identified using ultraviolet
 Discovery of supercritical fluids and its relevance to analytical separation science 3
absorption spectra [8]. These compounds contain a benzoquinone and a long isoprenoid tail
group. The repeat unit of these oligomers is the isoprenoid tail. The ubiquinones in this extract
contained 8–13 repeating units with base line resolution achieved in less than 8 min.
Around the same time period, Lee and Novotny’s research groups together illustrated the
first use of open tubular SFC using silica capillary columns of 50–100 μm internal diameter
[9,10]. This technology provided considerable chromatographic efficiency, but the flow rate
and pressure could not be separately controlled due to the use of fixed restrictors (typically
smaller i.d. capillaries) to control pressure and flow within the column. These restrictors were
placed after the detector if a spectroscopy detector was used and before the detector if flame
ionization or mass spectrometric (MS) detectors were used. The early advances of SFC in the
1980s were closer to open-tubular capillary GC with CO2 as the carrier gas. Accordingly, Lee
Scientific provided open tubular SFC instrumentation [11] that was commercially released in
1985. As open tubular SFC did not have separate control of the flow rate and column pressure,
it was not widely used, especially for pharmaceutical compounds where accuracy and pre-
cision of measurements is critical.
An important innovation to allow the development, and application of packed column SFC
in the pharmaceutical industry was the back pressure regulator. The first back pressure reg-
ulator that provided pressure control independent of mobile phase control was developed in
1988, and this device concept continues to be applied in all commercial SFC instrumentation
today [12]. Terry Berger led the development of Hewlett-Packard’s (now Agilent) first super-
critical fluid chromatograph [13] that focused on the use of packed column SFC.
A tabulated chronology reflecting the application of SFC to different molecule types and
milestones is shown in Table 1. For further detail on the history of the use of supercritical
fluids see references [37–40].

TABLE 1 Notable applications and milestone chronology in the development of SFC.

Year Application Milestone Author/vendor
1957 Lovelock proposes “critical state
chromatography”
1962 Metal porphyrins “High pressure GC.” First application of Klesper et al. [5]
open-tubular chromatographic separation
with fluids above their critical temperature
and pressures
1967 Term “supercritical fluid chromatography” Sie and Rijnders [14]
proposed
1968 Carotenoids, steroids, sterols, First report of analysis of small drug-like McLaren et al. [7]
nucleosides, amino acids, molecules
carbohydrates, polymers “Dense gas chromatography” terminology
coined

1969 Purines, nucleosides, Proposed solvent strength of CO2 was Giddings et al. [15]
nucleotides, steroids, sugars, similar to IPA
amino acids, proteins,
terpenes

Continued
4 1. Evolution of packed column SFC as a greener analytical tool

TABLE 1 Notable applications and milestone chronology in the development of SFC.—Cont’d

Year Application Milestone Author/vendor
1970 Poly aromatic hydrocarbons First use of pressure programming to Jentoft and Gouw [16]
(PAHs) control mobile phase density
1972 PAHs First automated preparative SFC Jentoft and Gouw [17]
Styrene oligomers
1972–1977 PAHs Automated/preparative SFC Hartmann and
Styrene oligomers with mobile phase modifiers (methanol in Klesper [18]
n-pentane)
1978 Coupling of SFC with MS detection Randall and
Wahrhaftig [19]
1982 Kit allowing conversion of Model 1084 Hewlett Packard
HPLC into SFC. Included first mechanical
back pressure regulator
1982 PAHs Use of 3 μm spherical silica stationary Gere et al. [20]
phases in SFC illustrated that decreasing
particle size increased chromatographic
efficiency
1985 Caffeine in coffee First report of hyphenated SFE-SFC and Sugiyama et al. [21]
utilization of DAD detection
1985 Phosphine oxides First chiral separation using SFC. First Mourier et al. [22]
mention of addition of water to mobile
phase
1986 D,L-Amino acids Chiral separation of pharmaceutically Hara et al. [23]
relevant compounds with methanol as Jasco
modifier
1986 Open tubular SFC instrument Lee Scientific
commercialized
1988 Development of electronic APBR Saito et al. [12]
Jasco
1990 Coupling of SFC with MS with atmospheric Huang et al. [24]
pressure ionization

1990 Carbon dioxide shown to be similar in Deye et al. [25]

polarity to hexane

1991 First publication of enhanced-fluidity Cui and Olesik [26]

liquid chromatography
1992 Release of “second generation” Hewlett Packard
instrumentation allowing independent Gilson
flow control under pressure gradient
condition. HP instrument included Peltier
pump head cooling

1995 Hewlett Packard SFC product line sold to

Berger instruments
 Discovery of supercritical fluids and its relevance to analytical separation science 5
TABLE 1 Notable applications and milestone chronology in the development of SFC.—Cont’d
Year Application Milestone Author/vendor
2000 Development of semi-prep SFC system Berger et al. [27]
with “next generation” sample collection Berger
2001 Development of 2-ethylpridine (2-EP) Princeton
phase allowing analysis of basic analytes Technologies
without mobile phase additives
2001 Drug molecules Mass directed semi-prep SFC Wang et al. [28]
2005 Estrogen metabolites in urine SFC/MS/MS Xu et al. [29]
2006 Polypeptides SFC/MS with methanol/TFA modifier Zheng et al. [30]

2009 Release of 1260 Infinity SFC system Aurora/Agilent

2010 Steroids, profens, xanthenes, Assessment of sub-2 μm particle stationary Berger [31]
sulfa drugs, nucleic acids phases in SFC

2011 Steroids, profens, sulfa drugs, Assessment of sub-2 μm superficially Berger [32]
nucleic acids porous stationary phases in SFC
2012 Release of 1260 Infinity SFC/HPLC system Agilent
Aurigemma et al. [33]
2012 Release of UPC2 system Waters

2014 Release of Nexerra (SFE-) SFC system Shimadzu

2015 Proprietary pharmaceutical Filing of GMP SFC methods for Hicks et al. [34]
compounds measurement of chiral purity

2018 Salbutamol First cross-institute SFC method transfer Dispas et al. [35]
(identical instrument platform)

2018–2021 Salbutamol First cross-institute SFC method transfer Dispas et al. [36]
(different instrument platforms)

Fundamentals of mobile phases

Carbon dioxide remains the most commonly used component in SFC mobile phases. The
carbon dioxide molecule has a zero-dipole moment. Carbon dioxide has a dielectric constant
lower than those of hydrocarbons, but many hydrocarbons are not miscible with CO2. Carbon
dioxide is a charge separated molecule, meaning it has two polar bonds with nonzero bond
dipole moments that results in a significant quadrupole moment [41]. The charge separation
and electronic structure allow CO2 to act as both a weak Lewis base (electron-pair donor) and
Lewis acid (electron-pair acceptor). Strong interactions with carbonyl oxygen groups have
been noted with CO2 acting as a Lewis base. CO2 is also capable of forming conventional hy-
drogen bonds with protic donors that assist in the solvation of hydrogen bonding molecules.
Even though for specific compounds such as ketone and aldehydes significant intermolecular
interactions occur, under most conditions, CO2 is generally considered nonpolar [25]. Other
more polar substances that may form supercritical fluids have been considered, but their
critical temperatures and pressures are too high to be used for practical application.
6 1. Evolution of packed column SFC as a greener analytical tool

A major advancement in the understanding of solvation at the molecular level in super-

critical fluids was density variation near a solute [42,43]. From numerous experimental
and theoretical studies, the conclusions are that density enhancements near a solute occur
out to long molecular distances (10–30 Å) away from the solute [43]. Accordingly, the molec-
ular level view of solvation in supercritical fluids is higher density clustering around the
solute with substantial free volume in the fluid at greater distances. However, specific
intermolecular interactions between the bonds of CO2 are not usually the cause of this clus-
tering and therefore nonpolar interactions are controlling this solvation. Conversely, polar
solutes such as sugar acetates are highly miscible with supercritical and liquid CO2 based
on specific intermolecular interactions [41], which illustrates the varied solvating capabilities
of CO2.
For the chromatographic separation of many polar analytes, polar modifiers are necessary
to provide high efficiency through increased desorption kinetics from the stationary phase.
A large range of polar modifiers has become popular, especially the addition of low-
molecular-weight alcohols, such as methanol, ethanol, and isopropyl alcohol, which are fa-
vored due to their significant polarity and hydrogen-bonding ability. Also, the polarity of
the solvent shell around a solute allows prediction of the solubility and chromatographic re-
tention. Reports on the polarity of mixtures of supercritical/liquid CO2 combined with polar
modifiers began in the 1980s using solvatochromic studies, which involves monitoring shifts
in the wavelength maxima in the optical spectra of indicator dyes and prediction of the po-
larity of the mixtures based on these shifts [44–46].
In 1990, the solvent strength of pure supercritical CO2 and CO2 with polar modifiers was
compared to the solvent strength of common liquids. Fig. 2 compares the transition energy of
the absorption maximum for Nile Red, a solvatochromic dye introduced by Deye et al. [25], to
that of a commonly used set of solvatochromic dyes, and the Reichardt’s dyes that are a com-
pound family of pyridinium betaine dyes. Reichardt dyes were not highly soluble in CO2 and
that was the reason for introducing a new solvatochromic dye. Nile Red is soluble in CO2, and
it is absorbance maxima shifts substantially with increased polarity as shown in Fig. 2. Nile
Red is also stable in all studied solvents. Fig. 2 compares the shift in the transition energy for
Nile Red and the specific Reichardt dye (E30) for common dyes and for CO2. The transition
energy for both dyes was calculated using this equation, E ¼ 28,591.44/(wavelength maxi-
mum in nanometers) ¼ transition energy in kcal/mol [47]. Using this analysis, CO2 was
documented to have a low solvent strength similar to that of low-molecular-weight hydrocar-
bons (Fig. 2) [25]. In this same study, the solvent strength of 10 v/v% methanol/CO2 was ob-
served to be closer to the solvent strength expected for an equal mixture (i.e., 50:50) of
methanol/CO2 assuming ideal solution conditions and indicates significant clustering of
methanol around the polar solvatochromic dyes.
From an early stage, many researchers observed that the addition of a small proportion of
modifier in the mobile phase absorbed onto both polar and nonpolar stationary phases or ad-
sorbents. For slightly higher density conditions than at the critical condition (i.e., increased
pressure above the critical pressure), a disproportionate amount of the modifier adsorbs to
the stationary phase, even for a nonpolar stationary phase such as ODS. For example, with
2% methanol/CO2 mixed mobile phase, the adsorbed layer on ODS was nearly 25 wt%
methanol [48]. Therefore, the addition of modifiers impacts the retention of polar analytes
through changes in the polarity, density, and critical condition of the mobile phase, and
 Discovery of supercritical fluids and its relevance to analytical separation science 7

60
Hexane CO2 (a.)

57.5
CCl4 Diethyl ether

55
C6H6 THF
E(NR) CH2Cl2 CH3CN
2-Propanol
1-Propanol
52.5 CHCl3 DMF EtOH
Pyridine 1-Butanol MeOH

50 Ethylene
Glycol

47.5
H2O

45
30 35 40 45 50 55 60
Et (30)

FIG. 2 Comparison of transition energies for Nile Red and E(30) solvatochromic dyes for a number of common liquids
and CO2. E(NR) transition energy for the absorption maximum of Nile Red. Et(30) is the transition energy for absorption
maximum of a Reichardt’s dye designated as 30. Reprinted with permission from J.F. Deye, T.A. Berger, A.G. Anderson,
Nile red as a solvatochromic dye for measuring solvent strength in normal liquids and mixtures of normal liquids with
supercritical and near critical fluids, Anal. Chem. 62 (1990) 615–622. Copyright 2022 American Chemical Society.

modification of the stationary phase (with the adsorptive interactions with surface silanols
typically having the most significant impact [48–50]).
Another advancement in SFC that allowed the separation of ionic pharmaceutical com-
pounds was the addition of ion pairing reagents into the mobile phase modifiers [51].
However, there are limitations to this method because many of the desired ion pairing
reagents were not soluble in conventional SFC mobile phases—even when modifiers were
added [51].
As modifiers were increasingly employed, the need to remain at supercritical conditions
became less relevant. For example, Sandra’s research group in 1994 showed sub- and super-
critical packed capillary conditions worked well for the separation of both basic and acidic
drugs and illustrated the importance of polar modifiers [52].
Alfred Francis was the first to study the solubility of a large range of compounds in liquid
carbon dioxide and to study ternary systems of liquid CO2 mixed with organic solvents [53].
In this work, he described 464 ternary mixture systems containing liquid CO2. In 1991, the
Olesik research group published the first study of enhanced-fluidity liquid chromatography,
EFLC, which mixes conventional liquids with small amounts of liquid carbon dioxide [26].
Soon thereafter, the capacity for enhanced-fluidity liquid mixtures to form solvent clustering
8 1. Evolution of packed column SFC as a greener analytical tool

or enhanced solvent density around analytes was reported. This was similar to that observed
in supercritical fluids except that an intermediate cluster size, between that observed
in supercritical fluids and conventional liquids, was noted [54]. Also, for many polar liquids,
as much as 40–50 mol w/w% CO2 can be added to the mixture before the solvent strength is
markedly decreased (i.e., these mixtures have a solvent strength similar to the original organic
solvent or solvent mixture [26,55,56]). This attribute was noted for all the enhanced-fluidity
liquid mixtures studied to date. Enhanced-fluidity liquid mixtures provide diffusion rates
approaching those of supercritical fluids (and lowered viscosities), but with high solvent
strength. Chemical engineers began studying similar liquid mixtures around the same time
for separations in processing conditions and coined the term “gas-expanded liquids” or
“GXLs” [57]. As numerous scientists have noted, SFC or chromatography containing lique-
fied gases provides a continuum condition in properties and applications between that of GC
and HPLC. Terms such as “unified chromatography” and “convergent chromatography”
emerged to describe these conditions [6,58–60].
In summary, supercritical fluids, subcritical liquids, or enhanced-fluidity liquids have
physicochemical properties that are intermediate between those of gases and liquids. These
fluids are typically compatible with detectors that are commonly used by gas chromato-
graphy and liquid chromatography. In the most common practice, which is to use carbon
dioxide in these mixtures, the fluids have low toxicity. Furthermore, the retention factors
in these chromatographic systems are primarily controlled by the solvent strength of the
CO2-solvent mixture. However, fine tuning of the solvent strength of a given solvent-CO2
mixture is controlled by system density (i.e., pressure). Finally, most chromatographic exper-
iments are accomplished at flow rates above the optimal flow velocity that means the band
dispersion is inversely related to the diffusion coefficient; and chromatographic efficiencies
are directly proportional to the diffusion coefficient, which explains why the efficiencies of
gas chromatography are higher than those for SFC, which are in turn higher than EFLC
and conventional reversed and normal-phase liquid chromatography. Table 2 provides a
comparison of the diffusion coefficients and viscosities of gases, supercritical fluids,
enhanced-fluidity liquids, and conventional liquids. Fig. 3 shows a comparison of the band
dispersion for the same compound with the same chromatographic column using reversed-
phase HPLC conditions compared to reversed-phase SFC conditions. SFC shows marked
efficiency gains in line with the theory discussed earlier.

TABLE 2 Mass transport properties of mobile phases observed with different separation approaches.
GC SFC EFLC HPLC
Mobile phase Gas Supercritical fluid Compressible liquid Liquid
Analytes Volatile Nonpolar/moderately Polar/ionic Nonpolar/polar/ionic
compounds polar compounds compounds compounds
Diffusivity, D (cm2/s) 101 103 to 104 103 to 105 105
Efficiency, N α D
Viscosity, (g/cms) 104 103 to 104 102 to 103 101
Pressure drop, ΔP α
 Sustainable aspects of subcritical and supercritical chromatographic separation methods 9

FIG. 3 Variation plate height, H, vs mobile phase linear speed (u) for HPLC (squares) and SFC at 20°C (diamonds are
for 95/5 v/v% CO2/methanol and triangles are for pure CO2) using a 5 μm RP-C18, 250 mm
4.6 mm. i.d. column.
Reproduced with permission of Elsevier from E. Lesellier, L. Fougere, D.P. Poe, Kinetic behavior in supercritical fluid chroma-
tography with modified mobile phase for 5 μm particle size and varied flow rates, J. Chromatogr. A 1218 (2011) 2058–2064.

Sustainable aspects of subcritical and supercritical chromatographic

separation methods

Worldwide, annual solvent use and its disposal across all applications is approximately
30 million metric tons [61]. In the United States, the average human weighs 137 lbs (62 kg).
Thirty million metric tons is the weight of 482,763,504 humans, which is well over the total
population (327.2 million) of the United States. In addition, solvent usage continues to
increase worldwide—mainly due to expanding economies. The applications of solvents in
separation science contribute significantly to organic waste. Estimates of solvent use in liquid
chromatography (LC) are approximately 150,000 tons annually [62].
The pharmaceutical industry is a heavy user of liquid chromatography ranging from
applications in drug discovery to large-scale drug production. The industry is therefore a
significant contributor to global solvent use and disposal challenges. Accordingly, most
pharmaceutical manufacturers have made the use of green solvents a high priority. Many
pharmaceutical companies, including Pfizer [63], AstraZeneca [64], and GlaxoSmithKline
[65], have developed solvent selection guides to facilitate the use of less hazardous solvents
by their employees. These guides are useful; however, the characterization of a solvent’s
properties is not uniform. The American Chemical Society Green Chemistry Institute Phar-
maceutical Roundtable was established in 2007 [66]. The Roundtable has developed tools
[67,68] to provide a standardized assessment of solvent greenness. Recently, the Roundtable
published the Analytical Method Greenness Score (AMGS) calculator [69] to quantify the
“greenness” of commonly used separation techniques. The AMGS calculator includes
cumulative energy demand, instrumental energy demand, the mass of solvent consumed,
and environmental, health and safety (EHS) scores. Also, the AMGS tool is the first selection
guide that includes the information on the greenness of supercritical/subcritical fluid
chromatography.
10 1. Evolution of packed column SFC as a greener analytical tool

The AMGS calculator allows for detailed comparisons of separations methods based on
solvent waste, instrument, and solvent energy. Fast, efficient separations provide the greenest
conditions. Using the AMGS calculator, scientists from eight pharmaceutical companies
compared typical internal methods using HPLC, ultrahigh pressure HPLC (UHPLC), and
SFC for analytical and preparative separations. The general trend was that SFC methods
and UHPLC methods were greener than HPLC methods for analytical separations. For pre-
parative separations, preparative SFC, and aqueous-based HPLC had similar greenness
scores. However, when the energy needed to remove the water after the separation was
included, SFC was substantially greener [69]. The AMGS calculator is expected to provide
great value in method and solvent selection at both the analytical and preparative scale being
a one stop shop for solvent safety, solvent energy, instrument energy, and waste. Enhance-
ments to the AMGS calculator for supercritical and subcritical fluid chromatography using
CO2 were recently developed that include corrections to the EHS parameter, a correction
for the density of CO2 values used in typical SFC conditions, and a new parameter to add
to the AMGS that allows for the calculation of the flammability hazard of mobile phases if
a leak occurs [70]. This is particularly important as the scale of the separation increases from
analytical to semipreparative and preparative scale.
Using the updated AMGS calculator and a life cycle analysis (LCA), the greenness of
HPLC and SFC reversed-phase separations of pharmaceutical compounds were compared.
Both the AMGS and the LCA showed similar trends. A correct holistic analysis must include
the instrument energy consumption for both. When comparing HPLC separations using
alcohol/water mobile phase to SFC separations with alcohol/H2O/CO2, the results show
that the additional energy consumption of current SFC instruments must be counter-
balanced by taking advantage of higher flow rates. HPLC methods using acetonitrile-water,
which are considered the gold standard for HPLC separations, were significantly less green
than the SFC methods examined on both an analytical and preparative scale [71]. Due to the
low volumes used, commonly used acid and base additives to the mobile phases such as
formic acid or ammonium hydroxide did not significantly impact the greenness of the
separations.

Analytical scale subcritical SFC

As noted in the phase diagram (Fig. 1), there is no barrier to transitioning from supercritical
to subcritical conditions or vice versa. As early as the early 2000s, most of the reported
SFC functioned under subcritical conditions. Considerable progress in the SFC field was
driven by the advent of new SFC instrumentation that provided reproducible retention times
for analytes, reproducible control of column back pressure, and separate control of flow
rate. Furthermore, effective interfaces to mass spectrometers and commonly used HPLC de-
tectors became common place. These advances were quickly accepted by the pharmaceutical,
food characterization, and environmental researchers—greatly expanding the application
space of SFC.
Improvements in SFC instrumentation, including pumps, mixers, autosamplers, back
pressure regulators, and reduced extra column dispersion volumes, enabled high-performance
and high-speed separations of compounds in seconds that compare favorably to fast HPLC
 Analytical scale subcritical SFC 11
separations [34,72]. These enhancements allowed the introduction of multiple injection single
experimental run (MISER) high-throughput analysis of enantiopurity to be a viable option for
the pharmaceutical industry. Using MISER, the enantiopurity of samples in a 96-well plate was
achievable [73,74]. Fast SFC separations became possible with this latest generation of analytical
instrumentation. For example, subcritical fluid chromatographic conditions using 90/10 v/v%
CO2/methanol mobile phase with irregularly shaped 3 μm particles and trans-1,2 diamino-
cyclohexane derivatives as selectors, packed into a 50
4 mm i.d. column, provided the
separation of three conformational isomers of an aromatic hindered bis ketone in 15 s, while
enantiomers of aryloxypropionic acid methyl esters were separated with a 100
4 mm i.d.
packed with 5 μm particles in 40 s. In each case, the flow rate was 7.0 mL/min—far beyond
the capability of liquid chromatographic instruments [75]. The speed of the separation is a
function of the mobile phase not the stationary phase and ultrafast achiral separations have
been reported too (e.g., Ref. [76]).
For achiral small molecule (MW <500 amu) separations, a broad range of stationary phases
are viable for SFC applications, including reversed-phase, normal-phase, and hydrophilic
interaction chromatography (HILIC). Using the linear solvation energy relationship (LSER)
that correlate the log k (retention factor) to five descriptors (E, S, A, B, V), which are related
to charge-transfer, dipole-dipole interactions, hydrogen bond acidity and basicity, and
dispersion, respectively, the intermolecular interactions available with different stationary
phases are described. Fig. 4 shows a spidergram that illustrates the varied selectivities of
five different types of columns (Water HSS C18 SB, XSelect CSH Fluorophenyl, BEH (silica),
BEH 2EP, and BEH RP18 Shield). The selectivities of these stationary phases are clearly
described by their location near a coefficient in the LSER model (e, s, a, b, v), which indicates
the relative importance of the descriptor illustrated above for a given stationary phase. This
type of representation allows ready selection of stationary phases with markedly different
selectivities.
Numerous studies have shown that the chiral stationary phases used in liquid separa-
tions (Chiralpak AD, AS, and Chiralcel OD, OJ, Pirkle phases, macrocyclic antibiotic phases
such as Vancomycin, and Rifamycin B, protein phases, ion-exchange and cyclofructan
phases, and other polysaccharide phases) work well in super/subcritical chromatography.
Common SFC solvents combinations used for chiral separations include CO2 with polar
modifiers such as methanol, ethanol, isopropyl alcohol, and at times additives such as
diamines, e.g., diethylamine (DEA). Additionally, unconventional modifiers such as tetra-
hydrofuran, methylene chloride, and dimethyl sulfoxide have been used in combination
with CO2/alcohol modifier systems for chiral separations at both the analytical and prepar-
ative scale with bonded phases [77]. Nonconventional modifiers such as alcohol/methylene
chloride or methanol/tetrahydrofuran may significantly improve resolution compared to
HPLC conditions or SFC with alcohol modifier, especially for enantiomeric pairs with low mis-
cibility with alcohols. However, these nontraditional modifiers are not recommended for
screening purposing but highlighted there may be other options when more conventional
solvents do not work well [77].
Polar compounds added to the modifier and the injection solvent, such as ammonium ac-
etate [78], were found to sharpen the peak shape and decrease the retention time of polar
pharmaceutical compounds significantly in achiral and chiral separations. Tetramethy-
lammonium acetate or ammonium chloride was found to be useful additive to modifiers
12 1. Evolution of packed column SFC as a greener analytical tool

FIG. 4 Spidergram classification of some commonly used stationary phase in SFC. Reproduced with permission of
Elsevier from S. Khater, C. West, E. Lesellier, Characterization of five chemistries and three particle sizes using supercritical
fluid chromatography, J. Chromatogr. A 1319 (2013) 148–159.

for both acidic and basic compounds [79], while ammonium hydroxide, ammonium formate,
and ammonium carbonate were found to be useful additives to improve the separation of
basic compounds [80,81].

Analytical enhanced-fluidity liquid chromatography

During the 1990s, the addition of liquefied gases such as CO2 and fluoroform to conven-
tional liquids was shown to enhance the efficiency, selectivity, and speed of analysis for
moderately polar compounds under reversed-phase [82], normal-phase [83], chiral [84],
and size exclusion chromatographic [55,85] conditions. Unique polymer separations were
described as well [86]. Fundamental studies on the phase diagrams for enhanced-fluidity
liquid conditions were undertaken including CO2/methanol, CO2/acetonitrile, fluoroform-
methanol, fluoroform-acetonitrile, methanol-H2O-CO2, and acetonitrile-H2O-CO2 mixtures
[87,88]. Other fundamental studies included polarity measurements under enhanced-fluidity
conditions.
 Preparative and analytical scale instrument improvements 13
During the early 2000s, the ability to provide buffers of know pH was documented [89],
and buffered EFLC systems showed considerable promise for the separation of polar
compounds such as basic tricyclic antidepressants and benzoic acids [90]. In the early
2010s, EFLC was used for the separation of biological compounds using isocratic conditions
that included H2O in the mobile phase. Compounds, such as nucleosides and nucleotides,
were separated using EFLC conditions with HILIC [91,92]. Polar natural polymers, such as
humic acids, were separated under EFLC water-based critical chromatography and size
exclusion chromatographic conditions [93,94]. This work showed that highly polar com-
pounds could be separated under EFLC conditions and selectivity as well as efficiency advan-
tages relative to similar HPLC conditions were garnered in nearly all studies. For reviews of
the fundamental studies of enhanced-fluidity liquids see Olesik [95] and a review of applica-
tions, Olesik [96].
This early EFLC research was undertaken using home-built instrumentation because com-
mercial SFC or HPLC instrumentation was not capable of operating under these conditions.
However, the more recent commercial SFC instruments are compatible with EFLC condition
with minor modifications to software and with changes in polymer seals in the solvent pumps.
This allows expanded consideration of EFLC and therefore unified chromatography spanning
from the supercritical state seamlessly through to liquid state chromatography. Now that EFLC
can be used on commercial SFC instrumentation, moving EFLC methods to other users and
considering EFLC for preparative scale applications is viable. These advantages in expanding
the SFC continuum to larger biomolecules and peptides are now possible with commercial in-
strumentation. Gradient EFLC is readily possible that allows for applications in ion exchange
chromatography, HILIC, and hydrophobic interaction chromatography (HIC).

Preparative and analytical scale instrument improvements

In modern SFC, the pharmaceutical industry has found (1) the operable polarity range of
SFC separations coincides with the range of pharmaceutically relevant small molecules
between logP 2 and 10 [97], (2) SFC methods are generally faster than the comparable HPLC
method, and (3) organic solvent waste is drastically reduced along with the overall purifica-
tion time for target analytes (APIs, impurities, etc.). Due to these advantages, SFC was deter-
mined to be most effective in replacing chiral normal-phase analytical separations as well as
analogous preparative applications, i.e., flash and normal-phase preparative chromatogra-
phy. However, the chemical space for the pharmaceutical industry has broadened to other
modalities such as peptides, antibody drug conjugates, and modified nucleotides and nucleic
acids. While chiral SFC separations of small molecules are still valuable, many analytical
separations have been reclaimed by UHPLC/UPLC instruments with reversed phase, chiral
columns advances and sub-2 μm particle size columns. Widespread adoption has been made
possible by touting such instrumentation as a straightforward upgrade to existing HPLC in-
strumentation with a few caveats. SFC has likewise experienced its own upgrade to UHPSFC,
with augmentations to almost all components to also exploit the advantages of sub-2 μm sta-
tionary phases and wider mobile phase options. In this section, we explore these improve-
ments in terms of mobile phase delivery, column compartments, chromatograph/detector
interfaces, fractionation, and backpressure regulation.
Another random document with
no related content on Scribd:
the decks. These tents, moreover, were no longer water-tight, and
the sleeping-place in the damp boats was very small.
Our negroes generally managed to stow themselves away under
shelter somehow, often one on top of the other, but I should have
liked better weather for this last bit of the journey, so that they might
have been able to get over all they had gone through at Bussa. They
made up for their discomfort at night by getting up late in the
morning. All this, however, did not prevent us from making good
headway without any over pressure, borne on as we were by the
strong current. On the 13th we covered forty-five miles, going on
until eight in the evening, just in time to anchor before we were
overtaken by a tornado, and an awful one too. Fortunately we found
shelter in a little gulf, and escaped with a good ducking.

IGGA.

On the 14th, judging by the rate at which we went, the current

must have been yet stronger. We made some fifty miles, passed the
night near Igga, and arrived there at eight o’clock in the morning on
the 15th.
 The country between Geba and Igga is uninteresting; no villages,
or scarcely any, were passed, and there was no cultivation. The
appearance of the banks is much what it is between Say and Bussa;
a few karités occur here and there, that is all. We met a canoe now
and then only. The oil-palms, which had begun to appear beyond
Say, now became more numerous, but the country still appeared
deserted.
In a large plain near Igga there is a factory kept by a white man.
Just before we reached it we saw a big boat called the Nigritian,
which was formerly the pontoon of Yola. The Royal Niger Company
had just been driven from the Benuë and from the Adamawa; its
trading agents had been recalled, together with the pontoon they had
been authorized to use on the river. This must have been a very
severe blow to the Company, for much of the ivory exported through
their agents came from Adamawa and Muri.
The Ribago, a pretty little craft of from six to seven hundred tons,
is moored at Igga. She is the best boat belonging to the Company.
She brings down palm-oil in the nut before it is extracted, karités and
other articles for export. The oil is of a very fine quality indeed. It will
probably be the Ribago which will tow us down-stream if all is
satisfactorily settled with the Company about Bussa and Auru.
The agent at Igga thought we should find Mr. Wallace at Lokodja. I
was very anxious to see him, for it is with him I must get the
misunderstanding, if misunderstanding there were, explained. His
word alone would suffice to exonerate the Company from blame, and
only if he could give me that word, should I care to accept his good
offices on my behalf.
After passing an hour at our anchorage at Igga, we started for
Lokodja to look for Mr. Wallace, whom it was very difficult to catch.
Fortunately for us, the current was still very strong, but navigation
was very tiring, for with the banks inundated as they were, it was
difficult to find the bottom amongst the tall grass. Late in the evening
we at last anchored near the left bank, and landed to cook a hasty
meal. Fili, one of the coolies who looked after the kitchen
department, had cleared a corner of bushes and lit a fire when, all of
a sudden, the men made a rush for the boats screaming manians!
manians! They had been attacked by the black ants they call
manians, the bite of which is very severe. No cooking for us to-night,
no meal however simple! No sleep either for our poor men, for the
rain began to pour down again. Worse still, the terrible manians
began to climb on board by the anchor-chains, by the ropes of the
grappling-hooks, by everything, in fact, which held us to the bank.
They had come to storm the barges, and the ropes and chains
became black with their swarms. The only way we were able to
check this novel kind of invasion was by lowering the chains and
ropes into the water.
This horribly comfortless night over, we started again with almost
empty stomachs. The scenery was very picturesque, but although
the water was high we felt the boats grate on the rocks lining the bed
of the stream. Navigation must be generally far from easy here.
The vegetation now became denser, and the oil-palm of much
more frequent occurrence. There were, however, few villages, and
they became further apart, on the banks at least, as we advanced. At
last in the evening our pilot told us we were approaching Lokodja.
Picturesque hills, from about six hundred to a thousand feet high,
lined the right bank, whilst on the left we could see the mouth of the
Benuë, now greatly increased in width by inundations.
About six o’clock we came in sight of the huts of the village, rising
in tiers from the slopes of a hill, their zinc roofs shining amongst the
verdure in the glow of the setting sun. We were at Lokodja, and as it
was nearly night we anchored off the bank.
Here we found Mr. Drew, the executive officer of the Company for
the Lokodja-Geba district, for whom we had waited in vain at Geba,
and also another officer who spoke French.
We were received with all due etiquette and invited to dinner. We
talked about the river; and Mr. Drew, who did not allow himself to
show any surprise at our having passed safely down it, must really
have been astonished. He told us he had himself achieved the
arduous task of going over the rapids in a light canoe accompanied
by one man only. He had intended to go down to Bussa by the
channel used by the natives. He had even been capsized, and
dragged down into the whirlpool. He owed his life entirely to his
canoe-man, who had plunged after him and brought him up from the
bottom. He still had the scar of a wound he had got from the sharp
flints, amongst which he had been rolled over and over.
Major Festing, who came in to dessert, invited us to go to him the
next day. We cut but sorry figures beside our hosts in their
unimpeachable costumes, for our clothes were torn by our struggles
in the bush, our gold lace was tarnished, our breeches were
patched, our boots had been bought in the country, and our helmets
were terribly battered about.
I do not know which agent of the Company it was who refused to
receive the leader of a French expedition because of his
disreputable appearance, with untrimmed beard and clothes in rags.
Times are greatly changed since then, or rather perhaps the
instructions given have been modified.
The next day we had breakfast with Major Festing, and were most
cordially received. Our host was then Commander-in-chief of the
troops in the service of the Niger Company. Lokodja was his
headquarters, and his soldiers, who were Haussas, were well
lodged. Their cantonments are charming, and the Major’s house had
every English comfort that could possibly be expected. Big airy
rooms adorned with weapons, looking-glasses and hunting pictures,
etc., native mats on the ground, flowers growing in the copper pots
manufactured in the country. Everything very simple and suitable.
Music was going on whilst we were at breakfast, as if we were on
board an admiral’s flag-ship or at the Grand Hotel in Paris. Children
played to us on the flute, regaling us with the familiar airs of the café-
concerts of France. We had printed menus, dainty salt-cellars,
caviare, whisky-and-soda, good stout, etc. Oh, what a delight it was
to eat a well-served meal on a table-cloth decked with fresh flowers!
If only we had had a few ladies in light summer costumes to share it
with us, it would indeed have been complete.
Major Festing most courteously placed at our disposal as
interpreter, a Haussa sergeant of his from the Senegal, who had
been at one time in the service of Mizon, and also of De Brazza. He
spoke a little French, and had been one of the last to leave the
station of Yola. He told us of all his strange wanderings to and fro,
and piloted us about the town when we went to make our purchases,
for we did make some purchases at Lokodja. To begin with, we
supplemented our stores of provisions, which was very necessary, if
we wished suitably to return the hospitality we received. We had,
moreover, very little of the dinner service left which we had brought
from France three years before. We had, it will be remembered, sent
to the bottom of the river everything not absolutely indispensable,
and we wanted some claret and champagne-glasses badly.
The natives of Lokodja were very civilized, using table napkins,
basins, dishes with covers, china flower-pots, etc., sold to them by
the Company, or rather bartered for native productions, for there is
no money currency in the Niger districts. The wages of the troops,
labour, and raw material are all paid for in merchandise, such as salt,
stuffs or ware of different kinds. The Company seem to make
considerable profit on these transactions. As for us, we were rich
enough to be generous. Suleyman, our interpreter, received orders
to buy everything offered at the price asked, for we should only have
to throw the things which were too heavy to take on, into the water
later. So we gave silk drawers for a dozen eggs, and long strings of
pearls, false ones of course, for three bananas.
The generosity of Commandant Mattei, agent of the old French
Niger Company, whom we so clumsily allowed an English Company
to supplant, has become proverbial, and the natives often quote it
apropos of the stinginess of the Niger Company. I am very sure that
our stay at Lokodja did nothing to lessen the fame of French
liberality. The natives of the banks of the Niger still bemoan the loss
of French traders and the hauling down of the French flag.
Lokodja, which we were able to visit, is a fairly large village, very
picturesquely situated on a mountain. It is cut across by ravines and
shaded by banana and papaw-trees, with numerous oil-palms. There
is a splendid view of the meeting of the Benuë and the Niger. The
remains of the steam-boat Sokkoto, which was wrecked on a rock,
are still to be seen, and further down the river are other stranded
boats.
We were told that Lokodja is the principal town of an extensive
district numbering from twenty to thirty thousand inhabitants. The
town properly so called, however, does not contain more than from
four to five thousand at the very most. The market, which is very
extensive, is much frequented, and is held in the afternoon. All
manner of European articles are offered for sale in it. The only native
industries are the beating of copper and the manufacture of rather
peculiar drawers made of two pieces of stuff sewn together and
adorned with a kind of open work. The blacksmiths, who are very
skilful in a kind of repoussé work done with a pointed instrument on
copper, make vases, cups, and ewers of it, which are really very
original in design.
Most curious of all the specialities of Lokodja, however, are the
games and the tam-tams held there. In the former, the performers
are all young graceful girls who are perfectly nude. I have visited
many towns of low morality. I know Naples, Port Saïd, and Colombo.
I have seen the so-called flower-boats of China and the Japanese
yoshivaras in that Orient where everything is possible, but never did I
witness anything to be compared with what goes on at Lokodja.
The chief of the village is the well-known Abegga, and the name
calls up for us French all manner of memories. Abegga is really
almost a relation of mine, for he is a freed man who was bought at
Sokoto, and given his liberty by my Uncle Barth. Abegga followed his
master to England first and then to Germany. Back again in Africa,
he entered the service of Commandant Mattei as interpreter, and to-
day he is king of Lokodja. Such are the chances of life!
We were received by him with effusion, for we awoke all his old
memories. Taburet, who from his translations from Barth’s book
knew more about Abegga than Abegga did himself, had a long talk
with him in English. In the end we sent our royal friend, Baudry’s
hunting-piece as a present, by the hands of his envoy.
We expected every minute to hear of the arrival of Mr. Wallace,
but he did not come. I could not remain at Lokodja for ever, so I took
Mr. Drew’s word for it that neither he nor the Company had had
anything to do with our difficulties at Bussa and Auru, accepting the
offer made to me with so much urgency that we should be towed
down-stream by the Ribago, the steamboat we had met at Igga, and
which had now come down again to Lokodja.

MOUNT RENNEL ABOVE LOKODJA.

We were to start at two o’clock. After we had made our farewell

visits I went to Mr. Drew and said, “I have decided to accept your
offer of having us towed down-stream.” Then I added rather
awkwardly, “How much?”
“Five pounds for each white man, and one for each black, was the
reply.
A good price truly just for towing us down-stream! It would come
to 1450 francs altogether! I merely, however, said “Oh!” just to relieve
my feelings.
Now was not this rather sharp practice on the part of the
Company? After pressing me so much to accept a service, I had
imagined that it was offered gratuitously as between one friendly
government and another, and what had annoyed me was the thought
of being under an obligation to the Royal Niger Company. But I was
quite wrong; I was dealing with the traders of the Company only, and
that put me at my ease.
They may have thought that having come so far I should not have
money enough left to pay them, and that I should have to leave in
their debt, but I simply said to Mr. Drew—
“All right; I will come back and settle with you in a minute.”
A few moments later I arrived with my bag of crown-pieces. I had
not, however, brought enough after all, for by some
misunderstanding, no doubt, we really had to pay six pounds for
each white man, and twenty-five shillings for each black, which
mounted the sum-total up to 1800 francs. However, I was able to
make up the difference at once all but two sous, I think, and those I
sent by Digui.
No doubt Carrol foresaw all these mercenary dealings when we
were at Geba, when he made such a fuss about paying Taburet for
his attendance on the people who were ill at the station, and wanted
to give me money for the miserable little musical-box which I had
been so glad to leave with him as a token of my gratitude.
The Royal Niger Company had in fact treated our expedition as a
party of traders, and I preferred that both for myself and for France. I
do not therefore owe the members of that Company any more
gratitude than I should the conductor of an omnibus in Paris when I
have paid him my six sous.
The loading of the Ribago went on slowly, but at five o’clock we
started; the pipers of Major Festing came down to the quay and
played the Marseillaise, whilst the guns of the station fired a salute
as, towed by the Ribago, we left for Assaba.
Now for a couple of words about the Royal Niger Company. I will
say nothing of the treaties or of the constitutive acts which preceded
its formation, for I have not got to draw up an indictment against it. I
will confine myself to quoting what Naval Lieutenant Agoult said on
the subject—“The Company is but the screen behind which England
hides herself.”
To the great detriment of the shareholders, the Company tries to
create an Empire, and in view of its acquisitions of territory, to make
head against the revolts caused by its rapacity, it is obliged to
maintain an army relatively large. This necessity causes a
mischievous friction between the military and civilian officers in the
service of the Queen, they and the trading agents sometimes
carrying their animosity to each other so far as to come to blows.
Then again the officers are anything but well treated by the
Company. Like the agents, they are taxed and taxed again. Heaven
only knows what an arduous profession theirs is. Carrol was always
on the road, and Festing, when we saw him, was suffering horribly
from a liver complaint. He had just returned from a twenty days’
campaign against the villages in the bush on the left bank, and he
was so tired he could not remain in the saddle. We were told of
several officers having recently been killed by poisoned arrows, and
of one who had died from eating poison in a village on the banks of
the river.
Moreover, this armed force and all the courage and devotion of
those who command it, fail to secure peace. Whilst we were on our
voyage, the horsemen of Bidda had come down to pillage as far as
the bank opposite Lokodja. It is only in the immediate neighbourhood
of the stations that things are quiet. The steam-launches have to be
constantly going up and down the arms of the river, especially in the
delta, to keep the natives in awe with their riflemen and their
machine-guns. It is rare for a boat to go down the river without being
fired at. At Abo, lower down-stream, the people were astonished that
we had been able to come so far without any fighting. It may have
been the effect of the flag we carried, for the tricolour flag is still
beloved and regretted in these parts for the sake of the memory of
Commandant Mattei.
The Company does not hold the country beyond the banks of the
river. Then, again, there are no means of communication between
one place and another. Truly we French may be proud of our work in
the French Sudan. We have done better than the English on the
Upper Niger; our colonization is far superior to theirs. On the Lower
Niger they have neither telegraph wires, for these go no further than
Akassa and Brass, at the mouth of the river, no road at all to be
compared with our line of revictualling posts, and of course, need we
add? they have no railway!
It seems to me a fact that of all the Niger districts, the richest and
the most favoured by Nature from every point of view are those we
occupy in the French Sudan.
Assaba is the residence of the Agent-general of the Company,
and there is a hospital there for the use of the employés. When the
French mission of the Pères du Saint Esprit left Lokodja it
established itself at Assaba.
A missionary was waiting for us when we landed, and I went at
once to his house. The situation is beautiful enough, but what a hard
life the Fathers lead! They are, I believe, rather harassed by the
Company, as much because they are French as because they are
Catholic, and as a result their tale of converts is not very long. Some
Sisters of Charity work in connection with them, and make their way
on foot from village to village in the interior, marching at night to
avoid the heat of the sun, and visiting the Christian natives far away
from the river.
A few hours’ walk off, the Fathers told us, are some big, very big,
villages, into which alone they are able to penetrate, not without
considerable danger to life sometimes. Terrible scenes of human
sacrifice and cannibalism have been witnessed by the devoted
Sisters. Such atrocities would never be tolerated in the French
Sudan.
But what does all that matter to the Company as long as it can
buy its palm-oil at the market-price, a price fixed by force?
That evening we had to dine with us the only Father of the
mission just then at Assaba, and two Sisters, one the Superior,
Sister Damien, a pale-faced Italian, whose hands had become
almost transparent, and whose features were wasted through
successive attacks of fever. For all that she still eagerly pursued her
vocation. I know nothing finer than the life led by these women at the
extreme advance guard of civilization, exposed to the heat of the
sun, to fever, to all manner of fatigue, to the indifference of the
negroes, and sometimes, as if all that were not enough, to the malice
of the whites.
I imagine that it was long since the Father and the Sisters had
enjoyed themselves so much. Unfortunately a tornado burst upon us
in the middle of dinner, and at eight o’clock we had to take refuge in
Father Hacquart’s rooms, through the cracks in the roof of which,
however, the rain poured in torrents.
We escorted our guests back to the mission house through the
rain.
That same night the long-expected Mr. Wallace, Agent-general of
the Company, arrived on the launch Nupé. I went to call on him the
next day. After congratulating me on our successful journey, he
renewed the assurances already made to me by Carrol, Festing, and
Drew. I heard later that Mr. Flint, another important member of the
Company, was also on board the Nupé. But he preferred to avoid us.
When we left we were able to rejoice the hearts of the
missionaries of Assaba, with a few bales of stuffs and knick-knacks,
with which they could reward their faithful natives. We wanted to stop
at Onitcha, the cross of the mission of which we could already see,
to give a greeting to the Pères de Lyon stationed there, but the
captain of the Ribago told us he had been ordered not to go there,
although Mr. Wallace had assured me to the contrary only a minute
before.
Avoiding Onitcha, therefore, we went to anchor for a few
moments, first at Illuchi, and then at Abo, where the Ribago was to
leave us.
The Company, however, was determined to escort us to the very
threshold of their territories. Those who know what it is to be
suspected, will involuntarily compare this conduct to the way in
which, in certain shops, customers are escorted to the door lest they
should steal anything on their way out.
No doubt, without being exactly sharpers, we might have got a lot
of information, and have made observations on many things if we
had remained longer on the river. Would that have been altogether to
the advantage of the Company? D’Agoult says he saw the steamer
laden with spirits going by, yet all the time, according to the
Company, all its subjects, white or black, would, under its beneficial
influence, become teetotalers or total abstainers.
It was politic too, perhaps, to hide from us the troubled state of the
district all along the river, and the precarious position of the
Company. Do its members know, I wonder, how happy these
discontented regions once were under the French Company, and all
that would result from the mere presence once again of the French
flag?
As for me, however, I prefer to think simply that this
obsequiousness of the Company towards us, this insistence on our
accepting the offer of being towed down-stream, and paying for the
service rendered, this eagerness to see us off, had but one aim, and
that aim a humane one.
We were escorted to Wari to save us from another attack from the
Patanis. Our departure was hastened because we were tired, worn
out, eager to taste once more the joys of home and family life. All
serious thinkers, whose opinion is of any weight, and who know
anything about English ways, will agree with me, irony or no irony!
We dined at Abo, and when night had fallen, a launch arrived at
our anchorage, which was to take charge of us. On board was a
bright, jovial young officer, Lieutenant Aron by name, of Australian
birth. Judging from what we saw of him, Australia must be to
England what the south of France is to the French. Did he not tell us
one day that the Company had a post at Kano, another at Kuka, and
twelve big steamers on the river? But for these venial exaggerations
he was a charming companion, what the English call a very good
fellow, who made the hours we were in his company pass very
pleasantly. We shall all, Lieutenant Aron included, long remember
the dinner we had together on the Kano, as the Ganagana pontoon
is called, whilst a tornado was raging, and he sung at the top of his
voice all the comic songs in the Anglo-Franco repertory, to the
accompaniment of the flute and the harmonium, whilst quaffing the
whisky and the claret we still had left.
As is well known, the Niger flings itself into the sea in an immense
number of branches. Two of these branches, viz. that of Brass and of
Forcados, are more practicable for navigation than any others. The
first belongs to the Royal Niger Company, the second to the Niger
Protectorate, a regular colony governed directly from England, and I
was told that the competition in trade between the two was very
keen.
I had long intended to go down to the sea, not by the Brass, but
by the Forcados branch, which would enable me to get away from
the Royal Niger Company sooner, and pass a few days in the
English districts on the coast belonging to the Niger Protectorate.
I preferred to embark there than in a port belonging to the
Company. The two Companies are, as already stated, more or less
rivals, and those on the French despatch boat Ardent had cause to
speak in terms of high praise of the way in which they were treated
by the English of the Protectorate.
Lieutenant Aron accompanied us on the Forcados branch as far
as Wari, where resides an English vice-consul. We were
breakfasting on board the launch when we came in sight of the
houses of Wari. Our three barges were roped together, and their
three tricolour flags flying. The launch, however, could not hoist the
British flag, its gear having somehow got damaged.
The Dantec now brought us up to the stockade, where we awaited
the arrival of the officers of the Protectorate. Then between
ourselves and our guide began an animated and certainly very
curious colloquy; astonishment on one side, vehement explanations
on the other. What changes in the expressions of the faces of those
engaged in the conversation! What shouts of laughter! What were
they saying? This is what I thought I made out. Seeing our three
barges each flying a tricolour flag, and the launch with no colours at
all, the English of the Protectorate had thought we had retaliated on
the Company by a skilful manœuvre for the bad turn they had done
the French the year before. “The Company,” they said, “had intended
to confiscate our barges, but they being well manned and well
armed, had instead captured the launch and taken her down under
the French flag to Wari.”
No, I cannot have understood the conversation, I must have
dreamed it all! The English never could have believed us capable of
such a thing, and would never have suggested it, even in their own
language. And yet—!
Who was it told me that the Protectorate and the Company were
enemies at heart, and that the English of Wari are always brooding
on the damages paid to the Niger traders on account of a certain
attack on the people of Brass from Akassa?
No doubt all these are merely such calumnies as are always
circulating.
We shall, all five of us, always remember the welcome we
received at Wari from the agents of the Protectorate, and this
memory will be the more cherished because a few days after our
return to France we heard the terrible news of the death of several of
them, who, having gone on a mission to the interior almost unarmed,
were massacred by the natives of Benin.
We had the best of receptions at Wari; the officers even gave up
their rooms and their very beds to us, knowing how greatly we
should appreciate such comforts. We became much attached to our
new friends.
At Wari I got rid of all the rest of my stores, which would have
been an encumbrance to me on my return journey. There were
plenty for the missionaries and for the servants at the Consulate.
Suzanne, our bicycle, rejoiced the heart of a Sierra Leonese; the
Dantec, with a few bottles of claret, delighted Lieutenant Aron; even
the Aube we left as a token of our friendship with the agents at the
Consulate. We were generous, no doubt, but unless we had sunk
our barges when we got to the sea, what else could we have done
with them?
As for the Davoust, it took us two days to empty, dismantle, and
take her to pieces, after which she was embarked in sections on
board the Axim, a Liverpool steamer, which took her back to Europe.
Sold as old metal, and what she fetched debited to the credit of
the budget of our expedition, all that is left of the Davoust is now
circulating in fairs or figuring in shop-windows, in the form of light
match-boxes and other small articles such as are made of
aluminium.
And this was the end of all the three sturdy barks: Davoust, Aube,
and Dantec, which for twelve whole months were all the world to us!
The Dantec had often seemed likely never to get to the end of her
journey; the Aube certainly ought not to have arrived, judging by the
two or three occasions on which she had seemed done for; at the
end of the voyage you could put your fingers through her rotten
planking. If she had run aground but once more, or if she had got
another blow in passing the last rapid, all would have been over with
her worm-eaten keel, and also with her crew. The Davoust too had
received many wounds, and what was more serious still, oxidation
was beginning to work havoc in her sections.
Ten times at least, face to face with some specially bad rapids, I
had made up my mind to lose one of the three, if not all; but, as the
English said, they were gallant ships. Bravely, in spite of rapids,
whirlpools, and rocks, they had made for the appointed goal, the
mouth of the river, bringing there without faltering the whole
expedition: we white men, the coolies, all our goods, and the French
flag!
No doubt it was Aube, Dantec, and Davoust, their sponsors, our
comrades, who had died at the task of the conquest of the Niger,
who had brought good luck to our three boats.
Thanks to them, I had kept my oath of 1888.
It was not therefore without emotion, without a sadness which
may have been childish, but which many will understand, that we
parted finally with the companions of so many dangers.
Have not boats souls? Sailors love them like old friends, like
heirlooms. We must attach ourselves affectionately to something in
this life, must we not?
The Axim took us to Forcados; the Forcados to Lago; the Olinda,
chartered specially for us, to Porto Novo.
On November 1, at five o’clock in the morning, there was great
excitement at the house of the officers of Porto Novo. Some people
had suddenly arrived, and were banging against the shutters. The
door was soon half-opened and a voice inquired, “Who are
you?”—“Hourst!”—“Where do you come from?”—“Timbuktu”—and
the next moment, without any further questioning, we all fell into
each other’s arms.
After all I experienced in Dahomey and in the Senegal, I will not
dwell too much on the goodness the Governor-General, M. Chaudié,
showed to us on our return, on the kindness he lavished upon every
member of the expedition, or on the reception our friends of St. Louis
gave us later, but I can never thank any of them enough.
We dismissed our coolies at St. Louis, thus effecting an immense
economy. Abdulaye, the carpenter, at once changed his costume for
that of a private citizen. A soft hat, a frock-coat, and a cane with a
silver handle, converted the chrysalis into a butterfly; at the same
time our old servant began to make up for his long months of
sobriety and abstinence. It was, in fact, impossible to find him even
to give him an extra tip.
The rest of our coolies dispersed about the town, holding
receptions in all the public places of the Sarracolais quarter, telling
their adventures with much declamation, and eliciting considerable
applause.
The negroes also, it seems, have their mutual admiration for
geographical societies!
Later all the brave fellows who had been devoted to us to the
death, and some of whom we looked upon as real friends, dispersed
themselves once more amongst the Galam villages dotted along the
banks of the Senegal, and there at least I can confidently assert our
mission, or rather, as Digui called it, the Munition, was and still is
popular.
That is something, at all events.
On December 12, 1896, we landed from the steamer on the quay
of Marseilles, where men were spitting just as they had been when I
left Brest. Looking out of the window of my cab upon the deserted
street, I saw a little Italian boy in the drizzling rain which was falling,
holding in his arms a plaster statuette representing a nude woman
with graceful, supple limbs, probably meant for Diana resting on a
crescent of the moon. She and her bearer looked cold and
melancholy enough. This was my first sight of a really civilized
human being after my three years’ exile.
NATIVES OF AFRICA.
 EPILOGUE

I have now narrated all our adventures, and I leave my readers to

judge of our work. I think it necessary still just to jot down here the
practical conclusions I came to, which may be of use later in French
colonial policy.
To begin with, let us consider how to turn the Niger to account as
a highway for reaching the heart of the Western Sudan.
The French Journal Officiel of Western Africa has published a
report written by Baudry on the possible importations and
exportations, to which I have nothing to add. To every unprejudiced
mind he has clearly proved that there is great wealth of natural
produce to be found in these districts, such as india-rubber, gutta-
percha, skins, wool, wax, karité, cotton, etc., which can easily be
bought, and are, in fact, simply waiting to be developed.
Now which would be the best route to take these products to
France? This is the point we have to elucidate to begin with.
We brought home our hydrographical map of the Niger, from
Timbuktu to Bussa, on a scale of 16 miles to the inch, in fifty sheets.
One glance at it will suffice to show that the river is not really
practically navigable further than Ansongo: that is to say, 435 miles
below the last French port in the Sudan.
Further down than Ansongo the river is simply one hopeless
labyrinth of rocks, islands, reefs, and rapids; and although at the time
of our transit there seemed to be fewer obstacles between Say and
Tchakatchi than elsewhere, it must be remembered that we passed
when the water was at its maximum height. As for the Bussa rapids,
they are simply impassable for laden boats.
“You passed all right, though!” some one said to me; and so we
did, but I think the tour de force by which, thanks to our lucky star,
we achieved our passage under the greatest difficulties, would not
be successful once in three times. We might, however, go down
again once more, but to go up would be quite a different matter.
None but little boats, very lightly laden, or without any cargo, such
as the canoes of the natives, can venture without foolhardiness into
such passes as we came through.
This is certainly not the way in which a river can be
remuneratively navigated. Even if an attempt were made to employ
the primitive means alone likely to succeed, beasts of burden, such
as camels, could compete on disastrous terms with the waterway.
To attempt therefore to turn the river to account in supplying the
central districts with merchandise, or to bring down their products to
the coast, would simply result in failure. To take merchandise up to
Say by means of the lower branches of the river, is but a utopian
dream, which would but result in disaster to those traders involved in
the speculation.
Nature has, in fact, laid her interdict on the navigation of a great
part of the course of the Niger; but at least the 435 navigable miles
above Ansongo, and between it and Timbuktu, added to the 622
between the latter town and Kolikoro, form what may be
characterized as a safe mill-stream, well within the French districts.
We have not as yet nearly realized all the resources of those
districts.
How then shall we get to this mill-stream of ours, or, as we may
perhaps call it, this inland trading lake? A unique solution to the
problem presents itself: we must finish the line of railway uniting
Kayes to Kolikoro.
The first workers at the task of penetrating into Africa were right.
The project of Mungo Park, and Faidherbe, taken up and continued
by the Desbordes, the Gallieni, the Archinards, etc., should be
continued, pushed on and completed without delay.
All has already been explored. We are no longer discussing a
castle in the air, with no firm foundations. We know what that railway
will cost, its whole course has been decided on and surveyed; only