Professional Documents
Culture Documents
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
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 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
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, William Farrell Oncology Medicinal Chemistry,
Pharmaceutical Technology & Development, Pfizer, Inc., Worldwide Medicinal Chemistry,
Operations, AstraZeneca, Macclesfield, La Jolla Laboratories, San Diego, CA,
United Kingdom United States
Raffeal Bennett Merck & Co., Inc., MRL, Ana- Paul Ferguson New Modalities & Parenteral
lytical Research & Development, Boston, MA, Development, Pharmaceutical Technology &
United States Development, Operations, AstraZeneca, Mac-
Martin Beres Separation & Analysis Technology clesfield, United Kingdom
Team, Bristol Myers Squibb, Princeton, NJ, Alexandre Grand-Guillaume-Perrenoud Chem-
United States ical & Analytical Development, Novartis
Pharma AG, Technical R&D, Basel, Switzerland
Blair K. Berger Department of Chemistry & Bio-
chemistry, The University of Texas at Arling- Davy Guillarme School of Pharmaceutical Sci-
ton, Arlington, TX, United States ences; Institute of Pharmaceutical Sciences of
Western Switzerland, University of Geneva,
Astrid Buica Early Chemical Development,
CMU, Geneva, Switzerland
Pharmaceutical Sciences, BioPharmaceuticals
Research & Development, AstraZeneca, Julie M. Herniman Faculty of Engineering &
Gothenburg, Sweden Physical Sciences, School of Chemistry,
University of Southampton, Southampton,
Sergio Cancho-Gonzalez Faculty of Engineer-
United Kingdom
ing & Physical Sciences, School of Chemistry,
University of Southampton, Southampton, Michael Hicks Analytical Research & Develop-
United Kingdom ment, Merck Research Labs, Rahway, NJ,
United States
Adrian Clarke Chemical & Analytical Develop-
ment, Novartis Pharma AG, Technical R&D, Sean Hindley GlaxoSmithKline Medicine Re-
Basel, Switzerland search Centre, Stevenage, Hertfordshire,
United Kingdom
Rebecca Cross Global Chemical Development,
Pharmaceutical Technology & Development, Syame Khater Technologie Servier, Orleans,
Operations, AstraZeneca, Macclesfield, France
United Kingdom Jennifer Kingston Oncology Chemistry, Oncol-
Amandine Dispas University of Liege (ULiege), ogy Research & Early Development,
CIRM, Vibra-Sante Hub, Laboratory of Pharma- AstraZeneca, Cambridge, United Kingdom
ceutical Analytical Chemistry; University of Andy Knaggs GlaxoSmithKline Medicine Re-
Liege (ULiege), CIRM, Mas-Sante Hub, Labora- search Centre, Stevenage, Hertfordshire,
tory for the Analysis of Medicines, Liege, United Kingdom
Belgium G. John Langley Faculty of Engineering &
Brady W. Drennan Department of Chemistry & Physical Sciences, School of Chemistry,
Biochemistry, The University of Texas at University of Southampton, Southampton,
Arlington, Arlington, TX, United States United Kingdom
xi
xii Contributors
Hanna Leek Early Chemical Development, Emily Roddy Early Chemical Development,
Pharmaceutical Sciences, BioPharmaceuticals Pharmaceutical Sciences, Research &
Research & Development, AstraZeneca, Development, AstraZeneca, Macclesfield,
Gothenburg, Sweden United Kingdom
Eric Lesellier University of Orleans, CNRS, Matthew Sanders Oncology Chemistry, Oncol-
Institute of Organic & Analytical Chemistry ogy Research & Early Development,
(ICOA), UMR, Orleans, France AstraZeneca, Cambridge, United Kingdom
Gesa Schad Shimadzu Europa GmbH,
Gioacchino Luca Losacco School of Pharmaceu-
Duisburg, Germany
tical Sciences; Institute of Pharmaceutical Sci-
ences of Western Switzerland, University of Kevin A. Schug Department of Chemistry &
Geneva, CMU, Geneva, Switzerland; Analytical Biochemistry, The University of Texas at
Research and Development, MRL, Merck & Co, Arlington, Arlington, TX, United States
Inc., Rahway, NJ, United States Linda Thunberg Early Chemical Development,
€ en Early Chemical Development, Pharmaceutical Sciences, BioPharmaceuticals
Kristina Ohl
Research & Development, AstraZeneca,
Pharmaceutical Sciences, BioPharmaceuticals
Gothenburg, Sweden
Research & Development, AstraZeneca,
Gothenburg, Sweden Tim Underwood GlaxoSmithKline Medicine
Research Centre, Stevenage, Hertfordshire,
Susan Olesik Department of Chemistry, The United Kingdom
Ohio State University, Columbus, OH,
United States Jean-Luc Veuthey School of Pharmaceutical Sci-
ences; Institute of Pharmaceutical Sciences of
Rebecca Poulten Early Chemical Development, Western Switzerland, University of Geneva,
Pharmaceutical Sciences, Research & CMU, Geneva, Switzerland
Development, AstraZeneca, Macclesfield, Caroline West University of Orleans, CNRS,
United Kingdom Institute of Organic & Analytical Chemistry
Katie Proctor Oncology Chemistry, Oncology (ICOA), UMR, Orleans, France
Research & Early Development, AstraZeneca, Craig White Exscientia AI Ltd, Abingdon,
Cambridge, United Kingdom Oxfordshire, United Kingdom
Joanna Raubo Oncology Chemistry, Oncology A. Paige Wicker Department of Chemistry &
Research & Early Development, AstraZeneca, Biochemistry, The University of Texas at
Cambridge, United Kingdom Arlington, Arlington, TX, United States
Preface
xiii
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 who described the solubilization of cobalt
liquid’s boiling points. He had unknowingly chloride in supercritical ethanol [5]. These
discovered their transition points to a fundamental experiments set in place the
supercritical phase. In a related experiment, cornerstones for further pivotal works in
he heated a sealed glass vial of ethanol noting the 20th century (see Chapter 1).
it expanded to twice its original volume, and In 1957, James Lovelock, one of the pio-
the liquid appeared to vanish. On cooling, a neers of gas chromatography (GC), first pro-
cloud of mist appeared that corresponded to posed the use of supercritical fluids for
the solvent’s critical temperature [2]. In a sub- chromatographic mobile phases for the anal-
sequent paper, he reported the critical temper- ysis of nonvolatile compounds (“critical state
ature of a series of liquids when the interface chromatography”—[6,7]). In 1962, Ernst
tension vanished—visualized through the Klesper and coworkers presented the first
disappearance of the liquid’s meniscus [3]. study on the use of an inorganic gas as a chro-
More than 45 years later, Andrews authored matographic mobile phase [8]. They used
the first systematic study of gas–liquid criti- chlorofluorocarbons (CHClF2 and CCl2F2)
cal point of carbon dioxide (generated from above their critical points (pressures of
carbonic acid) and nitrous oxide [4]. The 800–2300 psi and 115°C) with a 30-in. packed
first application of “supercriticality” was column containing Carbowax 20 M (polyeth-
reported in 1879 by Hannay and Hogarth ylene glycol) on a 180–250 μm Chromosorb W
Preface xv
diatomaceous earth support to separate All the aforementioned experiments,
colored nickel–porphyrin complexes. He among others, generated a curiosity that
termed this “high pressure gas chromato- paved the way for numerous advances in
graphy.” The elution strength was found to the field. At the heart of these develop-
be proportional to the system pressure, ments were improvements in understanding
which was controlled through restriction the impact of organic solvent on the mobile
capillaries and mobile phase flow rate [6]. phase elution strength, the role of mobile
As the work was undertaken on modified phase additives, and the significant role
gas-chromatographic instrumentation (GC), stationary phase interactions play on analyte
the technique was coined “high pressure retention and selectivity. Alongside these
GC”. This is considered the first practical theoretical developments, improvements in
demonstration of what was later to become instrument design saw the transition from
known as SFC. early modification of gas chromatographic
The term SFC was first proposed by Sie instruments, to modified HPLC instrumen-
and coworkers at Shell Research Laboratories tation and then the concerted development
(Amsterdam, The Netherlands) in a series of of instrumentation dedicated to the task.
papers in the late 1960s. In these experiments, At each juncture, improved understanding
they investigated the effects of pressure and of CO2 fluid metering allowed for more
temperature in open-tubular (squalene and accurate delivery of volumes of the com-
glycerol coated, and later packed column) pressible gas mobile phase and cosolvent
gas chromatography with carbon dioxide as modifiers. These advances were supported
the carrier gas [9–11]. They described how in- by developments in thermostated high-
creasing the carbon dioxide pressure altered pressure UV flow cells for improved detec-
the density of the gas and reduced analyte tion sensitivity, improved pump check
retention. At a similar timepoint ca. 1968, valves materials, and advances in back
J. Calvin Giddings group proposed the term pressure system regulation.
“dense gas chromatography” [12] and soon
after reported the influence of high pressures
on analyte retention in GC with carbon diox- SFC in the pharmaceutical industry
ide and ammonia as mobile phase/carrier
gas [13]. Having been available for many decades,
As is often the case in early research, the late SFC is a chromatographic technique that
1960s through to the mid-1980s saw the rise of many analytical chemists are aware of, but
certain claims that ultimately hindered the may have not yet used. It is often considered
pace of development of the technique, notably a niche technique with only limited applica-
the assertion that the solvent strength of car- tion for preparative and chiral chromatogra-
bon dioxide was similar to isopropyl alcohol phy. SFC has a perceived complexity over
[14], publications highlighting the deficiencies more established techniques such as GC
of packed columns on chromatographic per- and liquid chromatography (LC), and these
formance [15], and the belief that modifiers perceptions have prevented its application
did not increase mobile phase solvent strength as a standard lab technique. However, if an
[16]. Similarly, there were limitations in the analyst has experience in liquid chromatog-
available instrumentation, which meant that raphy, then much of the knowledge and in-
the full benefits of supercritical mobile phases strumental understanding is relevant and
could not be exploited [17]. transferable to SFC, and application to SFC
xvi Preface
is not as high a barrier to overcome as one due to the low viscosity of the mobile
might believe. phase allowing high flow rates to be used;
SFC has several interesting aspects as a • Effective with a wide range of organic
separation technique that are pertinent to solvents commonly used in synthetic
the analysis of both small and large phar- chemistry reactions, and is particularly
maceutical molecules as well as related useful for analysis of water labile
materials such as certain formulation excipi- molecules;
ents. In many cases, SFC can provide a • Virtually no stationary phase limitations
better solution than LC for the characteriza- in terms of selection choice, using carbon
tion of pharmaceutically relevant molecules. dioxide–based mobile phases (see
Fig. P2 shows the number of publications Chapter 3);
reporting the use of SFC for pharmaceutical • Compatible with many sample diluents
analysis year on year from 1985. There is a ranging from fully organic to fully
clear exponential upward trend reflecting aqueous samples (depending on mobile
the growing interest and importance of the phase composition, see Chapter 8);
technique in the industry. In fact, application • As a result of the low mobile phase
of the technique is expanding so rapidly in viscosity, highly efficient columns with
pharmaceutical science that we found it chal- sub-2 μm stationary phases can be
lenging to propose possible future applica- successfully employed to achieve higher
tions areas (see Chapter 13). efficiency and resolution than analogous
Some of the aspects that make SFC attrac- LC separations (see Chapter 7);
tive for pharmaceutical analysis include the • Is becoming the industry technique-of-
following: choice for chiral and preparative separations
due to the technique using less solvent
• Typically faster analysis times and
and instrumental energy than the analogous
gradient reequilibration than (U)HPLC
LC separations (see Chapters 5 and 6);
FIG. P2 CAS Scifinder search using keywords “SFC” + “pharmaceutical” for the period 1985–2021.
Preface xvii
• Produces less solvent waste than LC and you enjoy the book and find it helpful in
uses a “recyclable” gas as the primary expanding your knowledge of SFC.
mobile phase component. In addition, Michael Hicks
alcohols are often used as a cosolvent that Rahway, NJ, United States
further improves its “green” credentials. Paul Ferguson
Macclesfield, United Kingdom
November 2022
Scope and relevance for this text
fluids under conditions of incipient turbulence, Sep. [14] J.C. Giddings, M.N. Myers, J.W. King, Dense gas
Sci. 2 (1967) 699–727. chromatography of pressures to 2000 atmospheres,
[11] S.T. Sie, G.W.A. Rijnders, High-pressure gas chro- J. Chromatogr. Sci. 7 (1969) 276–283.
matography and chromatography with supercriti- [15] M. Novotny, W. Bertsch, A. Zlatkis, Temperature
cal fluids. IV. Fluid-solid chromatography, Sep. and pressure effects in supercritical-fluid chroma-
Sci. 2 (1967) 755–777. tography, J. Chromatogr. 61 (1971) 17–28.
[12] L. McLaren, M.N. Myers, J.C. Giddings, Dense- [16] B.W. Wright, R.D. Smith, Investigation of polar
gas chromatography of nonvolatile substances modifiers in carbon dioxide mobile phases for
of high molecular weight, Science 159 (1968) capillary supercritical fluid chromatography,
197–199. J. Chromatogr. 355 (1986) 367–373.
[13] J.C. Giddings, M.N. Myers, L.M. McLaren, R.A. [17] T.A. Berger, Separation of polar solutes by packed
Keller, High pressure gas chromatography of non- column supercritical fluid chromatography,
volatile species, Science 162 (1968) 67–73. 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
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
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
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
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)
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.
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.
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].
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.