Biopharmaceutical Scaleup and Optimisation

Marcel Dekker, Inc. New York BaselTM

edited byAnurag S. RathorePharmacia CorporationChesterfield, Missouri, U.S.A.

Ajoy VelayudhanOregon State UniversityCorvallis, Oregon, U.S.A.

Scale-Up and Optimization inPreparative ChromatographyPrinciples and Biopharmaceutical Applications

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.


ISBN: 0-8247-0826-1

This book is printed on acid-free paper.

HeadquartersMarcel Dekker, Inc.270 Madison Avenue, New York, NY 10016tel: 212-696-9000; fax: 212-685-4540

Eastern Hemisphere DistributionMarcel Dekker AGHutgasse 4, Postfach 812, CH-4001 Basel, Switzerlandtel: 41-61-260-6300; fax: 41-61-260-6333

World Wide Webhttp:/ /www.dekker.com

The publisher offers discounts on this book when ordered in bulk quantities. For moreinformation, write to Special Sales/Professional Marketing at the headquarters addressabove.


Neither this book nor any part may be reproduced or transmitted in any form or byany means, electronic or mechanical, including photocopying, microlming, and re-cording, or by any information storage and retrieval system, without permission inwriting from the publisher.

Current printing (last digit):10 9 8 7 6 5 4 3 2 1

PRINTED IN THE UNITED STATES OF AMERICA


CHROMATOGRAPHIC SCIENCE SERIES

A Series of Textbooks and Reference Books

Editor: JACK CAZES

1. Dynamics of Chromatography: Principles and Theory, J. Calvin Giddings2. Gas Chromatographic Analysis of Drugs and Pesticides, Benjamin J. Gudzinowicz3. Principles of Adsorption Chromatography: The Separation of Nonionic Organic

Compounds, Lloyd R. Snyder4. Multicomponent Chromatography: Theory of Interference, Friedrich Helfferich and

Gerhard Klein5. Quantitative Analysis by Gas Chromatography, Josef Novk6. High-Speed Liquid Chromatography, Peter M. Rajcsanyi and Elisabeth Rajcsanyi7. Fundamentals of Integrated GC-MS (in three parts), Benjamin J. Gudzinowicz, Mi-

chael J. Gudzinowicz, and Horace F. Martin8. Liquid Chromatography of Polymers and Related Materials, Jack Cazes9. GLC and HPLC Determination of Therapeutic Agents (in three parts), Part 1 edited

by Kiyoshi Tsuji and Walter Morozowich, Parts 2 and 3 edited by Kiyoshi Tsuji10. Biological/Biomedical Applications of Liquid Chromatography, edited by Gerald L.

Hawk11. Chromatography in Petroleum Analysis, edited by Klaus H. Altgelt and T. H. Gouw12. Biological/Biomedical Applications of Liquid Chromatography II, edited by Gerald L.

Hawk13. Liquid Chromatography of Polymers and Related Materials II, edited by Jack Cazes

and Xavier Delamare14. Introduction to Analytical Gas Chromatography: History, Principles, and Practice,

John A. Perry15. Applications of Glass Capillary Gas Chromatography, edited by Walter G. Jennings16. Steroid Analysis by HPLC: Recent Applications, edited by Marie P. Kautsky17. Thin-Layer Chromatography: Techniques and Applications, Bernard Fried and Jo-

seph Sherma18. Biological/Biomedical Applications of Liquid Chromatography III, edited by Gerald L.

Hawk19. Liquid Chromatography of Polymers and Related Materials III, edited by Jack

Cazes20. Biological/Biomedical Applications of Liquid Chromatography, edited by Gerald L.

Hawk21. Chromatographic Separation and Extraction with Foamed Plastics and Rubbers, G.

J. Moody and J. D. R. Thomas22. Analytical Pyrolysis: A Comprehensive Guide, William J. Irwin23. Liquid Chromatography Detectors, edited by Thomas M. Vickrey24. High-Performance Liquid Chromatography in Forensic Chemistry, edited by Ira S.

Lurie and John D. Wittwer, Jr.25. Steric Exclusion Liquid Chromatography of Polymers, edited by Josef Janca26. HPLC Analysis of Biological Compounds: A Laboratory Guide, William S. Hancock

and James T. Sparrow27. Affinity Chromatography: Template Chromatography of Nucleic Acids and Proteins,

Herbert Schott28. HPLC in Nucleic Acid Research: Methods and Applications, edited by Phyllis R.

Brown29. Pyrolysis and GC in Polymer Analysis, edited by S. A. Liebman and E. J. Levy

30. Modern Chromatographic Analysis of the Vitamins, edited by Andr P. DeLeenheer, Willy E. Lambert, and Marcel G. M. De Ruyter

31. Ion-Pair Chromatography, edited by Milton T. W. Hearn32. Therapeutic Drug Monitoring and Toxicology by Liquid Chromatography, edited by

Steven H. Y. Wong33. Affinity Chromatography: Practical and Theoretical Aspects, Peter Mohr and Klaus

Pommerening34. Reaction Detection in Liquid Chromatography, edited by Ira S. Krull35. Thin-Layer Chromatography: Techniques and Applications. Second Edition, Re-

vised and Expanded, Bernard Fried and Joseph Sherma36. Quantitative Thin-Layer Chromatography and Its Industrial Applications, edited by

Laszlo R. Treiber37. Ion Chromatography, edited by James G. Tarter38. Chromatographic Theory and Basic Principles, edited by Jan ke Jnsson39. Field-Flow Fractionation: Analysis of Macromolecules and Particles, Josef Janca40. Chromatographic Chiral Separations, edited by Morris Zief and Laura J. Crane41. Quantitative Analysis by Gas Chromatography, Second Edition, Revised and

Expanded, Josef Novk42. Flow Perturbation Gas Chromatography, N. A. Katsanos43. Ion-Exchange Chromatography of Proteins, Shuichi Yamamoto, Kazuhiro Naka-

nishi, and Ryuichi Matsuno44. Countercurrent Chromatography: Theory and Practice, edited by N. Bhushan Man-

dava and Yoichiro Ito45. Microbore Column Chromatography: A Unified Approach to Chromatography, edi

ted by Frank J. Yang46. Preparative-Scale Chromatography, edited by Eli Grushka47. Packings and Stationary Phases in Chromatographic Techniques, edited by Klaus

K. Unger48. Detection-Oriented Derivatization Techniques in Liquid Chromatography, edited by

Henk Lingeman and Willy J. M. Underberg49. Chromatographic Analysis of Pharmaceuticals, edited by John A. Adamovics50. Multidimensional Chromatography: Techniques and Applications, edited by Hernan

Cortes51. HPLC of Biological Macromolecules: Methods and Applications, edited by Karen M.

Gooding and Fred E. Regnier52. Modern Thin-Layer Chromatography, edited by Nelu Grinberg53. Chromatographic Analysis of Alkaloids, Milan Popl, Jan Fhnrich, and Vlastimil

Tatar54. HPLC in Clinical Chemistry, I. N. Papadoyannis55. Handbook of Thin-Layer Chromatography, edited by Joseph Sherma and Bernard

Fried56. GasLiquidSolid Chromatography, V. G. Berezkin57. Complexation Chromatography, edited by D. Cagniant58. Liquid ChromatographyMass Spectrometry, W. M. A. Niessen and Jan van der

Greef59. Trace Analysis with Microcolumn Liquid Chromatography, Milos KrejcI60. Modern Chromatographic Analysis of Vitamins: Second Edition, edited by Andr P.

De Leenheer, Willy E. Lambert, and Hans J. Nelis61. Preparative and Production Scale Chromatography, edited by G. Ganetsos and P.

E. Barker62. Diode Array Detection in HPLC, edited by Ludwig Huber and Stephan A. George63. Handbook of Affinity Chromatography, edited by Toni Kline64. Capillary Electrophoresis Technology, edited by Norberto A. Guzman65. Lipid Chromatographic Analysis, edited by Takayuki Shibamoto66. Thin-Layer Chromatography: Techniques and Applications: Third Edition, Revised

and Expanded, Bernard Fried and Joseph Sherma67. Liquid Chromatography for the Analyst, Raymond P. W. Scott68. Centrifugal Partition Chromatography, edited by Alain P. Foucault69. Handbook of Size Exclusion Chromatography, edited by Chi-San Wu70. Techniques and Practice of Chromatography, Raymond P. W. Scott71. Handbook of Thin-Layer Chromatography: Second Edition, Revised and Expanded,

edited by Joseph Sherma and Bernard Fried

72. Liquid Chromatography of Oligomers, Constantin V. Uglea73. Chromatographic Detectors: Design, Function, and Operation, Raymond P. W.

Scott74. Chromatographic Analysis of Pharmaceuticals: Second Edition, Revised and

Expanded, edited by John A. Adamovics75. Supercritical Fluid Chromatography with Packed Columns: Techniques and Appli-

cations, edited by Klaus Anton and Claire Berger76. Introduction to Analytical Gas Chromatography: Second Edition, Revised and Ex-

panded, Raymond P. W. Scott77. Chromatographic Analysis of Environmental and Food Toxicants, edited by Taka-

yuki Shibamoto78. Handbook of HPLC, edited by Elena Katz, Roy Eksteen, Peter Schoenmakers, and

Neil Miller79. Liquid ChromatographyMass Spectrometry: Second Edition, Revised and

Expanded, Wilfried Niessen80. Capillary Electrophoresis of Proteins, Tim Wehr, Roberto Rodrguez-Daz, and

Mingde Zhu81. Thin-Layer Chromatography: Fourth Edition, Revised and Expanded, Bernard

Fried and Joseph Sherma82. Countercurrent Chromatography, edited by Jean-Michel Menet and Didier

Thibaut83. Micellar Liquid Chromatography, Alain Berthod and Celia Garca-Alvarez-Coque84. Modern Chromatographic Analysis of Vitamins: Third Edition, Revised and

Expanded, edited by Andr P. De Leenheer, Willy E. Lambert, and Jan F. VanBocxlaer

85. Quantitative Chromatographic Analysis, Thomas E. Beesley, Benjamin Buglio,and Raymond P. W. Scott

86. Current Practice of Gas ChromatographyMass Spectrometry, edited by W. M. A.Niessen

87. HPLC of Biological Macromolecules: Second Edition, Revised and Expanded,edited by Karen M. Gooding and Fred E. Regnier

88. Scale-Up and Optimization in Preparative Chromatography: Principles and Bio-pharmaceutical Applications, edited by Anurag S. Rathore and Ajoy Velayudhan

89. Handbook of Thin-Layer Chromatography: Third Edition, Revised and Expanded,edited by Joseph Sherma and Bernard Fried

ADDITIONAL VOLUMES IN PREPARATION

Chiral Separations by Liquid Chromatography and Related Technologies, HassanY. Aboul-Enein and Imran Ali

To Csaba HorvathMentor and Friend


Preface

Preparative chromatography is arguably the most widely used puricationtechnique in the pharmaceutical and biotechnological industries. Scaling up apreparative separation, however, continues to be seen as a difcult and time-consuming task. While this perception may have had some validity a genera-tion ago, it is not accurate today. In fact, many industrial separations can bescaled up using one of the simple methods that have been described in theliterature. The goals of this book are to bring wider recognition to these simplemethods and to show that they are effective in many practical problems.

Another perception that persists is the view that results obtained in aca-demia are either inapplicable or not applied to the day-to-day problems facedby industrial professionals. While there is no doubt that a greater degree ofcollaboration between academia and industry is desirable, it is neverthelesstrue that useful results obtained in academia have been, and are being, usedin industry. Equally, the results and experience gained by industrial expertshave informed and rened the academic approach to these problems. The rstsix chapters of this book describe the state of the art in methods and approachesto scale-up and optimization of preparative chromatography. These chaptersare followed by a set of industrial case studies, which show how scale-up iscarried out for a variety of important separations.

The editors contribute an overview that includes a simple quantitativeapproach as well as a discussion of the various practical aspects of scale-up.Signposts are provided to later chapters in which more detail is provided onspecic topics that are discussed in earlier chapters.


Lightfoot et al. contribute an incisive overview to mass-transfer effectsin the current chromatographic context. In addition to laying out problemsthat are currently being addressed, the chapter outlines fruitful lines of futurestudy.

There has been a great deal of work on numerical optimization of nonlin-ear chromatographic separations, aimed at making it possible to obtain theproduction rates and yields for every operational mode for a given separation,thereby allowing a rational choice of the best operational mode for that separa-tion. Felinger summarizes such work and indicates how these results can beused to advantage in guiding the scale-up process.

Simulated moving beds (SMBs) are becoming increasingly popular inpreparative work, and may even have become the method of choice for someenantiomeric separations. Antias chapter describes clearly and succinctly theissues involved in designing and controlling SMB units.

Watler et al. give a detailed analysis of the theoretical issues that mustbe addressed in successfully scaling up ion-exchange separations. The varioustheoretical issuesselectivity, bandspreading, optimizationare presented ina way that allows direct application to other separations.

Levison describes a variety of ways in which ion-exchange adsorbentsare used at large scale, going beyond batch and column modes to suspendedand uidized beds. This detailed assessment provides a valuable complementto Watler et al.s analysis of theory applicable to ion-exchange chromatog-raphy.

The second section of the book contains a set of industrial case studiesthat provide the practical approach taken in industry to scale-up of realisticseparations. Fahrner et al. describe in detail the development of the crucialcation-exchange step in the kilogram-scale purication of a recombinant anti-body fragment. Useful discussions of appropriate analytical methods andscouting for stationary phases are followed by the development of an opti-mized stepwise elution protocol. The purication of supramolecular assembl-ies is a topic of great current interest. Sagar et al. report their experienceswith large-pore adsorbents in the purication of plasmid DNA. The entirepurication sequence for producing monoclonal antibodies to tumor necrosisfactor (TNF) is detailed by Ng. A variety of ion-exchange steps as well as asize exclusion step were used in this process. Miller and Murphy report on anormal-phase separation of a synthetic intermediate that was scaled up to thekilogram level. Guhan and Guinn describe the removal of a trace impurityfrom the parenteral drug alatrooxacin. Rathore provides a detailed develop-ment of how a chromatographic process was implemented and optimized,based on the simple approach to scale-up described in Chapter 1.


The book as a whole is intended to provide a useful summary of themethods and approaches current in preparative chromatography, enabling thereader to transfer the insights gained here to her or his specic separationproblem. In addition to serving as a reference for industrial practitioners ofpreparative chromatography, the material should be readily accessible to grad-uate students in many disciplines including chemistry, pharmacy, bioengineer-ing, and chemical engineering.

As the editors of this book, we wish to express our indebtedness toProfessor Csaba Horvath of Yale University, in whose research group we bothreceived our doctorates, and to whom the book is dedicated. There is littledoubt that science is one of the areas in which apprenticeship to an expertcontinues to be the premier way to gain, both explicitly and osmotically, in-sight into the eld. Discussing scientic problems with Csabahow tochoose, approach, and solve themwas a vital part of our training in becom-ing independent researchers.

Anurag S. RathoreAjoy Velayundhan


Contents

PrefaceContributors

Part I Methods and Approaches

1 An Overview of Scale-Up in Preparative ChromatographyAnurag S. Rathore and Ajoy Velayudhan

2 Interaction of Mass Transfer and Fluid MechanicsEdwin N. Lightfoot, John S. Moscariello, Mark A. Teeters,and Thatcher W. Root

3 Optimization of Preparative SeparationsAttila Felinger

4 Engineering Aspects of Ion-Exchange ChromatographyPeter Watler, Shuichi Yamamoto, Oliver Kaltenbrunner,and Daphne N. Feng

5 A Simple Approach to Design and Control of SimulatedMoving Bed ChromatographsFiroz D. Antia


6 Large-Scale Ion-Exchange Chromatography: A Comparisonof Different Column FormatsPeter R. Levison

Part II Case Studies

7 Development and Operation of a Cation-ExchangeChromatography Process for Large-Scale Puricationof a Recombinant Antibody FragmentRobert L. Fahrner, Stacey Y. Ma, Michael G. Mulkerrin,Nancy S. Bjork, and Gregory S. Blank

8 Case Study: Capacity Challenges in Chromatography-BasedPurication of Plasmid DNASangeetha L. Sagar, Ying G. Chau, Matthew P. Watson,and Ann L. Lee

9 Case Study: Purication of an IgG1 Monoclonal AntibodyPaul K. Ng

10 Case Study: Normal Phase Purication of KilogramQuantities of a Synthetic Pharmaceutical IntermediateLarry Miller and James Murphy

11 Case Study: Development of Chromatographic Separationto Remove Hydrophobic Impurities in AltraoxacinSam Guhan and Mark Guinn

12 Case Study: Process Development of Chromatography Stepsfor Purication of a Recombinant E. coli Expressed ProteinAnurag S. Rathore


Contributors

Firoz D. Antia, Ph.D. Department of Chemical Engineering Research andDevelopment, Merck & Co., Inc., Rahway, New Jersey, U.S.A.

Nancy S. Bjork Department of Analytical Chemistry, Genentech, Inc.,South San Francisco, California, U.S.A.

Gregory S. Blank, Ph.D. Department of Recovery Sciences, Genentech,Inc., South San Francisco, California, U.S.A.

Ying G. Chau Department of Chemical Engineering, Massachusetts Insti-tute of Technology, Boston, Massachusetts, U.S.A.

Robert L. Fahrner Department of Recovery Sciences, Genentech, Inc.,South San Francisco, California, U.S.A.

Attila Felinger, Ph.D. Department of Analytical Chemistry, University ofVeszprem, Veszprem, Hungary

Daphne W. Feng, B.S. Department of Process Development, Amgen, Inc.,Thousand Oaks, California, U.S.A.

Sam Guhan, Ph.D. Department of Bioprocess Research and Development,Pzer Global Research and Development, Groton, Connecticut, U.S.A.


Mark Guinn, Ph.D. Department of Bioprocess Research and Development,Pzer Global Research and Development, Groton, Connecticut, U.S.A.

Oliver Kaltenbrunner, Dipl. Ing. Dr. Department of Process Develop-ment, Amgen, Inc., Thousand Oaks, California, U.S.A.

Ann L. Lee, Ph.D. Merck Research Laboratories, Merck & Co., Inc., WestPoint, Pennsylvania, U.S.A.

Peter R. Levison, B.Sc., M.B.A., Ph.D. Department of Science and Tech-nology, Whatman International Ltd., Maidstone, Kent, England

Edwin N. Lightfoot, Ph.D. Department of Chemical Engineering, Univer-sity of WisconsinMadison, Madison, Wisconsin, U.S.A.

Stacey Y. Ma, Ph.D. Department of Analytical Chemistry, Genentech, Inc.,South San Francisco, California, U.S.A.

Larry Miller Department of Global Supply Early Process Research and De-velopment, Pharmacia Corporation, Skokie, Illinois, U.S.A.

John S. Moscariello Department of Chemical Engineering, University ofWisconsinMadison, Madison, Wisconsin, U.S.A.

Michael G. Mulkerrin, Ph.D. Department of Analytical Chemistry, Genen-tech, Inc., South San Francisco, California, U.S.A.

James Murphy Pharmacia Corporation, Skokie, Illinois, U.S.A.

Paul K. Ng Purication Development, Department of Biotechnology, Phar-maceutical Division, Bayer Corporation, Berkeley, California, U.S.A.

Anurag S. Rathore, Ph.D. Department of Bioprocess Sciences, PharmaciaCorporation, Chestereld, Missouri, U.S.A.

Thatcher W. Root, Ph.D. Department of Chemical Engineering, Universityof WisconsinMadison, Madison, Wisconsin, U.S.A.

Sangeetha L. Sagar, Ph.D. Merck Research Laboratories, Merck & Co.,Inc., West Point, Pennsylvania, U.S.A.


Mark A. Teeters, B.S. Department of Chemical Engineering, University ofWisconsinMadison, Madison, Wisconsin, U.S.A.

Ajoy Velayudhan, Ph.D. Department of Bioresource Engineering, OregonState University, Corvallis, Oregon, U.S.A.

Peter Watler, B.S., M.S., Ph.D. Department of Process Development, Am-gen, Inc., Thousand Oaks, California, U.S.A.

Matthew P.Watson, B.S. Merck Research Laboratories, Merck & Co., Inc.,West Point, Pennsylvania, U.S.A.

Shuichi Yamamoto, Ph.D. Department of Chemical Engineering, Yama-guchi University, Ube, Japan


1An Overview of Scale-Up inPreparative Chromatography

Anurag S. RathorePharmacia Corporation, Chestereld, Missouri, U.S.A.

Ajoy VelayudhanOregon State University, Corvallis, Oregon, U.S.A.

I. INTRODUCTION

Preparative chromatography continues to be the dominant purication tech-nique in the production of biological compounds, especially in the pharmaceu-tical and biotechnological industries. However, the conceptual complexity ofa purely theoretical approach to preparative chromatography is formidable,because we are dealing with systems of highly coupled, nonlinear partial dif-ferential equations [1,2]. Although theoretical work is progressing, it can cur-rently capture predictively only a few aspects of realistic biotechnological sep-arations, especially given the extremely complex biochemical feedstocks oftenused in these applications. It is not entirely coincidental that the current ap-proach to scale-up and optimization in industry is highly empirical. Althoughthis is natural, especially given the constraints of process validation, the rstfew chapters of this book attempt to show that current theoretical understand-ing does give insight into the practical issues involved in scale-up and optim-ization. These chapters show that a careful combination of basic theory withexperiments can reduce the time needed to achieve an effective scale-up of arealistic chromatographic separation.


It will be convenient to introduce some terminology [3] to clarify theensuing discussion. The various kinds of physiochemical interactions that areused in chromatography to produce selectivity are called modes of interaction.Examples include electrostatic interactions in ion-exchange or ion chromatog-raphy, hydrophobic interactions in reversed-phase and hydrophobic interactionchromatography, and specic interactions in afnity chromatography. Once amode of interaction has been chosen, the various ways in which a separationcan be achieved (isocratic or gradient elution, stepwise elution, displacement,frontal analysis) are called modes of operation. For many separations, the bestmode of interaction is easily specied, and scale-up or optimization focuseson the choice of mode of operation.

Finally, when the concentrations of all adsorbable components are lowenough to lie within the linear or Henrys law region of their respective adsorp-tion isotherms, the separation is called linear. Even if one components con-centration reaches the nonlinear region of its (multicomponent) adsorptionisotherm for some fraction of the separation, the process is called nonlinear.

The basic ideas for scale-up and optimization given in the beginningchapters are applied to real separations in the subsequent chapters in whichindustrial case studies are presented. An issue of practical importance in aseparation sequence is that of how to achieve the global optimum in the param-eter of interest (typically maximum productivity or maximum recovery or min-imum cost; mixed or combined optimization criteria are also possible). Thisissue is not discussed in detail in this chapter, because Chapter 3 deals withit comprehensively. Further, the case studies in subsequent chapters often al-lude to constraints from one separation step limiting or otherwise affectingthe choice of conditions in other steps.

The structure of this chapter is as follows. An introductory section onmethod development places in perspective the various steps involved in arriv-ing at an effective separation protocol at the bench scale. This is, of course,a necessary preliminary to scale-up, which by denition seeks to maintainupon scale-up the quality of a separation that has already been developed.Section III begins with heuristic rules for scale-up and then develops a simplequantitative model that claries when such heuristic rules can be used withreasonable accuracy. The issue of bed heterogeneity and its implications forscale-up are also discussed. In Section IV practical considerations characteris-tic of the various modes of interaction and operation are described briey.Although considerations of space preclude the full discussion of all these is-sues, key points are brought out and important references in the literature arehighlighted.


II. METHOD DEVELOPMENT

Method development is a multistep process that precedes scale-up and opera-tion at large scale. The general practice is to perform optimization at smallscale due to relatively smaller requirements of material and resources as wellas the ease of performing several runs in a parallel fashion.

A. Decoupling of Thermodynamics from KineticsA variety of parameterschoice of stationary and mobile phases, the particlesize of the stationary phase, the column dimensions, the ow rate, the feedloadingaffect the production rate and recovery obtained in a preparativeseparation. Trying to understand the interplay of all of these parameters simul-taneously is a daunting task. In addition, testing all the various possibilitiesexperimentally is likely to be extremely tedious and is impractical under typi-cal industrial constraints. However, the following simplication is availableto us at little cost. It is likely that equilibrium parameters (the choice of station-ary and mobile phases, leading to selectivity) can be selected independent ofkinetic parameters such as ow rate, feed loading, and particle size. Sucha decoupling of thermodynamic and kinetic parameters is probably rigorouslyjustiable only in linear chromatography, but even in nonlinear chromatogra-phy it is likely that choosing the mobile and stationary phases rst does notsignicantly decrease the attainable production rates and recoveries.

The rst step is therefore to choose the most effective mode of interac-tion. This is often clear from the fundamental properties of the feedstock orthe product. Other important factors include the objective and nature of theseparation problem, literature precedents, and prior experience with the prod-uct. Inputs from the vendors of chromatographic media and instrumentationmay also be useful at this stage. The strategy is depicted schematically in Fig.1 [4].

B. Optimization of Thermodynamics at Bench ScaleWe present here a simple and rapid approach to the thermodynamic componentof method development. We take the view that for many separations the choiceof stationary phase is far more important than the choice of mobile phase (thisis particularly true of ion-exchange runs, where standard salts are used asmobile phase modulators). Of course, there are many cases where specicbinding of various kinds can require the use of special additives for the mobile


Figure 1. Strategy for optimization of a chromatographic separation.

phase, but we ignore these situations in order to make the general approachclear. We therefore intend to use a standard mobile phase, and we wish toscreen a variety of stationary phases rapidly and equitably, i.e., we have re-duced the problem to one of resin screening.

Once a list of resin candidates has been prepared, screening is performedto select the best resin to perform a particular separation. Selection of theresin for a chromatography step is perhaps the most important step in methodoptimization [49]. A resin screening protocol is illustrated in Fig. 2. In mostcases the primary criterion for resin screening is selectivity. However, otherscreening criteria may also be identied and used depending on the particularseparation problem.

The general approach is as follows (the specics in what follows arefor ion-exchange chromatography in the gradient mode of operation, but thearguments can easily be generalized to other contexts). The process takes placein two stages.

Stage 1. All stationary phases are packed into columns of identicalsize. If possible, all columns should be run at the same ow rate. This is notalways practical (e.g., if the particle sizes available for different stationaryphases are markedly different, then pressure drop constraints may limit therange of ow rates). Run a test gradient that spans a wide range of modulatorlevels, so that feed retention is facilitated. Make the gradient as shallow as


Figure 2. Resin screening protocol. (Reprinted courtesy LCGC North America, Ad-vanstar Communications, Inc.)

practicable, in order to simultaneously get as much resolution as possible un-der these conditions. Stationary phases that exhibit little or no retention of theproduct are excluded at this stage. In addition, if almost no resolution is foundbetween the product and the primary impurities, these stationary phases maybe excluded. This latter decision should be made carefully, because the testgradient may not be a fair indicator of a sorbents resolution. In other words,a sorbent may provide poor resolution of the product under the test gradientbut high resolution under another gradient. Thus, the latter decision is to bemade only if there is good reason to believe that this sorvent is unlikely tobe effective.

Stage 2. For each of the stationary phases remaining, determine atailored comparison gradient that is intended to show each sorbent under itsmost effective conditions for the given feed mixture. Parameters such as the


feed loading and equilibration buffer should be kept the same for all sta-tionary phases. If the ow rate was the same for all runs in stage 1, then itshould be maintained in this stage. If different sorbents were run with differ-ent ow rates in stage 1, then use the same ow rate for each sorbent in thisstage.

The comparison gradient is centered around the modulator concentrationat which the product eluted in the test gradient in stage 1. Then, making theassumption that the band spreading of the peaks is inversely proportional tothe gradient slope, all other parameters being constant, we have

w m (1)where and w are respectively the gradient slope and product peak width inthe test gradient, and and m are the corresponding parameters in the compari-son gradient. If we require that all comparison gradients have the same time(for standardization), then the starting and ending modulator concentrations(cx and cy, respectively) can be determined from the equations

cx celution ntwF2mV

(2)

and

cy celution ntwF2mV

(3)

where celution and tW are respectively the concentration and time at which thecenter of the product peak eluted in the test gradient, n is the number of columnvolumes over which the comparison gradient is run, F is the ow rate, and Vis the column volume. Note that if the beginning concentration cx is found tobe negative from Eq. (2), it is set to zero.

The comparison gradient provides an equitable way of comparing differ-ent stationary phases for the given feed mixture, because each stationary phaseis provided with a gradient that is optimized for its particular retention behav-ior. Now the usual quantitative parameters of production rates and recoveryand purity can be used to determine which stationary phase is best.

The simple approach described above provides a rapid way to choosethe best resin. However, if the chromatography step is intended for operationat preparative scale, particulary for commercial manufacture, several otherissues must be addressed before nal resin selection. These include the costof the resin, the physical and chemical stability of the resin at the bed heightand the number of cycles to be used at the manufacturing plant, media avail-ability with respect to the demand at commercial scale, resin lifetime, leaching


of ligands, regulatory support les offered by the vendor, batch-to-batch vari-ations in resin quality, etc.

It should be noted that the assumption that peak width is inversely propor-tional to gradient slope is an approximation and is not always expected to bevalid (e.g., signicant competition among the product and impurities for bind-ing sites on the adsorbent could cause the assumption to fail). However, it islikely to be a reasonable approximation for many realistic separations. Moredetailed methods of this kind can be established (see, e.g., Quarry et al. [10])but would usually require more data for each sorbent. Similarly, Jandera et al.[11] and Jandera [12] determined optimal gradients in normal and reversed-phase systems through numerical optimization of the governing equations; thisis a signicant advance in the eld but is not yet at the level of accessibilitywhere industrial practitioners would use it routinely. The method outlined herewas chosen for its simplicity and ease of use in an industrial context.

This approach to resin screening is now demonstrated in detail for apractical separation problem. Rathore [4] showed that the stationary phase ofchoice for an anion-exchange separation was found rapidly using this ap-proach.

1. Resin Screening for an Anion-Exchange Chromatography ColumnThis case study presents data obtained during the optimization of an anion-exchange chromatography column used in the process of purifying a proteinmolecule derived from microbial fermentation.

Nine anion-exchange resinsBioRad High Q, BioRad DEAE, Phar-macia DEAE FF, Pharmacia Q FF, Pharmacia Q HP, Whatman Q, WhatmanQA52, Whatman DE53, and TosoHaas Q650Mwere chosen for screening.All chromatography experiments were performed using an A kta Explorer(Amersham Pharmacia Biotech). The buffer and other operating conditionswere chosen on the basis of prior experience with the molecule. (Pre-equilibra-tion buffer: 1 M Tris, pH 8.5. Equilibration buffer: 50 mMTris, pH 8.5. Proteinloading: 10 mg/mL resin.) Because the objective of this chapter is to lay outan efcient resin screening protocol and not to recommend a particular resin,the resins that were used will be referred to as resins 19 (not in the order inwhich they are named above). The optimum resin is expected to vary withthe separation problem.

Columns were packed with 1 mL of resin and equilibrated for 30 minwith the equilibration buffer. Equilibration was followed by loading of proteinsolution containing 12 times the intended protein loading for the respectivecolumn (mg protein/mL resin). After 30 min, a wash was performed with 5mL equilibration buffer and the ow-through stream was collected. Protein


Table 1 Final Comparison of Resins Considered

Recoveryb,c Pool FinalResina Bindingb Selectivityb (mAU/mL) purityc (%) decisionb

Resin 1 XResin 2 XResin 3 XResin 4 XResin 5 XResin 6 XResin 7 75.4 95.6 XResin 8 25.3 100 XResin 9 81.5 97.7 a Resins included BioRad DEAE, BioRad High Q, TosoHaas Q650M, Whatman Q, WhatmanQA52, Whatman DE53, Pharmacia DEAE FF, Pharmacia Q FF, and Pharmacia Q HP, not num-bered in this order.

b Satisfactory column performance; X unacceptable column performance.c Based on measurements by anion exchange HPLC (AE-HPLC)Source: Reprinted courtesy of LCGC North America, Advanstar Communications Inc.

was eluted with 10 mL of elution buffer (1 M NaCl in the equilibration buffer),and the eluant was collected separately. The ow-throughs and the eluantswere analyzed for protein by UV absorbance at 280 nm and for its purityby anion-exchange high performance liquid chromatography (AE-HPLC). Asshown in Table 1, it was found that most resins showed satisfactory bindingcharacteristics with the product. Only resin 1 showed anomalous behavior inthat the product was not retained under these conditions, so resin 1 was notconsidered further.

Next, columns were packed with 10 mL of the remaining eight resins,and separations were performed using an identical test gradient of 0500 mMNaCl in 20 column volumes (CV) of equilibration buffer. Peak fractions wereanalyzed by AE-HPLC. Figure 3 illustrates the performance of resins 3, 4,and 5 under a test gradient of 0500 mM NaCl in 20 CV. The Y axis in Figs.3 and 4 denotes the peak area obtained upon analysis by AE-HPLC (mAU)per unit injection volume (L). The ow velocity and the fraction sizes aregiven in the gure legends. It is evident that running identical gradients withdifferent resins leads to very different elution proles in terms of the peak

Figure 3. Column performance under test gradients. (Reprinted courtesy LCGCNorth America, Advanstar Communications, Inc.)


Table 2 Calculation of the Comparison Gradients

Test ComparisonMigration gradient, Gradient Column Flow gradient,

time Elution startend volume volume velocity startendResina (min) (mM) (mM) (CV) (mL) (mL/min) mMResin 7 5.21 110 100200 20 10 1.7 90130Resin 8 6.61 109 100200 20 10 0.9 80130Resin 9 7.35 146 100300 20 10 1.7 100200a Same as in Table 1.Source: Reprinted courtesy of LCGC North America, Advanstar Communications Inc.

width and peak position in the overall gradient. The poor selectivity obtainedwith resins 35 led to their elimination from further consideration.

As listed in Table 1, it was found that only resins 69 showed satisfac-tory selectivity between the product and the impurity. Moreover, because resin9 exhibited better resolution than resin 6 and they had identical matrix andligand chemistry, the former was chosen over the latter for further consider-ation.

Comparison gradients were calculated according to the procedure de-scribed above for resins 79. Product recovery was dened as the sum ofproduct peak areas (in mAU) in the pooled fractions (having 90% purity byAE-HPLC) per milliliter of injected sample. Pool purity was dened as thepurity of the total pool formed by mixing the fractions that meet the poolingcriteria. Table 2 shows the calculation of the comparison gradient for thesethree resins, and Fig. 4 illustrated the protein and impurity proles that wereobtained after fraction analysis by AE-HPLC.

Figure 4 reinforces the understanding that performing separations withthe designed comparison gradients yields very similar elution proles withdifferent resins and leds to a fair comparison of resin performance. It alsofollows from Fig. 4 that resin 8 showed good purity but poor recovery. Resins7 and 9 showed comparable recovery and pool purity. However, because ofits better selectivity, resin 9 was chosen as the resin for this purication processand selected for further optimization of buffer pH, protein loading, feed owrate, elution ow rate, gradient slope, and column length.

Figure 4. Column performance under comparison gradients. (Reprinted courtesyLCGC North America, Advanstar Communications, Inc.)


C. Optimization of Kinetics (Operating Conditions)at Bench Scale

Once the stationary and mobile phases have been chosen, we turn to the deter-mination of optimal operating conditions, i.e., determination of the kinetic, asopposed to thermodynamic, contributions. Thus, the particle size and columndimensions are determined in these studies, along with the optimal gradientslope and feed loadings. The following general approach is suggested.

First, experiments are performed to evaluate the effect of various op-erating parameters that affect resin performance in terms of the selectivity andprotein loading. These parameters may include the mobile phase conditions(pH, organic content, buffer composition, etc.) and the gradient slope and de-sign. Optimum mobile phase conditions and the gradient design are chosenfrom the experimental data obtained.

Next, the effect of ow velocity and protein loading on the quality ofseparation is evaluated and, on the basis of resin performance, the bed height,protein loading, and ow velocity are chosen to obtain satisfactory resolutionand cycle time. It is desirable that laboratory experiments be done at the bedheight that will be used at pilot scale in order to obtain comparable columnperformance at large scale.

A detailed analysis of the interaction among these kinetic parameters iscomplicated and is not described here. Many of the underlying issues arebrought out clearly by Felinger in his chapter on optimization (Chap. 3). Inindustrial practice, a heuristic approach similar to the one just described isoften used. It is likely to produce effective, if not necessarily optimal, op-erating conditions in the hands of an experienced practitioner. More detailsof these practical approaches are given in several of the industrial case studiesin this book.

This separation of very large molecules and particles such as viruses isan important industrial topic and is beginning to be addressed in the literature[13,14]. However, the eld is still in its infancy and is likely to change rapidly.We therefore do not feel that it would be appropriate to attempt a summaryhere, and we refer the reader to the growing literature on this subject.

III. THEORETICAL CONSIDERATIONS IN SCALE-UPA. Physical OverviewThe performance of a chromatography column depends on a variety of designand operating factors. In order to have a successful scale-up it is desirable to


Figure 5. Three different scales of columns used frequently during process develop-ment.

maintain kinetic (particle size, pore size, ligand chemistry, temperature, mo-bile phase) and dynamic (bed height, ow velocity, packing density) equiva-lence between the chromatography columns used in the laboratory and thepilot plant. This objective can be accomplished by using identical stationaryand mobile phases in the two columns and operating them at identical bedheight, linear ow velocity, protein loading (mg protein per mL of resin), feedconditions, gradient length, and gradient slope [6]. To handle the increasedvolume of load at pilot scale, the most common procedure used to increasecolumn volume is to increase the column diameter so that the column volumeincreases proportionately [8,15]. This keeps the residence time of the productconstant and avoids causing any product stability issues.

Figure 5 is a schematic illustration of the three sizes of columns thatare often used at laboratory, pilot plant, and commerical scales. Scouting ex-periments in the laboratory are mostly done in small columns to conservethe materials and also because several experiments can be done in parallelsimultaneously at lab scale. However, as discussed above, it is extremely im-portant to maintain bed height constant while scaling up, so the best approachis to perform the nal optimization steps at the bed height that will later beused at the pilot plant and commercial scale. This approach is illustrated inFig. 6 and 7.

These general considerations are frequently used in industry as the basisfor scale-up. In the next section, a quantitative analysis is given that showswhen such simple volumetric scale-up can be used and describes alternativesthat are appropriate when the column length must be changed on scale-up.

The van Deemter equation is widely used to characterize band broaden-ing in a chromatography column and is expressed as


Figure 6. Scaling from laboratory (or bench) to pilot-plant scale.

Figure 7. Scaling up from pilot-plant scale.


H A B/u Cu (4)where u is the linear ow velocity, H is the plate height of the column, andA, B, and C are constants. The plate height H is equal to the length of thecolumn divided by the total number of plates N, so H is smaller for a moreefcient column. A reects the quality of the packing of the column and isindependent of the linear ow velocity. A is small when the column is packedwell and is homogeneous throughout its length. B is a measure of the bandbroadening due to longitudinal diffusion of the sample components along theedge of their respective bands as they travel across the column. It decreaseswith increasing linear ow velocity because the sample components spendless time undergoing diffusion inside the column. C includes contributionsfrom the binding kinetics (adsorption/desorption) as well as the mass transferof the sample components to and from the packing particles.

Preparative chromatography is usually carried out at high ow velocityin order to increase throughput. Then the C term usually dominates Eq. (4),leading to the simplied form

H Cu, or, equivalently, N L (1/Cu) (5)In an ideal case, when the column packing and operating conditions are

kept the same while scaling up (C is a constant), the scale-up involves just avolumetric increase in column dimensions. For such a case, Eq. (5) can berewritten as

L/u CN (6)To preserve the efcacy of separation, the total number of plates is to be keptconstant, so it follows from Eq. (6) that if the bed height needs to be increasedor decreased for some reason (e.g., pressure drops too high), the linear owvelocity might also be altered appropriately so as to keep the ratio of L/uconstant. This ensures a constant number of plates in the column (N), andthe column performance is maintained. This simple analysis is expanded andgeneralized in the following section. In particular, if the particle size needsto be changed upon scale-up (for economic or other reasons), the more generaltreatment must be used.

This very simple physical introduction to scale-up sets the scene for astraightforward quantitative analysis of the problem in the next section.

B. Simple Scale-Up CalculationThe basic idea behind scale-up is to preserve the quality of the separationachieved at small scale [16,17]. Implicit in this approach is the admission that


we are not yet able to determine optimal operating conditions a priori fordifferent scales of operation. Thus, we settle for determining effective, near-optimal operating conditions at bench sale. Effective scale-up rules shouldthen produce comparable results at larger scale. (The alternative approach ofnding optimal operating conditions is currently practicable for some impor-tant classes of separation problems; this approach is discussed in Chapter 3).

A typical scale-up from laboratory to pilot plant is on the order of 50100-fold. This is frequently followed by a 1050-fold scale-up from pilot plantto nal commercial manufacturing scale.

The usual approach is to hold the plate count constant upon scale-upand increase the feed volume and column volume proportionately. This ap-proach was originally based on the assumption of linear adsorption. Later inthis section we discuss how this assumption can be relaxed.

If the subscripts b and l are used to describe parameters at bench andlarge scale, respectively, we have

Nl Nb (7)

Vfeed,lVcolumn,l

Vfeed,bVcolumn,b

(8)

If band spreading is dominated by pore diffusion, as is often the casein realistic separations [1820], then the plate count can be described by

N Lud 2p

(9)

where L is the column length, u the mobile phase linear velocity, and dp theparticle diameter. The proportionality constant includes geometrical factorssuch as the phase ratio and thermodynamic factors such as the retention factor.This result can be derived from the van Deemter equation [16]. CombiningEqs. (7) and (9) we get

Lluld 2p,l

Lb

ubd 2p,b(10)

This represents one constraint on the three variables Ll, ul, dp,l. Recall that thisapproach is based on mimicking bench-scale results at large scale; the vari-ables Lb, ub, dp,b are therefore assumed to be known.

Quantifying the pressure drop across the columns give another result;we use Darcys law in the form


u k d2p

pL

(11)

where k is the permeability without the dependence on particle diameter, whichhas been factored out, and is the mobile phase viscosity. In general, bestresults are obtained at the maximum allowable pressure drop [2]. If the maxi-mum permissible pressure drop is different at bench and large scales, therefollows

ulLld 2p,l

PubLbd 2p,b

(12)

where P is the ratio of maximum pressure drops at large and bench sale. Inthe common case where P 1, the result becomes

ulLld 2p,l

ubLbd 2p,b

(13)

Dividing Eq. (10) by Eq. (13) gives the simple expression:ul ub (14)

Thus equality of plate counts and maximum pressure drops leads to equalityof mobile phase velocity across scales. Substituting Eq. (14) into either Eq.(10) or Eq. (13) gives the familiar result

d 2p,lLl

d 2p,bLb

(15)

Typically, the choice of particle size at the large scale is limited by cost oravailability. Once a particle size is chosen, equation (15) species the columnlength. An approximate theoretical calculation for the optimal d 2p/L is givenin Guiochon et al. [2]; this can also be used to give another estimate of thecolumn length, given the particle size. The chapter by Felinger (Chap. 3) dis-cusses such optimal calculations in detail.

Finally, in order to determine the column diameter at large scale, Eq.(8) can be rewritten as

Vfeed,lVfeed,b

Vcolumn,lVcolumn,b

LlLb D

2c,l

D2c,b (16)Here, is the scale-up factor (which must be specic before scale-up canbegin) and Dc is the column diameter. Because the column length at large


scale has been determined from Eq. (15), the column diameter is obtainedfrom the equation

Dc,lDc,b

LlLb1/2

(17)

When the particle diameter is kept constant on scale-up, we obtain the resultsfor scale-up by volume overloading: From Eq. (15), the column length re-mains constant (in addition to the particle diameter and the mobile phase ve-locity); thus, scale-up consists simply of increasing the column diameter bya factor of .

These results have been obtained for pore diffusion as the process thatdominates band spreading. Analogous results for the case where external masstransfer (lm diffusion) is controlling are given by Pieri et al. [21], Grushkaet al. [16], and Ladisch and Velayudhan [22]. The case where pore diffusionand lm diffusion are comparable has been considered briey by Ladisch andVelayudhan [22]. A more general treatment including the effects of axial dis-persion (which is typically negligible for liquid chromatography) is found inLee et al. [23].

Although the approach taken above was based on linear adsorption[which allowed the specialization of the van Deemter equation to obtainEq. (9)], extensions to nonlinear adsorption are possible. Knox and Pyper [24]showed that Eq. (15) applied for scale-up of noninteracting compounds whosebands in isocratic elution can be approximated as right triangles (i.e., bandspreading is dominated by isotherm nonlinearity). Wankat and Koo [25]showed that Eq. (6) holds for single compounds that have nonlinear single-component isotherms. Golshan-Shirazi and Guiochon [26,27] demonstratedthat Eq. (12) also holds for two compounds with binary Langmuirianisotherms. Wantak [28] presents a general scale-up argument based on theconstancy of plate count at both scales, which also results in Eq. (12) forpore-diffusion-controlled runs and applies even for arbitrary multicomponentisotherms. In fact, it is quite possible that similar results hold for multicompo-nent adsorption as long as the isotherms are locally concave downward (sothat the leading edge of a band is self-sharpening and the trailing edge showsa proportionate pattern). The approach described above is therefore a goodstarting point for scale-up, although variations due to more complex adsorptionbehavior should be kept in mind.

There has been some discussion in the past about whether small particlesare needed in scale-up, particularly when the column is highly overloaded.This is based on the view that under conditions of strong overloading the band


spreading caused by isotherm nonlinearity dominates that caused by kineticfactors such as lm and pore diffusion that depend on particle size. There isan important distinction to be drawn here. Under such conditions that theproduct band can be separated from its nearest impurities thermodynami-cally, i.e., when a sequential stepwise elution schedule can be found suchthat no impurity elutes in the step in which the product elutes, then it is clearthat plate count and its causal kinetic factors are unimportant. Such thermody-namic separations are sometimes possible when onoff binding of the feedcomponents occurs, i.e., the feed components are either very strongly boundor almost completely unbound. This kind of all-or-nothing adsorption oftenoccurs for macromolecules [29,30]. The same mechanism is often exploitedin solid-phase extraction protocols. Under these conditions, it is clear that largeparticles can be used with impunity, at both bench scale and process scale.The only problem in scaling up such a separation is the possibility of overload-ing the column too heavily. Because the product is selectively displaced frombinding sites by more retentive impurities, it is possible that the product maymove faster than expected and thus emerge in more than one step of the step-wise elution schedule. This problem can be avoided by reducing the loadingor increasing the column length.

However, there are many separations of practical importance for whichsuch stepwise elution protocols cannot be found. Then the co-migration ofthe product peak with one or more impurities must be considered, and nowkinetic factors play a vital role in determining the extent of mixing betweenadjacent peaks in the chromatogram [2,31] and thus recovery and productionrates. In such cases, isocratic elution, gradient elution, and displacement chro-matography are used and are reasonably well described by the general scale-up equations developed above. More detailed models, especially for gradientelution, are presented in Chapter 4 of this volume by Watler et al.

C. Constancy of Phase Ratio with ScaleAn important assumption in the calculations above was that the phase ratio(and therefore the interstitial and intraparticulate porosities) remained constantupon scale-up. This is not always a good assumption when the overall scale-up factor is very large (above 100). A recent example of such variations inphase ratio is given in Heuer et al. [32]. In practice, it is worthwhile to estimatethe phase ratio experimentally at each scale and use the results above withcaution if the phase ratio changes appreciably with scale.

Experimental techniques to avoid such changes in packing structure asthe column diameter increases include axial compression, radial compression,


and annular expansion (mixed radial and axial compression). Useful reviewsof these packing methods are found in Jones [33,34] and Colin [35]. In manycases, these approaches have results in improved column performance.

Fundamental and applied work on packing continues, and results fromsoil physics and other elds are being applied to the chromatographic problem(e.g., Bayer et al. [36] and Cherrak and Guiochon [37]). It is too early to sayto what extent packing heterogeneities can be removed by further improve-ments in technology, but such observations raise the question of whether alumped parameter like the phase ratio is sufcient to capture packing geome-try. Close collaboration between manufacturers, industrial users, and academ-ics is needed to arrive at highly effective designs for process columns. Thenext section includes some practical observations on packing properties as afunction of scale under the heading of bed stability.

An assessment of these and related issues from a fundamental viewpointis presented in Chapter 2 of this book by Lightfoot et al.

IV. PRACTICAL CONSIDERATIONS IN SCALE-UPA. Practical GuidelinesIn practice, several issues must be kept in mind when attempting to scale upa separation. In this subsection, these issues are dealt with briey. They willrecur constantly in the case studies discussed in subsequent chapters.

1. Bed Stability (Physical)In a laboratory scale column the column wall offers support to the columnbed and contributes to the stability of the column. However, when the col-umn is scaled up and its diameter increases, the wall support contributionto bed stability starts to decrease. For column diameters greater than 2530cm, the lack of wall support may become an issue and could cause redistri-bution of packing particles and settling of the bed. The total drag force onthe packing particles is a function of the liquid velocity, the liquid viscos-ity, and the bed height. The supporting force that keeps the particles inplace decreases with increasing column diameter as a smaller fraction ofthe particles are supported by the column wall. Therefore, for large-scalecolumns with compressible packings, the maximum velocities are restrictedand decrease with increasing column diameter under identical bed heightand pressure drop, and the situation worsens with duration of column use[3841]. This phenomenon is more prominent for nonrigid gel materials


and is often reversible within limits but almost always with a marked hys-teresis [39,42]. As illustrated in Fig. 6, the issue of physical stability of thecolumn bed becomes particularly signicant at large scale because the bedheight is larger than in the lab and the column is put to more frequent reuseduring commercial manufacturing. The bottom of the column is most vul-nerable because it feels the pressure drop across the column as well as thatdue to the weight of the column itself, which can be appreciable for largecolumns. These effects must be taken into account for robust column de-sign. For cases where bed compression is a problem and the maximum per-missible bed height is smaller than the minimum required to obtain satisfac-tory separation, a suggested solution is to use stacked columns [42]. Thisreduces the pressure difference across any single section of the column andpermits the use of smaller particles and nonrigid gels to obtain enhancedresolution.

2. Bed Stability (Chemical)Chemical stability of the packing material includes any factors that may resultin deterioration of the column performance over a period of use. It may be theleaching of ligands into the mobile phase as often in afnity chromatographydestruction of the matrix in the mobile phases used for column operation,regeneration, or storage (e.g., silica packings at high pH) or irreversible bind-ing at the packing surface [41]. The issue of chemical stability of the columnbed becomes particularly signicant when the column is reused many timesduring commercial manufacturing.

3. Product LoadingThe product loading (milligrams of product loading per milliliter of resin) isgenerally held constant during scale-up. In most cases the resolution is foundto decrease with increasing product loading after the loading has reached acertain level. Further, this behavior is more prominent when the paricle sizeis small. To ensure a successful scale-up and successful operation at largescale, studies must be conducted at lab scale to determine the maximum prod-uct loading with which satisfactory resolution can still be achieved. It is com-mon to operate the column at 8090% of this maximum loading.

4. Gradient SeparationsGradient elution is widely used owing to its ability to provide higher ef-ciency, reduced process times and solvent consumption, and a concentrated


product stream. However, as process scale increases, buffer volumes increasealso, and it becomes increasingly difcult to form accurate and reproduciblegradients. Thus, either the possibility of performing step gradients should beexplored or the ability of the chromatography skid to perform adequate buffermixing and form controlled gradients must be evaluated [15].

5. Flow DistributionFor large diameter columns, uniform ow distribution at the column head maybecome difcult to achieve. This may result in deviations from the desiredplug ow and lead to peak tailing. Use of a ow distributor at the columninset is generally found to be the most effective way of ensuring uniform owdistribution in the column [41]. The rational design of inlet and outlet headersto ensure uniform distribution is discussed at length in Chapter 2.

6. Packing QualityPacking large columns such that the resulting packed column is homogeneousis very critical for obtaining uniform ow distribution. Channeling inside thecolumn often leads to peak tailing and/or peak splitting. A variety of ap-proaches have been developed by the different chromatographic equipmentvendors to alleviate this problem. The most popular technique at preparativescale is axial compression and the use of self-packing columns [35]. Somedegree of compression has been known to enhance resolution [41]. The con-nection between packing and the phase ratio used in quantitative scale-up wastouched upon in the previous section.

7. System DesignThe system dead volume arising from the piping and other support equipmentfor chromatography columns such as the valves, ow meters, air sensors, andtubings is much larger at pilot and particularly manufacturing scale than atlab scale. This leads to dilution effects and higher pressure drops as well asadditional band broadening, so the impact these factors may have on the over-all column performance must be evaluated. The general guidelines are to keepthe system dead volume to the minimum, to have bypasses through devicessuch as air traps and lters for use during sample load, and to choose tubingdiameter to achieve turbulent ow so as the reduce undesirable axial mixing[43]. Further, the chromatographic system should be designed such that allinlet sources are at or above the level of the column, whereas all outlet sinks


are at or below the level of the column. This ensures that the column is notbeing operated against a hydrostatic pressure head.

8. Fraction CollectionThe peak width and shape as seen in the chromatogram depends on severalfactors such as the column dimensions, extracolumn effects, operating condi-tions, and sample volume [42]. Thus, even if the scale-up is carried out follow-ing a well-thought-out methodology, it is still very likely that the peak widthand shape may differ from that obtained at lab scale. Thus, the fraction collec-tion strategy must be revisited at preparative scale based on column perfor-mance at that scale. The critical monitoring device, e.g., the UV detector,should be placed as close to the fraction collector as possible to ensure goodrepresentation of the process stream. Also, it is good to have a fraction collectorthat can collect fractions based on several process parameters, such as UV ab-sorbance, conductivity, time, volume, rst or second derivative of the signal, etc.

9. Media AvailabilityAlthough selectivity for a separation may be the primary criterion for selectionof a resin for analytical separation, several other factors need to be consideredbefore a resin is selected for preparative separation. These include the avail-ability of large quantities of the resin, cost, continuity of supply, batch-to-batch consistency, column lifetime, and support documentation to aid in regu-latory ling [44].

10. CostingThe cost of the feedstock is generally not given adequate consideration whenthe process optimization is carried out at bench scale. However, as the processis scaled up, the process should be modeled and raw material and facility costsmust be examined. Resin costs usually account for the biggest contributionto the raw material costs. Thus, although a certain resin may offer the bestselectivity at bench scale, it may be too expensive, under linear scale-up, atprocess scale [8].

11. Sample PretreatmentOften cells or inclusion bodies undergo a rupture step by homogenization orsome other mechanism prior to a chromatography step with the purpose ofcapturing the product from the cell culture or fermentation broth. In thesesituations the process stream is replete with lipids, nucleic acids, proteins, and


other macromolecular contaminants. These impurities affect the passage ofthe product through the column and have a very signicant effect on columnperformance. Because most cell-rupture operations are more effective at largescale than at bench scale, pretreatment of the process stream at large scaleto remove all the interfering contaminants may become crucial to ensuringsatisfactory and robust column performance [8].

12. Scale-DownAlthough our focus is on scale-up, it is worth mentioning that some issuesrelated to validation are often addressed by scaling down. Thus, using a small-scale study for viral clearance is often faster, safer, and cheaper while being atleast as accurate as larger scale studies [45]. Many of the scale-up approachesdiscussed above apply, mutatis mutandis, to scale-down.

B. Other Modes of OperationAlthough isocratic and gradient elution are the most common modes of opera-tion and displacement chromatography has a small but signicant role, thereare other modes of interest at large scale. In this context, frontal chromatogra-phy has always played an important role under the guide of adsorption stepsin a variety of applications, especially in the chemical industries. The processof feed introduction in isocratic, gradient, and displacement runs is nothingmore than frontal chromatography, so it is clearly an important part of therun. Here, expanded bed chromatography and simulated moving bed (SMB)chromatography are discussed briey. Though the basic scale-up rules men-tioned above are applicable to both these techniques, there are some uniqueconsiderations that are worth highlighting.

Expanded Bed ChromatographyAn expanded bed consists of specially designed packing particles that are u-idized in a column with controlled ow distribution so as to provide a largenumber of plates (high mass transfer) with minimal back-mixing. Thus, al-though expanded bed chromatography (EBC) offers more plates than batchadsorption, it also allows better utilization of binding capacity of the adsorbent[46].

Scale-up in EBC is a little more complicated than in other modes ofchromatography because of the additional requirements of maintaining thestability of the expanded bed. The ow distribution, ow velocity, composi-


tion, and other physical/chemical properties of the feed and the particle distri-bution and the physical/chemical stability of the adsorbent can all have a sig-nicant effect on the bed stability. Conditions optimized for high productivitymay not be able to provide long-term stability, so a process optimized at smallscale should be carefully evaluated with these two objectives in mind [47].

Simulated Moving Bed ChromatographySimulated moving bed chromatography (SMB) is a continuous chromato-graphic process in which a mobile phase and the sample components are in-jected into and withdrawn from a ring of chromatography columns at pointsthat are rotating between the columns during the process. SMB is slowly be-coming the technique of choice for performing efcient enantiomeric separa-tions [48].

The key to successful SMB operation is proper selection of the operatingow rates and the valve-switching times of the feed and eluant streams [49].Pumps typically used in this technique provide control of ow rates of betterthan 1%. These are important both for stability of the different zones and forachieving satisfactory separation. Just as in other modes of chromatography,the efciency of an SMB separation is negatively impacted by extracolumnvolume, which in this case consists of the volume of the recycling pump, owmeter, pressure sensor, and valves. Antia (Chapter 5) describes a simple andeffective approach to the design and control of SMB separations.

C. Practical Considerations for Modes of InteractionAs with modes of operation, there are many practical issues peculiar to eachmode of interaction. The general features of each mode are discussed brieyin this section. Again, the relevance of these issues in the design of practicalseparations will be highlighted by the design choices made in the case studiesdescribed in the following chapters.

1. Ion-Exchange ChromatographySeparation in ion-exchange chromatography (IEC) takes place because of dif-ferential ionic interactions between the charged ligands on the stationary phaseand the charged sample components in the feed in the presence of aqueousbuffer solution. Elution is then performed with increasing salt concentrationby altering the pH of the mobile phase, resulting in weakening of the ionicforces, with the components eluting in the order of increasing binding strengthwith the stationary phase.


Ion-exchange chromatography is the most widely used mode of chroma-tography for protein separation. This is due to the high dynamic capacitiesand low relative costs of the IEC resins, simple buffers used, high usableow rates, and the robustness, scalability, and ease of operation of most IECmethods. Most IEC resins have large volumetric capacities, with operationlimited only by the total concentration of the product and contaminants inthe feed. Scale-up in IEC can be performed following the generic guidelinesmentioned above, and it is the technique of choice as an early processing stepdue to the large qualities of material that have to be handled [50,51].

Watler et al. (Chap. 4) and Levison (Chap. 6) provide a clear and com-prehensive discussion of the use of IEC in preparative separations.

2. Hydrophobic Interaction ChromatographySeparation in hydrophobic interaction chromatography (HIC) is based on dif-ferences in hydrophobicity of the different sample components. At high saltconcentrations, the solubility of the hydrophobic product is reduced and thehydrophobic side chains of the sample component associate with the hy-drophobic ligands on the stationary phase [52]. Elution can be performed byreducing the polarity of the mobile phase in a continuous or stepwise manner.Because of the differences in the mechanisms of separation of HIC and IEC,HIC is frequently used to process the eluant stream from an IEC column.

Scale-up in HIC can be performed by following the generic guidelinesoutlined above [5355]. However, there are several considerations peculiarto HIC that must be kept in mind. Denaturation of the product molecule mayoccur at high salt concentrations during the separation process; the mobilephase may be too viscous and thus impair the accuracy and reproducibility ofgradient formation, apart from limiting the usable bed height; and temperaturevariations may change column performance considerably.

3. Size Exclusion ChromatographyThe separation in size exclusion chromatography (SEC) is due to differencesin the sizes and shapes of the sample components as they are carried by themobile phase through the three-dimensional porous structure of the stationaryphase. In most cases the maximum bed height and diameter that can be usedare limited by the physical stability of the gel. This is resolved by using aseries of smaller columns instead of a single long column during scale-up.Silica and agarose matrices are most commonly used in SEC and often containnegatively charged moieties at the surface that may cause adsorption of posi-tively charged proteins and the exclusion of negatively charged proteins atlow ionic concentrations.


In SEC, although the separation time increases linearly with columnlength as in other forms of chromatography, the resolution increases only asthe square root of the length. Also, the resolution of the sample componentsis more sensitive to the width of the sample band than the ow velocity. Oneof the suggested strategies for SEC optimization and scale-up is to select thebed height and linear ow velocity to obtain the desired cycle time and satis-factory resolution [56,57]. Next, the width of the sample zone is varied tone-tune resolution. Finally, the column diameter is changed to meet the pro-ductivity requirements for the column.

Feed volume in SEC normally varies between 1% and 5% of the totalcolumn volume to obtain satisfactory resolution for most large-scale separa-tions and up to 30% for desalting. Resins used in SEC have low binding capac-ities and are often compressible, so the separations are generally performedat very low ow velocities, and all the issues mentioned earlier in SectionIV.A.1 apply. Therefore, SEC is not considered by most an ideal techniquefor scale-up except for desalting or buffer exchange or as a nal polishing stepwhen the volumes and sample quantities are small enough (as in purication ofhuman serum albumin).

4. Reversed-Phase ChromatographySeparation in reversed-phase chromatography (RPC) is based on the differ-ences in the hydrophobicities of the different sample components. The hy-drophobic groups on the surface of the sample molecules bind with the veryhydrophobic stationary phase particles in the presence of an apolar mobilephase. Elution occurs when a mobile phase is used with increasing concentra-tion of an organic phase such as acetonitrile, methanol, or isopropanol. Thecomponents elute in the order of their hydrophobicity, and a separation isachieved.

Reversed-phase chromatography is very popular in the separation of smallmolecules. Its limited use in protein separation, particularly at large scale, is dueto the various issues that are associated with the use of the organic solvents inthe purication process. These include protein denaturation and/or unfolding,waste handling, and the need for special explosion-proof handling due to solventvolatility. Scale-up in RPC follows the guidelines outlined above.

5. Afnity ChromatographyThe separation of a target molecule from a mixture of species in afnity chro-matography (AC) takes place by virtue of specic and reversible binding witha ligand that is immobilized on the matrix. This technique offers short process


times and high specicity and resolution and is particularly useful when thetarget is present in very small quantities in the complex mixture. The primaryissues in this technique are that the ligands may be expensive and unstableand may leach from the matrix and be present in the product. Fouling andregeneration of the ligand upon use may also offer challenges, so AC is gener-ally used as the nal polishing step in the purication scheme [58].

Scale-up in AC is carried out following the generic guidelines outlinedin the previous section [5962]. The bed height is kept constant, and columnproductivity is increased by increasing column diameter. However, resins usedin AC are often compressible, so all the issues mentioned earlier in SectionIV.A.1 apply and should be considered during scale-up.

Another approach to scaling up afnity separations is to increase thecapacity of the adsorbent. This may be accomplished by increasing the concen-tration of the binding ligand that is coupled to the support particles. However,the binding capacity does not increase linearly with the ligand concentrationand depends on the characteristics of the ligand as well as those of the bindingproduct. This route for scale-up is therefore sensitive to the problems of leach-ing of ligands and fouling of resins. In addition, there is a partical limit onthe maximum attainable ligand concentration on the resin surface. Thus, thenal decision should depend on the comparison of economics and issues asso-ciated with the two approaches highlighted above.

6. Metal (Chelate) ChromatographyThe separation and operation principles in metal (chelate) chromatography(MC) are very similar to those in afnity chromatography. Separation is basedon interactions between the afnity tail on the sample component (such asdihistidyl tag) with the complexed heavy metals such as zinc and nickel onthe stationary phase [9]. MC can be as efcient as afnity chromatography,particularly in the removal of endotoxins.

Scale-up issues for metal chromatography are very similar to those ofafnity chromatography. The resins are generally custom-made and are rela-tively very expensive. The afnity tail often needs to be clipped before theend of the purication process. Care must be taken to avoid denaturation ofprotein feedstocks by the metals used in MC.

V. CONCLUSIONS

In this chapter we have provided an overview of the basic principles and prac-tice of scale-up in the preparative chromatography. Modes of interaction and


modes of operation were dened to clarify the options available upon scale-up.Because it is necessary to optimize a bench scale separation before attemptingto scale it up, attention was rst focused on method development at bench scale.Thermodynamic issues (resin screening) and kinetic issues (determination of op-erating conditions) were decoupled. A simple but generally applicable approachto resin screening was described. After a physically motivated introduction toscale-up, a straightforward calculation based on pore diffusion as the controllingkinetic contribution was presented and should sufce for many realistic separa-tions. More detailed analyses of the fundamental chromatographic processes,modeling approaches to importantmodes of interactionand operation, and indus-trial case studies are given in the subsequent chapters.

REFERENCES

1. HK Rhee, R Aris, NR Amundson. First-Order Partial Differential Equations, VolII: Theory and Application of Hyperbolic Systems of Quasilinear Equations. En-glewood Cliffs, NJ: Prentice-Hall, 1989.

2. G Guiochon, S Golshan-Shirazi, AM Katti. Fundamentals of Preparative andNonlinear Chromatography. Boston: Academic Press, 1994.

3. A Lee, A Velayudhan, Cs Horvath. Preparative HPLC. In: G Durand, M Bobi-chon, J Florent (eds.) Proceedings of the 8th International Biotechnology Sympo-sium. Vol. 1. Paris: Societe Francaise de Microbiologie, 1988, pp 593610.

4. AS Rathore. Resin screening for optimization of chromatographic separations.LC-GC 19:616, 2001.

5. JC Janson, T Pettersson. Large-scale chromatography of proteins. In: G Ganetsos,PE Barker, eds. Preparative and Production Scale Chromatography. New York:Marcel Dekker, 1993, pp 559590.

6. G Sofer, L Hagel. Handbook of Process ChromatographyA Guide to Opti-mization, Scale-up and Validation. New York: Academic Press, 1997, pp 27113.

7. P Hedman, JC Janson, B Arve, JG Gustafsson. Large scale chromatography: opti-mization of preparative chromatographic separations with respect to columnpacking particle size, mass loading and ow rate. In: G Durand, L Bobichon, JFlorent, eds. Proc 8th Int Biotechnol Symp, Vol 1. Paris: Soc Franc Microbiol,1988, pp 623643.

8. S Fulton, T Londo. Systematic development of chromatographic processes usingperfusion chromatography technology. In: G Subramanian, ed. Bioseparation andBioprocessing, Vol 1. New York: VCH, 1994, pp 4164.

9. RWisniewski, E Boschetti, A Jungbauer. Process design considerations for large-scale chromatography of biomolecules. In: KE Avis, VL Wu, eds. Biotechnologyand Biopharmaceutical Manufacturing, Processing, and Preservation. BuffaloGrove, IL: Interpharm, 1996, pp 61198.


10. MA Quarry, RL Grob, LR Snyder. Prediction of precise isocratic retention datafrom two or more gradient elution runs. Analysis of some associated errors. AnalChem 58:901917, 1986.

11. P Jandera, D Kromers, G Guiochon. Effects of the gradient prole on the produc-tion rate in reversed-phase gradient elution overloaded chromatography. J Chro-matogr A 760:2539, 1997.

12. P Jandera. Simultaneous optimization of gradient time, gradient shape, and initialcomposition of the mobile phase in the high-performance liquid chromatographyof homologous and oligomeric series. J Chromatogr A 845:133144, 1999.

13. S Yamamoto, E Miyagawa. Retention behavior of very large biomolecules inion-exchange chromatography. J Chromatogr A 852:2530, 1999.

14. S Yamamoto, E Miyagawa. Dynamic binding performance of large biomoleculessuch as -globulin, viruses, and virus-like particles on various chromatographicsupports. Prog Biotechno 16:8186, 2000.

15. E Groundwater. Guidelines for chromatography scale-up. Lab Pract 34:1718, 1985.16. EL Grushka, LR Snyder, JH Knox. Advances in band spreading theories. J Chro-

matogr Sci 13:2533, 1975.17. MR Ladisch, PC Wankat. Scale-up of bioseparations for microbial and biochemi-

cal technology. In: M Phillips, S Shoemaker, R Ottenbrite, R Middlekauf, eds.Impact of Chemistry on Biotechnology. ACS Symp Ser No. 362. Washington,DC: Am Chem Sco, 1988, pp 7296.

18. DM Ruthven. Principles of Adsorption and Adsorption Processes. New York:Wiley, 1984.

19. S Yamamoto, M Nomura, Y Sano. Scaling up of medium-performance gel ltra-tion chromatography of proteins. J Chem Eng Jpn 19:227234, 1986.

20. S Yamamoto, M Nomura, Y Sano. Factors affecting the relationship betweenplate heights and the linear mobile phase velocity in gel ltration chromatographyof proteins. J Chromatogr 394:363371, 1987.

21. G Pieri, P Piccardi, G Muratori, L Caval. Scale-up for preparative chromatogra-phy of ne chemicals. Chim Ind 65:331340, 1983.

22. MR Ladisch, A. Velayudhan. Scale-up techniques in bioseparation processes. In:RK Singh, SSH Rizvi, eds. Bioseparation Processes in Foods. New York: MarcelDekker, 1995, pp 113138.

23. CK Lee, Q Yu, SU Kim, NHL Wang. J Chromatogr 483:85, 1989.24. JH Knox, HM Pyper. Framework for maximizing throughput in preparative liquid

chromatography. J Chromatogr 363:1, 1986.25. PC Wankat, YM Koo. Scaling rules for isocratic elution chromatography. AIChE

J 34:10061012, 1988.26. S Golshan-Shirazi, G Guiochon. Theory of optimization of the experimental con-

ditions in preparative chromatography: optimization of the column efciency.Anal Chem 61:1368, 1989.

27. S Golshan-Shirazi, G Guiochon. Optimization of the experimental conditions inpreparative liquid chromatography with touching bands. J Chromatogr 517:229,1990.


28. PC Wankat. Scaling rules and intensication of liquid chromatography: extensionto gradient elution and displacement chromatography. Prep Chromatogr 1:303322, 1992.

29. HP Jennissen. Evidence for negative cooperativity in the adsorption of phos-phorylase b on hydrophobic agaroses. Biochemistry 15:5683, 1976.

30. A Velayudhan, Cs Horvath. Preparative chromatography of proteins: analysis ofthe multivalent ion-exchange formalism. J Chromatogr 443:1329, 1988.

31. G Guiochon, AM Katti. Chromatographia 24:165, 1987.32. C Heuer, P Hugo, G Mann, A Seidel-Morgenstern. Scale up in preparative chro-

matography. J Chromatogr A 752:1929, 1996.33. K Jones. Process scale high-performance liquid chromatography, Part I: An opti-

mization procedure to maximise column efciency. Chromatographia 25:437446, 1998.

34. K Jones. A review of very large scale chromatography. Chromatographia 25:547559, 1998.

35. H Colin. Large-scale high-performance preparative liquid chromatography. In:G Ganetsos, PE Barker, eds. Preparative and Production Scale Chromatography.New York: Marcel Dekker, 1993, pp 1146.

36. E Bayer, E Banneister, U Tallarek, K Albert, G Guiochon. NMR imaging of thechromatographic process. J Chromatogr A 704:37, 1995.

37. DE Cherrak, G Guiochon. Phenomenological study of the bed-wall friction inaxially compressed packed chromatography columns. J Chromatogr A 911:147,2001.

38. S Katoh. Scaling-up afnity chromatography. Trends Biotechnol 5:328331,1987.

39. AW Mohammad, DG Stevenson, PC Wankat. Pressure drop correlations andscale-up of size exclusion chromatography with compressible packing. Ind EngChem Res 31:549561, 1992.

40. JC Janson. Large-scale chromatography. GBF Monogr Ser 7:1320, 1984.41. JC Janson, P Dunhill. Factors affecting scale-up of chromatography. Fed Eur

Biochem Soc Meet (Proc). In: B Spencer, ed. Industrial Aspects of Biochemistry,Part 1, Vol 30 1974, pp 81105.

42. JC Janson, P Hedman. Large-scale chromatography of proteins. Adv BiochemEng 25:4399, 1982.

43. P Zeilon, L Stack, RE Majors. System design for process-scale chromatography.LC-GC 10:736742, 1992.

44. PR Levison. Techniques in process-scale ion-exchange chromatography. In: GGanetsos, PE Barker, eds. Preparative and Production Scale Chromatography.New York: Marcel Dekker, 1993, pp 617626.

45. G Sofer. Ensuring the accuracy of scaled-down chromatography models. Bio-Pharm 10:3639, 1996.

46. R Hjorth, P Leijon, AB Frej, C Jagersten. Expanded bed adsorption chromatogra-phy. In: G Subramanian, ed. Bioseparation and Bioprocessing, Vol 1. New York:VCH, 1994, pp 199228.


47. AB Frej. Expanded bed adsorption. In: MC Flickinger, SW Drew, eds. Encyclo-pedia of Bioprocess Technology: Fermentation, Biocatalysis, and Bioseparation,Vol 1. New York: Wiley-Interscience, 1999, pp 2031.

48. J Strube, A Jupke, A Epping, H Schmidt-Traub, M Schultze, R Devant. Design,optimization, and operation of SMB chromatography in the production of enan-tiomerically pure pharmaceuticals. Chirality 11:440450, 1999.

49. RM Nicoud, RE Majors. Simulating moving bed chromatography for preparativeseparations. LC-GC 18:680687, 2000.

50. S Ostrove. Considerations for scaling up to process chromatography. LC-GC 7:550554, 1989.

51. S Cresswell. Scaling up in ion exchange chromatography. Biotech Forum Eur9:446452, 1992.

52. E Grund. Hydrophobic interaction chromatography of proteins. In: G Subramanian,ed. Bioseparation and Bioprocessing, Vol 1. New York: VCH, 1994, pp 6588.

53. S Ishida, J Saeki, E Kumazawa, G Kawanishi, E Sada, S Katoh. Scale-up ofhydrophobic interaction chromatography for purication of antitumor antibioticSN-07. Bioprocess Eng 4:163167, 1989.

54. P Karsnas. Hydrophobic interaction chromatography. In: MC Flickinger, SWDrew, eds. Encyclopedia of Bioprocess Technology: Fermentation, Biocatalysis,and Bioseparation, Vol 1. New York: Wiley-Interscience, 1999, pp 602612.

55. K Vorauer, M Skias, A Trkola, P Schulz, A Jungbauer. Scale-up of recombinantprotein purication of hydrophobic interaction chromatography. J Chromatogr625:3339, 1992.

56. JC Janson. Process scale size exclusion chromatography. In: G Subramanian, ed.Process Scale Liquid Chromatography. New Yor

Documents

Biopharmaceutical Scaleup and Optimisation