Merck Exhibit 2050, Page 1
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`INTERNATIONAL UNION OF CRYSTALLOGRAPHY
`MONOGRAPHS ON CRYSTALLOGRAPHY
`
`IUCr BOOK SERIES COMMITTEE
`
`A. A. Chernov, Russia
`P. Coppens (Chairman), USA
`G. R. Desiraju, India
`J. Drenth, The Netherlands
`A. M. Glazer, UK
`J.P. Glusker, USA
`J. R. Helliwell, UK
`
`Merck Exhibit 2050, Page 2
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`6
`
`IUCr Monographs on Crystallography
`1 Accurate molecular structures: their determination and importance
`A. Domenicano and I. Hargittai, editors
`2 P. P. Ewald and his dynamical theory of X-ray diffraction
`D. W. J. Cruickshank, H. J. Juretschke, and N. Kato, editors
`3 Electron diffraction techniques, Volume 1
`J. M. Cowley, editor
`4 Electron diffraction techniques, Volume 2
`J. M. Cowley, editor
`5 The Rietveld method
`R. A. Young, editor
`Introduction to crystallographic statistics
`U. Shmueli and G. H. Weiss
`7 Crystallographic instrumentation
`L. A. Aslanov, G. V. Fetisov, and J. A. K. Howard
`8 Direct phasing in crystallography: fundamentals and applications
`C. Giacovazzo
`9 The weak hydrogen bond in structural chemistry and biology
`G. R. Desiraju and T. Steiner
`10 Defect and microstructure analysis by diffraction
`R. L. Snyder, J. Fiala, and H. J. Bunge
`11 Dynamical theory of X-ray diffraction
`A. Authier
`12 The chemical bond in inorganic chemistry
`I. D. Brown
`13 Structure determination from power diffraction data
`W. I. F. David, K. Shankland, L.B. Mccusker, and Ch. Baerlocher, editors
`14 Polymorphism in molecular crystals
`J. Bernstein
`
`IUCr Texts on Crystallography
`1 The solid state: from superconductors to superalloys
`A. Guinier and R. Julien, translated by W. J. Duffin
`4 X-Ray charges densities and chemical bonding
`P. Coppens
`5 The basics of crystallography and diffraction, second edition
`C.Hammond
`6 Crystal structure analysis: principles and practice
`W. Clegg, editor
`7 Fundamentals of crystallography, second edition
`C. Giacovazzo, editor
`
`Merck Exhibit 2050, Page 3
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`Polymorphism in
`. Molecular Crys~als
`
`JOEL BERNSTEIN
`
`Department of Chemistry
`Ben-Gurion University of the Negev
`
`CLARENDON PRESS • OXFORD
`2002
`
`Merck Exhibit 2050, Page 4
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`OXFORD
`
`UNIVERSITY PRESS
`Great Clarendon Street, Oxford OX2 6DP
`Oxford University Press is a department of the University of Oxford.
`It furthers the University's objective of excellence in research, scholarship,
`and education by publishing worldwide in
`Oxford New York
`Auckland Bangkok Buenos Aires Cape Town Chennai
`Dar es Salaam Delhi Hong Kong Istanbul Karachi Kolkata
`Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi
`Sao Paulo Shanghai Taipei Tokyo Toronto
`with an associated company in Berlin
`
`Oxford is a registered trade mark of Oxford University Press
`in the UK and in certain other countries
`
`Published in the United States
`by Oxford University Press Inc., New York
`© Oxford University Press, 2002
`The moral rights of the author have been asserted
`Database right Oxford University Press (maker)
`
`First published 2002
`All rights reserved. No part of this publication may be reproduced,
`stored in a retrieval system, or transmitted, in any form or by any means,
`without the prior permission in writing of Oxford University Press,
`or as expressly permitted by law, or under terms agreed with the appropriate
`reprographics rights organization. Enquiries concerning reproduction
`outside the scope of the above should be sent to the Rights Department,
`Oxford University Press, at the address above
`
`You must not circulate this book in any other binding or cover
`and you must impose this same condition on any acquirer
`
`A Catalogue record for this title is available from the British Library
`
`Library of Congress Cataloging in Publication Data
`Bernstein, Joel.
`Polymorphism in molecular crystals/ Joel Bernstein.
`(IUCr monographs on crystallography; 14)
`Includes index.
`I. Polymorphism (Crystallography) 2. Molecular crystals. I. Title. IL International
`Union of Crystallography monographs on crystallography; 14.
`
`QD951.B572002
`
`548'.3-dc21
`
`2001047556
`
`ISBN O 19 850605 8
`
`10 9 8 7 6 5 4 3 2 I
`
`Typeset by
`Newgen Imaging Systems (P) Ltd., Chennai, India
`Printed in Great Britain
`on acid-free paper by
`T. J. International Ltd, Padstow
`
`Merck Exhibit 2050, Page 5
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`3
`
`Controlling the polymorphic form obtained
`
`Crystal growth is a science and an art. The scientist's role in the crystal growth process is that
`of an assistant who helps molecules to crystallize. Most molecules, after all, are very good at
`growing crystals. The scientific challenge is to learn how to intervene in the process in or4er
`to improve the final product. (Etter 1991)
`
`I am very sorry, that to the many ... difficulties which you meet with, and must therefore
`surmount, in the serious and effectual prosecution of Experimental Philosophy, I must add one
`discouragement more, which will perhaps as much surprise you as dishearten you; and it is,
`that besides that you will find ... many of the experiments published by Authors, or related
`to you by the persons you converse with, false or unsuccessful, ... you will meet with several
`Observations and Experiments, which though communicated for true by Candid Authors or
`undistrusted Eye-witnesses, or perhaps recommended to you by your own experience, may
`upon further tried disappoint your expectation, either not at all succeeding constantly, or at
`least varying much from what you expected.
`This Advertisement may seem of so discouraging a nature that I 'should much scruple the
`giving it to you, but that I suppose the trouble at that unsuccessfulnesse which you may meet
`with Experiments, may be somewhat lessened, by your being forewarned of such contingencies.
`And that you should have the luck to make an Experiment once, without being able to make
`the same thing again, you might opt to look upon such disappointments as the effect of and
`unfriendliness in Nature or Fortune to your particular attempts, as proceeding from a secret
`contingency incident to some Experiments, by whomever they be tried. (Boyle 1661)
`
`Control nevertheless ~ important in science-tremendously so-not as an end but as a
`component of proof. The ability to control is the strongest possible demonstration of true
`understanding. Many doubted whether Becquerel, Curie, Bohr, Oppenheimer and the rest
`really understood what causes what inside the atom. But after July 16, 1945, when the day
`dawned prematurely to the northwest of Alamagordo, at White Sands, New Mexico, no one
`could possibly doubt any more, for the atom bomb was plainer than the sun. With a demon(cid:173)
`strated ability to control, the good scientist may sign off like the mathematician at the end of a
`proof: Quod erat demonstrandum. (Huber 1991)
`
`3.1 General considerations
`
`Crystallization is a process that has fascinated both scientists and casual observers
`throughout the ages. It is indeed remarkable that upwards of 1020 molecules or ions,
`distributed essentially randomly throughout some fluid medium (gas, liquid, or solu(cid:173)
`tion) coalesce, very often spontaneously, to form a regular solid with a well-defined
`
`Merck Exhibit 2050, Page 6
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`AGGREGATION AND NUCLEATION
`
`67
`
`structure, or in the case of polymorphs, with a limited number of well-defined struc(cid:173)
`tures. Those structures are invariant across a wide variety of conditions, in some cases
`almost under any conditions for which crystals form. Two of the principal questions
`to be asked for such a process is how it begins and how it proceeds, especially in the
`context of polymorphic systems. A great deal of work has been devoted to attempts
`to answer _these questions, and in spite of considerable progress_ especially on exper(cid:173)
`imental and empirical fronts, there is still much to be learnt in developing current
`models. Historical treatments of the classic notions of crystallization and recrystal(cid:173)
`lization, including many important references, have been given by Tipson ( 1956) and
`van Hook (1961). A µiore recent thorough account may be found in Mullin's book
`(1993).·
`For any substance it is possible in principle to define experimentally the solvents,
`temperature range, rate of evaporation or cooling, and many of the other conditions
`under which it will crystallize. This collection of conditions has been called the
`occurrence domain (Sato and Boistelle 1984). That domain exists for any substance,
`but rarely, if ever, are its contents completely known. The contents of the occurrence
`domain for any material-in the present context, any polymorph-are not necessarily
`unique. In regions in which there is an intersection of domains, one may expect that
`two or more polymorphs would crystallize under essentially identical conditions. On
`the other hand, determining which regions of the domain are unique to a particular
`polymorph can be advantageous in determining crystallization strategy. This chapter
`deals with a number of the factors which should be considered in making such a
`determination, along with examples of the phenomena associated with competitive
`polymorphic crystallizations.
`
`3.2 Aggregation and nucleation
`
`The thermodynamics and kinetics outlined in Chapter 2 attempt to treat the question
`of crystallization on the macroscopic scale. On the microscopic scale we would like
`to be able to answer questions about the critical size and structure of a collection of
`molecules that will grow into the eventual crystal. In particular, how and when will
`polymorphs be obtained or be prevented from forming? Classically, the first stage of
`crystallization is viewed as nucleation, the spontaneous formation or introduction of a
`nucleus, or centre of crystallization, in the crystallization medium from which crystals
`may grow, although nuclei may also be destroyed before growing into larger crystals.
`The size of such nuclei has been a matter of considerable discussion. On the one
`hand, for instance, Ostwald (1902) claimed that particles containing between 108 and
`1012 molecules are not sufficiently large to induce crystallization from supersaturated
`solutions, but later work indicated that much more modest numbers (e.g. 10-105)
`may be considered a critical size to generate crystals (McIntosh 1919; Tamman and
`von Gronow 1931). This amounts to a cube of approximately 100 A on an edge and a
`'crystal' nucleus weighing as little as 10- 18 g. Additional aspects of the question of
`the size of a crystal nucleus are discussed by Mullin (1993).
`
`Merck Exhibit 2050, Page 7
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`68
`
`CONTROLLING THE POLYMORPHIC FORM OBTAINED
`
`In an attempt to avoid some of the confusion extant in the current literature on the
`nature of nucleation, Mullin has provided a useful schematic classification for various
`terms in use:
`
`SECONDARY
`(induced by crystals)
`
`(spontaneous)
`
`(induced by foreign particles)
`
`Primary nucleation refers to those systems that do not contain crystalline matter.
`When no foreign bodies are present (i.e. the crystallization results from the spon(cid:173)
`taneous formation of nuclei of the crystallizing material) the process is referred to
`as homogeneous. The presence (intentional or unintentional) of foreign particles can
`also induce nucleation, which is then termed heterogeneous.
`Secondary nucleation deals with the situation in which nuclei are generated in
`the vicinity of crystals of the solute already present in a supersaturated solution.
`The solute crystals may have resulted from primary nucleation or may be deliberately
`added. This subject has also been covered by Mullin, as well as 'in a number of reviews
`(Strickland-Constable 1968; Botsaris 1976; Delong 1979; Garside and Davey 1980;
`Garside 1985; Nyvlt et al. 1985).
`Mullin has argued that the minimum number of molecules in a stable crystal nucleus
`can vary from about ten to several thousand. A model based on the simultaneous
`collision of this number of molecules with the degree of order required for it to
`be recognized by additional molecules as a crystal is highly unlikely. A more likely
`scenario is that the nucleus would be generated by a sequence of bimolecular additions
`in which the so-called critical cluster would be built up stepwise:
`
`A+A¢2A
`
`A+2A ¢ 3A
`l)A ¢ nA (critical cluster A1)
`
`A+ (n -
`
`In Mullin's model, further molecular additions to the critical cluster results in
`nucleation.
`A solution or melt can contain a variety of clusters A1, ... , An, in a system of
`competing equilbria. Each cluster in tum is a potential critical cluster for the nucleus
`of one or more polymorphic crystal modifications. In the context of polymorphic
`structures, in particular those which crystallize under similar conditions, there must
`be a number of processes of this type, all involved in competing equilibria. This
`
`Merck Exhibit 2050, Page 8
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`AGGREGATION AND NUCLEATION
`
`69
`
`is the idea behind Etter's (1991) extension of this model, describing the clusters as
`aggregates,
`
`~ Polymorphic form 1
`
`/
`
`Aggregoto
`
`1 ~ Polymo~hic form 2
`
`Molecules
`
`Aggregate 2
`
`~ Polymorphic form 3
`~
`~
`Polymorphic form 4
`
`which must contain the structural essence of the eventual crystal structure(s), and are
`therefore likely to be dominated by the same intermolecular interactions. Because
`such a system involves multiple equilibria, once nucleation occurs for one of the
`polymorphic forms, the equilibrium will be displaced in favour of that form at the
`expense of other forms. On a qualitative basis, this demonstrates the competition
`between kinetic and thermodynamic factors. For instance, even if Polymorphic Form
`1 were the thermodynamically most stable one, Polymorph 3 might be the only one
`obtained if Aggregate 2 nucleated crystal growth faster than Aggregate 1. When these
`factors are equal, or very nearly so, then two or more modifications may result from
`the same aggregate or from different aggregates, leading to concomitant polymorphs
`(see Section 3.5).
`Twenty years before Etter's model for competing aggregate structures in the forma(cid:173)
`tion of polymorphs, Powers ( 1971) clearly stated the fundamental question regarding
`the challenge of understanding the nucleation process:
`
`It would appear almost certain that the development of these at least transient aggregates are
`the precursors to the development of phase transitions to the ordered solid nucleus, in harmony
`with the local thermodynamic conditions. Yet though a vast amount of study-conferences,
`books, etc.-have been devoted to nucleation, it is still uncertain how this last transition takes
`place.
`
`Some experimental evidence for the presence of different aggregates in solu(cid:173)
`tion leading to polymorphic structures has been presented recently by Nather et al.
`( 1996a,b ), and there have been attempts to relate nucleation rates with proposed struc(cid:173)
`tures in solution based on molecular modelling (Petit et al. 1994). The number of
`studies of this nature is sure to increase with increasing sensitivity and sophistication
`of both experimental and computational tools; those are the kinds of investigations
`that can provide answers to Powers' challenge.
`From a practical point of view, control over nucleation, and in cases of polymorphic
`systems, control over the polymorph obtained as a result of nucleation, has been
`
`Merck Exhibit 2050, Page 9
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`70
`
`CONTROLLING THE POLYMORPHIC FORM OBTAINED
`
`the concern of those industries for which crystallization is a crucial or final step
`in the production process: for example, sugars, amino acids, pharmaceuticals, and
`fatty acids. A number of examples of studies regarding polymorphic variation and
`preferences for nucleation may be found in the literature from those disciplines.
`The initiation of crystal growth has been a problem for the sugar industry since
`its infancy. Aqueous solutions of sugars often tend to form syrups-indeed, that has
`become one form of marketing, although clearly not the preferred one. As noted above,
`Powers has reviewed the role of nucleation in the sugar industry, including much of
`the accumulated experience involving sucrose. As in many industries, successful
`techniques developed over decades or centuries and were considered trade secrets or
`even commonplace practice without being scientifically recognized or understood.
`Thus, traditionally, sucrose crystallizations, carried out from the huge copper vats in
`which solutions were concentrated, were initiated (i.e. nucleated) by the mechanical
`shock of hammering on the vat (Fig. 3.1). Yet it was only in 1912 that Young (1911)
`described mechanical shock as a factor in nucleating supersaturated solutions (Powers
`1971). Two other curiosities relating to nucleation of sugars demonstrate some of
`the difficulties encountered. Turanose was long considered to be a liquid at room
`temperature (Powers 1971), until it spontaneously crystallized; following that event
`fresh batches of the material always crystallized. In another case more closely related
`to polymorphism, a-D-mannose had been prepared routinely until the appearance of
`,8-D-mannose, following which the a form could not be induc,ed to crystallize in the
`same laboratory (Levene 1935; Dunitz and Bernstein 1995). As Powers noted, both
`of these cases can be attributed to unintentional seeding, a topk treated in more detail
`in Section 3.6.
`Black and Davey (1988) describe a number of the interrelationships and practical
`aspects of the control of nucleation, crystal growth, and polymorphic transforma(cid:173)
`tion of amino acids. The factors described and demonstrated for primary nucleation
`of L-glutarnic acid include temperature, critical nucleus, relationship of interfacial
`tension to solubility, thermal history, induction time, agitation, and effect of additive.
`Kitamura (1989) studied many of these nucleation factors in the competitive crys(cid:173)
`tallization of the a and ,B forms of L-glutamic acid. He found that at 25 °Conly the a
`modification nucleates and grows. In this system, at least, the effect of temperature
`on the relative nucleation rates of the two polymorphs is more 'remarkable' than the
`effect of the supersaturation ratio: as the temperature is increased with a constant
`supersaturation ratio, the amount of a decreases, He also reported that the ,B form
`tends to nucleate in stagnant solutions, while at 25 °C essentially only a nucleated
`homogeneously.
`In an example from the pharmaceutical industry Sudo et al. (1991) studied the rel(cid:173)
`ative nucleation properties of forms A and B of cimetidine, which is reported to have
`four polymorphic non-solvated forms and three polymorphic monohydrates. Modi(cid:173)
`fication A is preferred for pharmaceutical formulations. The 'waiting time method'
`was used to study the primary nucleation process (Harano and Oota 1978), mainly for
`competitive crystallization of the A and B modifications. A is a thermodynamically
`metastable form and is more soluble than B in any solvent. At high supersaturation
`
`Merck Exhibit 2050, Page 10
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`AGGREGATION AND NUCLEATION
`
`71
`
`Fig. 3.1 Detail from an 1850 drawing of the London sugar refineries of Messrs. Fairrie and
`Co., showing a copper crystallizing pan for sugar. The worker to the right of the pan is holding
`a mallet which was used to bang on the pan to induce nucleation of the crystallization process.
`(Reproduced from Fairrie 1925, with permission.)
`
`(SA ::: 4.5) modification A is obtained, in the presence or absence of seeds. At
`SA ::: 3.6, A was obtained regardless of the form of seed. At SA :::: 2.0 the form of
`the seed determined the form obtained.
`Stearic acid is often considered a prototype for the long chain acids used in many
`processes and applications. Sato an.d Boistelle (1984) studied the occurrence and
`crystallization behaviour of three of the polymorphic modifications (A, B, and C) by
`varying conditions such as temperature, supersaturation, and solvent from which they
`determined occurrence domains for the existence of the three forms. Polymorph A
`is thermodynamically unstable at all temperatures studied; below 30 °C form B is
`most stable, while form C is more stable above 30 °C. Forms A and C nucleate
`
`Merck Exhibit 2050, Page 11
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`72
`
`CONTROLLING THE POLYMORPHIC FORM OBTAINED
`
`preferentially from non-polar solvents at high supersaturation, whereas polymorph B
`nucleates more readily at lower supersaturations. Nucleation of Form B is preferred
`at higher supersaturations from polar solvents. This solvent effect could be influenced
`by the rate of stirring. If stirring is sufficiently 'violent' it increases nucleation, which
`enhances the formation of Form B.
`The mere existence of polymorphic structures can be used as a. probe of the
`nucleation process. For instance in considering the aggregation process in super(cid:173)
`saturated solutions of 2,6-dihydroxybenzoic acid, Davey et al. (2000) found a direct
`link between the relative occurrence of two polymorphic forms (from toluene and
`chloroform solutions) and the solvent-reduced self-assembly (aggregation) of the
`;
`molecule.
`Nucleation from the melt has been studied for palm oil, composed of triglycerides
`of palmitic and oleic acids, and exhibiting at least three polymorphs (van Putte and
`Bakker 1987). Nucleation curves (induction timer vs temperature T) of palm oil and
`palm stearin show discontinuities at 297 and 306 °C respectively, indicating the onset
`of nucleation, and the demarcation of the occurrence of the polymorphs, as confirmed
`by isothermal Differential Scanning Calorimetry (DSC) studies (Ng 1990a,b).
`
`3.3 Thermodynamic vs kinetic crystallization conditions
`
`Physical organic chemists have long been accustomed to making the distinction
`between 'thermodynamic' and 'kinetic' conditions when referring to reactions and
`reaction mechanisms. In chemical parlance, thermodynamic conditions essentially
`means those conditions under which thermodynamic equilibrium is maintained or
`very nearly maintained. Kinetic conditions refer to situations that are far from
`equilibrium (van Hook 1961).
`In terms of crystallization (Ward 1997), thermodynamic conditions might refer to
`a slow evaporation, a very slow cooling, a slow crystallization from the melt at a
`constant temperature only slightly below the melting point, a slow sublimation for
`which there is only a small difference between the temperature of the solid and that
`of the cold finger on which the sublimate is crystallizing, etc. On the other hand,
`kinetic conditions might refer to a high degree of supersaturation, rapid cooling of a
`solution or melt, rapid evaporation of solvent, large temperature difference between
`the sample and cold finger in a sublimation, etc. A number of examples will serve to
`demonstrate how these principles have been applied to the crystallization of different
`modifications in polymorphic systems.
`
`p-chlorobenzylidene-N-p-chloroaniline 3-1 is dimorphic (Bernstein and Schmidt
`1972; Bernstein and Izak 1976). The thermodynamically more-stable orthorhombic
`
`3-1
`
`Merck Exhibit 2050, Page 12
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`THERMODYNAMIC VS KINETIC CRYSTALLIZATION CONDITIONS
`
`73
`
`form may be obtained by slow evaporation of a solution in which ethanol or methylcy(cid:173)
`clohexane is the solvent. Prismatic crystals usually grow in hours to days, depending
`on the initial concentration of the solution. The metastable triclinic form is obtained by
`dissolving the maximum amount of substance in a boiling ethanolic solution, which
`is then immediately placed in a desiccator freshly charged with calcium chloride.
`Needle-l*e crystals appear within minutes, and the desiccant aids in accelerating
`the rate of evaporation. The crystals are metastable and may begin to spontaneously
`transform to the orthorhombic modification in periods ranging from hours to days.
`Berman et al. (1968) noted that 'mannitol is unusual among carbohydrates in that
`exists in several polymorphic forms', indicating that a number of these are often
`obtained simultaneously. They describe the preparation of a number of these mod(cid:173)
`ifications. The a form is obtained by slow crystallization from 96 per cent ethanol,
`the a' form by evaporation from 100 per cent ethanol and the f3 form from aqueous
`ethanolic solutions, all apparently under thermodynamic conditions. On the other
`hand the y form is obtained kinetically by rapid cooling of a 1: 1 water-ethanol solu(cid:173)
`tion. An additional K form was obtained (unexpectedly) upon evaporation of a boric
`acid/methanol solution (Kim et al. 1968).
`Bock has studied a number of systems in which different polymorphs were obtained
`under thermodynamic and kinetic conditions. (2-pyridyl)(2-pyrimidyl)arnine 3-11 is
`dimorphic. Modification I is readily crystallized thermodynamically 'from any sol(cid:173)
`vent' (toluene was actually used) while modification II is obtained kinetically by fast
`evaporation of an ethereal solution or by resolidification of the melt (Bock et al. 1997).
`1
`((NY) 1-1,coxx:yyoct-1,
`~I~
`S
`OCH,
`
`1-1,CO
`
`3-11
`
`3-111
`
`In 2,3,7,8-tetramethoxythianthrene 3-111, the less stable (lower in both density and
`absolute value of lattice energy) monoclinic modification is obtained under kinetic
`conditions: rapid crystallization from polar diisopropyl ether, whereas the more stable
`(higher density and lattice energy) orthorhombic modification is thermodynamically
`obtained from a non-polar hydrocarbon solvent.
`In pharmaceutical applications the choice of polymorphic modification for formu(cid:173)
`lation depends very much on the robustness of the crystallization process as well
`as the properties and characteristics of the preferred modification. Hence, consider(cid:173)
`able effort is expended in gaining control over the polymorphic form obtained under
`various conditions. As noted above, up to four polymorphic modifications and three
`monohydrates have been reported for cimetidine (SmithKline Beecham's Tagamet®)
`(Bavin et al.1979; Prodic-Kojic et al. 1979; Shibata et al. 1983; Hegedus and Gorog
`1985). In experiments to selectively crystallize the A form in preference to the more
`stable B congener it was found that with isopropanol as a solvent, A crystallizes
`exclusively at high supersaturation, in the presence or absence of seeds (Sudo et al.
`1991).
`
`Merck Exhibit 2050, Page 13
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`74
`
`CONTROLLING THE POLYMORPHIC FORM OBTAINED
`
`In another example, an antiarrhythmic under development (McCauley et al. 1993)
`was shown to exist in two anyhydrous polymorphs, two dihydrated enantiotropic
`polymorphs, a monohydrate, and the solvates of several organic solvents. Following
`characterization of all of these modifications it was desired to selectively obtain one
`of the dihydrates, termed modification A, which is thermodynamically less stable at
`room temperature than another dihydrate, D, in contact with aqueous solutions, but A
`is more stable over a wider range of relative humidities. The enantiotropic transition
`point between these two crystal modifications is 37 °C. Procedures were developed for
`obtaining A preferentially. Above the transition point a thermodynamic crystallization
`is carried out at 50 °C, using type A seeds as an added precaution to force the crystal(cid:173)
`lization to type A. The desired type A can also be obtained under kinetic conditions'by
`spontaneous crystallization below the transition point followed by rapid filtration and
`removal of excess water. The latter procedure prevents a transformation from the A
`state (metastable below the transition temperature) to the D form in the crystallization
`medium. Similar considerations were applied to develop procedures for the selective
`crystallization of the a and f3 modifications of glutamic acid (Kitamura 1989).
`
`3.4 Monotropism, enantiotropism, and crystallization strategy
`
`The examples cited in the previous section involved distinguishing between thermo(cid:173)
`dynamic and kinetic conditions for a crystallization. For a practising chemist that
`distinction is often made instinctively rather than consciously. However, it can be
`related directly to the monotropic or enantiotropic relationship between two poly(cid:173)
`morphic forms, as expressed in the energy vs temperature diagrams (Sections 2.2.3
`and 2.2.4 ). Practical means for determining the monotropic/enantiotropic relationship
`of two phases are given in Chapter 4. This information in tum can be used to design
`strategies for attempts to obtain desired crystalline forms at the expense of the less
`desired ones (Ceolin et al. 1996). Here we summarize the ramifications of the par(cid:173)
`ticular monotropic/enantiotropic relationship on the crystallization strategy once that
`relationship has been determined, preferably by generating the energy vs temperature
`diagram. For a dimorphic system there are four possibilities:
`
`1. Obtaining the thermodynamically stable form in a monotropic system: no trans(cid:173)
`formation can take place to another form, and no precautions need be taken to
`preserve the stable form or to prevent a transformation.
`2. Obtaining the thermodynamically stable form in an enantiotropic system: pre(cid:173)
`cautions must be taken to maintain the thermodynamic conditions (temperature,
`pressure, relative humidity, etc.) at which the G curve for the desired polymorph
`is below that for the undesired one.
`3. Obtaining the thermodynamically metastable form in a monotropic system: a
`kinetically controlled transformation may take place to the undesired thermody(cid:173)
`namically stable form. To prevent such a transformation it may be necessary to
`employ drastic conditions to reduce kinetic effects (e.g. very low temperatures,
`very dry conditions, storage in the dark, etc.)
`
`Merck Exhibit 2050, Page 14
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`CONCOMITANT POLYMORPHS
`
`75
`
`4. Obtaining the thermodynamically metastable form in an enantiotropic system:
`the information for obtaining and maintaining this form is essentially found in
`the energy-temperature diagram.
`
`3.5 Concomitant polymorphs
`
`In situations where there is overlap between the occurrence domains of two or more
`polymorphs the modifications may appear simultaneously or in overlapping stages
`so that a particular procedure or process yields more than one form under identical
`conditions. This phenomenon is termed concomitant polymorphism and has been
`treated in considerable detail in a recent review (Bernstein et al. 1999).
`The situations in which polymorphs concomitantly crystallize are determined by
`the experimental conditions in relation to both the free energy-temperature relation(cid:173)
`ships and the relative kinetic factors. These situations may arise because either specific
`thermodynamic conditions prevail or the competing kinetic processes have equiva(cid:173)
`lent or very similar rates. In thermodynamic terms we have seen that polymorphs can
`coexist in true equilibrium only at the thermodynamic transition temperature (where
`the G curves cross). The chance of carrying out a crystallization precisely at such a
`temperature must be small with the inevitable conclusion if concomitant polymorphs
`are produced that kinetics play at least some role in the overall process of crystal(cid:173)
`lization. The final consequence of this situation is that a system of concomitantly
`crystallizing polymorphs will be subject to change in the direction favouring the for(cid:173)
`mation of the most stable structure. If the crystals have grown from and remain in
`contact with solution, then the most likely route for this transformation is via dis(cid:173)
`solution and recrystallization. This situation is c