`
`a.nVIII3t
`
`iOnaI
`
`analyt
`
`
`
`
`
`
`
`ical science journal
`
`.mAn
`
`Lupin Ex. 1037 (Page 1 of 28)
`Lupin Ex. 1037 (Page 1 of 28)
`
`
`
`’h°Analyst
`
`The Analytical Journal of The Royal Society of Chemistry
`
`Analytical Editorial Board
`
`Chairman: J. N. Miller (Loughborough, UK)
`
`M. Cooke (Sheffield, U/O
`C. S. Creaser (Nottingham, UtO
`A. G. Davies (London, UK) :
`A. G. Fogg (Loughborough, UK)
`J. M. Gordon (Cambridge, L~K)
`G. M. Greenway (Huff, UK}
`S. J. Hill (Plymouth, UK)
`
`D.L. Miles (Keyworth, UK)
`R.M. Miller (Gouda, The Netherlands)
`B."L. Sharp (Loughborough, UK)
`M.R. Smyth (Dublin, /roland)
`Y. Thomassen (Os/o, Norway)
`P. Vadgama (Manchester, UIO
`
`Advisory Board
`
`J. F. Alder (Manchester, UK)
`A. M. Bond (Victoria, Australia)
`J. G. Dorsey (Cincinnati, OH, USA)
`L. Ebdon (Plymouth, UK)
`A. F. Fell (Brad/ford, UK)
`J. P. Foley (Villanova, PA, USA)
`M. F. Gin~ (Sao Paulo, Brazil)
`T. P. Hadjiioannou (Athens, Greece)
`W. R. Heineman (Cincinnati, OH, USA)
`A. Hulanicki (Warsaw, Poland)
`I. Karube (Yokohama, Japan)
`E. J. Newman (Peele, UK)
`J. Pawliszyn (Waterloo, Canada)
`T. B. Pierce (Harwell, UK)
`
`E. Pungor (Budapest, Hungary)
`J. R~2idka (Seattle, WA, USA)
`R. M. Smith (Loughborough, UK)
`K. ~tulik (Prague, Czechoslovakia)
`J. D. R. Thomas (Cardiff, UK)
`J. M. Thompson (Birmingham, UK)
`K. C. Thompson (Sheffield, UK)
`P. C. Uden (Amherst, MA, USA)
`A. M. Ure (Aberdeen, UK)
`C. M. G. van den Berg (Liverpool, UK)
`A. Walsh, KB (Melbourne, Australia)
`J. Wang (Las Cruces, NM, USA)
`T. S. West (Aberdeen, UK)
`
`Regional Advisory Editors
`For advice and help to authors outside the UK
`Professor Dr. U. A. Th. Brinkman, Free University of Amsterdam, 1083 de Boelelaan, 1081 HV
`Amsterdam, THE NETHERLANDS.
`Professor P. R. Coulet, Laboratoire de G6nie Enzymatique, EP 19 CNRS-Universit~ Claude
`Bernard Lyon 1, 43 Boulevard du 11 Novembre 1918, 69622 Villeurbanne Cedex,
`FRANCE.
`Professor O. Osibanjo, Department of Chemistry, University of Ibadan, Ibadan, NIGERIA.
`Professor F. Palmisano, Universit~ Degli Studi-Bari, Departimento di Chimica Campus
`Universitario, 4 Trav. 200 Re David--70126 Bad, ITALY.
`’Professor K. Saito, Coordination Chemistry Laboratories, Institute for Molecular Science,
`Myodaiji, Okazaki 444, JAPAN.
`Professor M. Thompson, Department of Chemistry, University of Toronto, 80 St. George
`Street, Toronto, Ontario, CANADA M5S 1A1.
`Professor Dr. M. Valc~rcel, Departamento de Quimica Analitica, Facultad de Ciencias,
`Universidad de CSrdoba, 14005 Cbrdoba, SPAIN.
`Professor J. F. van Staden, Department of Chemistry, University of Pretoria, Pretoria 0002,
`SOUTH AFRICA.
`Professor Yu Ru-Qin, Department of Chemistry and Chemical Engineering, Hunan U niversity,
`Changsha, PEOPLES REPUBLIC OF CHINA.
`Professor Yu. A. Zolotov, Kurnakov Institute ef General and Inorganic Chemistry, 31 Lenin
`Avenue, 117907, Moscow V-71, RUSSIA.
`
`Editorial Manager, Analytical Journals: Janice M. Gordon
`
`Editor, The Analyst
`Harpal S. Minhas
`The Royal Society of Chemistry,
`Thomas Graham House, Science Park,
`Milton Road, Cambridge, UK CB4 4WF
`Telephone +44(0)1223 420066.
`Fax +44(0)1223 420247.
`E-MaihAnalyst@RSC.ORG(Internet)
`
`US Associate Editor, Fhe Analyst
`Dr Julian F. Tyson
`Department of Chemistry,
`University of Massachusetts,
`Box 34510 Amherst MA
`01003-4510, USA
`Telephone +1 413 545 0195
`Fsx +1 413 545 4846
`
`Senior Assistant Editor
`Caroline Seeley
`
`Assistant Editors
`Sarah Williams, Yasmin Khan
`Editorial Secretaries: Claire Harris, Frances Thomson
`
`Advertisements: Advertisement Department, The Royal Society of Chemistry, Burlington
`House, Piccadilly, London, UK WIV 0BN. Telephone +44(0)171-287 3091.
`Fax +44(0)171-494 1134.
`
`Information for Authors
`Full details of how to submit m.aterial for
`publication in The Analyst are g~ven in tile
`Instructions to Authors in the January issue,
`Separate copies are a.vailable on r.equest
`The Analyst publisnes origins! research
`papers, critical reviews, tut.oria~, reviews,
`perspectives, news artic es, DOOK reviews
`and a conference diary.
`Original research papers. The Analyst pub,
`lishes full papers on all aspects of the t.heory
`and practice of analytical chemistr.y, funds,
`mental and applied, inorganic ana organ e
`including chemical, physical, biochemica’,
`clinical, pharmaceutical, .biological: en.viron.
`mental, automatic and computer-based
`methods. Papers on new approaches to
`existing methods, new techniques and
`instrumentation, detectors and sensors, and
`new areas of application with due attentlol~
`to overcoming limitati.o.ns an.d to underly ng
`principles are all equally welcome.
`Full critical reviews. These must be s
`critical evaluation of the existing state of
`knowledge on a particular facet of analytical
`chemistry.
`Tutorial reviews. These should be infor-
`mally written although they should still be a
`critical evaluation of a specific topic area.
`Some history and possible future develop.
`monte should be given. Potential authors
`should contact the Editor before writing
`reviews.
`Perspectives. These articles should
`provide either a personal view or a philoso.
`phical look at a topic relevant to analytical
`science. Alternatively, they may be relevant
`historical articles. Perspectives are included
`at the discretion of the Editor.
`Particular attention should be paid to the
`use of standard methods of literature
`citation, including the journal abbreviations
`defined in Chemical Abstracts Service
`Source Index. Wherever possible, the
`nomenclature employed should follow
`IUPAC recommendations, and units and
`symbols should be those associated with SI.
`Every paper will be submitted to at least
`two’ referees, by whose advice the Editorial
`Board of The Analyst will be guided as to its
`acceptance or rejection. Papers that are
`accepted must not be published elsewhere
`except by permission. Submission ef a
`manuscript will be regarded as an under-
`taking that the same material is not being
`considered for publication by another
`journal.
`Regional Advisory Editors. For the benefit
`of potential contributors outside the UK and
`N. America, a Group of Regional Advisory
`Editors exists. Requests for help or advice o~
`matters related to the preparation of papers
`and their submission for publication in The
`Analystcan be sent to the nearest member of
`the Group. Currently serving Regional
`Advisory Editors are listed in each issue of
`The Analyst.
`Manuscripts (four copies typed in double
`spacing) should be addressed to:
`H. S. Minhas, Editor, or
`J. F. Tyson, US Associate Editor
`All queries relating to the presentation and
`submission of papers, and any correspon-
`dence regarding accepted papers and
`proofs, should be directed either to the
`Editor, or Associate Editor, The Analyst.
`Members of the Analytical Editorial Board
`(who may be contacted directly or via the
`Editorial Office) would welcome comments,
`suggestions and advice on general policy
`matters concerning The Analyst.
`There is no page charge.
`
`Fifty reprints are supplied free of charge.
`
`The Analyst (ISSN 0003-2654) is published monthly by The Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road,
`Cambridge, UK CB4 4WF. All orders, accompanied with payment by cheque n sterling, payab e on a UK clearing bank or in US dollars payable
`on a US clearing bank, should be sent directly to The Royal Society of Chemistry, Turpin Distribution Services Ltd., Blackhorse Road,
`Letchworth, Herts, UK SG6 1HN. Tu rpin Distribution Services Ltd., is wholly owned by the Royal Society of Chemistry. 1995 Annual su bscription
`rate EC £408.00, USA $749.00, Canada £428.00 (excl. GST), Rest of World £428.00. Purchased with Analytical Abstracts EC £807.00, USA
`$1472.00, Canada £841.00 (excl. GST), Rest of World £841.00. Purchased with AnalyticalAbstracts plus Analytical Proceedings EC £925.00, USA
`$1699.00, Canada £971.00 (excl. GST), Rest of.World £971.00. Purchased with Analytical Proceedings EC £492.00, USA $905.00, Canada £517.00
`(excl. GST), Rest of WorId £517.00. Air freight and mailing in the USA by Publications Expediting Inc., 200 Meacham Avenue, Elmont, NY 11003.
`USA Postmaster: Send address changes to: The Analyst, Publications Expediting Inc., 200 Meacham Avenue, EImont, NY 11003. Second class
`postage paid at Jamaica, NY 11431. All other despatches outside the UK by Bulk Airmail within Europe, Accelerated Surface Post outside
`Europe. PRINTED IN THE UK.
`© The Royal Society of Chemistry, 1995. 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, electronic, mechanical, photographic, recording, or otherwise, without the prior permission of the
`publishers.
`
`Lupin Ex. 1037 (Page 2 of 28)
`
`
`
`Analys~t, October 1995, Vol. 120
`
`.2435
`
`Analysis of Organic Polymorphs
`A Review
`
`Terence L. Thre:lfall,
`chemistry Department, University of Fork, Heslington, Fork, UK Y01 5DD
`
`Summary of Contents
`Introduction and Definition of Polymorphism
`Significance of Polymorphism
`D!Stinction From Related Phenomena
`Stability of Polymorphs
`Methods for the Examination of Polymorphs
`Microscopy
`Infrared Spectxoscopy
`Raman Spectroscopy
`Ultraviolet and Fluorescence Spectroscopy
`Solid-state Nuclear Magnetic Resonance and Nuclear
`Quadmpole Resonance Spectroscopy
`X-ray Crystallography
`Thermal Analysis
`Solubility and Density Measurement
`~olvates
`Quantitative Aspects
`Amorphous and Crystalline Solids
`References
`
`Keywords: Polymorphism; phase transitions; amorphous
`m~teriats; solvates; microscopy; thermal analysis; infrared
`spectroscopy; Raman spectroscopy; solid-state nuclear
`magnetic resonance spectroscopy; X-ray diffraction
`
`Introduction and Definition of Polymorphism
`
`P01ymorphism1-7 in the chemical sense of the word* is a
`phenomenon of the solid state, associated With the structure of
`the solid. It has’proved difficult to define precisely although the
`basic concept is readily understood. The definitions which have
`been offered vary in breadth but the implication of all of them
`is that polymorphs involve different packings of the same
`molecules il~ the solid.4 The question of how similar the same
`molecules must be and of how dissimilar the different packing
`arrangements must be in order to qualify as polymorphs is more
`than a matter of semantics but goes to the root Of our
`understanding of the organic molecular solid state.
`McCrone has defined a polymorph as ’a solid crystalline
`phase of a given compound resulting from the possibility of at
`least two crystalline arrangements of the molecules of that
`compound in the solid state’ and has listed those types of solid
`phenomena which are excluded from this definition,x Later
`writers who have accepted this definition have tended to
`substitute their own list of exclusions,5 if they have addressed
`the matter at all, Buerger’s tentative definition3 ’ideally, two
`polymorphs are different forms of the same chemical compound
`which have distinctive properties’ is broader and appears not to
`
`* An on-line search of Chemical Abstracts will reveal more tt~an 47000 entries under
`’polymorphism’, Over 90% of these relate to genetic polymoi:phism, which at least in
`its origins can claim the tree etymology of the word. Some selectivity between
`biological and chemical uses can be achieved, but there is no certain searching strategy.
`Searching under ’phasq transition’ and related concepts will generate a further 44000
`entries, most of which refer to inorganic systems, and cannot be easily disentangled.
`Nevertheless, these represent only a proporiion of the papers containing infolanatton on
`polymorphs and polymorphism. Hence it is not possible to state how many
`publications relate to thos~ aspects of polymorphism described here.
`
`accept th~ need for separate phases and to include amorphous
`forms. The nature of the amorphous stateS,9 will be discussed
`later.
`Polytypism~0 is one-dimensional polymorphism, referring to
`different stacking of the same layers. It is most familiar in
`inorganic systems, particularly silicon carbide, but has been
`recognized in o~ganic crystals, both as orde~ed~-~3 and as
`disordered stacking.~4 There is no special term for two-
`dimensional polymorphism, although some liquid crystal
`systems display it. Liquid crystals are notorious for their ability
`to exist in different phases both in the mesomorphie and in the
`solid statO5at7 and this has led to the suggestion that the term
`polymorphism should apply to liquids as well as solids,~ but il
`is only the solid dimensions of liquid’crystals which can adop~
`distinct packing arrangements. Liquid-crystal polymorphism
`will not be dealt with specifically in this review except where it
`ts related to the polymorphism of solids. The long standing
`question~9 of whether allotropy and polymorphism are dis-
`tinct~0 is not an issue in the case of organic compounds.
`Inorganic polymorphs have been excluded because the ex-
`tended structures of which most inorganic crystals are com-
`posed raise concepts not discussed here.?~,~2 Protein polymor-
`phism usually refers to minor molecular sequence changes23-24
`rather than to packing, but different crystal packing of protein
`molecules is also known.Z5 Polymorphism of thin films~6,27 and
`polymers, both Of biological?S,29 and of osynthetic3° origin,
`although of the same nature as t.he concept of polymorphism
`considered here, will not be discussed.
`There is a profusion of words in the English language forthe
`phenomena discussed in this !:eview, yet not enough because of
`the overlapping usage. ’Polymorph’. (din)orph, t~imorph)~form’
`and ’mddification’ are all used to describe polymorphidphases
`but ’form’ and ’modification’ are also used in reference to
`crystal habit. ’Polymorph’ and ’form’ have been Used to
`describe solvates, whilst ’pseudopolymorph’ doubles for both
`solvates and for those solids which are otherwise not conside~’ed
`true polymorphic forms. The term ’pseudopolymorphic solvate’
`applied to crystals losing solvent molecules without change of
`crystalline form offers yet another source of confusion in
`terminology Genetic polymorphism which is now the major use
`of the term is often described as ’polymorphisms’ but this is
`occasionaliy seen also in chemical senses. In view of the almost
`universal use of ’polymorphic’ as the appropriate adjective, the
`word ’polymoq~hous’ seems superfluous despite dictionary
`support. There is an urgent need for consistent usages so as to be
`able to delineate the phenomena under consideration.
`There is no clear choice as to the best method of designating
`polymorphs. Arb]trm3, systems-are to be discouraged, but
`numbering based either on order of melting point or of room
`temperature stability have been recommended. Both are
`susceptible to change as a result of later identification of new
`polymorphic forms. Numbering based on order of discovery is
`unchangeable,- but requires a lmowledge of the history of the
`compound. The addition of the crystal class, as has been
`suggested for minerals3~ is not very practicable, since crystal-
`lographic classes are rarely determined from optical micro-
`scopic or X-ray powder diffraction studies for organic com-
`pounds. The assignment of a space group is even less realistic.
`
`Lupin Ex. 1037 (Page 3 of 28)
`
`
`
`2436
`
`Analyst, October 1995, Vol. 120
`
`In any case the distribution of organic molecules amongst
`crystal,°¢la~s~s, 0nd spac~ g~o, ups is extremely limited, as is
`dlscussdd ~ater.~L3~3 T,l~e aifdttton;of~,meltmg or upper transition
`point to a Roman numeral is probably the best compromise,1
`although ca~d(i~t~st be ~ak~tt~t0~’distinguish the melting point of
`the polymoriJh and tha~"of~tfi~ ~ransformed product.
`
`Significance of Polymorphism
`
`The continuing investigation of polymorphism by the Innsbruck
`school (Kofler, Kuhnert-Brandst~ttter, Burger) over more than
`half a century has shown that around one2third of organic
`substancesshow crystalline polymorphism under normal pres-
`sure conditions.34,35 A further third are capable of forming
`hydrates and other solvates.
`Muchof the literature on the polymorphism of organic
`compounds relates to pharmaceutical products.1,36~° The
`incentive for this interest in polymorphism began with the need
`to satisfy regulatory authorities in various countries as to the
`bioavailability of formulations of new chemical entities.~6,37 Of
`the several contributory factors to the bioavailability of finished
`products, the inherent solubility and rate of dissolution of the
`drug substance itself are of major importance. The solubility is
`dependent on the polym0rphic state, as different polymorphs
`have different energies and therefore different solubilities.4° It
`has been pointed out, particularly by Burger,36 that the
`difference in solubility between polymorphs is likely to result in
`significant bioavailability differences, in practice, only in
`exceptional cases. Although some may think that this represents
`an extreme v~ew, the consequences of p01ymorphism on
`bioavailability are cbmmonly overstated. Chloramphenicol
`palmitate, overwhich the original concerns were voiced.4~ is
`unique in that the s01ubilit) is related to the rate of ehzymic
`attack on the solid.4z This and n~;Cobidein,4~ which involves
`consideration of the amorphous state, hre among the handful of
`examples ’of marketed products showing major bioavailability
`differences as a result of polymorphism.
`As formulations have become more sophisticated and as the
`tolerances on products have become tighter, the need to identify
`polymorphic behaviour at an early stage of development has
`become important in the pharmaceutical industry as a means of
`ensunng reliable and robust processes44 and conformity with
`good manufacturing practice. The aim is to avoid, inter alia,
`tabletting problems and subsequent tablet failure,45,46 crystal
`growth in suspensions47,4s and resultant caking, precipitation
`fr6m solutionS and problems with suppositories,49 as well as
`chemical production problems such as filtrabilityl and to ensure
`analytical reproducib.ility. By extension such considerations
`relate to the control of quality in manufacture and product
`reliability in any industry by ensuring that the processes are well
`Ufiderst00d and under control so that unpleasant surprises do not
`occur.5° This point is most dramatically illustrated in the
`explosives industry, where the wrong polymorph can have
`greatly increased Sensitivity to detonation.S1,5~ Pigment colour
`and solubility are polymorph dependent,~-59 as are photo-
`graphic and photolithographic sensitizers.60 The performance of
`industrial products, particularly those based on natural fats and
`waxes6L62 and derived soaps,63 and on petrolefim prodncts64,65
`is in many cases related to polymorphic composition and degree
`of crystallinity. The same is true of the processing, acceptability
`and deterioration of foods and confectionery containing
`fats,66’67 sugars,68-72 polysaccharides7~ and other constitu-
`ents.74-75 A comprehensive summary of the solid-state proper-
`ties of lipids has recently appeared.76
`It is also worth establishing the polymorphic behaviour of a
`compound for the sake of good order in documentation so that
`reference works, fo~: example, pharmacopoeias, do not contain
`conflicting data34,77 such as a spectrum of one polymorph, but
`the melting point of another.
`
`A nlajor incentive to the study o~ polymorphism in the
`pharmaceutical industry during development has become
`strikingly apparent recently in the use of subsidiary, patents on
`desirable polymorphic formsTM to prolong the patent life of
`major products Much recent pharmaceutical patent litigation
`has concerned polymorphs and particular interest has been
`taken in Glaxo’s patent on the polymorph, of ranitidine7o
`(Zantac) which if held valid will extend the patent prote, ction
`fi’om 1995 to 2002 in many countries,s° For a compound with
`annual sales of over 2 400 million pounds sterling,sl the
`financial incentives to investigate polymorphs are obvious.
`Finally, the very existence of polymorphism tells us some-
`thing about the solid-gtate. Investigation of polymorphie
`systems, especially those with a large number of forms can help .....
`in understanding solid-state and molecular behaviour, and
`intermolecular interactions8z and the relationship between
`crystal structure, crystal growth and crystal habits3 and their
`influence on bulk properties. Apart from lcnowledge for its own
`sake, this is of clear application in the development of organic
`electronics4,s5 and other specialty productss6-s8 and in under-
`standing the function of biological tnembranes,s9
`
`Distinction From Related Phenomena
`
`At one time polymorphism was regarded only as different
`arrangements of rigid molecules in the solid state.90.9~* A clear
`dichotomy existed between this and arrangements of molecules
`in different forms, such as could be imagined would occur,vith
`isomeric, tautomeric, zwitterionic and chiral structures and later
`with different conformers.9z The early crystallographic studies
`on rigid aromatic molecules tended to reinforce the distin~ion.
`This simple division could only be maintained whilst details of
`the rich variety of solid-state structures were inaccessible. The
`early examples of dynamic isomerism and tautomerism were
`few93,94 and the proposition that they could not be part of
`polymorphism was copied by reviewers until even the examples
`were forgotten.95 A quoted example of a tautomeric s01id-state
`structure, that of 3,5-dichloro-2,6-dihydroxy dimethyl tere-
`phthalic acid was shown in 1972 not to be tautomeric, but to
`involve conformational change with hydrogen bonding differ-
`ences.96 One would have expected examples of tautomerically
`related solid structures to be exceedingly numerous, since the
`molecular energetic requirements can easily be fulfilled as is
`shown by the widespread occurrence of tautomerism in
`solution.97 Tantomeric polymorphism is surprisingly rare, but a
`well investigated example is now loaown, that of 2-arffino-
`3 -hydroxy-6-phenylazopyridine.9s
`There are a few papers, in the literature either where
`tautomeric polymorphism is invoked99-~°5 or where examina-
`tion of the IR spectra is suggestive of forms whose difference
`resides in transfer of hydrogen between one part of the molecule
`and another,m6 The instances of 1,3-cyclohexadienone and
`squaric acid (3,4-dihydroxy-3-cyclobutene-l,2-dione are more
`difficult to place unambiguously in the category of tautomeric
`polymorphism. Proton transfer between donor and acceptor
`oxygen sites results in little change in over-all structure.1°7
`Both tantomeric equilibrium and the neutral < ~ zwitterionic
`equilibrium formally involve such .an intramolecular hydrogen
`transfer. The nominal, difference is that ~a charge separation is
`produced in zwitterions which cannot’ be extinguished intra-
`moleculafly by a double-bond rearrangement cascade. The
`difference may be even smaller in practice because charge
`stabilization of zwitterions can occur intermolecularly, for
`example, in solution through solvation, whilst tautomeric
`structures can retain a substantial part of their charge as shown
`hy dipole moment and IR spectroscoptc studies.mS,m9 Anthra-
`
`* Earli(r literature can be accessed via m~efences 1, 2 and 10.
`
`Lupin Ex. 1037 (Page 4 of 28)
`
`
`
`nilic acid exists as two metastable forms containing only
`uncharged molecules and a. form stable at room temperature,
`half the molecules of which have been shown from crystallo-
`graphic studies to be zwitterionic and half uncharged,lm A
`related phenomenon is the changing of allegiance of hyda’ogen-
`bonded hydrogens between electron donor atoms, which is a
`prolific source of polymorphism,m The role of hydrogen-
`bonding networks in determining crystal structure has been
`discussed extensively,x12 Conformational differences between
`molecules of different structures have been admitted, perhaps
`reluctantly, and distinguished by the title conformational
`polymorphism.~13 The original examples form one extremity
`where molecules in distinctive conformations pack similarly,92
`hut it is now obvious froln the plethora of crystal structures, as
`could always have been deduced from elementary considera-
`tions of energy minimization; that any change of packing will
`cause geometrical change in molecules and conversely that any
`change in geometry will invite different packing of the
`molecules.82 The extent will depend on the rigidity of the
`,molecules. Although some floppy ring systems maintain their
`shape in different forms~14,~t5 even nominally rigid structures
`such as the ring systems of steroids116 can show substantially
`different conformations in different polymorphs. Heteroaro-
`matic117-~21. and benzoquinone~zz planes are frequently bent
`and even benzene ringslz3 may be. Thus it seems pragmatic to
`accept conformational polymorphism as a normal sub-set of
`polymorphism and the term will only be used here when it is
`necessary to distinguish cases of substantial conformational
`change.
`The distinction between p01ymorphism and chirality is made
`in most accounts of polymorphism; yet it has recently been
`pointed out that if conformational polymorphism is accepted,
`then racemates and conglomerates of rapidly interconverting
`chiral systems are in fact polymorphs.5 Such systems are
`generally ones with al~ easy conformational change where the
`trivial distinguishing feature from other conformational poly-
`.morphism is that the result of such a change is a reflection of an
`asymmetrical structure across a mirror plane. Although this
`seems difficult to accept, the dextrorotatory and laevorotatory
`forms of such systems are then equally polymorphs.124 The
`narrow line of demarkation between polymorphism,, conforma-
`tional polymorphism and chirality first seems to have been
`recognized by Eistert et al..~25 Examples of rapidly inter-
`changing enantiomers in solution capable of independent
`existence in the solid state are known126,127 but uncommon.
`A further extension of the concept of conformational
`polymorphism is to be found where there is rapid interconver-
`sion between isomers.~8 As in the chiral examples, one
`molecular species or the other becomes exclusively incorpor-
`ated in the crystal because the mechanism of crystal growth acts
`as such an exquisitely discriminatory process.129
`Since a hydrate and an anhydrous form are constitutionally
`distinct, they cannot bear a strictly polymorphic relationship on
`the basis of any definition. However, the observation of material
`of differen~ lnelting point or other properties during re-
`crystallization may be due (apart from chemical reaction with
`solvent or decomposition) to solvation or polymorphism and the
`methods of examination are similar in either case. Hence the
`term ’pseudopolymorphism’ has become common~3° particu-
`larly in the pharmaceutical industry. The, term .seems un-
`necessary and could lead to confusion13~ with its use to describe
`all other phenomena related to pblymorphism~ and so will not
`be used here. It must be emphasized, however, that the
`distinction between solvates and polymorphs is not as clear-cut
`as might be imagined, either conceptually or practically.
`
`* In the Case of phenothiazines 121 the point of interest is not that the ring system is bent,
`but that the heteroatoms are out of the plane of the aromatic rings and in the opposite
`sense to expectation.
`
`Analyst, October 1995, Vol. 120
`
`. 2437
`
`The traditional narrow view of polymorphism, rigidly
`excluding chirality and isomerism, has caused considerable
`difficultylab to the investigators of the systems described above
`and it is suggested that the way to avoid these problems is to
`adopt the gloss originally proposed by McCrone and co-
`workers~,~7 on his definition of polymorphism, namely that the
`criterion is that the component molecules must have the same
`structure in solution irrespective of the polymorph from which
`they were derived; but, a:s has been suggested by Dunitz?
`without excluding tautomerism, isomerism or conformers per
`se. Thus, rapidly interconverting species would be accepted,
`whilst slowly interconverting species would be excluded, as
`was surely within the original contemplation. Despite appear-
`ances, this proposal is likely, to multiply examples of poly-
`morphism very little and it avoids what otherwise must be
`artificial situations of accepting phases as polymorphs based on
`impeccable polymorph behaviour until their crystal structure
`reveals excluded molecular forms.98,11°,~3~ !f, as asserted, the
`transition between polymorph I and polymorph II of 1,3-cyclo-
`hexadiene occurs by transfer of hydrogen from one oxygen to
`another, then this is nominally an example of tautomeric
`polymorphism,m7 If, on thc other hand, the same change occurs
`or can be made to occur.by a movement of the whole molecule
`then it is an example of regular polymorphism. The boundaries
`between the various alternative solid structural concepts are too
`subtle and too vague to be used to define polymorphism.
`Although the requirement of the same structure in solution
`has been canvassed above, one-component phase diagrams are
`constructed on the basis of equilibrium with vapour, rather than
`liquid. It is just in the instance of conformational, configura-
`tional or hydrogen mobility that molecular differences between
`vapour,~33,1~4 melt, solution126,135 and solid are found. The
`mobilities are inevitably of different magnitudes in different
`states. We shall be increasingly obliged to decide where to draw
`the boundaries of polymorphism as more comparative studies
`involving polymorphs and molecular structure in different
`states are undertaken.
`One negative consequence of accepting the wider view of
`polymorphism should be noted, namely that the thel’modynamic
`relationships: discussed later, are likely to be less certain for the
`wider polymorphic family.9° . ¯ ¯
`
`Stability of Polymorphs
`
`~
`
`.
`
`Polymorphs, or strictly dimorphs where only two forms are
`under consideration, may be in an enantiotropic or monotropic
`relationship.19,~ An enantiotropic relationship implies that
`each form has a range of temperature over which it is stable with
`respect to the other and a transition point at which the forms are
`equistable and in principle interconvertible.~37 Above that
`temperature the thermodynamic tendency is to the formation
`exclusively of the form stable at the higher temperature. Below
`the transition temperature the low- temperature form is the only
`stable one with respect to the other, although there is usually a
`greater tendency for the high temperature form to become
`frozenAn than for a low- temperature form to persist beyond its
`stability range.8 Forms outside their range of stability are
`described here as metastable138. In the case of a monotropic
`relationship one form is metastable with respect to another at all
`temperatures. There is no observable transition point, although
`the thermodynamic description implies a theoretical transition
`point above the melting point which is therefore unattainable.~9
`The use of the terms enantiotropic or monotropic in reference to
`a phase, as opposed to a transition, is ambiguous and likely to
`lead to confusion, since a polymorph can have a monotropic
`relationship to a second polymorph, but be enantiotropic in
`relation to a third polymorph. Flufenamic acid provides such an
`example.14o The distinction between thermodynamic and
`kinetic transition points also needs to be drawn.~41
`
`Lupin Ex. 1037 (Page 5 of 28)
`
`
`
`2438
`
`Analyst, October t995, Vol. 120
`
`Polymorphs only exist in the solid state: melting or
`dissolution destroys any distinctions. It is therefore important in
`examining polymorphs analytically not to submit them to
`conditions under which they melt, dissolve or are rendered more
`likely to interconvert. Heating and grinding142-~44 are obviously
`potentially hazardous operations in this context, but often
`cannot be avoided. The presence of solvent, even one in which
`the substance appears insoluble, will speed up .the inter-
`conversion.145 Trace moisture, acid or alkali on vessels can be
`similarly effective in interconverting polymorphs or in catalys-
`ins competing and confusing phenomena such as ring-opening
`reactions, for example, in 3,5-dihydroxy-3-methylvaleric acid
`derivatives,146, o~ group transfer reactions.~47
`It might be s