USF Libraries
USF Digital Collections

Crystal engineering of organic compounds including pharmaceuticals

MISSING IMAGE

Material Information

Title:
Crystal engineering of organic compounds including pharmaceuticals
Physical Description:
Book
Language:
English
Creator:
Bis, Joanna A
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Supramolecular chemistry
Supramolecular synthon
Hydrogen bond
Co-crystal
Polymorphism
Dissertations, Academic -- Chemistry -- Doctoral -- USF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
Neutral or charge-assisted hydrogen bonds occurring between organic molecules represent strong and directional forces that mediate the molecular self-assembly into well defined supramolecular architectures. A proper understanding of hydrogen bonding interactions, their types, geometries, and occurrence in supramolecular motifs, is a prerequisite to crystal engineering, i.e. to the rational design of functional solid materials.Multiple-component organic crystals represent ideal systems to study the intermolecular interactions between the constituent molecules that can be pre-selected for their hydrogen bonding sites and geometrical capabilities. In particular, the systematic structural analysis of supramolecular systems that are comprised of simple molecules facilitates the development of strategies for the rational design of new multiple-component compounds involving more complex components such as drug molecules.The work presented herein shows a combination of systematic database and experimental studies in the context of reliability and hierarchy of several hydrogen bonded supramolecular synthons that exist in a series of model co-crystals and organic salts. The acquired paradigms are ultimately utilized in crystal engineering of pharmaceuticals. In addition, the viability of a mechanichemical approach toward supramolecular synthesis in the context of its efficacy and the effect on polymorphism in multiple-component compounds is also addressed.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2006.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Joanna A. Bis.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 199 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001787550
oclc - 124066589
usfldc doi - E14-SFE0001424
usfldc handle - e14.1424
System ID:
SFS0025744:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Crystal Engineering of Organic Compounds Including Pharmaceuticals by Joanna A. Bis A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemistry College of Arts and Sciences University of South Florida Major Professor: Michael J. Zaworotko, Ph.D. Julie P. Harmon, Ph.D. Edward Turos, Ph.D. Matthew L. Peterson, Ph.D. Date of Approval: April 12, 2006 Keywords: Supramolecular chemistry, Supramolecular Synthon, Hydrogen Bond, Cocrystal, Polymorphism Copyright 2006, Joanna A. Bis

PAGE 2

Dedication To my parents and grandparents

PAGE 3

Acknowledgements I would like to thank my advisor, Pr ofessor Michael J. Zaworotko, for the opportunity to conduct research under his su pervision, and for his advice and guidance throughout the Graduate Program. I would also like to thank Dr. Julie P. Harmon, Dr. Edward Turos, and Dr. Matthew L. Peterson, my committee members, for their helpful comments and encouragements. I wish to express my appreciation to Dr. Elisabeth Rather, Dr. Brian Moulton, and Dr. Victor Kravtsov for their invaluable cont ribution to my crystallography experience, and Dr. Peddy Vishweshwar for his helpful di scussions related to the preparation of scientific manuscripts. In addition, I would like to acknowledge all members of my research group, as well as Faculty and Staff of the Chemistry Department of University of South Florida, for their friendly accommodation. At last, I would like to express my deepes t thanks to my closest family and friends who constantly supported me th roughout the years of studies.

PAGE 4

i Table of Contents List of Tables vii List of Figures xix List of Schemes xvi Abstract xvii Chapter 1 Introduction 1 1.1. Supramolecular Chemistry 1 1.1.1. Fundamentals 1 1.1.2. Supramolecular Interactions and the Role of Hydrogen Bonds 2 1.2. Crystal Engineering 3 1.2.1. Fundamentals 4 1.2.2. Hydrogen Bonded Supramolecular Synthons 6 1.2.3. The Cambridge Structural Database 8 1.3. Hydrogen Bonded Organic Co-crystals 9 1.3.1. Co-Crystals in the Context of Investigation of Supramolecular Heterosynthons 12 1.3.2. Co-Crystals in the Context of Green Chemistry 14

PAGE 5

ii1.3.3. Co-Crystals in the Context of Polymorphism 16 1.3.4. Co-Crystals in the Context of Pharmaceuticals 18 1.4. Summary 22 Chapter 2 Recurrence of hydroxylaroma tic nitrogen supramolecular heterosynthon in the presence of a competing cyano acceptor 24 2.1. Focus 24 2.2. Results and Discussion 25 2.2.1. CSD Analysis 26 2.2.2. Structural Features of Neutral and Ionic HydroxylAromatic Nitrogen Interaction 29 2.2.3. Crystal Structure Descriptions 30 2.3. Conclusions 46 2.4. Experimental 47 2.4.1. Syntheses 47 2.4.2. Single Crystal X-ray Crystallography 51 Chapter 3 Methods of Preparation of Co-c rystals and Polymorphism in Cocrystals 56 3.1. Focus 56 3.2. Results and Discussion 57 3.2.1. Methods of Preparations of Co-crystals 57 3.2.2. Polymorphism in 4-cyanophenol trans-1,2-bis-(4pyridyl)ethylene co-crystal 60

PAGE 6

iii3.2.3. Polymorphism in (4-cyanopyridine)2 4,4-biphenol cocrystal 63 3.3. Conclusions 74 3.4. Experimental 76 3.4.1. Syntheses 76 3.4.2. Single Crystal X-ray Crystallography 77 Chapter 4 Robustness of Supramolecular Heterosynthons: 2-Aminopyridinium Carboxylate 80 4.1. Focus 80 4.2. Results and Discussion 81 4.2.1. CSD Analysis 82 4.2.2. Structural Features of Neutral and Ionic 2AminopyridineCarboxylic Acid Interaction 86 4.2.3. Crystal Structure Descriptions 88 4.3. Conclusions 103 4.4. Experimental 104 4.4.1. Syntheses 104 4.4.2. Single Crystal X-ray Crystallography 107 Chapter 5 Crystal Engineering of Pharmaceuticals 110 5.1. Focus 110 5.2. Results and Discussion 111 5.2.1. CSD Analysis and Literature Overview 111

PAGE 7

iv5.2.2. Bicalutamide 116 5.2.3. Indomethacin 125 5.3. Conclusions 129 5.4. Experimental 132 5.4.1. Syntheses 132 5.4.2. Single Crystal X-ray Crystallography 133 Chapter 6 Summary and Future Directions 136 6.1. Summary 136 6.2. Future Directions 139 Appendices 159 Appendix 1. Experimental data for compound 1 160 Appendix 2. Experimental data for compound 2 161 Appendix 3. Experimental data for compound 3 162 Appendix 4. Experimental data for compound 4 163 Appendix 5. Experimental data for compound 5 164 Appendix 6. Experimental data for compound 6 165 Appendix 7. Experimental data for compound 7 166 Appendix 8. Experimental data for compound 8 167 Appendix 9. Experimental data for compound 9 168 Appendix 10. Experimental data for compound 10 169 Appendix 11. Experimental data for compounds 11a and 11b 170 Appendix 12. Experimental data for compound 12 171

PAGE 8

vAppendix 13. Experimental data for compound 13a and 13b 172 Appendix 14. Polymorphism screen data for compound 1 173 Appendix 15. Polymorphism screen data for compound 2 174 Appendix 16. Polymorphism screen data for compound 3 175 Appendix 17. Polymorphism screen data for compound 4 176 Appendix 18. Polymorphism screen data for compound 5 177 Appendix 19. Polymorphism screen data for compound 6 178 Appendix 20. Polymorphism screen data for compound 7 179 Appendix 21. Polymorphism screen data for compound 8 180 Appendix 22. Polymorphism screen data for compound 9 181 Appendix 23. Polymorphism screen data for compound 10 182 Appendix 24. Polymorphism screen data for compound 11a and 11b 183 Appendix 25. Polymorphism screen data for compound 12 185 Appendix 26. Polymorphism screen data for compound 13a and 13b 186 Appendix 27. Experimental data for compound 14 187 Appendix 28. Experimental data for compound 15 188 Appendix 29. Experimental data for compound 16 189 Appendix 30. Experimental data for compound 17 190 Appendix 31. Experimental data for compound 18 191 Appendix 32. Experimental data for compound 19 192 Appendix 33. Experimental data for compound 20 193 Appendix 34. Experimental data for compound 21 194

PAGE 9

viAppendix 35. Experimental data for compound 22 195 Appendix 36. Experimental data for compound 23 196 Appendix 37. Experimental data for compound 24 197 Appendix 38. Experimental data for compound 25 198 Appendix 39. Experimental data for compound 26 199 About the Author End Page

PAGE 10

vii List of Tables Table 1.1. Comparison of selected chem ical bond types and their features 3 Table 1.2. The occurrence of polymorphism in organic compounds 18 Table 1.3. Occurrence of hydroge n bonding moieties in APIs 19 Table 2.1. CSD statistics related to s upramolecular synthons that occur in structures containing onl y OH, Narom, and CN 28 Table 2.2. Comparison of the melting points of co-crystals 1-12 and the corresponding components 51 Table 2.3. Crystallographic data and stru cture refinement parameters for cocrystals 1-12 53 Table 2.4. Geometrical parameters of supramolecular heterosynthon I present in co-crystals 1-12 55 Table 3.1. Crystallographic data and stru cture refinement parameters for cocrystals 11b, 13a, and 13b 78 Table 3.2. Geometrical parameters of supramolecular heterosynthon I present in co-crystals 11b 13a, and 13b 79 Table 4.1. Percentage occurrence, distan ce ranges, and average distance for supramolecular synthons IV-VII 84

PAGE 11

viiiTable 4.2. Crystallographic data and st ructure refinement parameters for compounds 14-22 108 Table 4.3. Geometrical parameters of select ed intermolecular interactions present in compounds 14-22 109 Table 5.1. Comparison of the melting points of co-crystals 23-25 and the corresponding components 125 Table 5.2. Crystallographic data and st ructure refinement parameters for compounds 23-26 134 Table 5.3. Geometrical parameters of s upramolecular heteros ynthons present in compounds 23-26 135

PAGE 12

ix List of Figures Figure 1.1. Examples of supramolecula r synthons: carboxylic acid homosynthon (left) and carboxylic acidpyr idine heterosy nthon (right) 7 Figure 1.2. Crystal structure of th e triclinic form of quinhydrone 10 Figure 1.3. The Hoogsteen base-pairing in the structure of 9-methyladenine 1methylthymine co-crystal 10 Figure 1.4. Distribution of hydr ogen bonded co-crystal structures archived in the CSD between 1978-2004 11 Figure 1.5. Itraconazole succinic acid co-crystal 21 Figure 1.6. Carbamazepine saccharin co-crystal 22 Figure 2.1. Histograms of contacts for s upramolecular synthons: a) OHNarom in I; b) OHN C in II; and c) OHO in III 29 Figure 2.2. Histograms representing th e distribution of carbon-oxygen bond lengths in a) neutral phenolic moie ties and b) deprotonated phenolic moieties 30 Figure 2.3. Crystal structure of 3-cyanophenol 4-phenylpyridine, 1 31 Figure 2.4. Crystal struct ure of (3-cyanophenol)2 1,2bis -(4-pyridyl)ethane, 2 32 Figure 2.5. Crystal struct ure of (3-cyanophenol)2 t-1,2bis -(4-pyridyl)ethylene, 3 33

PAGE 13

xFigure 2.6. Crystal structure of 3-cyanophenol t-1,2bis -(4-pyridyl)ethylene, 4 35 Figure 2.7. Crystal structure of phenazine hydroquinone, FOQHEY 35 Figure 2.8. Crystal structure of 4-cyanophenol 4-phenylpyridine, 5 36 Figure 2.9. Crystal struct ure of (4-cyanophenol)2 4,4-bipyridine, 6 37 Figure 2.10. Crystal struct ures of (4-cyanophenol)2 1,2bis -(4-pyridyl)ethane, 7 37 Figure 2.11. Crystal struct ure of (4-cyanophenol)2 t-1,2bis -(4-pyridyl)ethylene, 8 38 Figure 2.12. Representation of the relativ e orientation of the adjacent 2:1 adducts: a) down the a axis in 6 and b) down the b axis in 8 38 Figure 2.13. Crystal struct ure of (3-cyanopyridine)2 4,4-biphenol, 9 39 Figure 2.14. Crystal struct ure of (4-cyanopyridine)2 resorcinol, 10 40 Figure 2.15. Crystal struct ure of (4-cyanopyridine)2 4,4-biphenol, 11a 41 Figure 2.16. Crystal struct ure of (4-cyanopyridine)3 phloroglucinol, 12 42 Figure 2.17. Crystal structur e of (4,4-bipyridine)3 phloroglucinol, TEKKOJ 43 Figure 2.18. Supramolecular hetrosynthon I pr esent in the crystal structure of 3hydroxypyridine 44 Figure 3.1. Solid-to-liquid conversion occu rring in a mixture of 3-cyanopyridine and 1-napthol 60 Figure 3.2. Crystal structure of 4-cyanophenolt-1,2bis (4-pyridyl)bipyethylene form I, 13a 61 Figure 3.3. Crystal structure of 4-cyanophenolt-1,2bis (4-pyridyl)bipyethylene form II, 13b 62

PAGE 14

xiFigure 3.4. X-ray powder diffract ion patterns of 4-cyanophenol t-1,2bis (4pyridyl)bipyethylene polymorphs: bulk sample (green) and simulated patterns of form I (black) and form II (red) 63 Figure 3.5. Concomitant polymor phs of (4-cyanopyridine)2 4,4-biphenol cocrystal: a) form I irregular hexa gons, b) form II parallelepiped plates 64 Figure 3.6. ORTEP plot of supramolecu lar adducts in (4-cyanopyridine)2 4,4biphenol drawn at 50% probability level for non-hydrogen atoms: (a) form I, (b) form II 66 Figure 3.7. Crystal packi ng in (4-cyanopyridine)2 4,4-biphenol form II, 11b 66 Figure 3.8. Histogram representing the tors ion angle distribution in 22 crystal structures of 4,4-biphenols 68 Figure 3.9. a) Distribution of torsion angles in the 623 crystal structures that contain biphenyl moieties; b) Distri bution of torsion angle deviations within pairs of conformers pr esent in 67 crystal structures 68 Figure 3.10. Polymorphic conversions between form I and form II of the (4cyanopyridine)2 4,4-biphenol co-crystal 69 Figure 3.11. PXRD patterns of (4-cyanopyridine)2 4,4-biphenol co-crystals: (a) simulated form I and (b) experimental form I obtained via slow evaporation from acetone; (c) simulated form II and (d) experimental form II obtained via acetone-drop grinding 70

PAGE 15

xiiFigure 3.12. Crystal structures of the pol ymorphic forms of quinhydrone: a) form and b) form 72 Figure 3.13. Crystal structures of the polymorphic forms of caffeine glutaric acid co-crystal: a) form I and b) form II 73 Figure 4.1. Examples of supramolecu lar homosynthons: a carboxylic acid homosynthon IV and a 2-aminopyridine homosynthon V 81 Figure 4.2. Histograms of contacts for cr ystal structures containing both 2aminopyridine and carboxylic acid mo ieties: a) N(py)O contacts in supramolecular heterosynthon VI or VII, b) N(am)O contacts in supramolecular heterosynthon VI or VII, c) N(am)N(py) contacts in supramolecular homosynthon V 85 Figure 4.3. Scatter plot of carbon-oxygen bond lengths in: a) neutral carboxylic acids, b) carboxylate anions 86 Figure 4.4. Histograms that pres ent distribution of the C NC angle in a) neutral 2-aminopyridines, and b) protonated 2-aminopyridines 87 Figure 4.5. Supramolecular interactions in 2-aminopyridinium 4-aminobenzoate, 14 89 Figure 4.6. Crystal structure of 2-aminopyridinium 4-aminobenzoate, 14 1D hydrogen bonded chains are intercon nected via a benzoate amine N HOinteractions to form 2D corrugated supramolecular sheet 89 Figure 4.7. The angle between core plan es parallel to the interactions OANA CANAOA and OBNBCBNBOB 90

PAGE 16

xiiiFigure 4.8. Hydrogen bonding interactions in 2-aminopyridinium isophthalate, 15 91 Figure 4.9. Supramolecular interactions in bis (2-aminopyridinium) terephthalate, 16 92 Figure 4.10. Crystal structure of bis (2-aminopyridinium) terephthalate, 16 92 Figure 4.11. Supramolecular adducts in 2amino-5-methylpyridinium benzoate, 17, are sustained via charge-assisted heterosynthons VII that form 1D chains 94 Figure 4.12. Crystal structure of 2-am ino-5-methylpyridinium benzoate, 17 viewed along the c axis (down the intercalating 1D chains which are colored green, red, blue and gray) 94 Figure 4.13. Hydrogen bonding in bis (2-amino-5-methylpyridinium) 5tertbutylisophthalate, 18 96 Figure 4.14. Crystal structure of bis(2-amino-5methyl pyridinium) terephthalate, 19 1D hydrogen bonded chains ar e cross-linked via NHOinteraction to form 2D sheets 97 Figure 4.15. Crystal packing of bis (2-amino-5-methylpyridinium) 2,6napthalenedicarboxylate, 20 98 Figure 4.16. Crystal structures of a) 2-amino-5-methylpyridinium adipate adipic acid, 21 100 Figure 4.17. Crystal structure of 2-aminopyrid inium adipate adipic acid dihydrate 100

PAGE 17

xivFigure 4.18. Crystal structure of b is(2-amino-5-picolinium) 2,5thiophenedicarboxylate 2,5-th iophenedicarboxylic acid, 22 : a) the supramolecular adduct reveals that the expected supramolecular heterosynthon VII is absent; b) a vi ew of the chain that is generated by three 1-point supramolecular heterosynthons 102 Figure 5.1. Co-crystal of barbital and N,N bis (4-bromophenyl)melamine, JICTUK10, sustained by 3-point recognition supramolecular heterosynthon 111 Figure 5.2. Amide supramolecular homosynt hon present the crystal structure of pure CBZ 112 Figure 5.3. Carboxylic acidamide supram olecular heterosynt hon present in cocrystals of a) CBZ aspirin and b) piracetam gentisic acid 113 Figure 5.4. Carboxylic acidpyridine supr amolecular heterosynthon present in (ibuprofen)2 4,4-bipyridine co-crystal 113 Figure 5.5. Charge-assisted supramolecu lar heterosynthons are present in Prozac succinic acid co-crystal 114 Figure 5.6. (a) JATMEW, 3-[2-( N,N -dimethylhydrazino)-4thiazolylmethylthio]N2-sulfamoylpropionamidine maleic acid. Structural parameters suggest formation of a salt, (b) SAGQEW a propionic acid solvate of mebendazole 115 Figure 5.7. Representation of the crystal pack ing of bicalutamide in a) form I and b) form II 118

PAGE 18

xvFigure 5.8. 2:2 supramolecular adducts formed between bicalutamide and 4,4bipyridyl in 23 120 Figure 5.9. Histogram repr esenting the NHNarom contact distribution in the crystal structures containing both NH and Narom moieties 121 Figure 5.10. Crystal stru cture of bicalutamide 4,4bipyridyl, 23 121 Figure 5.11. 2:2 supramolecular adducts fo rmed between bicalutamide and t-1,2bis (4-pyridyl)ethylene in 24 122 Figure 5.12. Crystal structure of bicalutamide t-1,2bis (4-pyridyl)ethylene, 24 123 Figure 5.13. 5-member supramolecular aggr egate formed between bicalutamide, t-1,2bis (4-pyridyl)ethylen e and acetone in 25 124 Figure 5.14. Crystal structure of (bicalutamide)2 t-1,2bis (4pyridyl)ethylene (acetone)2, 25 124 Figure 5.15. Representation of the crystal packing of bicalutamide in a) two triclinic forms and b) monoclinic form 126 Figure 5.16. Methanol solvate of indomethacin, BANMUZ 127 Figure 5.17. Supramolecular heterosynthon VII exhibited in 2-aminopyridinium indomethacin, 26 128 Figure 5.18. Crystal structure of trimethopr im benzoate is sustained by the 2aminopyridine-carboxylate supa rmolecular heterosynthon VII 129

PAGE 19

xvi List of Schemes Scheme 2.1. Co-crystal formers used in the investigation of the hierarchy of O HNarom and OHN C supramolecular heterosynthons 26 Scheme 2.2. Supramolecular synthons that can form when OH, N and CN are present in the same structure 27 Scheme 4.1. Molecular structures of components present in complexes 14-22 82 Scheme 4.2. Supramolecular heterosynt hons that can be formed between carboxylic acids and 2-aminopyridin es: 2-aminopyridine-carboxylic acid supramolecular heterosynt hon VI and 2-aminopyridiniumcarboxylate supramolecula r heterosynthon VII 83 Scheme 5.1. Molecular structure of bicalutamide 117 Scheme 5.2. Molecular structure of indomethacin 126

PAGE 20

xvii Crystal Engineering of Organic Compounds Including Pharmaceuticals Joanna Bis ABSTRACT Neutral or charge-assisted hydrogen bonds occurring between organic molecules represent strong and directional forces that mediate the molecular self-assembly into well defined supramolecular architectures. A proper understanding of hydrogen bonding interactions, their types, geometries, and occurrence in supramolecular motifs, is a prerequisite to crystal engineer ing, i.e. to the rational design of functional solid materials. Multiple-component organic crystals represent ideal systems to study the intermolecular interactions between the constitu ent molecules that can be pre-selected for their hydrogen bonding sites and geometrical cap abilities. In particul ar, the systematic structural analysis of supramolecular system s that are comprised of simple molecules facilitates the development of strategies for the rational design of new multiplecomponent compounds involving more complex com ponents such as drug molecules. The work presented herein shows a comb ination of systematic database and experimental studies in the context of re liability and hierarch y of several hydrogen bonded supramolecular synthons that exist in a series of model co-crystals and organic salts. The acquired paradigms are ultimately utilized in crystal engineering of

PAGE 21

xviii pharmaceuticals. In addition, the viability of a mechanichemical approach toward supramolecular synthesis in the context of its efficacy and the effect on polymorphism in multiple-component compounds is also addressed.

PAGE 22

1 Chapter 1 Introduction 1.1. Supramolecular Chemistry Beyond molecular chemistry based on the covalent bond there lies the field of supramolecular chemistry whose goal it is to gain cont rol over the intermolecular bond Jean-Marie Lehn 1.1.1. Fundamentals Supramolecular chemistry ,1-3 known also as chemistry beyond the molecule ,4 is based on the underlying phenomena of mutual affinity and selective recognition of molecules interacting via a variety of noncovalent forces to form well organized assemblies. The origin of supramolecular ch emistry can perhaps be traced back to the 19th century, when the concepts of lock-and-key ,5 Corpora non agunt nisi fixata (agents cannot act unle ss they are bound),6 and bermolecle (supermolecule)7 emerged. Much of this discipline has been delin eated by 1960s and 1970s research involving host-guest systems8 for selective binding of small alkali metal cations by macrocyclic receptors, which encompassed mostly crown ethers9 and cryptands.10 The continued fascination of molecular recognition phenomena illustrated by Nature (the self-assembly of DNA, antigen-antibody recogni tion, protein folding, etc.), in spired chemists to further explore supramolecular systems in the context of weaker intermolecular interactions such as hydrogen bonds and stacking.11-17 In particular, the mani pulation of intermolecular

PAGE 23

2 interactions, that leads to spontaneous, but controllable self-assembly of complementary moieties, has become one of the major interests in supramolecular design. The growing interest in the novel supramolecular approach can be considered as the manifestation of a conceptual change toward chemical synthesis.18 The emphasis on interaction, rather than reaction between molecules,19 has opened a great opportunity to generate a novel class of s upramolecular structures, or supermolecules ,20 with a wide range of complexity. Although supramolecular s ynthesis has not yet reached the level of sophistication represen ted by advanced organic synthe ses (e.g. those of vitamin B12 21 and taxol22), the influx of progress in this field has indicated its potential to generate recognition-directed assemblies in simple pro cedures and without the need of making or breaking covalent bonds. 1.1.2. Supramolecular Interactions and the Role of Hydrogen Bonds Generally, chemical bonds are considered to fall into two categories: short-range and long-range.23 While the short-range forces (e.g. covalent) are responsible for the formation of molecular systems, the long -range interactions, including dipole-dipole interactions, stacking, and hydrogen bonds, are those that contribute to the association of molecules into supramolecular structures with defined stoichiometries. Thus, a thorough knowledge of the non-covale nt bonds, their various types and their relative strengths is of crucial importance in the context of cont rolling supramolecular assemblies. A general comparison of selected chemical bond types is presented in Table 1.1.

PAGE 24

3Table 1.1. Comparison of selected ch emical bond types and their features Chemical Bonds Bond Energies (kJ/mol) Building blocks Products Features Covalent 200 400 Atoms Molecules H > T S MW: 1 1000 Da Solvent effect: Secondary Hydrogen Bond Dipole-Dipole stacking Van der Waals 4 120 5 50 < 50 < 5 Molecules Supermolecules H T S MW: 1 100 kDa Solvent effect: Primary I believe that as the methods of stru ctural chemistry are further applied to physiological problems it will be f ound that the significance of the hydrogen bond for physiology is greater than that of an y other single stru ctural feature. Linus Pauling In particular, hydrogen bonds24-26 are very important in the context of molecular recognition and, as anticipated at early stages by L. Pauling,27 they are responsible for numerous phenomena occurring in biological systems. Due to their strength and directionality, the role of hydrogen bonds in the formation of supramolecular assemblies has been studied with respect to molecu lar association in both solution and solid state.14,25,28-39 Specifically, the pre-determined fo rmation of supramolecular species mediated via hydrogen bonds in the solid stat e, has become a foundation for generating novel materials with well defined stru ctures and useful properties. 1.2. Crystal Engineering Crystal engineering is the understanding of intermolecular in teractions in the context of crystal packing a nd in the utilization of such understanding in the design of new solids with desired physical and chemical properties Gautam Desiraju

PAGE 25

4 1.2.1. Fundamentals In an ultimate case of molecular recogniti on, molecules associate into a perfectly organized, single chemical entity held by periodic, three dimensional arrays of noncovalent forces, thereby generating a crystal As the formation of crystalline architectures involves a series of complex molecular recogni tion events that occu r at a high level of precision, crystals have b een described as the supermolecules par excellence .40 The idea of applying the principles delineated by supramolecular chemistry to solid state for the rational design of novel crys talline materials, has led to the development of the new field of crystal engineering.41-43 The term crystal engineering was originally introduced in 1955 by Pepinsky,44 who demonstrated that crystalliz ation of organic ions with metalcontaining complexes results in structures with controllable cell dimensions and symmetries. Subsequently, from an important work of Schmidt45 related to the solid-state photodimerization, it became clear that a crys tal can be thought as a self-assembly resulting from a series of molecular rec ognition events and that the physicochemical properties of a crystal depend upon the intern al arrangement of the molecules in the crystal lattice. Now, as illustrated by Desirajus definition,41 crystal engineering has become synonymous with a broader discipline of making crystals by design 42 toward the utilization of thei r specific properties. In particul ar, the area of organic compounds has witnessed a remarkable expansion in re sult of crystal engineered materials for specific applications, e.g. : non-linear optics (NLO),46 porous materials,47 photographic materials,48 and pharmaceuticals.49

PAGE 26

5 Although the field of crystal e ngineering is indirectly re lated to the prediction of the ultimate crystal structure of the designe d compound, it should be noted that there are fundamental differences betw een the two research areas.43 Crystal structure prediction (CSP) requires specification of molecular geom etry and orientation, unit cell dimensions and the space group. In contrast, crystal engi neering is much less restrictive in the perspective that it involves the design of crystals with well defined non-covalent connectivities and networks based upon preselected molecular components that possess specific moieties. One of the continuing scandals in the physical sciences is that it remains in general impossible to predict th e structure of even the simplest crystalline solids from a knowledge of their chemical composition John Maddox A comprehensive prediction of the lowe st energy crystal structure through computational methods50 still remains elusive,51-58 although the field is advancing rapidly.59-61 As highlighted by Maddox,62 the prediction of an unknown crystal structure in a complete fashion remains a formidable challenge, and supramolecular synthesis of crystals has alternatively been directed to ward empirical approaches, based upon analysis of crystal packing modes present in selected sets of existing crystal structures. With the aid of extended databases, the broader antici pation of structural patterns resulting from molecular recognition has become an inherent practice in crystal engineering. In this respect, the suitability of hydrogen bonds to generate predetermined motif s has already been mentioned. In particular, a solid knowledge of the hydrogen bonding capabilities and geometrical complementarities exhibited by specific moieties as well as a rational

PAGE 27

6 selection of molecular building blocks can afford novel compounds with pre-defined composition and supramolecular architectures. 1.2.2. Hydrogen Bonded Supramolecular Synthons Supramolecular synthons are structural units within supermolecules which can be formed and/or assembled by known or c onceivable synthetic op erations involving intermolecular interactions Gautam Desiraju Given that the ultimate organizatio n of molecules in a crystal, or crystal packing results de facto from a series of molecular recogn ition phenomena, crystal engineering has naturally emerged around the idea of estab lishing, and later utiliz ing, intermolecular connectivities that are strong and directiona l enough to govern crystal packing with a reasonable degree of predictability. In this respect and based upon the analogy to covalent synthesis,63 the term supramolecular synthon64 has been introduced. Supramolecular synthons, also called motifs65 or patterns,66 can be regarded as regions within a crystal structure where the recognition between the constituent functional groups occurs, and this retrosynthetic approach67 helps to simplify the intrinsically difficult task of analyzing supramolecular architectures in the solid state.51,54 Considering that neutral and chargeassisted68-74 hydrogen bonds have been recognized as the most important non-covalent interactions in solid state supramolecular chemistry,25,39 it is not surprising that the utilization of hydrogen bonded supramolecular s ynthons in crystal engineering of organic solids has become ubiquitous.41,75-79 Supramolecular synthons can be separated into two distinct categories: supramolecular homosynthons,80 that result from the interaction

PAGE 28

7 between alike, self-complemen tary functional groups and supramolecular heterosynthons ,80 composed of different but comple mentary functional groups. Examples of supramolecular homosynthons include the carboxylic acid81,82 and amide dimers,83 whereas supramolecular heterosynth ons include carboxylic acidamide,84-87 hydroxylpyridine,88-90 and carboxylic acidpyridine.80,91-95 Examples of a supramolecular homosynthon and heteros ynthon is presented in Figure 1.1. Figure 1.1. Examples of supramolecular synthons: carboxylic acid homosynthon (left) and carboxylic acidpyridin e heterosynthon (right) Supramolecular homosynthons tend to exist in structures of single-component compounds, although their existence has also been observed in several crystals comprised by, for example, two different carboxylic acids.96-98 On the other hand, if multiple functional groups are present, it is more likely that they would engage in supramolecular heterosynthons. Furthermore, if a prefer ential supramolecular heterosynthon can be formed between functional groups that ar e located on different molecules, a multicomponent compound would be generated. In th is respect, a better understanding of the interplay between supramolecu lar synthons facilitated by th e interpretation of existing crystal structures would help in the design of new multiple-component crystals. Specifically, the general trends obs erved in a series of relevant crystal structures, in terms of the prevalence of specific supramolecu lar synthons over others, would provide a

PAGE 29

8 valuable insight for crystal engineering strategies toward the generation of new multiplecomponent materials comprised of polyfunctional (therefore more complex) molecules. 1.2.3. The Cambridge Structural Database studies of individual struct ures are of limited value: if an unexpected structural feature is observed, it may not be statistica lly significant and may well be ascribed to experimental errors or pack ing effects. () Thus, the systematic analysis of large numbers of related structures is a powerful research techniqu e, capable of yielding results that could not be obtained by any other method. Frank Allen, et al The identification and analysis of speci fic supramolecular synthons existing in a large set of correlated crystal structures in order to assess their general organizational tendencies is now a prerequisi te for crystal engineering. In this regard, investigating hundreds of thousands of crystal structures repor ted to date is current ly facilitated by the Cambridge Structural Database, CSD.99 As a collection of X-ray and neutron diffraction data for over 350,000 organic and organomet allic compounds (ConQuest V1.7, August 2005), the CSD provides a great amount of stat istically valuable information regarding the molecular and supramolecular structure of these compounds.100-103 In the context of the later, the quantitative (frequency of occurrence) and qualit ative (geometrical attributes: distances, angles, etc.) analyses of intermolecular interactions allow for a comprehensive evaluation of the robustness of supramolecular synthons. In particular, earlier reports from Etter66,75,76 and Desiraju64,104 have concentrated on the investigation of hydrogen bonding patterns with the aid of th e CSD and they have contributed to the proliferation of hydrogen bonded supramolecula r synthons as design tools in crystal

PAGE 30

9 engineering. 1.3. Hydrogen Bonded Organic Co-crystals There is still plenty of room at the bottom Richard Feyman The physicochemical properties of a crystall ine material are inherently dependant upon the chemical nature of its constituents and the crystal packing. The properties of a compound can therefore be changed if the inte rnal arrangement of molecules is altered.105 An alternative to influence the crystal packing of a compound can be based on the manipulation of the non-covalent forces that hold the cons tituents together, by introduction of another, rationally preselected for its hydrogen bonding sites, component. In effect, a multi-component compound, or a co-crystal, results. The meaning of the term co-crystal is cu rrently a subject of debate.106,107 A broad definition of a cocrystal given by Dunitz: a crystal containing two or more components together107 would include molecular adducts, salts, solvat es/hydrates, inclusion compounds, etc. In a more specific perspective taken by others49,108 a co-crystal is perceived as a multiple component crystal formed between compounds that are solid under ambient conditions: at least one component is molecular and fo rms a supramolecular synthon with the remaining components That all components of a co-crystal ( co-crystal formers ) are solids under ambient conditions has important imp lications with respect to the stability of a co-crystal and its susceptibility for preparation in the solid-state. In light of the above description, co-c rystals have been encountered in the

PAGE 31

10 literature for a long time under vari ous terms, e.g. molecular compounds,109 organic molecular compounds,110 addition compounds,111 molecular complexes,112 solid-state complexes,113 or heteromolecular crystals.114 The first co-crystal s appeared during the 19th century. A prototypal example of co-cryst allization is perhaps the synthesis of pbenzoquinone and hydroquinone (quinhydrone), reported in 1844 by Whler, which was then followed by studies of ha logen derivatives of quinhydrone.109,115 Structural information about quinhydrone (Figure 1.2), however, was not available until the 1960s.116,117 Inspired by the elucidation of DNA structure 118,119 through X-ray analysis, numerous nucleobase complexes were reported in the 1950s and 60s.120-123 The molecular recognition between me thyl derivatives of adenine and thymine is presented in Figure 1.3.122 Figure 1.2. Crystal structure of the triclinic form of quinhydrone Figure 1.3. The Hoogsteen base-pairing in the structure of 9-methyladenine 1methylthymine co-crystal

PAGE 32

11 0 50 100 150 200 250 300 350 4001978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004Year N Figure 1.4. Distribution of hydrogen bonded co -crystal structures archived in the CSD between 1978-2004 Although long known, co-crystals are not as widely studied as single-component crystals or solvates. There are ca. 1, 487 hydrogen bonded molecular co-crystals which constitute only ca. 0.42 % of all structures archived in the CSD, as compared to 35,882 hydrates (ca. 10%). However, based upon the in creasing number of relevant literature, it is clear that the interest in co-crystals is growing, Figure 1.4. The salient feature of cocrystals is that they can be designed from first principles. An a ppropriate knowledge of the supramolecular chemistry of the functiona l groups present in a given molecule can facilitate selection of an appropriate co-c rystal fromer which will form supramolecular heterosynthon(s) with the target molecule. In summary, co-crystals constitute a particularly attractive class of compounds that can be studied toward fundament al aspects such as: modification of physichochemical properties, understanding non-c ovalent interactions, viability toward green chemistry preparation, polymorphism, etc.

PAGE 33

12 1.3.1. Co-Crystals in the Context of Invest igation of Supramolecular Heterosynthons The challenge of qualitative classifi cation of hydrogen bonded motifs was addressed by Etter et al. based upon a graph-set system.75 In this approach, hydrogen bonded patterns can be described as chains (C), dimers (D), rings (R), or intramolecular hydrogen bonds (S). Each specific descriptor is then followed by the number of proton acceptors (superscript), number of proton donor s (subscripts) and the number of atoms involved in a particular motif. For instance, R2 2(8) notation is used to describe an eightmembered ring with two hydrogen bond acceptors and two hydrogen bond donors, and can be exemplified by carboxylic acid and amide homosynthons, or carboxylic acidamide and carboxylic acid pyridine heterosynthons. Gr aph-sets are useful to evaluate the frequency of a given hydr ogen bonding pattern, however, they do not provide information related to the types of proton donors an d proton acceptors engaged in the pattern. Therefore the frequency of s upramolecular heterosynthons composed of specific hydrogen bond donors/acceptors could not be addressed via the graph-set systematization. As already mentioned, a thorough u nderstanding of supramolecular heterosynthons is a prerequisite for developi ng crystal engineering of co-crystals. On the other hand, co-crystals represent ideal system s for systematic studies of non-covalent heterointeractions, as the majority of co -crystals is sustained by supramolecular heterosynthons rather than supramolecular homosynthons.124-127 Based on earlier studies of Robertson and Donohue,128 and from the work in the context of utilizing co-crystals to delineate hydrogen bonding preferences, Ette r proposed several hydrogen bonding rules,

PAGE 34

13 one of which states: the best hydrogen-bond donor and th e best hydrogen-bond acceptor will preferentially form hydrogen bonds to one another .66,76 For instance, the studies based upon co-crystallizations of 2-aminopyr imidine (hydrogen bond acceptor) with two carboxylic acids of different strengths (hydr ogen bond donors) revealed selective binding of the 2-aminopyrimidines to the stronge r acid. These and related results were rationalized based upon the differences in pKa values of the interacting molecules.129-131 It should be noted, however, that there have been inconsistencies with the above rule, as exemplified by co-crystals, in which carboxylic acid moieties interact with the weaker, rather than the stronger sites of a basic co-crystal former.132 Such observations suggest that the differences in pKa may not reliably predict th e interactions between the components in co-crystals. Considering that the pKa is a property of solution that is not defined in crystals, such rela tionships may not be simply tr ansferred to the solid state.39 In the context of systematization of different hydrogen bond ed synthons, Aakery et al presented systematic stud ies of the competition betw een three distinct hydrogen bonding moieties: primary amide, pyridine, and carboxylic acid. Th e study involved cocrystals of iso -nicotinamide (4-pyridinecarboxamide ) and a range of aromatic and aliphatic acids.91,133 The generated co-crystals re vealed consistent hydrogen bonding patterns comprised of two robust supramol ecular synthons: acid pyridine heterosynthon and self-complementary amide homosynthon. The reproducibility of the hydrogen bonded motifs suggests a dominant tendency of the acidpyridine heterosynthon over the acidamide heterosynthon, that is formed in acid/amide-containing compounds in the absence of pyridines.81,84,134-137 Further examples based on these results include rational

PAGE 35

14 design of ternary co-crystals of iso -nicotinamide with two different carboxylic acids.138 Notwithstanding the valuable contribution of the presented work, delineation of the hierarchies of supramolecular heterosynt hons that can occur within a variety of functional groups in a competitive environm ent represents a crystal engineering challenge. Furthermore, the utilization of the CSD to a ssess the prevalence of one supramolecular heterosynthon over another can be addressed only in a few instances, i.e. the most ubiquitous functional groups.139 Thus, the relative ranking involving numerous supramolecular hetrosynthons such as hydroxylpyridine vs. hydroxylamine, hydroxylpyridine vs. hydroxylcyano, hydroxylamine vs. hydroxyl-cyano, etc. still remains an issue that needs to be addressed experimentally.140 1.3.2. Co-Crystals in the Context of Green Chemistry The advantage of obtaining well defined single crystals of a synthesized product is inherently related to the po ssibility of direct determina tion of its crystal structure. Additional benefits linked to the speed and accuracy of single crystal X-ray diffraction instrumentation have directed supramolecular synthesis of crystall ine materials toward solution-based methods, e.g. slow evaporati on, heating/cooling, addition of anti-solvent (solvent in which the components are not or sparingly soluble), etc. However, other aspects associated with solution co-crystalli zation, namely, mismatched solubility of the reactants, the possibility of unexpected solvate formation, or uncontrollable effects of solvents on polymorphic behaviors, could re present significant challenges for a crystal engineer. With this viewpoi nt, using less conventional me thods of co-crystallization141

PAGE 36

15 such as growth from the melt or grinding the co-crystal formers not only overcomes aforementioned problems, but also can be of interest from the perspectives of green chemistry,142 which seeks to reduce and preven t pollution via implementation of environment-friendly chemical processe s. In particular, the mechanochemical143 approach to supramolecular reactions, repres ented by grinding of two or more solids entirely eliminates some of the aspects accompanied with solution crystallizations, e.g. the recovery, storage and dis posal of organic solvents. Whereas grinding has been a commonplace in the context of classic covalent synthesis,144,145 its utilization with respect to co-c rystallization has not been as common. Even though grinding by mortar and pestle to induce co-crystal formation has been known since late 19th century,109 it was implemented nearly on e century later by Etter, in the research originally di rected toward the understandi ng of preference of hydrogen bonds, and their role in the structures of a va riety of organic co-cry stals that involved nucleobases and acentric organic molecules.95,146-150 Recent advances in the area have introduced a novel grinding technique, termed solvent-drop grinding .151 The addition of small amounts of solvent to a grinding proc edure in order to accel erate the reaction has been found to be a method especially effectiv e for the preparation of multiple-component compounds, such us inclusion compounds, salts and co-crystals.152-154 Although, the recent reports illustrate that the grinding approach is a viable means of preparing hydrogen bonded co-crystals, me hcanochemical techniques have not been routinely used on the academic level of res earch, and their further exploration in the context of co-crystal reproducibility, stab ility, and polymorphism still remains to be

PAGE 37

16 addressed. 1.3.3. Co-Crystals in the Context of Polymorphism A polymorph is a solid crystalline phas e of a given compound resulting from the possibility of at least two different arrange ments of the molecules of that compound in the solid state Walter C. McCrone Polymorphism,155-158 a phenomenon recognized in 1822,159 can be described as the existence of a substance in more than one crystalline form. The inability to reliably predict the existence of polymor phs or, for that matter, crysta l structures in general, has important intellectual property and scientific implications.156,160 For example, the appearance of an undesired polymorph can invoke problems during the formulation process of a commercial compound a nd lead to patent litigations.156 On the other hand, a novel polymorph can offer an opportunity in terms of better physicochemical performance and new product development. Furthermore, since physicochemical properties of a compound can differ critically from one form to another, inducing and controlling a specific polymor ph is of utmost importance in the chemistry of pharmaceuticals,156,161 explosives,162,163 pigments,164,165 etc. Although the awareness among chemists of polymorphism increases, the frequency of occurrence of polymorphic compounds is not entirely obvious. The generality of McCrones statement that the number of forms known for a given compound is proportional to th e time and money spent in research on that compound remains unclear, despite the indications that the frequency of polymorphism represented

PAGE 38

17 by the CSD may be underestimated.56,166 As suggested by Desiraju,167 polymorphism may not be equally apparent in different categories of co mpounds, and it tends to be prominent in molecules that contain multiple hydrogen bonding moieties (thereby forming multiple supramolecular synthons), and/or possess conformational flexibility.18,168 Considering that these two featur es are inherently exhibited by drug molecules, polymorphism, as well as solvate formation (pseudopolymorphism),169,170 in pharmaceuticals is well documented.61,161,171-173 Specific examples of the existence of polymorphism in popular compounds include, for instance, ROY,174,175 aspirin,61 piracetam,176,177 and virazole.178 Conformational flexibility of molecule s can also lead to conformational isomorphism,156,168,179,180 where more than one molecular conformer exists in the same crystal structure. In addition, simultaneous crystallization of polymorphs, known as concomitant polymorphism, can occur under certain conditions.181 Despite the fact that most of the pol ymorphs have been observed in singlecomponent compounds, co-crystals also exhibit polymorphism.182,183 A CSD survey reveals ca. 94,900 single-component organic compounds, of which ca. 1,600 are polymorphic.184 On the other hand, there are onl y ca. 1,487 hydrogen bonded molecular co-crystals (comprised of components that ar e solid at room temperature), of which 21 are polymorphic; yet only 11 have 3D coordinates determined for two or more forms.185 The percentage occurrences of polymorphi sm in single-component compounds and cocrystals suggest that its extent is comparable (1.7 % vs. 1.5%).

PAGE 39

18Table 1.2. The occurrence of polymorphism in organic compounds. Data adapted from Zaworotko M. et al J. Pharm. Sci., 2006 Category Total Polymorphic compounds Single-component compounds 94,900 1600 (1.7 %) Co-crystals 1,487 21 (1.4 %) Interestingly, a comparative structural study of co-crystal polymorphs (albeit based upon limited data), revealed that the su pramolecular synthons exhibited in all cases are persistent and polymorphism is related to rather subtle conformational or crystal packing variations.182,183 In general, the existence of polymorphism is not yet entirely understood, however, there have been advances in controlling th is phenomenon. Specificall y, recent literature revealed that the utilization of solvent-drop grinding approach can be an efficient way to achieve selective transforma tions between specific polymorphs in both single-component compounds and co-crystals.186-188 1.3.4. Co-Crystals in the Cont ext of Pharmaceuticals The idea of utilizing co-crystals to modify physicochemical properties of compounds has attracted considerable atten tion with respect to pharmaceuticals. The majority of active pharmaceutical ingredients (A PIs) occur as solids and their crystalline forms are highly preferred over the amor phous forms, due to the physicochemical stability considerations.171 Although amorphous APIs often exhi bit enhanced solubility in aqueous systems, they are thermodynamically unsta ble and tend to revert into more stable crystalline products. The rejec tion of impurities and the ease of the isolation of single

PAGE 40

19 crystals are additional benefits associated with dealing with crystalline materials. The problems with the utilization of crystalline APIs are related to their poor solubility which in turn negatively affects the bioavailability of an API. Therefore the development of crystalline forms of API that exhibit optimized physicochemical performance is of utmost importance in the field of pharmaceuticals. Table 1.3. Occurrence of hydrogen bonding moieties in APIs Functional Group Top 100 Prescription Drugs % Alcohol 39 3 amine 37 Carbonyl 35 Ether 33 2 amine 31 Carboxylic acid 30 Ester 22 Aromatic N 12 2 amide 11 Sulfonamide 3 The inherent nature exhibited by APIs, na mely their hydrogen bonding sites amenable to engage in supramolecular heterosynthons, co nstitutes a particularly suitable system for crystal engineering studies. Indeed, the 100-top selling prescr iption drugs contain hydrogen bond donors and acceptors, e.g. hydrox yl (39%) and carboxylic acid (30%) moieties, Table 1.3.189 In this respect, the interest in developing new forms of APIs resulted in the emergence of a new class of APIs, pharmaceutical co-crystals A pharmaceutical co-crystal can described as a multiple component crystal in which at

PAGE 41

20 least one component is molecu lar and a solid at room temperature (the co-crystal former) and forms a supramolecular synt hon with a molecular or ionic API .49,190 Crystalline APIs have trad itionally been limited to so lvates/hydrates, polymorphs, and salts.171 Solvates/hydrates are usually discovered as a result of adventitious uptake of solvent/water upon crystallizati on and, like polymorphs, they ar e difficult to be rationally designed. Additional complications of the usag e of API solvates/hydrates may be related to the possibility of desolvation/dehydr ation, followed by formation of amorphous material, which may occur as a function of tim e and storage conditions In this respect, co-crystals are less likely to exhibit such behavior as their com ponents are solids. Salt forms of APIs are commonplace and their role in the optimization of API properties, including solubility, has been established.191-194 Nevertheless, it is im portant to note, that salt formation is targeted in the case of drugs possessing ionizable (basic or acidic) moieties. In contrast, co-cry stallization of APIs can be expanded over molecules that possess a broader range of hydrogen bonding mo ieties. Thus, API co-crystals exhibit advantages over API solvates/hydrates, polymor phs, and salts in the way that based upon rational design it is possible to generate di verse range of API forms with optimized physicochemical properties, e.g. solubility, th ermal stability, and hygr oscopicity, without the need of covalent modificati on and without altering their origin al biological activity. Co-crystals of several important APIs ha ve been reported in the scientific and patent literature, and they include: carbam azepine (CBZ), aspirin, profens, piracetam, caffeine, loracarbef, cephalexin, cefaclor, con azoles, topiramate, modafinil, phenytoin, olanzapine, nabumetone, fluoxetine, theophyllin e, sulfadimidine, trimethroprim, and

PAGE 42

21 paracetamol.61,80,86,132,154,186,195-208 Detailed investigations of physicochemical performance of some of the reported APIs showed that their co-crystallization is inherently related to the m odification of the physical prop erties, such us solubility, dissolution rate, thermal stability, etc. For instance, the dissolution studies in aqueous medium of co-crystals of itraconazole (Sporanox) and dicarboxylic acids, e.g. itraconazole succinic acid (Figure 1.5), indicate that the co-crystals achieve and sustain from 4 to 20-fold higher concentrations as compared to the highly insoluble pure itraconazole.132 Figure 1.5. Itraconazole succinic acid co-crystal Another example, carbamazepine (Tegretol ) co-crystal with saccharin, Figure 1.6, in addition to its enhanced solubility, was found to be resistant to undesired hydrate formation and has not exhibited pol ymorphism based upon 1200 high-throughput (HT)209,210 screening experiments.211

PAGE 43

22 Figure 1.6. Carbamazepine saccharin co-crystal In summary, the application of the co ncepts of supramolecular chemistry and crystal engineering to the de velopment of pharmaceuticals offers an opportunity towards the generation of novel API form s. Specifically, the alteration of the physical properties of the solid dosage form with simultaneous retention of the th erapeutic attributes of an API represents an attractive approach to ba lance the bioavailability, stability and other performance characteristics. 1.4. Summary The presented work will focus on applying the concepts of crys tal engineering to the design and generation of novel multi-com ponent compounds with pre-determined composition and intermolecular interactions. Pa rticular emphasis wi ll be placed upon cocrystals and their further explora tion toward the following aspects: Delineation of the reliability of hydrogen bonded supramolecular heterosynthons and their hierarchies in a competitive enviroment. Advances in

PAGE 44

23 understanding the mechanisms that govern molecular recognition in the crystalline state will lead to insights towa rd rational design of co-crystals of more complex molecules. In addition, the inves tigation of neutral and charge-assisted supramolecular heterosynthons (co-crysta l vs. salt) will be addressed. Viability of mechanochemical met hods toward preparation of multiplecomponent compounds (organic salts a nd co-crystals). Reduction or entire elimination of organic solvents from the experimental co-crystallization procedures can be advantageous from the perspectives of green chemistry principles, as well as from the viewpoi nt of reproducibility of the products obtained from solution crystallizations. Polymorphism in co-crystals. The su sceptibility of co-crystals toward polymorphism or solvate formation w ill be addressed based upon traditional solution techniques and innovative solv ent-drop grinding screening methods. Pharmaceutical co-crystals. Rational co -crystallization of API molecules, which represent more complex systems due to their multiple hydrogen bondong sites, with judiciously selected co-cry stals formers will be demonstrated as a result of the acquired knowledge from the preceding model compound investigations.

PAGE 45

24 Chapter 2 Recurrence of hydroxylaromatic nitrogen supramolecular heterosynthon in the presence of a competing cyano acceptor 2.1. Focus As already mentioned, the CSD contai ns enough information to evaluate competitiveness of some supramolecular homosynthons vs supramolecular heterosynthons.65,137,212 However, the prevalence of one supramolecular heterosynthon over another can be addressed only in a few in stances, i.e. the most ubiquitous functional groups.139,140 Therefore, assessing the hierarchies that exist within a given set of supramolecular heterosynthons still represen ts a challenge and complicates crystal engineering of compounds that possess multiple functionalities. In an effort to explore pharmaceutical co-crystals, studies on model co-crystals that contain hydroxyl (OH), aromatic nitrogen (Narom), and cyano (CN) moieties have been conducted. A combination of such functiona lities is present in a range of vitamins and APIs, e.g. vitamin B1, cimetidine, bicalutamide, etc. An analysis of the crystal structures of the model co-crystals is expect ed to facilitate the delineation of hierarchies between hydroxylaroma tic nitrogen (OHNarom) and hydroxylcyano (OHN C) supramolecular heterosynthons.

PAGE 46

25 2.2. Results and Discussion The co-crystallization experiments involved co-crystal formers that possess the OH, Narom, and/or CN moieties, and these functiona l groups: (a) are sterically accessible, (b) are not involved in intramolecular intera ctions, and c) are not accompanied by other competing hydrogen bond donors and acceptors. In an individual experiment two cocrystal formers were combined. Within a pair one co-crystal former possessed two of the three moieties (e.g. Narom/CN) and the second co-crystal former possessed the remaining moiety (e.g. OH). According to this strategy, the i ndividual pairs of co-crystal formers are combined as follows: Narom/CN with OH, OH/CN with Narom, and OH/Narom with CN. Such an approach to delineate the hierar chies of two supramolecular heterosynthons relies on the idea that a co-crystal can result only if the favored supramolecular heterosynthon is formed between the co-crystal formers. Conversely, a co-crystal is not expected to be formed if a dominant supram olecular heterosynthon al ready exists in one of the pure components. The formation of a co-crystal is determined by multiple techniques: melting point measurements DSC, IR spectroscopy, powder X-ray diffraction (PXRD), and single crystal X-ray diffraction. The crystal structures of all obtained co-crystals were analyzed in th e context of the existence of specific supramolecular heterosynthons. Small organic molecules that contain OH, Narom, and CN moieties used in this study are shown on Scheme 2.1. Co-crystalliza tions of these chemicals afforded the following co-crystals: 3-cyanophenol 4-phenylpyridine, 1 ; (3-cyanophenol)2 1,2bis (4pyridyl)ethane, 2 ; (3-cyanophenol)2 trans-1,2bis (4-pyridyl)ethylene, 3 ; 3-

PAGE 47

26 cyanophenol trans-1,2bis (4-pyridyl)ethylene, 4 ; 4-cyanophenol 4-phenylpyridine, 5 ; (4cyanophenol)2 4,4-bipyridine, 6 ; (4-cyanophenol)2 1,2bis (4-pyridyl)ethane, 7 ; (4cyanophenol)2 trans-1,2bis (4-pyridyl)ethylene, 8 ; (3-cyanopyridine)2 4,4-biphenol, 9 ; (4-cyanopyridine)2 resorcinol, 10 ; (4-cyanopyridine)2 4,4-biphenol, 11 ; and (4cyanopyridine)3 phloroglucinol, 12 O H OH OH CN O H CN N N N N N N N CN N NC N OH O H OH O H OH N OH OH N OH OH CN CN NC CN CN 4-Phenylpyridine 4,4'-Bipyridine 1,2-Bis(4-dipyridyl)ethane Trans-1,2-bis (4-dipyridyl)ethylene (phenpy) (bipy) (bipyeta) (bipyete) 1-Hexadecanol 1-Naphthol 4,4'-Biphenol 1,3-Dihydroxybenzene 1,3,5Trihydroxybenzene (hexdec) (naphth) (bphe) (res) (phlgl) 3-Cyanophenol 4-Cyanophenol 3-Cyanopyridine 4-Cyanopyridine 3-Hydroxypyridine 5-Hydroxyisoquinol ine (3cyph) (4cyph) (3cypy) (4cypy) (3hypy) (5hyquin) 1-Cyanonaphthalene (cynaphth) 1,3-Dicyanobenzene (o-cyben) 1,4-Dicyanobenzene (p-cyben) Scheme 2.1. Co-crystal formers used in th e investigation of the hierarchy of OHNarom and O HN C supramolecular heterosynthons 2.2.1. CSD Analysis There are three possible supramolecular sy nthons that can be formed when OH, Narom, and CN moieties are present in the same crystal structure: a hydroxylpyridine supramolecular heterosynthon I a hydroxylcyano supr amolecular heterosynthon II, and

PAGE 48

27 a hydroxyl supramolecular homosynthon III (Scheme 2.2). N R O H N R OH R O H R OH R I II III Scheme 2.2. Supramolecular synthons that can form when OH, N and CN are present in the same structure A CSD survey of compo unds that contain OH, Narom, and CN moieties was conducted to evaluate the frequency of occurrence of supramolecular synthons I II and III .213 In order to determine appropria te distance ranges, within which I II and III exist, distance distribution plots were generated, Figure 2.1.214 Based on visual inspection of the resulting histograms, the lower and higher cut o ffs for hydrogen bonds were determined. The histograms reveal th at the supramolecular heterosynthon I and II occur within the ranges of 2.50 3.00 a nd 2.70 3.20 respectively, whereas supramolecular homosynthon III exhibits range of 2.50 3.00 .215 Based upon these limits the number of entries that exhibit the targeted supramolecular synthons was determined, Table 2.1. It should be noted that the frequencies of occurrence of supramolecular synthons can be influenced by the presence of other hydrogen bond donors and acceptors, therefore the competing mo ieties, such us carboxylic acids, amines, amides, sulfonamides, carbonyls, water, chlori de and bromide ions, etc. were removed from the analyzed sets of structures.

PAGE 49

28 Table 2.1. CSD statistics related to supram olecular synthons that occur in structures containing only OH, Narom, and CN Moieties present in a structure No. of structures Supramolecular synthon Structures with synthon DA [] Mean ( ) [] OH and Narom 136 OHNarom OHO 135 ( 99% ) 37 ( 27% ) 2.50-3.00 2.50-3.00 2.77(8) 2.78(8) OH and CN 61 OHN C OHO 56 ( 92% ) 18 ( 29% ) 2.70-3.20 2.50-3.00 2.9(1) 2.78(8) OH, Narom and CN 3 OHNarom OHN C OHO 3 0 0 Supramolecular heterosynthon I is well established in crystal engineering.88,89,216 The CSD analysis reveals that in th e structures cont aining only OH and Narom I exists in 135 of 136 crystal structures. Therefore I is favored over III as its occurrence reaches ca. 99 % vs. 27% of the occurrence of III .217 On the other hand, of 61 crystal structures that contain only OH and CN, 56 (92%) entrie s exhibited supramolecular heterosynthon II whereas 18 (29%) structures exhibited III which indicates the dominance of heterosynthon II over homosynthon III as shown in Table 2.1. It should be noted, that 13 structures contained both II and III due to the presence of multiple OH moieties. However, the number of crystal structures ar chived in the CSD that possess all three OH and Narom and CN moieties (competing moieties are absent) resulted in only 3 entries, thereby precluding a meaningf ul statistical evaluation.218 Thus, the competition between I and II could only be assessed based upon experiment al results. Herein, a series of cocrystals that can help to eval uate the relative hierarchy of I and II is presented.

PAGE 50

29 0 10 20 30 40 50 60 70 80 902.5 2.56 2.62 2.68 2.74 2.8 2.86 2.92 2.98 3.04 3.1 3.16 3.22 3.28 3.34 3.4 3.46OHNarom distance []number of entrie s 0 5 10 15 20 252.5 2.56 2.62 2.68 2.74 2.8 2.86 2.92 2.98 3.04 3.1 3.16 3.22 3.28 3.34 3.4 3.46OHNC distance []number of entrie s a) b) 0 200 400 600 800 1000 12002.5 2.58 2.66 2.74 2.82 2.9 2.98 3.06 3.14 3.22 3.3 3.38 3.46OHO distance []number of entrie s c) Figure 2.1. Histograms of contacts for supramolecular synthons: a) OHNarom in I; b) O HN C in II; and c) OHO in III 2.2.2. Structural Features of Neutral and Ioni c HydroxylAromatic Nitrogen Interaction Phenols and pyridines can form neut ral co-crystals or organic salts.219 The neutral nature of heterosynthon I was confirmed by spectroscopy, proton location in the difference Fourier map, and structural parame ters of ancillary groups, namely the CNC angle in the pyridine moieties and C O bond lengths in the phenolic moieties.220,221 The CNC angle in pyridines is known to be se nsitive to protonation and its cationic form exhibits higher values (ca. 121 ) than that of the correspond ing neutral molecules (ca. 116 ).222,223 The histograms representing carbon -oxygen bond lengths distribution in neutral and ionic phenolic moieties were ge nerated using the CSD and are shown in Figure 2.2 (only good quality crys tal structures: ordered, erro r free, nonpolymeric with

PAGE 51

30 3D coordinates determined and R<5%, were ch osen for the analysis). The CSD analysis of 2588 crystal structures that contain neutral phenolic moieti es reveals that the average COH bond length is 1.36(2) There are 260 crystal structures that contain deprotonated phenolic moieties. The calculated average for ionic CObond length in such structures is 1.28(3) .224 0 200 400 600 800 1000 12001.2 1.22 1.24 1.26 1.28 1.3 1.32 1.34 1.36 1.38 1.4 1.42 1.44 1.46 1.48 1.5C-O distancenumber of entrie s a) 0 10 20 30 40 50 601.2 1.22 1.24 1.26 1.28 1.3 1.32 1.34 1.36 1.38 1.4 1.42 1.44 1.46 1.48 1.5C-O distance [A]number of entrie s b) Figure 2.2. Histograms representing the distri bution of carbon-oxygen bond lengths in a) neutral phenolic moieties and b) deprotonated phenolic moieties 2.2.3. Crystal Structure Descriptions The crystal structure of 3-cyanophenol 4-phenylpyridine 1 reveals discrete 1:1 supramolecular adducts sustained by OHNarom supramolecular heterosynthon I also coded as D graph-set.75 In addition to IR spectroscopic evidence, the neutral nature of I is supported by structural data: the CO distance is 1.363(6) and the CNC angle within the pyridine ring is 115.6(2) The OHNarom hydrogen bond distan ce (D: 2.708(3) ) is within the expected range for hydroxylpyridine intera ctions (Table 2.1). In this structure the phenpy is twisted, with a torsion angle of 24.6 between the aromatic rings. The dihedral angle formed by the 3cyphe and phenpy rings is 79.7 The glide related supramolecular adducts are stabilized via stacking occurring between the adjacent

PAGE 52

31 phenpy molecules with an interplanar separati on of ca. 3.79 Such organization of molecules results in a columnar arrangeme nt of the supramolecular adducts along the c axis. The adjacent columns are related by translation along the b axis and are connected via weak CHN C interactions, which results in 2D molecular sheets (Figure 2.3). Figure 2.3. Crystal structure of 3-cyanophenol 4-phenylpyridine, 1 The crystal structure of (3-cyanophenol)2 1,2bis -(4-pyridyl)ethane 2 reveals centrosymmetric 2:1 adducts sustaine d by two supramolecular heterosynthons I (D: 2.691(2) ). The CO distance is 1.352(2) an d the CNC angle within the pyridine ring is 116.6(2) indicating a neutral OHNarom hydrogen bond. The dihedral angle formed by 3cyphe and the bipyeta rings is 132.5 The adjacent adducts interact via weak CHN C forces forming 1D zigzag chains Translation related chains are connected via weak CHN C interaction along the b axis, thereby forming 2D

PAGE 53

32 molecular sheets (Figure 2.4). Figure 2.4. Crystal structure of (3-cyanophenol)2 1,2bis -(4-pyridyl)ethane, 2 The asymmetric unit of (3-cyanophenol)2 trans-1,2bis -(4-pyridyl)ethylene 3 consists of two 3cyphe molecules and one bipyete molecule. The components form noncentrosymmetric supramolecular adduct s sustained by two supramolecular heterosynthons I (D1: 2.727(2) D2: 2.730(2) ). The CO dist ances are 1.351(2) and 1.357(2) and the CNC angles within the two bipyete rings are 116.2(2) and 116.1(2) respectively. The dihedr al angles between the 3cyphe and the bipyete rings are 145.4 and 66.9 The adjacent supramolecular adduc ts are stabilized by face-to-face stacking occurring between th e aromatic moieties of 3cyphe and bipyete molecules. Such molecular assembly affords a column ar alignment of the adducts along the b axis. The molecular columns are further interconnected by weak CHN C interactions, thereby forming 2D sheets (Figure 2.5).

PAGE 54

33 Figure 2.5. Crystal structure of (3-cyanophenol)2 t-1,2bis -(4-pyridyl)ethylene, 3 3-cyanophenol trans-1,2bis -(4-pyridyl)ethylene co-crystal 4 consists of 1:1 discrete supramolecular adducts sustained by I (D: 2.663(3) ). The CO distance is 1.361(3) and the CNC angle of the hydrogen bonded bipyete pyridyl ring is 116.2(3) The dihedral angle between the 3cyphe and the bipyete ring is 98.8 The inversion related dimers are stabilized by stacking along b axis. The stacking occurs between parallel oriented bipyete molecules with the interplanar separation of ca. 3.44 The presence of the weak CHN C interaction between the adjacent 3cyphe molecules leads to the formation of centrosymmetric pa irs of the supramolecu lar adducts, which in turn directs the molecular assembly into 2D layers (Figure 2.6). It should be noted that the alignment of the two bipyete in 4 satisfies the topochemical principle for [2+2] photodimerization in the solid st ate, which states that the olefins should be parallel and separated by less than 4.2 .45 In this context, successful co valent synthesis of new cyclic molecules in high yields a nd in a solvent-free manner225 based upon bipyete and

PAGE 55

34 resorcinol co-crystals has been recently demonstrated.89,226 Furthermore, it should be pointed out that the 1:1 stoichiometry of the components in 4 is somewhat unexpected, consider ing that the ratio of hydrogen bond acceptor : donor (in this case Narom : OH) between the two components is 2:1. While the OH moieties of 3cyphe are utilized in a strong OHNarom hydrogen bond, only one of the two Narom sites of bipyete acts as an OHNarom acceptor. The remaining Narom moiety participates in a weak CHNarom interaction227-230 with the neighboring 3cyphe molecules. In effect, the adjacent bipyete molecules are aligned on top of each other and stabilized via continuous stacking. A literature search reveals that similar molecular arrangements are exhibited in several co-crystals of phenazines.231,232 For instance, the crystal structure of phenazine hydroquinone, (CSD refcode FOQHEY),231 Figure 2.7, is reminiscent of the crystal structure of 4 with respect to the unexpected 2: 1 stoichiometry of the components interacting via a similar set of intermolecula r interactions. These observations suggest a significant contribu tion of CHNarom and aromatic stacking to the overall crystal packing of the presented co-crystals. In particular, the existence of non-bonded Narom is not uncommon, as revealed by the CSD surve y. In the set of 135 crystal structures comprised only by OH and Narom moieties (Table 2.1) ther e are 19 entries (14%) that exhibit non-bonded weak CHNarom interactions, in addition to the primary supramolecular heterosynthon I

PAGE 56

35 Figure 2.6. Crystal structure of 3-cyanophenol t-1,2bis -(4-pyridyl)ethylene, 4 Figure 2.7. Crystal structure of phenazine hydroquinone, FOQHEY Similarly to 1 co-crystal of 4-cyanophenol 4-phenylpyridine 5 is comprised of 1:1 discrete supramolecula r entities sustained via I (D: 2.695(4) ). The CO distance is 1.345(3) and the CNC angle within the phenpy rings is 116.9(3) and the dihedral angle between the 4cyphe and the phenpy rings is 108.8 In this crystal structure the phenpy is twisted, and the torsion angle is 33.6 which is similar to phenpy in 1 (24.6 ). The glide related supramolecula r adducts are stabilized via stacking (ca. 3.74 )

PAGE 57

36 occurring between the phenpy molecules along the b axis. The adjacent columns of the 1:1 adducts are connected by centrosymmetric CHN C dimers thereby generating 2D sheets (Figure 2.8). Figure 2.8. Crystal structure of 4-cyanophenol 4-phenylpyridine, 5 The crystal structure of (4-cyanophenol)2 4,4-bipyridine 6 reveals 2:1 centrosymmetric supramolecular adduct sustained by I (D: 2.718(3) ). The CO distance is 1.346(3) and the CNC angle is 116.3(3) The dihedral angle between the 4cyphe and the pyridine ring is 30.1 The bipy molecules are flat and stack on top of each other along the a axis with an interp lanar separation of ca. 3.59 The packing of supramolecular adducts is further extended into 2D through centrosymmetric CHN C dimer formed between the neighboring 4cyphe molecules (Figure 2.9).

PAGE 58

37 Figure 2.9. Crystal structure of (4-cyanophenol)2 4,4-bipyridine, 6 In the crystal structure of (4-cyanophenol)2 1,2bis -(4-pyridyl)ethane 7 the 2:1 supramolecular aggregates are sustai ned by supramolecular heterosynthons I (D: 2.698(4) ), Figure 2.10. The CO distance is 1.359( 4) and the CNC angle is 117.2(3) The dihedral angle between the 4cyphe and the bipyeta ring is ca. 66.8 The adducts interact via CHO dimer along the c axis, thereby forming 1D ch ains. These chains are then interconnected via weak CHN C thus generating 2D molecular sheets. Figure 2.10. Crystal struct ures of (4-cyanophenol)2 1,2bis -(4-pyridy l)ethane, 7 The crystal structure of (4-cyanophenol)2 trans-1,2bis -(4-pyridyl)ethylene 8 is reminiscent of that in co-crystal 6 It is comprised by 2:1 centrosymmetric supramolecular adducts sustained by I (D: 2.714(4) vs 2.718(3) in 6 ). The CO distance is 1.349(3) and the CNC angle of the bipyete ring is 114.8(3) The

PAGE 59

38 dihedral angle between the 4cyphe and the bipyete rings is 142.6 ( vs 30.1 in 6 ). The supramolecular adducts are stabilized by interactions occurring between the bipyete molecules (ca. 3.66 ) along b axis and they further extended into 2D through a centrosymmetric CHN C dimer formed between the neighboring 4cyphe molecules (Figure 2.11). Although the prim ary (stacked 2:1 adducts) and secondary (2D network resulting from the presence of the CHN C dimers) architectures of 6 and 8 are similar, the two crystal structures differ in the rela tive orientations of the 2:1 adducts down the a and b axes for 6 and 8 respectively. In 6 the 2:1 adducts are pa cked side by side, and generate planar layers, Figure 2.12a. In 8 the adjacent adducts, ra ther than adopting coplanar orientation, are aligned at ca. 40 with respect to each other, Figure 2.12b. Figure 2.11. Crystal stru cture of (4-cyanophenol)2 t-1,2bis -(4-pyridyl)ethylene, 8 a) b) Figure 2.12. Representation of the relative orientation of the adjacent 2:1 adducts: a) down the a axis in 6 and b) down the b axis in 8

PAGE 60

39 In the crystal structure of (3-cyanopyridine)2 4,4-biphenol 9 a 2:1 centrosymmetric supramolecular aggregate exis ts that is sustained by two supramolecular heterosynthons I (D: 2.765(3) ). The CO distance is 1.381(3) and the CNC angle of the hydrogen bonded 3cypy ring is 117.79(2) suggesting the neutral nature of the O HNarom hydrogen bond. The dihedral angle between the flat bphe and the 3cypy rings is 54.1 The adducts assemble further through CHN C dimer into 2D molecular sheets (Figure 2.13). Figure 2.13. Crystal struct ure of (3-cyanopyridine)2 4,4-biphenol, 9 The components of (4-cyanopyridine)2 resorcinol co-crystal, 10 assemble to form 2:1 discrete adducts sustained by I (D1: 2.856(2) D2: 2.763(2) ). The CO distances in a res molecule are 1.361(2) and 1.360(2) and the CNC angles of the hydrogen bonded 4cypy rings are 117.1(1) and 117.2(1) respectively. The dihedral angle between the res and the 4cypy rings are 68.6 and 75.1 The hydroxyl moieties of res molecules adopt a convergent orie ntation and interact with two 4cypy molecules (ca.

PAGE 61

40 3.88 ) which are stacked in face-to-face manner. These discrete 2:1 entities are extended into 1D chain via weak CHO233,234 and CHN C forces along c axis. The translation related 1D chains are then packed side by side through weak CHN C interactions to form 2D layers (Figure 2.14). Figure 2.14. Crystal struct ure of (4-cyanopyridine)2 resorcinol, 10 The asymmetric unit of (4-cyanopyridine)2 4,4-biphenol 11 consists of half a bphe molecule and one 4cypy molecule and the resulting 2:1 centrosymmetric supramolecular adducts are sustained by I (D: 2.733(4) ; CO distance : 1.353(3) ; C NC angle: 116.3(3) ). The bphe molecules are flat (similarly to bphe in 9 ) and the dihedral angle between the bphe and the 4cypy rings is 49.2 The 2:1 adducts are connected through CHN C dimers forming 1D chains. The packing is further stabilized by C HO and CHN C interactions existing between chains which are aligned side by side along the b axis (Figure 2.15). Based upon further investigations

PAGE 62

41 involving solid state co-cry stallization, co-crystal 11 (thereafter form I, 11a ) was found to exhibit additional polymorphic modification.235 The comparative structural analysis of both forms I and II will be discussed in more details in chapter 3. Figure 2.15. Crystal struct ure of (4-cyanopyridine)2 4,4-biphenol, 11a Co-crystal of (4-cyanopyridine)3 phloroglucinol 12, consists of discrete 3:1 entities sustained by I (D1: 2.721(4) D2: 2.784(3) D3: 2.795(3) ), Figure 2.16. The CO distances are 1.353(3) 1.359(3) 1.367(3) and the CNC angle of the hydrogen bonded 4cypy rings are 116.3(3) 116.8(3) and 116.3(3) respectively. The dihedral angles between the phlgl and the 4cypy rings are 94.9 98.4 and 99.3 Despite that phgl can adopt a 3-fold geometry,216 in this structure, the phlgl molecules exhibit convergent orientation. In effect, two of the three 4cypy molecules interact via face-to-face stacking (ca. 3.84 ). Such a ggregates self-assemble via CHN C dimers into centrosymmetric pairs of adducts. These pairs are further extended through a C HN C interactions to form 1D tapes. Such ta pes are then packed side by side along the b axis forming 2D molecular layers.

PAGE 63

42 The CSD analysis contain several co-crystals of phlgl with pyridines, of which one is particularly relevant from the cr ystal structure perspective, namely (4,4bipyridine)3 (phloroglucinol)2 (CSD refcode: TEKKOY),236 Figure 2.17. Its crystal structures are also su stained by heterosynthon I and the OHNarom hydrogen bond lengths are similar to those observed in 12 (D1: 2.71(1) D2: 2.75(1) D3: 2.794(9), D4: 2.71(1) D5: 2.73(1) D6: 2.76(1) ). In the crys tal structure of TEKKOJ, phlgl molecules also adopt convergent c onformation. In effect, the adjacent bipy molecules stack in face-to-face manner (ca. 4.02 ). Si milar arrangement of molecules is observed in 12 ( 4cypy molecules indicated by the capped stic k style). Such aggregates are then linked together by the remaining bipy molecule into 1D chains The adjacent chains are related by 21 screw axis and stabilized via weak CHO interactions occurring between the phlgl molecules, thereby generati ng 2D molecular sheets. Figure 2.16. Crystal struct ure of (4-cyanopyridine)3 phloroglucinol, 12

PAGE 64

43 Figure 2.17. Crystal structure of (4,4-bipyridine)3 phloroglucinol, TEKKOJ While the presented co-cryst al structures resulted fr om successful experiments, attempts that involved co-cryst allization of 1-hexadecanol ( hexdec) with 3-cyanopyridine ( 3cypy ) and 4-cyanopyridine ( 4cypy) were ineffective. The unsuccessful attempts were also confirmed by grinding, solvent-drop grinding and melting precedures (more details in chapter 3). Since both ( 3cypy ) and ( 4cypy ) are capable of forming co-crystals with other molecules, e.g. res, bphe, phlgl the lack of success with hexdec perhaps can be contributed to the molecular features of hexdec The existence of long hydrophobic chains that are stabilized by a seri es of cooperative van der Waal forces237 can possibly inhibit the dissociation of hexdec structure and the fo rmation of hydrogen bonded adducts with other molecules. An analysis of the CSD showed that of the 135 structures that are sustained by I (Table 2.1), none contain long chain aliphatic monoalcohols. In summary, the crystal structures of the 12 co-crystals demonstrate the recurrence of supramolecular heterosynthon I in the presence of the CN moiety in all

PAGE 65

44 reported cases. Such a high reliability of I is further confirmed by the unsuccessful cocrystallizations of 3hypy and 5hyquin with CN-containing compounds: cynaphth ocyben and p-cyben These negative results can be rationalized by the existence of OH/Narom moieties in the molecular structures of 3hypy and 5hyquin For instance, the crystal structure of 3hypy238 exhibits the OHNarom supramolecular heterosynthon (Figure 2.18), which, as shown by 1 12 can not be easily altere d by the introduction of a CN moiety. Figure 2.18. Supramolecular hetrosynthon I present in the crystal structure of 3hydroxypyridine The hydrogen bond lengths of I in all co-crystals correspond to the expected values of a typical OHNarom interaction (Table 2.1) and th e structural parameters (CO lengths and CNC angles) of the ancillar y groups suggest a neutral character of I The supramolecular chemistry of I can also be predicted to a certain extent. Depending on the geometrical distribution and the ratio of the OH and Narom moieties, the formation of I leads to 1:1 or 1:2 discrete supramolecular entities, which due to the presence of CN moieties, can be further extended to 1D chains or 2D sheets sustained by weak C HN C interactions. The existence of co-crystal 4 and the CSD statistical data that reveals the presence

PAGE 66

45 of the CHNarom interaction in ca. 14% of the crys tal structures sustained primarily by I, indicates the importance of these w eak interactions in the molecular association.104,233,239-242 Although inherently soft, these in teractions combined with other weak forces, e.g. stacking, prove to be directional e nough to lead to the formation of co-crystals that would otherwise be unexp ected if only strong hydr ogen bonds would be taken into consideration.243 Furthermore, this result suggest s that, while it is possible to anticipate the existence of some primary inte ractions, due to the c ontribution of weaker forces, it is difficult to predict the ultimate crystal packing of a substance based upon the knowledge of molecular structure of the constituent(s). Co-crystals 1-12 were examined in the contex t of their thermal stability. Recent reports of Aakery et al and Nangia et al ., related to systematic studies of cocrystals that contain dicar boxylic acid as one of the components, suggest that their melting point alteration can be rationalized based upon the thermal properties of the acid component; i.e. the alternating trend exhibited in the aliphatic carboxylic acid series (in a homologous series, a carboxylic acid with odd number of carbon atoms has a relatively lower melting point than those with an even number of carbon atoms)244 is maintained in co-crystals.133,245 However, whether a co-crystal will melt at a temperature higher, lower, or in between the melting temperatures of the co-crystal constituents, could not be predicted. The analysis of the th ermal behavior of co-crystals 1-12 (Table 2.3) shows no correlation with respect to the melting points of the corresponding co-crystal formers. Co-crystals 1 and 5 melt at temperatures lower than their constituents, while co-crystals 6 and 7 melt at temperatures higher than thei r components. The remaining co-crystals 2, 3,

PAGE 67

46 4, 8, 9, 10, and 12 melt at temperatures within the ranges indicated by the melting points of the respective co-crystal formers. Theref ore, although the rationalization of the melting trends in a specific class of co-crystals can be possible,94,246 the anticipation of the melting point for a given co-crystal still remains elusive.91,98,247 2.3. Conclusions In summary, the study presented herein involves a series of model co-crystals adds to the limited amount of the CSD inform ation related to the frequency of occurrence of supramolecular heterosynthon I in the presence of the comp eting acceptor, CN moiety. That I occurs reliably in the presence of a CN moiety, suggests that OHNarom hydrogen bond is favored over the possible OHN C hydrogen bond II Considering that I and II are favored over III the relative ranking of th ese supramolecular synthons can be presented as: I > II > III Therefore, I can be particularly suitable for crystal engineering of co-crystals comprised by the OH, Narom, and CN moieties. Furthermore, if the robustness of I remains intact in the presence of broader range of hydrogen bonding donors and acceptors, I could perhaps, be considered as a remarkably predictable interaction, which can be reliably utilized in co-crystallization of more complex compounds that possess multiple hydrogen bonding sites. This conclusion is particularly relevant to co-crystals of APIs, since they are relatively complex molecules that often contain either OH or Narom moieties, or can interact with OHor Narom-containing cocrystal formers. Modification of the physicoche mical properties that is associated with the co-crystal formation may lead to inte resting opportunities toward new formulations

PAGE 68

47 for the improved performance of an API. 2.4. Experimental 2.4.1. Syntheses All reagents were purchased from Aldric h and used without further purification. Single crystals of compounds 1-12 were obtained via slow eva poration of stoichiometric amounts of starting materials in appropriate solvents and were isolated from solution before complete evaporation of the solvents. Co-crystal 1: 3-Cyanophenol4-phenylpyridine. To 3-cyanophenol (0.015 g, 0.13 mmol) was added 4-phenylpyridine (0.020 g, 0.13 mm ol) and 2 mL of 1:1 of acetone and ethyl acetate solvent mixture. Slow evaporati on of the solution afforded colorless crystals of 1 (0.027 g, 0.10 mmol, 77%), mp = 54-57 C after 3 days. Co-crystal 2: (3-Cyanophenol)21,2-bis(4-pyridyl)ethane. To 3-cyanophenol (0.026 g, 0.22 mmol) was added 1,2bis(4-pyridyl)ethane (0. 020 g, 0.11 mmol) and the mixture was dissolved in 2 mL of acetone After 3 days colorless crystals of 2 (0.035 g, 0.083 mmol, 75%), mp = 106-108 C, were observed. Co-crystal 3: (3-Cyanophenol)2trans-1,2-bis(4-pyridyl)ethylene. Co-crystal 3 was obtained using the reagents in 4:1 molar ratio. To 3-cyanophenol (0.052 g, 0.44 mmol) was added trans-1,2bis (4-pyridyl)ethylene (0.020 g, 0.11 mmol) and 2 mL of 1:1 acetone and ethyl acetate mixture. After 2 days colorless crystals of 3 (0.031 g, 0.074 mmol, 67%), mp = 112-114 C, were obtained.

PAGE 69

48 Co-crystal 4: 3-Cyanophenoltrans-1,2-bis(4-py ridyl)ethylene. Co-crystal 4 was obtained using the starting materials in 2: 1 ratio. To 3-cyanophenol (0.026 g, 0.22 mmol) was added trans-1,2bis (4-pyridyl)ethylene (0.020 g, 0.11 mmol) and the mixture was dissolved in 2 mL of ethyl acetate. Colorless crystals of 4 (0.026 g, 0.086 mmol, 79%), mp = 124-125 C, appeared after 2 days. Co-crystal 4 can also be obtained by using the reagents in 1:1 molar ratio. Co-crystal 5: 4-Cyanophenol4-phenylpyridine. To 4-cyanophenol (0.015 g, 0.13 mmol) was added 4-phenylpyridine (0.020 g, 0.13 mmol). The mixture was dissolved in 2 mL of chloroform and left to evaporate slowly at 4 C. After 4 days co lorless crystals of 5 (0.030 g, 0.11 mmol, 85%), mp = 65-66 C, were observed. Co-crystal 6 : (4-Cyanophenol)24,4-bipyridine To 4-cyanophenol (0.031 g, 0.26mmol) was added 4,4-bipyridine (0.020 g, 0.13 mmol). The solid mixture was dissolved in 2 mL of metha nol and the solution was left undisturbed to ev aporate under ambient conditions. After 12 days yellow needles of 6 (0.042 g, 0.097 mmol, 0.75%), mp = 143-146 C, were formed. Co-crystal 7: (4-Cyanophenol)21,2-bis(4-pyridyl)ethane To 4-cyanophenol (0.026 g, 0.22 mmol) was added 1,2bis (4-pyridyl)ethane (0.020 g, 0.11 mmol) and the mixture was dissolved in 2 mL of metha nol. After 8 days colorless crystals of 7 (0.031 g, 0.073 mmol, 67%), mp = 138-139 C, were obtained. Co-crystal 8: (4-Cyanophenol)2trans-1,2-bis(4-pyridyl)ethylene. To 4cyanophenol (0.026 mg, 0.22 mmol) was added 1,2bis (4-pyridyl)ethylene (0.020 mg, 0.11 mmol). The solid mixture was dissolved in 2 mL of acetonitrile. After 2 days

PAGE 70

49 colorless crystals of 8 mp = 141-142 C were observed. Co-crystal 9: (3-Cyanopyridine)24,4-biphenol. To 3-cyanopyridine (0.040 g, 0.38 mmol) was added 4,4-biphenol (0.036 g, 0.19 mmol) and the mixture was dissolved in 2 mL of methanol. After 6 days colorless crystals of 9 (0.052 g, 0.13 mmol, 68%), mp = 250 C (followed by decomposition), were observed. Co-crystal 10: (4-Cyanopyridine)2resorcinol. To 4-cyanopyridine (0.040 g, 0.38 mmol) was added resorcinol (0.021g, 0.19 mmol ) and 2 mL of acetonitrile. The solution was left to evaporate at ambient temperatur e and after 8 days colorless crystals of 10 (0.055 g, 0.16 mmol, 84%), mp = 93-94 C, were formed. Co-crystal 11a: (4-Cyanopyridine)24,4-biphenol To 4-cyanopyridine (0.040 g, 0.38 mmol) was added 4,4-biphenol (0.036 g, 0.19 mmol). To the solid mixture was added 2 mL of methanol and the solution wa s left to evaporate at ambient conditions. After 4 days, yellow crystals of 11a (0.044 g, 0.13 mmol, 67%) were formed. Melting point of 11a was not determined due to its d ecomposition followed by melting of 4,4biphenol (more details related to this co-crystal and its polymorphs are included in chapter 3). Co-crystal 12: (4-Cyanopyridine)3phloroglucinol. To 4-cyanopyridine (0.041 g, 0.39 mmol) was added phloroglucinol ( 0.016 g, 0.13 mmol) and the mixture was dissolved in 2 mL of acetone. Af ter 3 days yellow needles of 13 (0.045 g, 0.10 mmol, 77%), mp = 116-117 C were observed. Additional data:

PAGE 71

50 Solution co-crystallization attempts to obtain 3-cyanopyridine 1-hexadecanol and 4-cyanopyridine 1-hexadecanol were unsuccessfu l; the IR spectroscopy and PXRD spectra of the solid obtained from soluti on evaporation of the corresponding components in equimolar ratio revealed a mixture of starting materials. Solution co-crystallizations that involv ed stoichiometric amounts of the following pairs: 3-cyanopyridine and 1-naphthol, 3-cy anopyridine and resorcinol, 3-cyanopyridine and phloroglucinol, and 4-cyanopyridine a nd 1-naphthol, resulted in liquid products, unsuitable for solid state characterizations. Solution co-crystallization atte mpts to obtain: 3-hydroxypyridine 1cyanonaphthalene, (3-hydroxypyridine)2 1,3-dicyanobenzene, (3-hydroxypyridine)2 1,4dicyanobenzene, 5-hydroxyisoquinoline 1-cyanonaphthalene, (5hydroxyisoquinoline)2 1,3-dicyanobenzene, a nd (5-hydroxyisoquinoline)2 1,4dicyanobenzene resulted in mixtur es of the star ting materials. All co-crystals were analyzed by infrare d spectroscopy using a Nicolet Avatar 320 FTIR instrument. The purity of bulk samples was confirmed by X-ray powder diffraction. Co-crystals 1 5 6 8 9 and 11a were analyzed on a Rigaku Miniflex Diffractometer using Cu K ( = 1.54056 ), 30 kV, 15 mA. The data was collected over an angular range of 3 to 40 2 in continuous scan mode using a step size of 0.02 2 and a scan speed of 2.0/min. Compounds 2-4 7 10 and 12 were analyzed on Bruker AXS D8 discover X-ray diffract ometer equipped with GADDSTM (General Area Diffraction Detection System), a Bruker AXS HI-STAR area detector at a distance of

PAGE 72

51 15.05 cm as per system calibration, a copper source, automated x-y-z stage, and 0.5 mm collimator. Data were collected over 2.1-37.0 2 range at a step size of 0.02 2. Melting points of compounds 1 10 and 12 were determined on a MEL-TEMP apparatus, and the comparison of the melting points of 1-12 with the corresponding constituents is summarized in Table 2.2. Table 2.2. Comparison of the melting points of co-crystals 1-12 and the corresponding components Co-crystal Mp of co-crystal [ C] Mp of component 1 [ C] Mp of component 2 [ C] 1 54 57 81 82 (3cyph) 68 71 (phenpy) 2 106 108 81 82 (3cyph) 110 112 (bipyeta) 3 112 114 81 82 (3cyph) 150 153 (bipyete) 4 124 125 81 82 (3cyph) 150 153 (bipyete) 5 65 66 110 113 (4cyph) 68 71 (phenpy) 6 143 146 110 113 (4cyph) 110 114 (bipy) 7 138 139 110 113 (4cyph) 110 112 (bipyeta) 8 141 142 110 113 (4cyph) 150 153 (bipyete) 9 250 (dec) 49 50 (3cypy) 283 (bphe) 10 93 94 76 79 (4cypy) 109 111 (res) 11 76 79 (4cypy) 283 (bphe) 12 116 117 76 79 (4cypy) 216 (phlgl) 2.4.2. Single Crystal X-ray Crystallography Co-crystals 1-12 were examined under a microsc ope and suitable single crystals were selected for X-ray diffraction. Data were collected on a BrukerAXS SMART APEX CCD diffractometer with monochromatized Mo K radiation ( = 0.71073 ) connected to KRYO-FLEX low temperature device. Data for 2 3 6 7 and 9 were collected at 100 K. Data for 1 4 5 8 and 10-12 were collected at 298 K. Lattice parameters were determined from least s quare analysis, and reflection data were

PAGE 73

52 integrated using the program SAINT. Lorent z and polarization corre ctions were applied for diffracted reflections. In addition, the data of all compounds, except one ( 12 ) was corrected for absorption using SADABS.248 Structures were solved by direct methods and refined by full matrix least squares based on F2 using SHELXTL.249 All non-hydrogen atoms were refined with anisotropic disp lacement parameters. All H-atoms bonded to carbon atoms, were placed geometrically a nd refined with an isotropic displacement parameter fixed at 1.2 times Uq of the atoms to which they were attached. The O bonded protons were located from Fourier differen ce map and refined isotropically based upon the corresponding O atom (U(H)=1.2Uq(O)). Crystallographic data for 1 12 are presented in Table 2.3 and selected hydrogen bond distances are listed in Table 2.4.

PAGE 74

53 Table 2.3. Crystallographic data and struct ure refinement parameters for co-crystals 1-12 1 2 3 4 5 6 7 8 9 Chemical formula C7H5NO C11H9N (C7H5NO)2 C12H12N2(C7H5NO)2 C12H10N2C7H5NO C12H10N2C7H5NO C11H9N (C7H5NO)2 C10H8N2 (C7H5NO)2 C12H12N2(C7H5NO)2 C12H10N2(C6H4N2)2 C12H10O2 Formula .wt. 274.31 422.48 420.46 301.34 274.31 394.42 422.48 420.46 394.42 Crystal system MonoclinicMonoclinicMonoclinicMonoclinicMonoc linic Triclinic MonoclinicMonoclinicMonoclinic Space group P 21/ c P 21/ c P 21/ n C 2/ c C 2/ c P 1 P 21/ c P 21/ c P 21/ c a () 9.90(1) 13.223(3) 16.542(2) 18.715(8) 26. 22(1) 3.848(2) 11.195(3) 14.660(3) 20.866(6) b () 21.65(3) 6.197(1) 7.506(1) 7.228(3) 7.481(5) 8.755(4) 7.335(2) 4.110(1) 7.437(2) c () 7.590(9) 14.630(3) 18.709(2) 23.208(9) 19 .41(1) 14.364(6) 13.530(4) 18.290(4) 6.766(2) ( ) 90 90 90 90 90 96.647(7) 90 90 90 ( ) 112.14(2) 113.661(3)106.763(2)90.48(1) 128.31 (1) 94.498(8) 99.456(7) 90.74(3) 98.142(5) ( ) 90 90 90 90 90 95.474(7) 90 90 90 Volume (3) 1508(3) 1098.1(4) 2224.2(4) 3139(2) 2987(3) 476.6(4) 1095.9(6) 1101.9(4) 1039.4(5) Dcalc ( g cm-3) 1.208 1.278 1.256 1.275 1.220 1.374 1.280 1.267 1.260 Z 4 2 4 8 8 1 2 2 2 range 1.88-24.991.68-26.731.95-25.001.75-24.991.98-25.00 1.43-26.371.84-25.002.61-23.350.99-26.73 Nref./Npara. 2607/191 2314/145 3798/290 2748/208 2585/190 1888/136 1818/145 1565/145 2181/136 T (K) 298 100 298 100 298 100 100 298 100 R1 0.0558 0.0597 0.0504 0.0688 0.0630 0.0659 0.0672 0.0502 0.0685 wR2 0.1461 0.1550 0.1501 0.1492 0.1759 0.1926 0.1752 0.1307 0.1747 GOF 0.924 1.103 1.069 1.002 0.805 1.042 1.044 0.814 1.082 Abs coef. 0.076 0.083 0.082 0.081 0.077 0.090 0.083 0.083 0.083

PAGE 75

54Table 2.3. Crystallographic data and struct ure refinement parameters for co-crystals 1-12 (continued) 10 11a 12 Chemical formula (C6H4N2)2 C6H6O2 (C6H4N2)2 C12H10O2(C6H4N2)3 C6H6O3 Formula .wt. 318.33 394.42 438.44 Crystal systemTriclinic MonoclinicTriclinic Space group P 1 C 2/ c P 1 a () 9.171(4) 39.87(2) 7.889(3) b () 9.941(5) 7.586(5) 8.156(3) c () 10.265(5)6.915(2) 19.673(7) ( ) 95.674(8)90 83.866(7) ( ) 94.987(8)91.97(2) 85.627(6) ( ) 114.784(8)90 63.512(5) Volume (3) 836(7) 2090(2) 1125.9(6) Dcalc ( g cm-3)1.263 1.253 1.293 Z 2 4 2 range 2.01-24.991.02-25.091.04-25.00 Nref./Npara. 2875/2171830/136 3915/298 T (K) 298 298 298 R1 0.0506 0.0682 0.0586 wR2 0.1397 0.2121 0.1676 GOF 1.058 0.875 0.938 Abs coef. 0.086 0.082 0.089

PAGE 76

55Table 2.4. Geometrical parameters of supramol ecular heterosynthon I present in co-crystals 1-12 Hydrogen bond d () D () () 1 O HNarom 1.63 2.708(3) 168.9 2 O HNarom1.77 2.691(2) 176.3 O HNarom 1.69 2.727(2) 173.3 3 O HNarom 1.75 2.729(2) 173.6 4 O HNarom1.64 2.663(3) 165.6 5 O HNarom 1.60 2.695(4) 175.7 6 O HNarom1.84 2.718(3) 175.4 7 O HNarom1.80 2.698(4) 175.5 8 O HNarom 1.74 2.714(4) 159.5 9 O HNarom1.77 2.765(3) 166.3 O HNarom 1.96 2.856(2) 169.9 10 O HNarom 1.86 2.763(2) 175.0 11a O HNarom1.80 2.733(4) 174.1 O HNarom 1.84 2.784(3) 161.6 O HNarom 1.77 2.721(4) 167.7 12 O HNarom 1.77 2.795(3) 176.0

PAGE 77

56 Chapter 3 Methods of Preparation of Co-c rystals and Polymorphism in Cocrystals 3.1. Focus As demonstrated in chapter 2, the single crystal analysis of model co-crystals revealed the existence of the OHNarom supramolecular heterosynthon ( I ) in all described structures. The reliabil ity of supramolecular synthons can be related to their recurrence in co-crystals prepared via diffe rent methods and under variable conditions. Considering the recent reports related to th e successful utilization of mechanochemical approach toward co-crystallization,141,151 model co-crystals 1-12 have been investigated in the context of their reproducibility via dry grinding and solvent-drop grinding. Grinding co-crystallization procedures invol ve grinding stoichiometric amounts of two (or more) co-crystal formers using mortal and pestle or in a mechanical grinder.143 Solvent-drop grinding relies on the addition of a small amount of solvent to the grinding procedure.151 In addition, growth of co-crystal s from melt was utilized and was based upon melting stoichiometric amounts of co-cryst al formers to a temperature slightly higher than the melting point of the higher melting co-crystal former and allowing the melted mixture to cool down to room temperature. Due to the increasing awareness of pol ymorphism in organic compounds and its

PAGE 78

57 important implications in terms of both fundamental and commercial interest, it was appropriate to address the existence of this phenomenon in the set of model co-crystals presented herein, with special emphasis on the origin of polymorphism. Solvent-drop grinding method has been utilized herein, due to its efficiency in the polymorphism screen and control, as repor ted in recent literature.186,188,208 3.2. Results and Discussion 3.2.1. Methods of Preparations of Co-crystals The reproducibility of co-crystals obtained from solvent evaporation approach was evaluated via co-crystallization methods that involved dry grinding, solvent-drop grinding, and growth from melt. To search fo r other possible polymorphs of co-crystals 1-12 solvent-drop grinding involving seven solven ts of different polarities: cyclohexane, toluene, chloroform, ethyl acetate, metha nol, dimethyl sulfoxide (DMSO), and water were utilized. It was observed that whereas 4-minute grinding with solvent-drop was efficient enough to obtain pure co-crystals, th e dry grinding approach has not always led to complete conversions. To achieve comple te co-crystallizations, the time of dry grinding needed to be extended to 20 minutes. As confirmed by PXRD and FT IR analysis, co-crystals 1 2 5 6 7, 9, 10, and 12 were reproduced using the above procedures and no additional forms were observed. However, the mechanical co-crystallization was found to have additional effects on the remaining co-crystals 3 4 8, and 11

PAGE 79

58 Co-crystallization of 3-cyanophenol and trans-1,2-bis(4-pyridyl)ethylene. 3cyphe and bipyete can form co-crystals with 2:1 stoichiometry ( 3 ) and with 1:1 stoichiometry ( 4 ). It was observed that dr y grinding or melting the com ponents in 2:1 ratio result in mixtures of 3 and 4, whereas solvent-drop grinding with all the applied solvents affords pure 3 When the two components are combined in 1:1 ratio, the dry grinding and melting, also afford mixtures of 3 and 4 The solvent-drop grinding with cyclohexane, toluene, chloroform, ethyl acetate, methanol, and water solvents afforded mixtures of 3 and 4 however, DMSO-drop grinding revealed the formation of pure 4 Co-crystallization of 4-cyanophenol and trans-1,2-bis(4-pyridyl)ethylene. The PXRD patterns obtained based upon melti ng, dry grinding, solvent-drop grinding (cyclohexane, toluene, chloroform, et hyl acetate, methanol, and water) of 4cyphe and bipyete in 2:1 ratio revealed additional peaks and shifts as compared to the PXRD pattern of 8 Of the seven solvents only DMSO-drop gri nding resulted in a product of which the PXRD pattern matched the pattern of 8 Interestingly, when the components in 1:1 ratio were subjected to melting, grinding and solvent-drop grinding, similar peaks and shifts were observed. Search for additional forms of this co-crystal based upon a series of solution crystallizations of the components in various ratios (1:1, 2:1, and 4:1), followed by single crystal X-ray analys is, revealed the existence of a 1:1 co-crystal of 4cyphe and bipyete which exhibit two concomitant polymorphic modifications, form I ( 13a ) and from II ( 13b ). Detailed discussion and structural comparison of 13a and 13b will be presented in section 3.2.2. Co-crystallization of 4cyanopyridine and 4,4-biphenol Dry grinding and

PAGE 80

59 solvent-drop grinding of 4cypy and bphe in 2:1 ratio revealed consistent, however drastically different PXRD patterns, as compared to the XPD pattern of 11a (the crystal structure of 11a was introduced in chapter 2). In s earch for the unknown forms, a series of solution crystallization experiments, followed by single crystal X-ray analysis, revealed the existence of an additional polymorph form II, 11b The two polymorphs appeared concomitantly when a mixture of MeOH and EtOAc in 1:1 ratio was used. It was established that form II was exclusivel y afforded from the solvent-drop grinding experiments involving cyclohexane, toluene, chloroform, ethyl acetate, methanol, and water, whereas the utilization of DMSO l ead to the formation of a reported DMSO solvate of bphe (CSD refcode ECELON01, see appendix 24).250 Detailed discussion and structural comparison of 11a and 11b will be presented in section 3.2.3. The attempts to obtain 3-hydroxypyridine 1-cyanonaphthalene, (3hydroxypyridine)2 1,3-dicyanobenzene, (3-hydroxypyridine)2 1,4-dicyanobenzene, 5hydroxyisoquinoline 1-cyanonaphthalene, (5-hydroxyisoquinoline)2 1,3dicyanobenzene, and (5-hydroxyisoquinoline)2 1,4-dicyanobenzene via mechanochemical and melting co-crystalliz ations resulted in mixtures of the corresponding starting materials, which conf irmed, mentioned in chapter 2, unsuccessful solution co-crystallizations of the listed pairs of co-crystal formers. These results were expected based upon the observation that CN-contaning compounds would not interact with the OH/Narom-contaning compounds. A series of co-crystallization experiments that involved stoichiometric amounts of the following pairs: 3-cyanopyridine and 1naphthol, 3-cyanopyridin e and resorcinol, 3-

PAGE 81

60 cyanopyridine and phloroglucinol, and 4-cyanopyr idine and 1-naphthol, resulted in liquid products. Additional co-crystallizati on experiments carried out at -4 C also resulted in liquid products. An illustration of the so lid-to-liquid conversion when 3-cyanopyridine and 1-napthol are combined is presented in Figure 3.1. Based upon the high reliability of heterosynthon I (chapter 2) and the fact that co-crystals 9-12 (comprised of 3cypy 4cypy and phenols ) exist, it could be speculated that th e listed co-crystal formers reacted, and that the melting points of the co-crystalli ne products are much lower than room temperature. However, to prove or disprove this hypothesis additional experimentations are needed. Figure 3.1. Solid-to-liquid conversion occurring in a mixture of 3-cyanopyridine and 1napthol 3.2.2. Polymorphism in 4-cyanophenol trans-1,2-bis-(4-pyridyl)ethylene co-crystal 4-cyanophenol t-1,2bis -(4-pyridyl)ethylene co-crystal, 13 exhibits two concomitant polymorphs: monoclinic form I ( 13a ) and triclinic form II ( 13b ). The asymmetric unit of 13a consists of two 4cyphe molecules and two bipyete molecules. Interestingly, the assembly of the two com ponents results in two distinct supramolecular

PAGE 82

61 entities (Figure 3.2). The firs t entity is a non-centrosymm etric 2:1 adduct sustained by I (D1: 2.721(3) D2: 2.669(4) ; CO distances: 1.336(3) 1.355(4) ; CNC angles: 116.6(3) and 116.1(3) ) with the dihedral angle between the 4cyphe and the bipyete rings being ca. 76.9 and 79.4 The second entity is a non-hydrogen bonded bipyete residing in between the 2:1 a dducts. The free and hydrogen bonded bipyete molecules alternate through stacking along the b axis. The stacking involves face-to-face aromatic interactions of re latively short interplanar dist ance of ca. 3.61 which can make the unsaturated moieties of bipyete suitable for [2+2] photodimerization in the solid state. The crystal packing is further exte nded into 2D planar sheets in result of the formation of CHN C dimer between the stacked columns. Figure 3.2. Crystal structure of 4-cyanophenol t-1,2bis (4-pyridyl)bipyethylene form I, 13a

PAGE 83

62 Figure 3.3. Crystal structure of 4-cyanophenol t-1,2bis (4-pyridyl)bipyethylene form II, 13b The asymmetric unit of 13b also consists of two 4cyphe molecules and two bipyete molecules. However, unlike in 13a the components form two crystallographically independent 1:1 supramolecular adducts sustained by I (D1: 2.689(4) D2: 2.717(4) ; CO distances: 1.327(5) 1.347(4) ; CNC angles: 117.2(3) 116.6(4) ). The dihedral angles between the 4cyphe and the bipyete rings in the respective adducts are ca. 74.5 and 79.7 The 1:1 adducts stack on top of each other in face-to-face fashion along the a axis, and the distance between the adjacent bipyete molecules is ca. 3.48 Such columns of th e stacked adducts expand into 2D planar sheets through centrosymmetric CHN C dimer formed between adjacent 4cyphe molecules (Figure 3.3). The simila rity of crystal packings of 13a and 13b is manifested on the calculated powder diffraction patterns (Figure 3.4.). The major peaks of the two forms overlap, while the difference between the patterns lay in the small intensity peaks

PAGE 84

63 at ca. 16-17 2 and ca. 26 2 051015202530354045 0 200 400 600 800 1000 1200 1400 1600 1800 experimental Form 2 Form 1 Figure 3.4. X-ray powder diffraction patterns of 4-cyanophenol t-1,2bis (4pyridyl)bipyethylene polymorphs: bulk samp le (green) and simulated patterns of form I (black) and form II (red) 3.2.3. Polymorphism in (4-cyanopyridine)2 4,4-biphenol co-crystal An examination by IR spectroscopy, powde r X-ray diffraction and single crystal X-ray diffraction of the irregular hexagons and parallelepiped plates (Figure 3.5) obtained from solution crystallization confirmed the presence of two concomitant polymorphs of (4cypy)2 bphe form I ( 11a ) and form II ( 11b ), respectively (numbering is based on order of discovery).

PAGE 85

64 a) b) Figure 3.5. Concomitant polymorphs of (4-cyanopyridine)2 4,4-biphenol co-crystal: a) form I irregular hexagons, b) form II parallelepiped plates The single crystal X-ray structures of form I and form II confirm the expected 2:1 stoichiometry and reveal the presen ce of supramolecular heterosynthon I between the 4cypy and bphe molecules. However, the crystal p acking patterns of the two forms are distinct due to conforma tional differences in the bphe molecules (Figures 3.6 and 3.7). Form I crystallizes in C 2/ c and the asymmetric un it consists of half a bphe molecule (residing on crystallographic inversion center) and one 4cypy molecule. The bphe molecules are flat (torsion angl e between two phenol rings is 180.0 ), sustain centrosymmetric (4cypy)2 bphe supramolecular adducts (Figure 3.6a). The dihedral angle between 4cypy and bphe molecules (planes of the ar omatic rings represented by atoms C11N11C15 and C2C1C6) is 128.5 The OHNarom hydrogen bond distance (D: 2.733(4) ) is within the expected range for alcohol-aromatic n itrogen interactions (Table 2.1). The crystal structure of 11a was already described in chapter 2 (co-crystal 11a ). Form II ( 11b ) crystallizes in P 21/ n The asymmetric unit consists of four bphe

PAGE 86

65 molecules and two bphe molecules, which differ in their torsion angles: ca. 145.7(3) (torsion represented by C3C4C7C12) in bphe and ca. 160.2(3) (torsion represented by C23C24C27C32) in bphe Figure 3.5b. The dihedral a ngles between the planes of the aromatic rings of 4cypy and bphe are 153.1 (planes represented by C61N61 C65 and C9C10C11) and 112.0 (planes represented by C41N41C45 and C2C1 C6). The dihedral angles between th e planes of the aromatic rings of 4cypy and bphe are 141.1 (planes represented by C51N51C55 and C22C21C26) and 104.1 (planes represented by C71N71C75 and C29C30C31). This range of dihedral angles in the (4cypy)2 bphe adducts in form II (Figure 3.6b) is uns urprising given that supramolecular heterosynthon I is a one-point recognition interac tion with rotational freedom. The O HNarom hydrogen bond distances within the supr amolecular adducts are 2.693(3) and 2.789(4) in (4cypy)2 bphe and 2.838(3) and 2.766(3) in (4cypy)2 bphe (a)

PAGE 87

66 (b) Figure 3.6. ORTEP plot of supramolecular adducts in (4-cyanopyridine)2 4,4-biphenol drawn at 50% probability level for non-hydrogen atoms: (a) form I, (b) form II. Note that bphe molecules are flat in form I, whereas twisted in form II The crystal structure of 11b is shown in Figure 3.7. The adjacent 2:1 adducts are connected via non-centr osymmetric CHN C dimers forming 1D chains. The chains are packed side by side along the a axis, thereby generating a 2D network. Figure 3.7. Crystal packing in (4-cyanopyridine)2 4,4-biphenol form II, 11b (4cypy)2 bphe (4cypy)2 bphe

PAGE 88

67 The CSD contains only 22 structures involving bphe.251 The number of crystallographically independent bphe molecules in these stru ctures is 33. Figure 3.8 reveals that one molecule is flat (torsion angle is 0 ), 16 molecules are slightly twisted (05 ), and torsion angles of the remaining 16 molecules are spread over a wide range (570 ). Furthermore, 3 of the 22 structur es exhibit conformational isomorphism252 and one compound exhibits conformational polymorphism.253 To our knowledge (4cypy)2 bphe form II is the first bphe -containing compound that exhibits both phenomena. A similar analysis involving biphenyl deri vatives and their co mplexes reveals 623 structures with 1,026 crystallograp hically independent molecules. The histogram of torsion angles (Figure 3.9a) show s a maximum corresponding to a 1-5 torsion angle range (189 molecules), however there is a s econd maximum occurri ng within the torsion angle range of 25-45 (651 molecules). These statistics are in agreement with previous reports and suggest that such molecules ar e especially influential upon crystal packing effects because of a small rotationa l barrier around the central CC bond.254,255 The torsion angles encountered in bphe (form I), bphe and bphe (form II), ca. 0.0 34.3 and 19.8 respectively, are consistent with the statistical data Interestingly, 92 of the 623 structures were found to exhibit conformati onal isomorphism. Furthermore, 19 of the 623 exhibit conformational polymorphism, of which 11 also exhibit conformational isomorphism. This subset of 92 structures wa s analyzed to determine the torsion angle differences between conformers that co-exist in the same crystal structure. To simplify the analysis, only structures that contain tw o conformational isomorphs were taken into consideration. This resulted in a subset of 67 entries. In 57 of these structures the torsion

PAGE 89

68 angle differences are rather subtle (ca. 1-10 ), whereas the remaining 10 pairs of conformational isomorphs exhibit highe r torsion angle deviations (10-40 ), Figure 3.9b. 0 2 4 6 8 10 12 14 16 18 051015202530354045505560657075808590Torsion angle [deg] N Figure 3.8. Histogram representing the torsion an gle distribution in 22 crystal structures of 4,4-biphenols 0 20 40 60 80 100 120 140 160 180 2000102030405060708090Torsion angle [deg] N 0 5 10 15 20 25 30 35 4051015202530354045505560657075808590Torsion angle [deg] N a) b) Figure 3.9. a) Distribution of torsion angles in the 623 crystal structures that contain biphenyl moieties; b) Distribution of torsion angle deviations within pairs of conformers present in 67 crystal structures The calculated density of form I is sligh tly lower than the calculated density of form II (1.253 Mg/m3 vs. 1.266 Mg/m3). The melting points of both co-crystal forms was not recorded because (4cypy)2 bphe decomposes upon heating. The DSC thermograms

PAGE 90

69 (see appendix 11) of the two polymorphs show two endothermic peaks; one corresponding to the co-crystal decomposition (ca. 94.3 C in form I and ca. 96.1 C in form II) and another corresponding to melting of bphe (ca. 284.1 C), respectively. Polymorphic transformations between the two forms of (4cypy)2 bphe were also investigated, as illustrated in Figure 3.10. When form I is dry or solvent-drop ground with methanol, acetone, or ethyl acetate, conversion to form II is observed. Conversely, slow evaporation of methanol, ethyl acetate, or acetone solutions of form II yield form I. Slurry experiments involving 1:1 mixtures of form I and form II in methanol, ethyl acetate or acetone afford form II. Experime ntal and simulated PXRD patterns of both polymorphs are presented in Figure 3.11. (4-Cyanopyridine)2+ 4,4'-Biphenol MeOH: EtOAc(1:1)Co-crystal Form I + Form II Form IMeOH, EtOAc, or Acetone G r i n d i n g o r s o l v e n t d r o p g r i n d i n gForm II Grinding or solvent drop grinding MeOH, EtOAc, or Acetone Slurry in MeOH, EtOAc, or Acetone Figure 3.10. Polymorphic conversions be tween form I and form II of the (4cyanopyridine)2 4,4-biphenol co-crystal

PAGE 91

70 Figure 3.11. PXRD patterns of (4-cyanopyridine)2 4,4-biphenol co-crystals: (a) simulated form I and (b) experimental form I obtained via slow evaporation from acetone; (c) simulated form II and (d) experimental form II obtained via acetone-drop grinding In summary, the combination of various methods of co-crystallization, namely grinding, solvent-drop grinding, gr owth from melt, and solution co-crystallization, have led to new discoveries. The first finding is re lated to the existence of a dimorphic cocrystal of 4cyphe bipyete ( 13a and 13b ) which was observed in result of grinding, solvent-drop grinding and melting co-crystallization me thods. The two forms of 4cyphe bipyete exhibit 1:1 stoichiometry and they occur concomitantly. The formation of this co-crystal is somewhat surprising from the perspective that a co-crystal of 4cyphe and bipyete with the expected 2:1 stoichiometry exists ( 8 ). Interestingly, 2:1 and 1:1 stoichiometries, although polymorphic forms we re not detected, also exist in the cocrystals of 3cyphe and bipyete ( 3 and 4 ). Similar molecular arrangement of bipyethe in co-crystals 4 13a and 13b may suggest that the crystal p ackings are strongly stabilized by aromatic stacking and CHNarom interactions, in a ddition to the primary supramolecular heterosynthon I The second finding is related to the ex istence of an additional polymorph of 1015202530354045 1015202530354045 a) b) c) d)

PAGE 92

71 (4cypy)2 bphe form II, which was discovered based on grinding and solvent-drop grinding co-crystallizations. Form II of (4cypy)2 bphe ( 11a ) exhibits concomitant and conformational polymorphism and confor mational isomorphism. To our knowledge, concomitant or conformational polymorphism or both phenomena, have been observed in several co-crystals,186,256,257 whereas all three phenomena have only been reported for the triphenylsilanol 4,4-dipyridyl co-crystal.256 It is interesting to note that the supramolecular heterosynthon, I which sustains both (4cypy)2 bphe and 4cyphe bipyete co-crystals remains unchanged in their polymorphic modifications. The origin of polymorphism in 11a and 11b can be related to the conformational flexibility of the bphe molecules, while polymorphism in 13a and 13b results from the variations in the crystal p acking. A detailed structural analysis of the 11 polymorphic co-crystals reported in the CSD reveals, that within the set of polymorphs of a given co-crystal, supr amolecular heterosynthons also persit.182,183,185 For instance, the crystal structures of two quinhydrone polymorphs, form and form Figure 3.12, are sustained by alternating p -benzoquinone and hydroquinone molecules that interact via OHO hydrogen bonds. Such a molecula r arrangement leads to the generation of infinite linear chains.116,258 The two crystal packings differ in the relative orientation of these chains. In form (Figure 3.12a) the chains are oriented nearly orthogonally with respect to each other, whereas in form (Figure 3.12b) the chains are packed side by side. Although concomitant pol ymorphism was not repo rted, a review of the relevant literature, reveal ed that both forms of quinhydrone can be crystallized from acetone.

PAGE 93

72 a) b) Figure 3.12. Crystal structures of the po lymorphic forms of qu inhydrone: a) form and b) form The persistency of supramolecular heterosy nthons is also seen in the dimorphic co-crystal of caffeine glutaric acid.186 In form I and form II of this co-crystal the components interact via carboxylic acid imidazole supramolecular heterosynthon, Figure 3.13. The difference between the two pol ymorphs is related to rather subtle conformational variations of the methylen e groups of glutaric acid (conformational polymorphism).179 Form I and form II occur concomitan tly when the co-crystallization is carried out in chloroform. Interestingly, th e polymorphic outcome of co-crystallization can be controlled by applying solvent-dr op grinding methodology. Form I could be obtained when non-polar solvents (e.g. hexane ) were added to the grinding procedure, whereas form II could be obtained upon the addi tion of polar solvents (e.g. acetonitrile). Further structural analysis of the 9 remaining co-crystals shows that the supramolecular heterosynthons also persist within a give n set of polymorphs. This indicates that the differences in crystal p ackings that result from the engagement of molecules in different hydrogen bonding modes, observed in single-component compounds, do not occur in co-crystals.

PAGE 94

73 a) b) Figure 3.13. Crystal structures of the polymorphic forms of caffeine glutaric acid cocrystal: a) form I and b) form II The existence of concomitant polymorphism in co-crystals 11 and 13 implies that the free energy difference between the crystalline forms is small.56,181 Despite the fact that the phenomenon of concomitancy of polymorphs has been long known,259 there has been little information rela ted to the frequency of its occurrence in organic compounds.260 This perhaps can be explained by the gene ral difficulty to quantitatively assess polymorphism. An overview of the literature related to the 11 existing polymorphic cocrystals revealed that in four cases concomitant polymorphism was apparent.261 By including (4cypy)2 bphe and 4cyphe bipyete to the total number of structurally characterized polymorphic co-crystals, it b ecomes clear that in nearly 50% of cases concomitancy is present. This is a relativel y high frequency of occurrence, which perhaps could be higher considering that not all polymorphs may have been reported or discovered yet. With the emergence of X-ra y diffraction and the development of more sophisticated methods of analysis, as well as the increasing awareness of the

PAGE 95

74 phenomenon, more discoveries of concomitant polymorphism could perhaps be expected. 3.3. Conclusions The presented series of co-crystallization experiments have demonstrated that the utilization of grinding, solvent-drop gr inding and melting techniques in the cocrystallization procedures offers a viable means for supramolecular synthesis. Such synthetic approaches can be advantageous from the perspective of green chemistry principles. Additionally, the so lvent-drop grinding approach showed to be useful in addressing the issue of stoichiometry and pol ymorph control in co-crystals. Although the role of a solvent in the nucleation process is still not entirely understood, the results obtained based upon solvent-drop grinding suggest that it is generally possible to find experimental conditions, under which a specific co-crystal form exists.187 In the total number of co-crystals, (1 3 presented herein and 1487 archived the CSD), 23 have been found to exhibit polymorphi c behavior (2 presented herein and 21 archived in the CSD).185 Thus, the percentage occurrenc e of polymorphic co-crystals is 1.5%, as compared to the 1.7% of occu rrence of polymorphic single component compounds (Table 1.2).183 It can be therefore concluded that the extent of polymorphism in both classes of compounds is comparab le. It should be noted, however, that the information related to polymorphism in co-c rystals is limited (23 polymorphic co-crystals vs. 1600 polymorphic single component compo unds) and that further investigations conducted on a broader range of co-crystal s would enhance the significance of the presented results. Furthermore, drawing a m eaningful conclusion from such studies is

PAGE 96

75 complicated by the fact that the absence of a polymorphic behavior in a given co-crystal at given experimental conditi ons does not mean that polymorphism can not be observed at other conditions. From this perspective, assessing the frequency of concomitant polymorphism is also complicat ed, although based on the availabl e data (albeit limited) it could be determined that at least ca. 50% of known co-crystal polymorphs can exist concomitantly. This result may be relevant from a commercial perspective, as concomitant crystallizations need to be avoide d because they can lead to materials that do not meet prescribed norms.181 In this context, utilizati on of high-through put screening within a wider range of experimental c onditions (e.g. range of solvent systems, temperature variations) offers a suitable mean s in the identification of polymorphs. If the persistency of supramolecular hete rosynthons within a set of polymorphs holds true over a broader range of co-crystal s it may become of impor tance in the context of polymorphism control. In particular, the variety of factors that influence polymorphic behaviors of organic compounds (supramolecu lar synthon, conformational, and crystal packing variations) could be reduced by eliminating the possibility for molecular selfassembly through different supramolecular s ynthons. Advancements in the context of understanding and controlling polymorphism is of high relevance from the perspective of crystal engineering, where the control over the design and prep aration of desired crystal structures is fundamental. Polymorphism is al so of utmost importance in the context of preparation of solid forms of APIs and the opportunity to reduce its extent co-crystal formulations would represent a particular ly attractive approach.

PAGE 97

76 3.4. Experimental 3.4.1. Syntheses Co-crystallization via grinding: Stoichiometric amounts of the starting materials were ground with a mortar and pestle for ca. 20 minutes. Co-crystallization via solvent-drop grinding: Stoichiometric amounts of the starting materials were ground with a mortar and pestle for ca. 4 minutes with the addition of seven solvents (10 L per 50 mg of co-crystal): cyclohexane, toluene, chloroform, ethyl acetate, me thanol, DMSO, and water. Co-crystallization via melting: Stoichiometric amounts of the starting materials were heated until melt and the mixture was left to crystallize at ambient conditions. Cocrystals 11 and 12 could not be obtained by this proce dure due to decomposition of 4,4biphenol and sublimation of phloroglucinol upon heating. Co-crystallization via solution evaporation. Co-crystal 11a and 11b: (4-cyanopyridine)24,4-biphenol, form I and form II. A solution of 4-cyanopyridine ( 0.040 g, 0.38 mmol) and 4,4-biphenol (0.036 g, 0.19 mmol) in 2 mL of 1:1 methanol and ethyl acetate was allowed to evaporate slowly at ambient conditions. Yellow crystals of tw o distinct morphologies, irregular hexagonal plates and parallelepiped pl ates (total: 0.056 g, 0.14 mmol, 74 %), appeared within four days. Co-crystal 13a and 13b: 4-cyanophenoltrans-1,2-bis(4-pyridyl)ethylene, form I

PAGE 98

77 and form II. To 4-cyanophenol (0.026 g, 0.22 mmol) was added trans-1,2-bis(4pyridyl)ethylene (0.040 g, 0.22 mmol) and the mixture was dissolved in 2 mL of methanol. After 14 days colorless crystal line product (total: 0.029 g, 0.096 mmol, 88%), mp = 153-156 C was observed. Single crystal X-ray diffraction analysis performed on several selected crystals revealed the presence of two concomitant polymorphs. 3.4.2. Single Crystal X-ray Crystallography Suitable crystals of the polymorphs studied herein were examined under a microscope and suitable single crystals were selected for X-ray analysis. Data were collected on a BrukerAXS SMART APEX CC D diffractometer with monochromatized Mo K radiation ( = 0.71073 ). Data were collected at 298 K. Lattice parameters were determined from least square analysis, and reflection data were integrated using the program SAINT. Lorentz and polarization corrections were applied for diffracted reflections. In addition, the data was corrected for absorption using SADABS.248 Structures were solved by direct methods and refined by full matrix least squares based on F2 using SHELXTL.249 All non-hydrogen atoms were refined with anisotropic displacement parameters. All H-atom s bonded to carbon atoms were placed geometrically and refined with an isotropic displacement parameter fixed at 1.2 times Uq of C atoms. OH protons were located from Fourier difference map inspection and refined isotropically with thermal paramete rs based upon the corresponding O atom (U(H)=1.2Uq(O)). Crystallographic data for 11b 13a and 13b are presented in Table 3.1 and selected hydrogen bond distances are list ed in Table 3.2. Crysta llographic data and

PAGE 99

78 selected hydrogen bond distances of 11a were included in the e xperimental section of chapter 2). Table 3.1. Crystallographic data and struct ure refinement parameters for co-crystals 11b, 13a, and 13b 11b 13a 13b Chemical formula (C6H4N2)2 C12H10O2C7H5NO C12H10N2C7H5NO C12H10N2 Formula .wt. 394.42 301.34 301.34 Crystal systemMonoclinicMonoclinicTriclinic Space group P 21/ n Cc P 1 a () 12.787(2) 18.719(8) 7.616(2) b () 7.2838(9) 7.602(3) 10.003(3) c () 44.445(6) 22.09(1) 22.497(6) () 90 90 87.342(5) () 90.895 90.762(9) 81.024(6) () 90 90 69.737(6) Volume (3) 4139.1(9) 3144(2) 1588.1(8) Dcalc ( g cm-3)1.266 1.273 1.260 Z 8 8 4 range 0.92-25.002.18-27.000.92-22.50 Nref./Npara. 7284/541 5352/417 4046/415 T (K) 298 298 298 R1 0.0510 0.0505 0.0502 wR2 0.1654 0.1390 0.1291 GOF 0.812 1.058 0.791 Abs coef. 0.083 0.081 0.080

PAGE 100

79Table 3.2. Geometrical parameters of supramol ecular heterosynthon I present in co-crystals 11b 13a, and 13b Hydrogen bond d () D () () O HNarom 1.72 2.838(3) 173.7 O HNarom 1.69 2.766(3) 164.2 O HNarom 1.56 2.693(3) 163.2 11b O HNarom 1.79 2.789(4) 172.3 O HNarom1.73 2.721(3) 160.9 13a O HNarom1.72 2.669(4) 161.3 O HNarom 1.52 2.689(4) 164.9 13b O HNarom 1.52 2.717(4) 172.0

PAGE 101

80 Chapter 4 Robustness of Supramolecular Het erosynthons: 2-Aminopyridinium Carboxylate 4.1. Focus In the previous sections, it was illustra ted that the understand ing of the interplay of intermolecular interactions in the solid st ate can be facilitated by studies of model cocrystals comprised of components preselec ted for their chemical and geometrical attributes. Thus far the presented investigatio ns have been related to 1-point recognition supramolecular synthons,262 such us hydroxylpyri dine, hydroxylcyano, and hydroxylhydroxyl. It is generally accepted that the probability of formation of a supramolecular motif increases proportionally to the number of involved hydrogen bonds.39 In this context, carboxylic acids and 2-aminopyrdine s represent a suitable pair of moieties that engage in a 2-point recognition supramolecular synthon262 with high probability of formation, as determined by the pioneering work of Allen et al based upon extensive CSD studies.65 Carboxylic acids play a significant role in crystal engineering because they are self-complementary and they can form supramolecular heterosythons with a wide range of other functional groups.80,85,86,91-95,137 In fact, a CSD search reveals that there are over 7,500 crystal structures of carboxylic acids. Ho wever, only ca. 25% of total structures exhibit homosynthon IV (Figure 4.1), whereas the rema ining 75% of compounds exhibit

PAGE 102

81 a variety of supramolecular heterosynthons. 2-aminopyridines and their derivatives such as 2-aminopyrimidines and melamines, wh ich are often encount ered in biological systems, are likewise capable of forming either supramolecular homosynthon V (Figure 4.1)263 or supramolecular heterosynthons.264,265 IV V Figure 4.1. Examples of supramolecular homosynthons: a carboxylic acid homosynthon IV and a 2-aminopyridine homosynthon V In this section, the presented resear ch is focused on the ability of 2aminopyridines and carboxylic acids to form a reliable supramolecular heterosynthon, with an ultimate view to employing this supramolecular heterosynthon in a rational design of new multiple-component compounds th at also contain structurally more complex APIs. 4.2. Results and Discussion 2-aminopyridine and 2-amino-5-methylpyrid ine were used in this study. They were reacted with a series of monoand di carboxylic acids (Scheme 4.1) to form the following compounds: 2-aminopyridinium 4-aminobenzoate, 14 ; 2-aminopyridinium

PAGE 103

82 isophthalate, 15 ; bis (2-aminopyridinium) terephthalate, 16 ; 2-amino-5-methylpyridinium benzoate, 17 ; bis (2-amino-5-methylpyridinium) 5-tertbutylisophthalate, 18 ; bis (2-amino5-methylpyridinium) terephthalate, 19 ; bis (2-amino-5-methylpyridinium) 2,6napthalenedicarboxylate, 20 ; 2-amino-5-methylpyridinium adipate adipic acid, 21 ; and 2amino-5-methylpyridinium 2,5-thiophenedica rboxylate 2,5-thiophenedicarboxylic acid, 22 N NH2 N NH2 C H3 COOH COOH NH2 COOH COOH COOH HOOC S COOH HOOC COOH COOH COOH COOH t-Bu COOH HOOC 2-aminopyridine 2-amino-5-methylpyridine benzoic acid 4-aminobenzoic acid terephthalic acid 2,6-naphthalene dicarboxylic acid 2,5-thiophenedicarboxylic acid isophthalic acid 5-t-butylisopthalic acid adipic acid Scheme 4.1. Molecular structures of components present in complexes 14-22 4.2.1. CSD Analysis There are several possible tw o-point recognition supramol ecular synthons that can be formed between 2-aminopyridines and carboxylic acids: supramolecular homosynthon IV ; a 2-aminopyridine supramolecular homosynthon V ; a 2-aminopyridine-carboxylic acid supramolecular heterosynthon VI ; the ionic form of VI i.e. supramolecular

PAGE 104

83 heterosynthon VII (Scheme 4.2). As will become clear, both statistical and experimental data indicate that the supramolecular heterosynthons VI or VII are favored over the related supramolecular homosynthons IV and V N N H H N N H H H O O R O O R H VI VII Scheme 4.2. Supramolecular heterosynthons that can be formed between carboxylic acids and 2aminopyridines: 2-aminopyridine-carboxylic acid supramolecular heterosynthon VI and 2aminopyridinium-carboxylate supramolecular heterosynthon VII A CSD study of compounds that contai n at least one 2-aminopyridine and one carboxylic acid moiety was conducted in orde r to determine the occurrence of 2-point molecular recognition supramolecula r heterosynthon, also coded as R2 4(8) graph-set.66,75 It should be noted that ambiguity can arise wi th regards to the posit ion of carboxylic acid proton in such complexes. However, the char acter of the interac tion does not influence whether or not a supramolecular heterosynt hon occurs. The proton position was therefore omitted from the search parameters, and the resulting data covers both neutral and charge-assisted hydrogen bonds. The percenta ge occurrence and hydrogen bond distances of supramolecular synthons IV-VII are presented in Table 4.1.

PAGE 105

84 Table 4.1. Percentage occurrence, distance ranges, and average distance for supramolecular synthons IV-VII Number of entries Distance range [] Mean ( ) [] Both groups 123 Synthon IV 0 OO 2.50-3.00 Synthon V 40 ( 33% ) NN 2.90-3.25 3.03(6) N(py)O 2.50-2.85 2.66(6) Synthon VI or VII 95 ( 77% ) N(am)O 2.70-3.10 2.87(8) The CSD contains 123 crystal structur es with both 2-aminopyridine and carboxylic acid groups.266 In order to determine appropr iate ranges for defining contact limits, distance distribution plots were gene rated. Based on visual inspection of the resulting histograms, the lower and higher cut offs for hydrogen bonds were determined. The histograms (Figur e 4.2) reveal that VI or VII exhibit ranges of 2.50 2.85 (average 2.66(6) ) for N(py) O and 2.70 3.10 (average of 2.87(8) ) for N(am)O, respectively. Based upon these limits, IV is not found in any of the 123 compounds, while 40/123 structures (33 %) exhibit the 2-aminopyridine supramolecular homosynthon V 95/123 structures (77%) exhib it supramolecular heterosynthons VI or VII It should be noted that 12 structures that exhibit VI or VII also contain V due to the presence of multiple 2-aminopyridine moieties. In the remaining 27/123 structures that do not exhibit VI or VII the carboxylic acid functionality form s hydrogen bonds with other competitive proton donors or acceptors such as amines amides, imidazoles, water molecules and chloride ions. The same search was perf ormed in the absence of other strong donors

PAGE 106

85 and/or acceptors that can compete with ei ther aminopyridine or carboxylic acid moieties, e.g. alcohols, 1 and 2 amides, 1 and 2 sulfonamides, imidazoles, carbonyls, nitriles, nitro-compounds, phosphine oxides, chloride ions, bromide ions and water molecules. The number of structures containing both 2-aminopyridine and car boxylic acid moieties is thereby reduced from 123 to 34. In this subset, the percenta ge occurrence of VI or VII increased to 97% (33/34), although 7/33 com pounds also contained the 2-aminopyridine supramolecular homosynthon V These statistics suggest that supramolecular heterosynthons VI or VII are robust even in the presen ce of competing hydrogen bonding moieties. This observation is tested ba sed upon study of the new compounds presented herein. 0 5 10 15 20 25 302.2 2.28 2.36 2.44 2.52 2.6 2.68 2.76 2.84 2.92 3 3.08 3.16 3.24 3.32 3.4 3.48N(py)---O distance [A]number of entries a) 0 2 4 6 8 10 12 14 162.2 2.28 2.36 2.44 2.52 2.6 2.68 2.76 2.84 2.92 3 3.08 3.16 3.24 3.32 3.4 3.48N(am)---O distance [A]number of entries b) c) 0 1 2 3 4 5 6 7 82.22 2.3 2.38 2.46 2.54 2.62 2.7 2.78 2.86 2.94 3.02 3.1 3.18 3.26 3.34 3.42 3.5N---N distance [] number of entries Figure 4.2. Histograms of contacts for crystal structures containing both 2-aminopyridine and carboxylic acid moieties: a) N(py)O contacts in supramolecular heterosynthon VI or VII, b) N(am)O contacts in supramolecular heterosynthon VI or VII, c) N(am)N(py) contacts in supramolecular homosynthon V

PAGE 107

86 4.2.2. Structural Features of Neutral and Ionic 2-AminopyridineCarboxylic Acid Interaction The ionic nature of the supramolecula r heterosynthon present in compounds 1422 was confirmed by spectroscopy, proton location, and structural parameters of ancillary groups. It is well known that the geometrical features of neutral carboxylic group are different from those of a carboxylate anion.267 The scatter plot for CO vs. C=O bond distances in carboxylic acids and carboxylate an ions is presented in Figure 4.3 (only good quality neutral carboxylic acid structures containing ordered, error free and nonpolymeric organic compounds with 3D coordina tes determined and R < 5% were chosen for the analysis). 1827 carboxylic acid crystal structures reveal that CO distances average is 1.31(2) whereas C=O distances average is 1.21(2) On the other hand, the scatter plot of the 1696 compounds containing at least one carboxylat e moiety indicates that the CO distances average is 1.25(2) a) 1.1 1.15 1.2 1.25 1.3 1.35 1.4 1.45 1.11.151.21.251.31.351.41.45 C=O distance []C-O distance [] b) 1.1 1.15 1.2 1.25 1.3 1.35 1.4 1.45 1.11.151.21.251.31.351.41.45 C-O distance []C-O distance [] Figure 4.3. Scatter plot of carbon-oxygen bond lengths in: a) neutral carboxylic acids, b) carboxylate anions

PAGE 108

87 The CNC angle in pyridines is known to be sensitive to protonation,222,223,268,269 and the cationic form exhibits higher values than that of the corresponding neutral molecules. A graphical repres entation of the CNC angle distribution in both protonated and unprotonated 2-aminopyridines is pr esented in Figure 4.4. Histograms were generated from good quality crystal struct ures (only ordered, error free and nonpolymeric organic compounds with 3D coordina tes determined and R < 5%). In order to distinguish protonated 2-aminopyridines fr om neutral 2-aminopyridines specific restrictions were applied dur ing the CSD searches. For ne utral 2-aminopyridines, the aromatic nitrogen was defined to be uncha rged and the number of bonded atoms was set to 2. In the case of protonated 2-aminopyridines, hydrogen atoms were placed on the aromatic nitrogen, the charge was set to +1 and the coordination number was set to 3. The average CNC angle encountered in 213 neutral 2-aminopyridines is 116(2) In comparison, the set of 127 cationic 2-aminopyrid ines exhibits a higher CNC angle with an average value of 121(2) a) 0 5 10 15 20 25 30 35 105108110113115118120123125128130 C-N-C angle [] numebr of entries b) 0 5 10 15 20 25 30 105108110113115118120123125128130 C-N-C angle [] number of entries Figure 4.4. Histograms that present distri bution of the CNC angle in a) neutral 2aminopyridines, and b) protonated 2-aminopyridines

PAGE 109

88 4.2.3. Crystal Structure Descriptions The crystal structure of 2-aminopyridinium 4-aminobenzoate, 14 reveals the expected 1:1 cation:anion supramolecula r complex, sustained by supramolecular heterosynthon VII (Figure 4.5). The hydrogen atom of the primary amine moiety involved in the formation of the R2 2(8) supramolecular heterosynthon is assigned as syn oriented, and the exterior hydrogen atom as antioriented. This terminology is used herein. In addition to IR spectroscopic evid ence, the presence of heterosynthon VII is supported by structural data: the CO bond distances of the carboxylate group are 1.273(2) and 1.256(2) ; the CNC angle within the 2-aminopyr idinium cation is 122.3 The hydrogen bond distances of supramolecular heterosynthon VII, 2.611(2) and 2.814(2) for N+ (py)Oand N(am)O-, respectively, are within the excepted ranges. The carboxylate group of the 4-ami nobenzoate is oriented at 19.2 with respect to the benzene ring plane and the di hedral angle between the 2aminopyridinium cation and the carboxylate group is 15.1 However, the 1:1 supramolecu lar adduct is almost planar (6.8 ). The anti oriented NH of the amine of the 2-ami nopyridinium cation is involved in an additional N(am)Ohydrogen bond with an oxygen atom of the adjacent carboxylate (D: 2.800(2) ), thereby genera ting a 1D chain of supramolecular heterosynthons along the c axis (Figure 4.6).

PAGE 110

89 Figure 4.5. Supramolecular interactions in 2-aminopyridinium 4-aminobenzoate, 14 Figure 4.6. Crystal structure of 2-aminopyridinium 4-aminobenzoate, 14 1D hydrogen bonded chains are interconnected via a benzoate amine NHOinteractions to form 2D corrugated supramolecular sheet Within the chain of supramolecular heterosynt hons the supramolecular adducts align in a zigzag fashion. The angle between core pl anes parallel to the interactions OANACA NAOA in heterosynthon VIIA and OBNBCBNBOB in adjacent heterosynthon VIIB is 101.6 (Figure 4.7). This pattern has b een called the sha llow glide motif,270 and is also observed in crystal packing of primary amides.83,271 The aminopyridiniumcarboxylate heterosynthons VII are related by a glide plane a nd are also inclined with A nti Syn 1D hydrogen bonded chain of supramolecular heterosynthons

PAGE 111

90 respect to each other. A similar networking pattern occurs in several of the compounds reported herein. Figure 4.7. The angle between core pl anes parallel to the interactions OANACANAOA and OBNBCBNBOB It is interesting to note th at supramolecular heterosynthon VII occurs even in the presence of a functional group with hydrogen bonding capabilit y, a primary amine. The NH2 of the 4-aminobenzoate anion acts as a hydrogen bond donor to one of the lone pairs of the carboxylate from an adjacent chain (NO-: 3.087(2) ), thereby connecting the chains in the direction of the a axis. The overall crystal packing of 14 can therefore be described as a corrugated 2D network resu lting from interconnected 1D hydrogen-bonded chains of supramolecular heterosynthons (Figure 4.6). Isophthalic acid possesses two carboxylic groups that are amenable to the formation of VI or VII but a 1:1 complex is formed in 2-aminopyridinium isophthalate, 15 since the acid molecule only undergoes single deprotonation. The carboxylate and aminopyridinium moieties form the expected R2 2(8) supramolecular heterosynthon VII O1 Heterosynthon VIIA N1 C1 O1 N1 N2 O2 Heterosynthon VIIB C2 N2 O2

PAGE 112

91 The CO bond distances in the carboxylate moiety are 1.266(1) and 1.262(1) and the CNC angle of the 2-aminopyridinium is 122.5 The CO bond lengths in the acid moiety are 1.210(1) and 1.330(2) Th e hydrogen bond distances of supramolecular heterosynthon VII are 2.703(1) and 2.836 (1) for N+ (py)Oand N(am)O-, respectively. The anti oriented NH of the amine group interacts with another carboxylate (NO-: 2.942(1) ), thereby bridging the adjacent supramolecular heterosynthons. The carboxylic acid group is involved in a charge-assisted O-HO(OO-: 2.623(2) ) interaction with a neighbor ing carboxylate group (Figure 4.8). Figure 4.8. Hydrogen bonding interactio ns in 2-aminopyridinium isophthalate, 15 The components of 15 form almost planar (dihedral angle is 5.8 ) adducts and the angle between core planes parallel to the interactions OANACANAOA in heterosynthon VIIA and OBNBCBNBOB in heterosynthon VIIB is 69.6 The existence of both carboxylic and carboxylate groups in 15 was confirmed by IR spectroscopy which revealed absorption bands at 1556 cm-1 and 1378 cm-1 (corresponding to carboxylate

PAGE 113

92 ions) and 1682 cm-1 and 1261 cm-1 (corresponding to C=O and CO stretches, respectively). Compound 16 bis (2-aminopyridinium terephthalate ), is sustained by threecomponent aggregates consisting of two ami nopyridinium cations a nd one terephthalate anion (Figure 4.9). Figure 4.9. Supramolecular interactions in bis (2-aminopyridinium) terephthalate, 16 (b) Figure 4.10. Crystal structure of bis (2-aminopyridinium ) terephthalate, 16 The hydrogen bonds within supramolecular heterosynthon VII are 2.624(2) and 2.797(2) for N+ (py)Oand N(am)O-, respectively. The CO bond distances in the

PAGE 114

93 carboxylate group (1.271(2) and 1.244(2) ) and CNC angle in the 2aminopyridinium (122.8 ) support the proposed ionic character of 16 The anti oriented NH of the amine group hydrogen bonds to the carboxylate of an adjacent supramolecular adduct (D: 2.809(2) ) and the dihedral angle between core planes of neighbouring heterosynthons is 97.9 A chain of supramolecular heterosynthons is thereby formed and it is connected to anothe r chain through the terephthalate anion. The overall hydrogen-bonding pattern in 16 can therefore be descri bed as a corrugated 2D network consisting of interconnected trimer ic supramolecular adducts aligning nearly perpendicularly with respect to each other (Figure 4.10). 2-amino-5-methylpyridinium benzoate, 17 consists of 1:1 supramolecular adducts sustained by supramolecular heterosynthon VII (Figure 4.11). Protonation occurs on the aromatic nitrogen of 2-amino5-met hylpyridine as evidenced by the CNC angle of 122.8 and CO distances of 1.271(2) and 1 .242(2) Both benzoate and 2-amino5-methylpyridinium ions are flat (the maximum deviati ons from the plane are 0.95 and 0.71 for benzoate and aminopyridinium, respec tively) but their plan es are twisted at 9.9 Supramolecular heterosynthons VII are connected in one di rection as the result of an additional hydrogen bond involving the anti -oriented NH of the amine (D: 2.846(2) ). The dihedral angle between the plan es parallel to the interactions, OANACA NAOA in heterosynthon VIIA and OBNBCBNBOB in adjacent heterosynthon VIIB is 102.5

PAGE 115

94 Figure 4.11. Supramolecular adducts in 2-amino-5-methylpyridinium benzoate, 17, are sustained via charge-assisted hetero synthons VII that form 1D chains Figure 4.12. Crystal structure of 2-amino-5-methylpyridinium benzoate, 17 viewed along the c axis (down the intercalating 1D chains wh ich are colored green, red, blue and gray) Several 2-aminopyridinium monocarboxyl ate supramolecular compounds have been already reported in the litera ture, e.g. 2-aminopyridinium benzoate272 and 2aminopyridinium salicylate,273 2-aminopyridinium butynoate.274 Their crystal structures are also sustained by 1D H-bonded chains. The N+ (py)Oand N(am)Ohydrogen bonds in 17 and the three related compounds are similar: 2.688 and 2.801 in 17 ; 2.699 and 2.868 in 2-aminopyridinium benzoate; 2.699 and 2.845 in 2-aminopyridinium salicylate; 2.670 and 2.889 in 2-aminopyr idinium butynoate. These four compounds exhibit similar crystal packing in which the 1D chains intercalate w ith each other along the b axis (Figure 4.12). One dimensional chains of supramolecular heterosynthons also occur in 2-aminopyridinium-sulfonates.275 Similarly, the aminopyridinium-sulfonate

PAGE 116

95 supramolecular heterosynthon exhibits charge -assisted character with proton transfer occurring to the aromatic nitrogen of the 2-aminopyridine moiety. In bis (2-amino-5-methylpyridinium) 5-tertbutylisophthalate, 18 2-amino-5methylpyridinium ions interact with both car boxylate groups of the 5-t-butylisophtalate anions via supramolecular heterosynthon VII The 2:1 adducts extend along the c axis via two centrosymmetric hydrogen bonds formed between the anti oriented NH moiety of the amine groups and adjacent carboxylates (D: 2.914(2) ), Figure 4.13. The inversion center between two adjacent heterosynthons l eads to the formation of a four component supramolecular unit consisting of three cyclic supramolecular heterosynthons (graph-set notation: R2 2(8), R2 4(8), R2 2(8)). This type of motif has also been observed in related aminopyrimidinium-carboxylate salts276,277 and several carboxylic acid-amide cocrystals.86,87 The CO bond distances are 1.258(2) 1.261(2) and 1.261(2) 1.262(2) within the carboxylate groups and the CNC angles in the 2-aminopyridinium cations are 122.7 and 122.5 The carboxylate groups are twisted from the plane of the aromatic moiety at 20.0 and 6.2 respectively. The intermol ecular hydrogen bonds of the heterosynthons VII are within anticipated ranges : 2.623(2) and 2.619(2) for N(py)O, and 2.788(2) and for 2.799(2) for N(am)O-.

PAGE 117

96 Figure 4.13. Hydrogen bonding in bis (2-amino-5-methylpyridinium) 5tertbutylisophthalate, 18 The crystal structure of bis (2-amino-5-methylpyridinium) terephthalate, 19 compares closely to that of 16 2-amino-5-methylpyridinium ca tions interact with both carboxylate moieties via charge-assist ed supramolecular heterosynthons VII thereby affording 2:1 supramolecular adducts. The adduc ts are further interconnected into a chain of supramolecular heterosynthons via N(am)Ohydrogen bonds formed between the anti oriented NH of amine groups and adjacent carboxylates (Figure 4.14). The CO bond lengths are 1.242(4) and 1.272(4) and the CNC angle in the 2-amino-5-methylpyridinium residue is 123.3 The torsion angle of the carboxylate group with respect to the aromatic mo iety of terephthalate anion is 18.6 (vs. 17.4 in 16 ) and the angle between core planes parall el to the supramolecular heterosynthons, OANACANAOA in VIIA and OBNBCBNBOB in adjacent supramolecular heterosynthon VIIB is 100.8 (vs. 97.9 in 16 ). The hydrogen bond distances in supramolecular heterosynthon VII are 2.679(4) and 2.812(4) for N+ (py)Oand

PAGE 118

97 N(am)O, respectively. The anti oriented NH of the amine group hydrogen bonds with the carboxylate of the adjacent heterosynt hon (D: 2.800(2) ). All three H-bond distances are within excepted ranges and they corre spond closely to the di stances exhibited by 16. Figure 4.14. Crystal structure of bis (2-amino-5methylpyridinium) terephthalate, 19 1D hydrogen bonded chains are cross-linked via NHOinteraction to form 2D sheets Therefore, compounds 16 and 19 are related in terms of their composition, hydrogen bond motifs and packing modes. The differen ce in the molecular structure of the components in 16 and 19 (a methyl substituent in the aminopyridinium residue) is not big enough to disrupt the 2D corr ugated packing mode. Indeed, 16 and 19 are isostructural, crystallizing in space group P 21/ n with similar unit cell dimensions: a =5.1991(8) b =14.606(2) c =11.190(2) =95.094(3) and a =5.627(2) b =14.480(6) c =11.351(4) =99.622(7) for 16 and 19 respectively, and a unit cell similarity index ( )278 of 0.013. Bis (2-amino-5-methylpyridinium) 2,6-napthalenedicarboxylate, 20 consists of three-component adducts sustained by supramolecular heterosynthon VII The CO bond

PAGE 119

98 distances are 1.241(3) and 1.277(3) and CNC angle is 122.5 The carboxylate functionalities ar e twisted at 36.7 with respect to the aromatic core of the anion and the angle between planes parall el to the interactions OANACANAOA in A and OBNB CBNBOB in adjacent supramolecular heterosynthon B is 60.9 The hydrogen bonds within VII are 2.631(3) and 2.897(3) for N+ (py)Oand N(am)Orespectively. The anti oriented NH of the amine group hydr ogen bonds with the carboxylate of the adjacent supramolecular heterosynthon (D : 2.837(3) ). The crystal packing of 20 is remarkably similar to 16 and 19 (Figure 4.15). Figure 4.15. Crystal packing of bis (2-amino-5-methylpyridinium) 2,6napthalenedicarboxylate, 20 1D hydrogen bonded chains are cross-linked via NHOhydrogen bonds to form 2D sheets The asymmetric unit of bis (2-amino-5-methylpyridinium) adipate adipic acid, 21 consists of one 2-amino-5-methylpyridinium cat ion, half an adipate anion and half a free adipic acid. The crystal structure reveal s formation of supramolecular heterosynthon VII

PAGE 120

99 between the 2-aminopyridinium cation and both sides of the adipate anion. The ions further self-assemble via centrosymmetric hydrogen bonds between the anti oriented N Hs of the amine groups and adjacent carboxylates (D: 2815(2) ), Figure 4.16. The fourcomponent core is extended into two dimensions via charge-assisted O-HOhydrogen bonds between neutral adipic acid molecu les and adjacent carboxylate anions (OO-: 2.557(1) ). The hydrogen bonds of VII are 2.719(2) and 2.816(2) for N+ (py)Oand N(am)Orespectively. The four-component supramol ecular adducts are nearly flat, with the maximum deviation from planarity being 3.8 The crystal packing of 21 can be described as 2D infinite sheets parallel to th e plane with an inter-p lanar distance of ca. 3.3 The presence of both COOH and COOgroups is supported by structural data and IR spectroscopy. The CO bond distances in the carboxylate groups are 1.243(2) and 1.280(2) whereas those in the acid are 1.316( 2) and 1.217(2) The CNC angle in the aminopyridinium residue is 122.9(1) IR spectroscopy reveal s a strong band at 1681 cm-1 and moderate band at 1265 cm-1 corresponding to C=O and CO stretches, respectively. The bands assigned to COOstretches are observed near 1550 cm-1 and 1395 cm-1. The CSD contains several compounds th at exhibit a 2-amin opyridinium cation, dicarboxylate anions, and dicarboxyli c acids. One of these compounds, 2aminopyridinium adipate adipic acid dihydrate,279 exhibits similar crystal packing to that of 21 (Figure 4.17).

PAGE 121

100 Figure 4.16. Crystal structures of a) 2-amino-5-methylpyridinium adipate adipic acid, 21 (b) Figure 4.17. Crystal structure of 2-aminopyridin ium adipate adipic acid dihydrate. Herein, the voids between adipic acid molecule s are occupied by water molecules In the dihydrate, water molecules occupy the voids between two adjacent adipic acids. The acid molecules can therefore form bridges between adjacent four-component cores. Unexpectedly, the pr esence of methyl group in 21 does not lead to changes in crystal packing. The si milarity in packing be tween the dihydrate and 21 might result from the fact that the methyl groups of 2amino-5-methylpyridine occupy the space of the

PAGE 122

101 two water molecules. Both compounds crystall ize in P-1 and their unit cell dimensions are: a =5.070(7) b =7.208(1) c =18.388(3) =88.468(2) = 85.015(2) = 72.373(2) for 21 and a =5.019(7) b =7.369(1) c =18.025(3) =86.481(2) = 88.999 (2) = 72.231(2) for the dihydrate. The isostructurality descriptor, is 0.078. The structures of 2-aminopyridiniu m succinate succinic acid (2:1:1)280 and 2aminopyridinium fumarate fumaric acid281 exhibit different supramolecular networks since the components are non-planar. The asymmetric unit of b is(2-amino-5-picolinium) 2,5-thiophenedicarboxylate 2,5-thiophenedicarboxylic acid, 22 consists of one 2-amino-5-methylpyridinium cation, half a 2,5-thiophenedicarboxylate anion and half a 2,5-thiophened icarboxylic acid. Although the acid molecule possesses two functional groups capable of two-point recognition supramolecular heterosynthon VII is absent (Figure 4.18). Instead, the carboxylate group is involved in three 1-point supramolecular hete rosynthons: one with the protonated nitrogen at om of 2-amino-5-methylpyrid inium (D: 2.686(2) ), a second with the adjacent carboxylic acid molecu le (D: 2.545(1) ), and a third with the anti oriented NH of the amine moiety of anot her 2-amino-5-methylpyridinium residue (D: 2.806(2) ). The CO bond distances ar e 1.251(2) 1.267(2) and 1.219(2) 1.318(2) for the carboxylate and carboxylic acid moiety, respectively. The CNC angle of the aminopyridinium is 123.3(2) The IR spectrum suppor ts the existence of both ionic and neutral functional groups in 9 as it exhibits absorption bands at 1673 cm-1 and 1237 cm-1 for C=O and CO, respectively and a COOasymmetric stretch is present at 1627 cm-1. Compound 9 is the only example presented herein that does not exhibit

PAGE 123

102 supramolecular heterosynthon VII (a) (b) Figure 4.18. Crystal structure of b is(2-amino-5-picolinium) 2,5thiophenedicarboxylate 2,5thiophenedicarboxylic acid, 22 : a) the supramolecular adduct reveals that the expected supramolecular heterosynthon VII is absent; b) a view of the chain that is generated by three 1-point supramolecular heterosynthons The interacting components are not represented stoichiometrically for clarity In summary, the CSD survey and mo del compound studies reported herein indicate that supramolecular heterosynthons VI and/or VII will occur reliably when 2aminopyridine and carboxylic acid moieties are present in the same compound. The CSD analysis suggests 77% probability of the supramolecular heterosynthon, which means that from an empirical perspectiv e it is strongly favored over th e related carboxylic acid, IV or 2-aminopyridine V supramolecular homosynthons. In the absence of other competing functionalities the probabili ty of the occurrence of VI or VII is ca. 97%. The statistical reliability of these supramolecular he terosynthons was confirmed by compounds 14-22 in which 8/9 were found to exhib it supramolecular heterosynthon VII The supramolecular chemistry of supramolecular heterosynthon VII is also predictable to a certain extent. It is see Figure 12 b

PAGE 124

103 capable of further self-assembly into chains and sheets and it is noteworthy that very similar crystal packing was observed in several compounds despite the presence of methyl groups. That supramolecular heterosynthon VI was not observed in 14-22 does not mean it is not relevant. Indeed, several co -crystal structures have been reported that are based upon 2-aminopyrimidine.127,264 Whether or not a salt or co-crystal forms seems to be related to the ancillary groups that are bonded to the 2-aminopyridine moiety. For example, all 2-aminopyridines that interact w ith carboxylic acids appear to be in the form of salts whereas 2-aminopyrimidines form either neutral or ionic supramolecular heterosynthons and melamines tend to exist as monoprotonated salts. Taking into consideration the strength of the three bases (2-aminopyrimidine < melamine < 2aminopyridine), one can speculate that the pKa value of the 2-aminopyridine moiety influences whether a supramolecular heterosynth on is neutral or ionic. Prediction of what will happen for a given carboxylic acid is fu rther complicated by factors such as the pKa value of the acid, the nature of s ubstituents and the fact that pKa values are determined in solution. 4.3. Conclusions In conclusion, the charge assisted 2-aminopyridinium-carboxylate or neutral 2aminopyridine-carboxylic acid supramolecular heterosynthons occur in 77% of the compounds in which these two functional groups are present. This is a high level of probability when one considers that such we ll-known 2-point supramolecular synthons as carboxylic acid dimers only occur in ca. 25% of the compounds in which carboxylic

PAGE 125

104 acids are present because of competing supr amolecular heterosynthons. This level of predictability makes the 2-aminopyridinecarboxylic acid supramolecular synthon particularly suitable for crysta l engineering of networks, sa lts and/or co-crystals. The modification of the physicochemical propertie s that occurs with salt or co-crystal formation is relevant to the pharmaceutical industry, where formulation of APIs for optimal solubility, bioavailability or stab ility of drugs are of great importance. 4.4. Experimental 4.4.1. Syntheses All reagents used to synthesize 14-22 were purchased from Aldrich. Compounds 14-22 were prepared by dissolving stoichiometr ic amounts of starting materials in an appropriate solvent. Crystals suitable fo r single crystal X-ray diffractometry were obtained by slow evaporation of th e solvent under ambient conditions. Compound 14: 2-aminopyridinium 4-aminobenzoate A solution of 2aminopyridine (0.010 g, 0.11 mmol) and 4-ami nobenzoic acid (0.015 g, 0.11 mmol) in 2 mL of ethanol was allowed to evaporate slowly at room temperature. Colorless crystals of 14 (0.025 g, 0.065 mmol, 59%), mp=152-153 C, were obtained after 7 days. Compound 15: 2-aminopyridinium isophthalate. The crystallization of 2aminopyridine (0.029 g, 0.31 mmol) with isophthalic acid (0.051 g, 0.31 mmol) in 2 mL of ethanol afforded colorless crystals of 15 (0.042 g, 0.16 mmol, 52%), mp=198-201 C, within 7 days.

PAGE 126

105 Compound 16: bis(2-aminopy ridinium) terephthalate. A solution of 2aminopyridine (0.060 g, 0.64 mmol) and terepht halic acid (0.053 g, 0.32 mmol) in 2 mL of methanol evaporated slowly at room temperature. Colorless crystals of 16 (0.056 g, 0.16 mmol, 50%), mp=294 C dec., suitable for X-ray crystallography appeared within 6 days. Compound 17: 2-amino-5-me thylpyridinium benzoate. A solution of 2-amino-5methylpyridine (0.019 g, 0.18 mmol) and benz oic acid (0.22 g, 0.18 mmol) in 2 mL of ethanol was left undisturbed to evaporate slowly under ambient conditions. Colorless crystals of 17 (0.033 g, 0.014 mmol, 77%), mp=140-141 C, were obtained within 7 days. Compound 18: bis(2-amino-5-methyl pyridinium) 5-tertbutylisophthalate. Colorless crystals of 18 mp=174-175 C, were obtained from the reaction of 2-amino-5methylpyridine (0.020 g, 0.18 mmol) with 5tertbutylisophthalic acid (0.021 g, 0.090 mmol) in 2 mL of ethanol. The solution was left to evaporate slowly at room temperature and yielded 0.025 g (0.057 mmol, 63%) of the product within 8 days. Compound 19: bis(2-amino-5-meth ylpyridinium) terephthalate. Compound 19 was formed via reaction of 2-amino5-methylpyridine (0.030 g, 0.28 mmol) with terephthalic acid (0.023 g, 0.14 mmol) in 2 mL of methanol. Colorless crystals (0.038 g, 0.099 mmol, 72%), mp=250 C dec., appeared within 6 days. Compound 20: bis(2-amino-5-methylpy ridinium) 2,6-napthalenedicarboxylate. A solution of 2-amino-5-methyl pyridine (0.030 g, 0.28 mmol) and 2,6napthalenedicarboxylic acid (0.030 g, 0.14 mmol) in 2 mL of dimethyl formamide, was left undisturbed to evaporate slowly at ambient conditions. Colorless crystals of 20 (0.036

PAGE 127

106 g, 0.083 mmol, 60%), mp=390 C dec., were obtained within 9 days. Compound 21: bis(2-amino-5-methyl pyridinium) adipate adipic acid. 2-amino5-methylpyridine (0.030 g, 0.28 mmol) and adip ic acid (0.041 g, 0.28 mmol) dissolved in 2 mL of ethanol afforded colorless crystals of 21 (0.036 g, 0. 14 mmol, 50%), mp=151152 C, within 7 days. Compound 22: bis(2-amino-5-methylpy ridinium) 2,5-thiophenedicarboxylate 2,5-thiophenedicarboxylic acid. Compound 22 was obtained via reaction of 2-amino-5methylpyridine (0.030 g, 0.28 mmol) and 2,5thiophenedicarboxylic acid (0.048 g, 0.28 mmol) in 2 mL of ethanol. Colorless crystals (0.062 g, 0.22 mmol, 80%), mp=221-222 C, appeared within 7 days. All compounds were analyzed by infrared spectroscopy using a Nicolet Avatar 320 FTIR instrument. The purity of bulk samples was confirmed by X-ray powder diffraction analysis conducted on a Riga ku Miniflex Diffractometer using Cu K ( = 1.540562 ), 30 kV, 15 mA. The data were co llected over an angular range of 3 to 40 2 in continuous scan mode using a step size of 0.02 2 and a scan speed of 2.0/min. The syntheses of compounds 14 22 were also accomplished via solvent-drop grinding using the same solvents as those in the solution crystallizations. Stoichiometric amounts of starting materials were processed for 4 minutes in a ball mill. The IR and XPD spectra of the products obtained from so lvent-drop grindings, matched those of the products obtained from slow evaporation.

PAGE 128

107 4.4.2. Single Crystal X-ray Crystallography Compounds 14-22 were examined under a microsc ope and suitable si ngle crystals were selected for X-ray analysis. Data were collected on a BrukerAXS SMART APEX CCD diffractometer with monochromatized Mo K radiation ( = 0.71073 ) connected to KRYO-FLEX low temperature device. Data for 14-22 were collected at 100 K. Lattice parameters were determined from least s quare analysis, and reflection data were integrated using the program SAINT. Lorent z and polarization corre ctions were applied for diffracted reflections. In addition, th e data was corrected for absorption using SADABS.248 Structures were solved by direct me thods and refined by full matrix least squares based on F2 using SHELXTL.249 All non-hydrogen atoms were refined with anisotropic displacement parameters. All H-atoms bonded to carbon atoms, except methyl groups, were placed geometrically a nd refined with an isotropic displacement parameter fixed at 1.2 times Uq of the atoms to which they were attached. N or O bonded protons, as well as H-atoms of methyl groups were located from Fourier difference map and refined isotropically based upon the corresponding N, O or C atom (U(H)=1.2Uq(N, O)). Crystallographic data for 14-22 are presented in Table 4.2, whereas selected hydrogen bond distances ar e listed in Table 4.3.

PAGE 129

108Table 4.2. Crystallographic data and struct ure refinement parameters for compounds 14-22 14 15 16 17 18 19 20 21 22 Chemical formula C5H7N2 C7H6NO2 C5H7N2 C8H5N2O4(C5H7N2)2 C8H4O4 C6H9N2 C7H5O2 (C6H9N2)2 C12H12O4 (C6H9N2)2 C8H4O4 (C6H9N2)2 C12H6O2 (C6H9N2)2 C6H8O4 C6H10O4 (C6H9N2)2 C6H4O4S C6H2O4S Formula .wt. 231.25 260.25 354.36 230.26 438.52 382.42 432.47 508.57 560.59 Crystal system orthorhombicmonoclinicmonoclinicmonoclinictriclinic monoclinicmonoclinictriclinic orthorhombic space group P ca21 P 21/c P 21/n P 21/c P -1 P 21/n P 21/n P -1 P nma a () 18.932(2) 12.562(1)5.1991(8)9.678(3) 8.5420(1) 5.627(2) 5.834(2) 5.0700(7) 14.735(1) b () 5.6894(7) 8.0768(9)14.606(2)10.838(4) 11.050 (2) 14.480(6) 16.260(7) 7.2081(10)18.752(1) c () 10.415(1) 12.236(1)11.190(2)11.836(4) 13. 096(2) 11.351(4) 10.985(4) 18.388(3) 9.0740(6) () 90 90 90 90 101.086(2) 90 90 88.468(2) 90 () 90 109.831(2)95.094(3)111.210(6)103.007(2) 99.622(7) 92.141(7) 85.015(2) 90 () 90 90 90 90 98.983(2) 90 90 72.373(2) 90 volume (3) 1121.8(2) 1167.8(2)846.4(2) 1157.3(7) 1155.9(3) 911.9(6) 1041.3(7) 638.0(2) 2507.3(3) Dcalc ( g cm-3) 1.369 1.48 1.39 1.322 1.26 1.393 1.379 1.324 1.485 Z 4 4 2 4 2 2 2 1 4 range 2.15-24.95 1.72-25.012.30-27.082.26-25.151.64-25.37 2.30-25.112.24-24.701.11-26.392.17-27.07 Nref./Npara. 1954/155 2056/1721855/1182067/154 4593/289 1615/127 1771/145 2532/163 2846/175 T (K) 100 100 100 100 100 100 100 100 100 R1 0.0284 0.0331 0.0434 0.0553 0.0582 0.0743 0.0536 0.0433 0.0339 wR2 0.0721 0.0959 0.1091 0.1625 0.1681 0.1826 0.1424 0.1324 0.0907 GOF 1.087 0.966 1.029 1.058 1.056 1.103 1.047 1.121 1.064 abs coef. 0.096 0.112 0.101 0.091 0.087 0.099 0.096 0.100 0.270

PAGE 130

109Table 4.3. Geometrical parameters of select ed intermolecular interactions present in compounds 14-22 Interactiona d() D () (deg) N+ H(py)O1.74 2.611(2) 171.7 N HsO1.99 2.814(2) 175.2 N HaO2.01 2.800 (2) 162.7 14 NHO2.30 3.087(2) 162.2 O HO1.76 2.623(1) 173.4 N+ H(py)O1.84 2.703(1) 166.0 N HsO1.98 2.836 (1) 169.6 15 N HaO2.10 2.942(1) 167.0 N+ H(py)O1.75 2.624(2) 172.3 N HsO1.85 2.797(2) 178.8 16 N HaO1.92 2.809(2) 179.3 N+ H(py)O1.81 2.688(2) 172.4 N HsO1.87 2.801(2) 168.2 17 N HaO1.92 2.846(2) 165.4 N+ H(py)O1.77 2.623(2) 161.9 N HsO1.95 2.788(2) 177.2 N HaO2.20 2.914(2) 144.1 N HsO1.74 2.619(2) 172.2 N HaO1.91 2.799(2) 176.6 18 N HaO2.18 2.940(2) 146.5 N+ H(py)O1.80 2.679(4) 174.5 N HsO1.92 2.812(4) 172.7 19 N HaO1.86 2.800(4) 168.3 N+ H(py)O1.75 2.631(3) 175.2 N HsO2.02 2.897(3) 170.0 20 N HaO1.95 2.837(3) 170.6 O HO1.68 2.557(1) 173.9 N+ H(py)O1.84 2.719(2) 176.3 N HsO1.96 2.816(2) 178.9 21 N HaO2.07 2.815(2) 145.0 O HO1.67 2.545(1) 172.1 N+ H(py)O1.84 2.686(2) 159.4 22 N HaO1.93 2.806(2) 174.0 a The synand antiNH groups of the 2-am inopyridiniums are referred with subscript s and a, respectively.

PAGE 131

110 Chapter 5 Crystal Engineering of Pharmaceuticals 5.1. Focus Thus far it has been demonstrated that the hydroxylaromatic nitrogen ( I ) and 2aminopyridinium-carboxylate ( VII ) supramolecular heterosynthons are reliable interactions and that they persist regardless of the method of prepar ation of the model cocrystals and salts. However, it should be noted that the model compounds have been prepared using relatively simple and rigid molecules. The conclusions drawn from the presented studies would gain more significan ce if the acquired observations held true over more complex multiple-component systems. In this respect, the chemical nature of drug molecules related to their biological activity is determined by their multiple hydrogen bonding sites, which also makes them suitable for crystal engineering studies. The research presented in this sectio n is a strategic extension of the model compound studies and will demonstrate how the supramolecular heterosynthons I and VII can be exploited via crystal engineering of two drug molecules, bicalutamide and indomethacin, both of which contain multiple hydrogen bonding sites. Furthermore, the viability of solid-state co-crystallization and susceptibility toward polymorphism or solvate formation of the obtained AP I co-crystals will be addressed.

PAGE 132

111 5.2. Results and Discussion 5.2.1. CSD Analysis and Literature Overview The examination of the 1487 co-crystals arch ived in the CSD, has revealed that only ca. 5% of the whole dataset contain API molecules, and these include: barbital (32 entries),282 sulfonamide drugs (14),283 phenathiazines (8),284 carbamazapine (5),285 theophyllines (5),286 caffeine (6),287 flurbiprofen (2),288 ibuprofen (1),289 itraconazole (1),290 diphenylhydantoin (1),291 and trimethoprim (1).292 Whereas some of the examples tend to be the result of sere ndipity, perhaps the first crysta l engineered API co-crystals appeared from the extensive research of Whitesides et al concerning supramolecular assemblies sustained by 3-poi nt recognition hydrogen bonds262 formed in co-crystals of melamine derivatives and barbital, a centr al nervous system depressant, Figure 5.1.125,293296 Although the studies were not originally oriented toward pharmaceutical applications, the resulting series of co-crystals delineat ed the enormous potential represented by the APIs with respect to dive rsity of the co-crystal co mpositions and the inherent modification of their physi cochemical properties. Figure 5.1. Co-crystal of barbital and N,N bis (4-bromophenyl)melamine, JICTUK10, sustained by 3-point recognition supramolecular heterosynthon

PAGE 133

112 Due to the growing interest in the subj ect, several examples of designed binary co-crystals of APIs have app eared in the recent literature.61,80,86,87,132,154,195,204,208 Crystal engineering approach to APIs based upon rational utilizati on of reliable supramolecular heterosynthons have been exemplified by a series of co-crystals that involve carbamazepine, CBZ, an anti-epileptic drug.61,86,87 The primary amide supramolecular homosynthon present in pure CBZ crystal structures,297-300 Figure 5.2, is replaced by carboxylic acidamide heterosynthon, thr ough an introduction of a carboxylic acid component. As an example, CBZ aspirin co-crystal is pr esented in Figure 5.3a. The successful strategic approach founded on u tlilization of the same supramolecular heterosynthon has been employed in co-cryst allization of several carboxylic acids with another amide-containing drug, pirace tam, a nervous system stimulant176,177,301 The crystal structure of piracetam gentisic acid,208 sustained by carboxylic acidamide heterosynthon is shown in Figure 5.3b. Figure 5.2. Amide supramolecular homosynthon pr esent the crystal structure of pure CBZ

PAGE 134

113 a) b) Figure 5.3. Carboxylic acidamide supramolecu lar heterosynthon presen t in co-crystals of a) CBZ aspirin and b) piracetam gentisic acid Further examples concerning crystal engi neering of APIs based upon utilization of reliable intermolecular in teractions include co-cryst als sustained by carboxylic acidaromatic nitrogen supr amolecular heterosynthon, as ex emplified by co-crystals of itraconazole,132 ibuprofen and flurbiprofen,80 and caffeine.154 Crystal structures of ibuprofen 4,4-bipyridine is pr esented in Figure 5.4. Figure 5.4. Carboxylic acidpyridine supramolecular heterosynthon present in (ibuprofen)2 4,4-bipyridine co-crystal Rational design of API co-crystals has b een also based upon the utilization of charge-assisted O HCland N HClhydrogen bonds,137,302,303 as demonstrated

PAGE 135

114 recently by Childs et al .204 The study involved successful co-crystallizations of an antidepressant, fluoxetine hydrochloride (Prozac ) with several pharmaceutically acceptable carboxylic acids, of which one, Prozac succinic acid co-crystal, is presented in Figure 5.5. An important trait result ing from this study is that depending on the aqueous solubility of the utilized co-cry stal former, it is possible to fine-tune the dissolution rate the API. In addition, it was observed that the solubili ty of the Prozac succinic acid cocrystal is doubled as compared to the fluoxetine hydrochloride salt. Figure 5.5. Charge-assisted supramolecular heterosynthons are present in Prozac succinic acid co-crystal It should be noted, that due to the lack of a generally accepted definition of a cocrystal, there may be ambigu ity concerning whether or not a compound is a co-crystal, a salt or a solvate. The distinction between a co-crystal and a salt can be especially problematic if X-ray crystallography is th e only method of characterization and the difference between the two extremes is ca. 1 in a hydrogen atom position. For instance, 3-[2-( N,N -dimethylhydrazino)-4-thiazolylmethylthio]N2-

PAGE 136

115 sulfamoylpropionamidine maleic acid (CSD refcode JATMEW), was reported as a neutral complex.304 However, the structural parameters (C O bond lengths and C N C bond angles) suggest the formation of a ma leate anion and a propionamidinium cation, therefore denoting a salt, Figure 5.6a. When s earching for co-crystals, the physical state of the components must also be taken into consideration. For in stance, a molecular complex of mebendazole and propionic acid (SAGQEW),305 shown in Figure 5.6b should be classified as a solvate rather than a co-cry stal, as the propionic ac id exists as a liquid under ambient conditions (mp 21 C). a) b) Figure 5.6. (a) JATMEW, 3-[2-( N,N -dimethylhydrazi no)-4-thiazolylmethylthio]N2sulfamoylpropionamidine maleic acid. Structur al parameters suggest formation of a salt, (b) SAGQEW a propionic acid solvate of mebendazole Additionally, the CSD mining for co-crystal s may be complicated by the database errors; the CSD searches retrieve ionic com pounds despite limiting the searches to neutral compounds. For example, salts EBIBEW PIKLEA, QAWNAD, VAPBAP, VENLUV, etc. are all retrieve d as neutral compounds.306-310 These findings suggest that the identification of co-crystals archived in the CSD should be supported by inspection of the

PAGE 137

116 structural parameters of co-crystal compone nts, and/or revision of the corresponding publications. In summary, design strategi es that target reliable s upramolecular heterosynthons (determined by preceding CSD searches or model compound studies), which can be formed between an API and a co-crystal form er, represents an attractive approach to discovering new crystalline forms of APIs. Considering the profound implications of developing new forms of APIs in the c ontext of both intellectual property and physicochemical properties, it is somewhat surprising that a rational design and generation of API co-crystals has onl y been endeavored in recent years.49 The applicability of a crystal engi neering approach toward gene rating new forms of APIs has been demonstrated by the diverse set of the reported API co-crystals, although some of them are sustained by co-crystal formers that are not pharmaceutically acceptable. However, the limited number of examples indi cates the need for further exploration in order to achieve a better understanding of th e intermolecular forces that influence the formation of API co-crystals, as well as the factors that determine their physicochemical properties. 5.2.2. Bicalutamide Bicalutamide (propanamide, N-[4-cya no-3-(trifluoromet hyl)phenyl]-3-[4fluorophenyl)sulfonyl]2-hydroxy-2-methyl-( )), is a non-steroidal antiandrogen used in the treatment of prostate cancer.311 From a supramolecular perspective, bicalutamide is a relatively complex molecule due to the conf ormational flexibility and the presence of

PAGE 138

117 multiple hydrogen bonding moieties: hydroxyl (O H), 2 amine (N H), carbonyl (C=O), cyano (C N), and sulfonyl (O=S=O), Scheme 5.1. The anticipation of the supramolecular chemistry of this API is therefore complicat ed by both conformational variations and the possibility of existance of various hydrogen bonded synthons. Indeed, the multiplecomplementary nature of bicalutamide is mani fested by the existence of two polymorphic modifications, Figure 5.7.312,313 Form I of bicalutamide is sustained by O HO (D: 3.145(3) ) hydrogen bonds occurring between the hydroxyl and carbonyl moieties. Bicalutamide form II exhibits two primary supramolecular heterosynthons. The first heterosynthon, O HN C (D: 2.905(3) ), occurs between the hydroxyl and cyano moiety and the second heterosynthon, N HO (D: 3.104(2) ), ex ists between the N H of the 2 amide and the sulfonyl moieties. N F F F S N F O O OH O H Scheme 5.1. Molecular structure of bicalutamide To date, no examples of bicalutamide co-cry stals, solvates, or hydrates have been deposited in the CSD. Therefore, bicalutami de represents a suitable candidate for a crystal engineering case study that addresses its feasibility toward rational design of cocrystals, considering the presence of mu ltiple hydrogen bonding f unctionalities in its

PAGE 139

118 molecular structure. a) b) Figure 5.7. Representation of the crystal packin g of bicalutamide in a) form I and b) form II The first step of the crystal engine ering experiment was based upon careful examination of molecular stru cture of bicalutamide to id entify the hydrogen bonding sites capable of forming reliable supramolecular synthons. The next step, however, which involves a CSD analysis of the existing struct ures that are chemically related to the targeted compound to find how they engage in molecular associati on, was not successful. There are no compounds in the database that exhibit exac tly the same set of hydrogen bond donors and acceptors as the set present in bicalutamide (O H, N H, C N, C=O,

PAGE 140

119 and O=S=O). With this viewpoint it become ap parent that the design strategy needed to rely upon a simplified approach, in which one or two moieties would be targeted at a time for their interactions with other functiona l groups. The formation of supramolecular heterosynthons was considered to occur w ith co-crystal formers that possess moieties different than those already present in bical utamide. With the perspective of the model co-crystal studies presented in chapter 2, that delineated the remarkably high reliability of hydroxylaromatic nitrogen (O HNarom) supramolecular heterosynthon I in the presence of cyano moiety, and considering that these two moieties (OH and CN) are present in bicalutamide, the OH moiety wa s selected as the primary target for heterosynthon formation with Narom-based co-crystal former s. Taking into account the complexity of the API, the initial selecti on of co-crystal formers was aimed toward simple Narom-containing heterocycles. Two co-cry stal formers were chosen: 4,4bipyridyl ( bipy ) and trans-1,2-bis(4-pyridyl)ethylene ( bipyete ), and although these molecules are not pharmaceutically acceptabl e, they represent good model candidates from the viewpoint of a crystal engineering strategy. From the co-crystallization experiment s, two co-crystals were obtained: bicalutamide 4,4bipyridyl ( 23 ) and bicalutamide trans-1,2-bis(4-pyridyl)ethylene ( 24 ). In addition, it was observed that co-crystal 24 tends to be solvated in the presence of acetone. The crystal structure of (bicalutamide)2 trans-1,2-bis(4pyridyl)ethane (acetone)2 ( 25 ) is also presented. Crystal structure of bicalutamide bipy 23 reveals discrete 2:2 centrosymmetric supramolecular adducts sustained by the targeted heterosynthon I (O HNarom) and a N

PAGE 141

120 HNarom heterosynthon, Figure 5.8. Th e hydrogen bond distance of O HNarom is 2.759(5) which corresponds to the average length of I observed in the model cocrystals (chapter 2) and other compounds repor ted in the CSD (Table 2.1). The distance of the second hydrogen bond, N HNarom, is 3.499(7) which is relatively long, as compared to a typical N HNarom interaction.314 The contacts distribution for the N HNarom heterosynthon retrieved from the CS D, (Figure 5.9) reveals that the N HNarom interaction occurs in the range of 2.75 3.30 (average of 3.0(1) ). Figure 5.8. 2:2 supramol ecular adducts formed between bica lutamide and 4,4bipyridyl in 23 In this structure the bipy molecule is twisted at 26.14 Although the bipy molecule in co-crystal 6 is flat, the torsion angle observed in 23 corresponds closely to the twist angles of the phenpy molecules in co-crystals 1 and 5 (24.6 and 30.1 respectively). The supramolecular adducts are stabilized by continuous aromatic stacking occurring between the adjacent bipy molecules, along the a axis. The stacked adducts are further connected via weak C HN C interactions and form 2D sheets Figure 5.10. Similar crystal packing, ch aracterized by the columnar arrangement of the

PAGE 142

121 supramolecular adducts, was al so observed in co-crystals 1, 4, 5, 6 and 8 0 5 10 15 20 25 30 35 40 45 50 2.52.62.72.82.933.13.23.33.43.5NHNarom distance []number of entrie s Figure 5.9. Histogram re presenting the NHNarom contact distribution in the crystal structures containing both NH and Narom moieties Co-crystal 23 can also be prepared by solven t-drop grinding. The screen for polymorphs of 23 based upon solvent-drop grinding with cyclohexane, toluene, chloroform, ethyl acetate, methanol, DMSO, and wa ter has not revealed additional forms. Figure 5.10. Crystal structure of bicalutamide 4,4bipyridyl, 23 Crystal structure of bicalutamide bipyete 24 is reminiscent to that of 23 The bicalutamide and bipyete interact via heterosynthon I (D: 2.811(3) ). The adjacent

PAGE 143

122 aggregates are related by a center of i nversion and engage in formation of 2:2 supramolecular adducts via N HNarom (D: 3.119(3) ) interac tion, and are additionally stabilized by stacking occurring between the bipyete molecules (ca. 3.76 ), Figure 5.11. Such supramolecular dimers ar e further translated along the a axis forming 1D columns of continuously stacked 2:2 adducts. The columns are related by translation and interconnected by weak C HN C interaction, thereby generating 2D layers, Figure 5.12. Similar crystal packing was observed in model co-crystals 6 and 8 Figure 5.11. 2:2 supramolecular adducts formed between bicalutamide and t-1,2bis (4pyridyl)ethylene in 24 Co-crystal 24 can also be prepared by solv ent-drop grinding. The screen for polymorphs of 24 based upon solvent-drop grinding with: cyclohexane, toluene, chloroform, ethyl acetate, methanol, DMSO, and water has not revealed any additional polymorphs. However, it was observed that when co-crystallization of bicalutamide and bipyete is carried out in acetone, a solvate of the co-crystal ( 25 ) results. As revealed by XPD and FTIR analysis, 25 can be reproducibly obtained from acetone-drop grinding.

PAGE 144

123 Figure 5.12. Crystal structure of bicalutamide t-1,2bis (4-pyridyl)ethylene, 24 The crystal structure of 25 reveals that the crystallization of bicalutamide and bipyete from acetone ( acet ) affords new composition of the co-crystal, (bicalutamide)2 bipyete (acet)2, and the components assemble into five-member supramolecular aggregates. Two bicalutamide molecules interact with one bipyete molecule via the heterosynthon I (D: 2.739(6) ) and with two acet molecules via a N HO(acet) heterosynthon (D: 3.235(7) ), Figur e 5.13. In this structure, one bipyete molecule is replaced by two acet molecules. In effect, the continuous stacking observed in 24 is disrupted by the insertion of the acet molecules in between the 2:1 adducts formed by bicalutamide and bipyete molecules, Figure 5.14.

PAGE 145

124 Figure 5.13. 5-member supramolecular aggre gate formed between bicalutamide, t-1,2bis (4pyridyl)ethylene and acetone in 25 Figure 5.14. Crystal structure of (bicalutamide)2 t-1,2bis (4-pyridyl)ethylene (acetone)2, 25 The melting points of 23-25 (Table 5.1.) are in between the melting points of the corresponding constituents. Interestingly, the melting points of 24 and 25 are very close (161-163 C vs 163-164 C) despite the difference in the compositions. The solvent loss in 25 is observed at 98 C (crystals become opaque and the phase change is confirmed by DSC) and the remaining co-crystal exhibits 2:1 stoichiometry of bicalutamide and bipyete which differ from the 1:1 stoichiometry of bicalutamide and bipyete in 24 No correlation was observed when the melting points of 23-25 were compared to the melting

PAGE 146

125 points of the related model co-crystals. Th e melting points distribution in co-crystals 1-13 and 23-25 highlight the difficulty in correlati ng the effect of the composition and structural variations with the resu lting change of thermal behavior. Table 5.1. Comparison of the melting points of co-crystals 23-25 and the corresponding components Compound Mp of co-crystal [ C] Mp of Bicalutamide [ C] Mp of component 2 [ C] 23 157-159 191-192 110 114 (bipy) 24 161-163 191-192 150 153 (bipyete) 25 164-165 191-192 150 153 (bipyete) 94 (acet) 5.2.3. Indomethacin Indomethacin (1-(p-chlorobenzoyl)-5-met hoxy-2-methylindole-3-acetic acid) is known for its analgesic and anti -inflammatory activity. It possesses three functional groups that are capable of forming h ydrogen bonds: carboxylic acid (COOH), methoxy (C O CH3), and carbonyl (C=O), Scheme 5.2. Indo methacin exists in three polymorphic modifications: a monoclin ic and two triclinic.315-317 The triclinic structures are sustained by supramolecular homosynthon VII whereas the monoclinic form, exhibits a carbonylcarboxylic acid supramolecular heterosynthon, in addition to VII Figure 5.15.

PAGE 147

126 Cl O N O OH O Scheme 5.2. Molecular structure of indomethacin a) b) Figure 5.15. Representation of th e crystal packing of bicalutamide in a) two triclinic forms and b) monoclinic form In addition to the three polymorphs, only tw o solvates of indomethacin have been so far structurally characterized.318,319 The crystal structures of the reported methanol (CSD refcode: BANMUZ) and tert-butanol (CSD refcode: BANMOT) solvates are sustained by a cyclic hydrogen bonding motif fo rmed due to an insertion of two hydroxyl moieties of the alcohol molecule in between the carboxylic acid dimer, Figure 5.16. Up to date, no co-crystals or salts of indomethaci n have been deposited in the CSD.

PAGE 148

127 Figure 5.16. Methanol solvate of indomethacin, BANMUZ. Similar hydrogen bonding motif exists in tert-butanol solvate of indomethacin, BANMOT Indomethacin was chosen for this study b ecause it represents a suitable candidate for the extension of the model compound series (chapter 4) toward cr ystal engineering of APIs based upon supramolecular heterosynthons I and VII Indomethacin possesses several hydrogen bonding sites, of which one is carboxylic acid, the target moiety. Considering the high probability of formation of heterosynthon VII demonstrated based upon simple model molecules, it is anticipat ed that the COOH moiety of indomethacin will also be utilized in th e 2-point recognition supramolecular heterosynthon with 2aminopyridines. 2-amino-5-methylpyridine was se lected as a simple counterpart for cocrystallization with the relatively comp lex indomethacin. Although not pharmaceutically acceptable, 2-amino-5-methylpyridine is a good candidate from the perspective of its chemical and geometrical features that matc h with the features of the COOH functional group, as proved in the preceding studies i nvolving model compounds (chapter 4). Crystallization of indomethacin with 2-amino-5-methylpyridine afforded compound 26 which is sustained by supramolecular heterosynthon VII (Figure 5.17).

PAGE 149

128 The crystal structure of compound 26 exhibits the expected 1:1 stoichiometry. The hydrogen bond distances in the charge-a ssisted supramolecu lar heterosynthon VII are within expected ranges: 2.728(4) and 2.790(5) for N+ (py)Oand and N(am)O-, respectively. The CO bond lengths are 1.228( 5) 1.258(5) and the CNC angle in the 2-aminopyridinium cation is 123.1 The angle between core planes parallel to the interactions OANACANAOA in synthon VIIA and OBNBCBNBOB in synthon VIIB is 92.4 (Figure 4.7). The anti oriented NH of the amine moiety forms H-bond with carboxylate (D: 2.843(4) ) of the ne ighboring anion thereby bridging adjacent supramolecular heterosynthons. Similar hydrog en bonding motif has also been seen in compounds 14, 16, 17, 19, and 20 The reproducibility of 26 was confirmed by solvent-drop grinding involving seven solvents: cyclohexane, toluene, chloroform, ethyl acetate, methanol, DMSO, and water no additional forms. The susceptibility of 26 toward solvent-drop grinding suggests that solid-state methods can be efficiently u tilized in screening fo r crystalline salts of APIs.152 Figure 5.17. Supramolecular heterosynthon VII exhibited in 2-aminopyridinium indomethacin, 26

PAGE 150

129 In summary, the application of heterosynthon VII has been demonstrated in a successful crystal engineering experiment of indomethacin, a carboxylic acid API, with 2-amino-5-methylpyridine salt former. Conversily, VII can be targeted in crystal engineering of 2-aminopyridinecontaining APIs and carboxylic acid co-crystal formers. For instance, trimethoprim, an antibacterial ag ent, forms organic salts with a range of carboxylic acids: formic acid,320 acetic acid,321 trifluoroacetic acid,322 malonic acid,323 glutaric acid,277 benzoic acid,324 3-chlorobenzoic acid,325 2-nitrobenzoic acid,326 terephthalic acid,327 through the 2-point r ecognition interaction VII Figure 5.18. Figure 5.18. Crystal structure of trimethoprim benzoate is sustained by the 2aminopyridine-carboxylate supa rmolecular heterosynthon VII 5.3. Conclusions In summary, the use of two r obust supramolecular heterosynthons I and VII has been demonstrated in crystal engi neering of two APIs: bicalutamide and

PAGE 151

130 indomethacin. From a supramolecular perspectiv e, the molecular structures of both drugs are relatively complex due to their multiple hydrogen bonding sites. The successful cocrystallization of bicalutamide with pyridines confirms the robustness of I not only in the presence of the C N moiety, but also suggests that I can persist in th e presence of other hydrogen bonding groups: carbonyl (C=O), 2 amine (N H), and sulfonyl (O=S=O). Similarly, the applicat ion of heterosynthon VII toward modification of crystal structure of indomethacin also proved that VII is a reliable interacti on even in the presence of carbonyl (C=O) and ether (C O C) moieties. The reliability of I and VII was also determined based upon reproducibility of 23, 24, and 26 in solvent-drop grinding experiments. In addition, that polymorphism was not observed in 23, 24, and 26 can be an important observation considering that both bicalutamide and indomethacin are polymorphic in their pure states. It should be noted that in several cases, API co-crystals comprised by polymorphic components have not yet exhibited polymorphism,208,211 which may have an important implications fr om the viewpoint of polymorphism control. With the perspective of the necessity of better understanding and control of crystalline forms of APIs, pharmaceutical co-c rystals appear to represent a significant class of compounds. It is important to note th at the crystal engineering approach leaves the molecular structure of an API intact, wh ile diverse range of new compositions with modified physicochemical properties is readily accessible. Furtherm ore, they may offer more opportunities than, for example, salts, which are formed only when the API possesses moieties that are sufficiently basi c/acidic for protonation/deprotonation.In cocrystallization, molecules po ssessing wider range of hydrogen bonding moieties can be

PAGE 152

131 targeted. In particular, utilization of GRAS (generally regarded as safe)328 compounds, food additives or even sub-therapeutics, such us aspirin or acetaminophen, as the cocrystal formers is feasible. Nevertheless, there still remains a need fo r further exploration of API co-crystals in order to address some fundamental issues For instance, while the role of API salt forms in the optimization of API propertie s, e.g. solubility, has been established,191,193,194 the role of API co-crystals in this context has been addressed only in a few cases.132,204,211 Another question could be related to whether API co-crystals are more or less prone to polymorphism as compared to the pure API. It is not a trivial task to assess the frequency of occurrence of polymorphism in either of the mentioned categories as the absence of polymorphism is not synonymous with its non-e xistence. However, if one considers that co-crystallization is based upon satisfying the molecular recognition sites of a targeted API by matching it with a comple mentary co-crystal former, it is possible to anticipate a decreased tendency of the co-cry stal to exhibit polymorphism as compared to the pure components. The fact that polymorphism seen in the 11 reported co-crystals has not been related to hydrogen bonding variations, coul d support these arguments, however more research needs to be conducted in this context. In conclusion, it is appropriate to hi ghlight that despit e many questions and challenges, API co-crystals represent valu able category of compounds. They offer many opportunities in the context of their viability to rational de sign and large diversity of composition and physicochemical properties, which can be utilized in new API formulations.

PAGE 153

132 5.4. Experimental Bicalutamide was used as received fro m Transform Pharmaceuticals Inc., MA. Indomethacin was used as received from the Department of Pharmaceutical Sciences, College of Pharmacy, Un iversity of Michigan. 5.4.1. Syntheses Co-crystal 23: bicalutamide 4,4bipyridyl. To bicalutamide (0.050 g, 0.115 mmol) was added 4,4-bipyridine (0.018 g, 0.115 mmol). To the solid mixture was added acetone (1 mL) and the solution was left to evaporate at ambient temperature. After 3 days colorless plates of 23 were formed, mp=157-159 C. Co-crystal 24: bicalutamide t-1,2-bis(4-pyridyl)ethylene. To bicalutamide (0.100 g, 0.23 mmol) was added t-1,2-bi s(4-pyridyl)ethene (0.021 g, 0.115 mmol). To the solid mixture was added DMSO (0.5 mL) and the so lution was left to evaporate at ambient temperature. After 3 days colorless plates of 24 were formed, mp=161-163 C. Co-crystal solvate 25: (bicalutamide)2 trans-1,2-bis(4pyridyl)ethane(acetone)2. To bicalutamide (0.100 g, 0.23 mmol) was added t-1,2-bis(4pyridyl)ethene (0.021 g, 0.115 mmol) in 2:1 mola r ratio. To the solid mixture was added 1mL hexane / acetone (1:1) and the solution was left to evaporate at ambient temperature. After 3 days colorless plates of 25 were formed, mp=163-164 C (at 98 C the crystal became opaque). Compound 26: 2-amino-5-meth ylpyridinium indomethacin. A solution of 2-

PAGE 154

133 amino-5-methylpyridine (0.010 g, 0.093 mmol ) and indomethacin (0.033 g, 0.093 mmol) in 2 mL of ethanol was left undisturbed to ev aporate slowly at room temperature. Yellow crystals of 26 mp=146-148 C, appeared after 7 days. Co-crystallization via solvent-drop grinding: Stoichiometric amounts of the starting materials were ground with a mortar and pestle for ca. 4 minutes with the addition of seven solvents (10 L per 50 mg of product): cyclohexane, toluene, chloroform, ethyl acetate, methanol DMSO, and water. In case of 25 acetone was also used. 5.4.2. Single Crystal X-ray Crystallography Compounds 23-26 were examined under a microsc ope and suitable si ngle crystals were selected for X-ray analysis. Data were collected on a BrukerAXS SMART APEX CCD diffractometer with monochromatized Mo K radiation ( = 0.71073 ) connected to KRYO-FLEX low temperature device. Data for 23-25 were collected at 298 K, whereas data for 26 were collected at 100 K. Lattice parameters were determined from least square analysis, and reflection data were integrated usi ng the program SAINT. Lorentz and polarization corrections were a pplied for diffracted reflections. In addition, the data was corrected for absorption using SADABS.248 Structures were solved by direct methods and refined by full matrix least squares based on F2 using SHELXTL.249 All non-hydrogen atoms were refined with anisotr opic displacement parameters. All H-atoms bonded to carbon atoms, except methyl groups, were placed geometrically and refined with an isotropic displacement parameter fixed at 1.2 times Uq of the atoms to which they

PAGE 155

134 were attached. N or O bonded protons, as well as H-atoms of methyl groups, were located from Fourier difference map and refined is otropically based upon the corresponding N, O or C atom (U(H)=1.2Uq(N, O)). Crystallographic data for 23-25 are presented in Table 5.2, whereas selected hydrogen bond distances are listed in Table 5.3. Table 5.2. Crystallographic data and struct ure refinement parameters for compounds 23-26 23 24 25 26 Chemical formula C18H14F4N2O4S C10H8N2 C18H14F4N2O4S C12H10N2 C18H14F4N2O4S C12H10N2 C3H6O C6H9N2 C19H15NO2Cl Formula wt. 1173.11 612.59 1159.12 465.92 Crystal system triclinic tric linic triclinic orthorhombic space group P-1 P-1 P-1 P ca21 a () 8.298(2) 8.274(3) 11.214(2) 12.564(2) b () 11.434(3) 10.172(3) 11.724(2) 11.338(1) c () 15.299(4) 18.601(7) 12.198(3) 30.501(4) () 74.506(4) 87.521(7) 73.721(4) 90 () 75.852(5) 78.256(6) 83.866(4) 90 ( ) 79.963(4) 71.067(7) 66.461(4) 90 volume (3) 1347.0(6) 1449.4(9) 1411.4(5) 4344.7(1) Dcalc ( g cm-3) 1.446 1.404 1.364 1.425 Z 1 2 1 8 range 1.41-25.00 1.12-26.73 1.74 to 25.00 1.34-28.29 N ref./Npara. 4680/ 370 5985/ 388 4786 / 363 10289/595 T (K) 298 298 298 100 R1 0.0724 0.0554 0.1004 0.0718 wR2 0.2208 0.1498 0.3434 0.1837 GOF 0.786 1.028 1.011 1.038 abs coef. 0.190 0.180 0.182 0.215

PAGE 156

135Table 5.3. Geometrical parameters of su pramolecular heterosynthons present in compounds 23-26 Interactiona d() D () (deg) O HNarom 1.68 2.759(5) 163.1 23 N HNarom 2.55 3.499(7) 157.4 O HNarom 1.93 2.811(3) 157.5 24 N HNarom 2.25 3.119(3) 157.7 O HNarom 1.81 2.739(6) 154.1 25 N HOacet 2.33 3.235(7) 147.9 N+ H(py)O1.87 2.728(4) 163.3 N HsO1.82 2.790(5) 165.1 26 N HaO1.76 2.843(4) 169.3

PAGE 157

136 Chapter 6 Summary and Future Directions 6.1. Summary The goal of the presented work is to illustrate the application of crystal engineering principles toward a generation of multiple-component organic crystalline materials with pre-determined composition and intermolecular interactions in a rational and controllable manner. Specifically, the sy stematic studies invol ving preparation and structural analysis of hydrogen bonded mu lti-component compounds have afforded a basis for a better understanding and control of the supramolecular synthons in the solid state. In particular, the know ledge acquired from the investig ation of organic co-crystals and salts sustained by simple components has le d to the delineation of the reliability of two supramolecular synthons, namely hydroxylaromatic nitrogen ( I ) and 2aminopyridinecarboxylate ( VII ) heterosynthons. The determination of the reliable reccurence of I and VII subsequently lead to a developmen t of strategies for the design of multiple-component compounds that involve more complex molecules possessing multiple functional groups, such as APIs. The exploitation of these strategies toward drug molecules that possess several hydrogen bonding sites has proved to be successful in the case of bicalutamide and indometh acin. Both bicalutamide and indomethacin formed co-crystals and a salt, respective ly, with pre-determined composition and predictable intermolecular interactions. Th e presented results can be particularly

PAGE 158

137 attractive, from the viewpoint that the trad itional API forms, such as polymorphs and solvates/hydrates tend to a ppear serendipitously, rather than based upon a rationally designed experiment. In addition, it has been illustrated that organic co-crystals a nd salts exhibit high susceptibility toward solid-state preparati on methods, such us growth form melt, dry grinding, and solvent-drop grinding. In part icular, the later has been confirmed to constitute a reliable technique for a re producible formation of multiple-component compounds. It has been determined that the utilization of an ap propriate solvent can direct co-crystallization toward specific co-cry stal stoichiometries. In particular, DMSOdrop grinding of 3-cyanophenol and trans -1,2bis (4-pyridyl)ethylene in 1:1 ratio affords complete conversion of the starti ng materials into a 3-cyanophenol trans -1,2bis (4pyridyl)ethylene co-crystal, which otherw ise occurs concomitantly with the (3cyanophenol)2 trans -1,2bis (4-pyridyl)ethylene co-crystal when other solvents are used in the grinding or solution evaporation methods. In addition, the mechanochemical approach to supramolecular synt hesis is inherently relevant in the generation of a bulk material according to green chemistry pract ices: directly from starting materials and based upon clean and high yieldi ng procedures. The investigations in the context of polymorphism pr ovided new insights concerning the origin of this phenomenon in co-crystals. Specifically, the structural analysis of the two dimorphi c co-crystals, (4-cyanopyridine)2 4,4-biphenol and 4cyanophenol t-1,2-bis(4-pyridyl)ethylene, revealed the existence of identical hydrogen bonded heterosynthons in both forms of the corresponding co-crystal s. These results

PAGE 159

138 support the observations formulated based upon the analysis of the existing polymorphic co-crystals: polymorphism in co -crystals is related to conf ormational and crystal packing variations rather than su pramolecular synthons. Although th ese conclusions are made from the study of a limited number of exampl es, a generality of this observation over a broader set of co-crystals may have important implications in the context of controlling polymorphism. Furthermore, the use of so lvent-drop grinding to obtain specific polymorphs of co-crystals has been successful, as illustrated based on (4cyanopyridine)2 4,4-biphenol co-crystal. In summary, the presented research has c ontributed to the overall progress of the field of crystal engineering, whose ultimate goal is the understanding of intermolecular interactions and the ability to rationall y design new crystallin e solids for useful applications. Specifically, it has been dem onstrated that crystal engineering of pharmaceuticals is possible with an appropria te understanding of their supramolecular chemistry and the interplay of a supramolecular synthons that can potentially exist when other components are introduced. Consideri ng the dependability of physichochemical properties of APIs on the molecular arrangement in their crystals, the advantage of crystal engineered APIs is inherently related to the control of their physicochemical performance: solubility, bioavail ability, stability, etc. While th e role of salt forms of APIs can be considered as established, the role of co-crystal forms of APIs still remains to be explored and there are many questions and ch allenges that will need to be addressed. These challenges can be linked to scale up pr ocesses, evaluating properties of the bulk co-crystalline material, utilization of automatized methodologies, such as high-

PAGE 160

139 throughput co-crystallization, issues related to regulatory procedures etc. Nevertheless, the need for further exploration of AP Is remains clear because the value and opportunities of a successful deve lopment of co-crystallizatio n strategies of APIs is significant in the context of both drug deve lopment and intellectual property. 6.2. Future Directions In conclusion, several directions for fu ture research in the field of crystal engineering of can be highlighted. Considering the viabilit y of multi-component compounds, co-crystals or organic salts, toward the investigati on of robustness and hierarchie s of various supramolecular synthons, the potential of further studies that focus on an even broader range of hydrogen bonding moieties has become apparent. In partic ular, a systematic investigation of the competition of various hydrogen bonds in the presence of three, four, or even more functional groups should be addressed. In the context of API co-c rystals, the studies should be expanded to utilization of pharmaceutically acceptable co-crystal formers and further evaluation of their physicochemical performance, i.e. solubility, bioavailability, stabilit y, toxicology, etc. In addition, the experimental prot ocols related to API co-cryst allization on a large scale and determination of bulk properties may be important. Since the existence of polymorphism in API co-crystals has attracted attention, exhaustive screen involving an HT approach could provide more insight toward better understanding the phenomenon and evaluating it s frequency. Specifically, addressing

PAGE 161

140 whether or not co-crystals of APIs are more or less prone to polymorphism may lead to important scientific and intellectual propert y implications. Additionally, the observation related to the persistence of supramolecular synthons within a set of polymorphs perhaps requires further investigation. Considering the multi-disciplinary charac ter of crystal engineering, it would be interesting to utilize other classes of co-cry stal formers, for instance, molecules that mimic biologically active com pounds. When coupled with APIs, such biomolecular cocrystals could be of use to study drug-DNA or drug-enzyme interactions in the solid state. The stability of co-crystals in aqueous envi ronment could be evaluated based upon slurry experiments.

PAGE 162

141 Reference List 1. Lehn, J. M. Science 1985 227 849. 2. Lehn, J. M.; Atwood, J. L.; Davies, J. E. D.; MacNicol, D. D.; Vgtle, F. Comprehensive Supramolecular Chemistry ; Pergamon: Oxford, 1996. 3. Vgtle, F. Supramolecular Chemistry ; Wiley, J.: New York, 1991. 4. Lehn, J. M. Supramolecular Chemistry: Concepts and Perspectives ; VCH: Weinheim, 1995. 5. Fischer, E. Ber. Deutsch. Chem. Ges. 1894 27 2985. 6. Ehrlich, P. Studies on Immunity ; Wiley: New York, 1906. 7. Wolf, K. L.; Wolff, R. Angew. Chem. 1949 61 191. 8. Cram, D. J.; Cram, J. M. Science 1974 183 803. 9. Pedersen, C. J. J. Am. Chem. Soc. 1967 89 7017. 10. Dietrich, B. ; Lehn, J. M.; Sauvage, J. P. Tetrahedron Lett. 1969 2889. 11. Zerkowski, J. A.; Seto, C. T. ; Wierda, D. A.; Whitesides, G. M. J. Am. Chem. Soc. 1990 112 9025. 12. Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990 112 5525. 13. Rebek, J. Acc. Chem. Res. 1990 23 399. 14. Whitesides, G. M.; Simanek, E. E.; Ma thias, J. P.; Seto, C. T.; Chin, D. N.; Mammen, M.; Gordon, D. M. Acc. Chem. Res. 1995 28 37. 15. Sijbesma, R. P.; Meijer, E. Curr. Opin. Colloid Interface Sci. 1999 4, 24. 16. Barthelemy, P.; Lee, S.; Grinstaff, M. W. Pure and Appl. Chem. 2005 77 2133. 17. MacGillivray, L. R.; Atwood, J. L. Nature 1997 389 469.

PAGE 163

142 18. Whitesides, G. M.; Mathias, J. P.; Seto, C. T. Science 1991 254 1312. 19. Lindoy, L. F.; Atkinson, I. M. Self-Assembly in Supramolecular Systems ; Royal Society of Chemistry: Cambridge, 2000. 20. Lehn, J. M. Angew. Chem. Int. Ed. 1990 29 1304. 21. Woodward, R. B. Pure Appl. Chem. 1968 17 519. 22. Nicolaou, K. C. Angew. Chem. Int. Ed. 1996 35 589. 23. Maitland, G. C.; Rigby, E. B.; Smith, E. B.; Wakeham, W. A. Intermolecular Forces: Their Origin and Determination ; Oxford University Press: Oxford, 1981. 24. Jeffrey, G. A. An Introduction to Hydrogen Bonding. ; Oxford University Press: Oxford, 1997. 25. Prins, L. J. ; Reinhoudt, D. N.; Timmerman, P. Angew. Chem. Int. Ed. 2001 40 2383. 26. Jeffrey, G. A.; Saenger, W. Hydrogen Bonding in Biological Structures ; Springer: Berlin, 1991. 27. Pauling, L. The Nature of Chemical Bond ; Cornell University Press: New York, 1948. 28. Rotello, V. M.; Viani, E. A.; Desl ongchamps, G.; Murray, B. A.; Rebek, J. J. Am. Chem. Soc. 1993 115 797. 29. Hirschberg J. H. K. K.; Brunsveld, L.; Ramzi, A.; Vekemans Jajm; Sijbesma, R. P.; Meijer, E. W. Nature 2000 407 167. 30. Lehn, J. M.; Mascal, M.; Decian, A.; Fischer, J. J. Chem. Soc. 1990 479. 31. Amaya, T.; Rebek, J. J. Am. Chem. Soc. 2004 126 14149. 32. Biros, S. M.; Ullrich, E. C. ; Hof, F.; Trembleau, L.; Rebek, J. J. Am. Chem. Soc. 2004 126 2870. 33. Karle, I. L.; Ranganathan, D.; Haridas, V. J. Am. Chem. Soc. 1997 119 2777. 34. Zafar, A.; Geib, S. J.; Hamuro, Y.; Carr, A. J.; Hamilton, A. D. Tetrahedron 2000 56 8419. 35. Yang, J.; Fan, E. K.; Geib, S. J.; Hamilton, A. D. J. Am. Chem. Soc. 1993 115 5314.

PAGE 164

143 36. Rebek, J. Angew. Chem., Int. Ed. Engl. 1990 29 245. 37. Murray, T. J.; Zimmerman, S. C. J. Am. Chem. Soc. 1992 114 4010. 38. Fan, E.; Vanarman, S. A.; Kincaid, S.; Hamilton, A. D. J. Am. Chem. Soc. 1993 115 369. 39. Steiner, T. Angew. Chem. Int. Ed. 2002 41 48. 40. Dunitz, J. D. Pure Appl. Chem. 1991 63 177. 41. Desiraju, G. R. Crystal Engineering: the Design of Organic Solids ; Elsevier: Amsterdam, 1989. 42. Braga, D. Chem. Commun. 2003 2751. 43. Moulton, B.; Zaworotko, M. J. Chem. Rev. 2001 101 1629 44. Pepinsky, R. Phys. Rev. 1955 100 971. 45. Schmidt, G. M. Pure Appl. Chem. 1971 27 647. 46. Panunto, T. W.; Urbanczyk-Lipkowska, Z.; Johnson, R.; Etter, M. C. J. Am. Chem. Soc. 1987 109 7786. 47. Russell, V. A.; Evans, C. C.; Li, W. J.; Ward, M. D. Science 1997 276 575. 48. Taylor, L. D.; Warner, J. C. US 5338644 A 19940816 Cont. of US 5,177,262. 49. Almarsson, .; Zaworotko, M. J. Chem. Commun. 2004 1889. 50. Gdanitz, R. J. Curr. Op. Solid State Mater. Sci. 1998 3 414. 51. Gavezzotti, A. Acc. Chem. Res. 1994 27, 309. 52. Ball, P. Nature 1996 381 648. 53. Desiraju, G. R. Nature Mater. 2002 1 77. 54. Dunitz, J. D. Chem. Commun. 2003 545. 55. Lommerse, J. P. M.; Motherwell, W. D. S.; Ammon, H. L.; Dunitz, J. D.; Gavezzotti, A.; Hofmann, D. W. M.; Leusen, F. J. J.; Mooij, W. T. M.; Price, S. L.; Schweizer, B.; Schmidt, M. U.; Van Eijck, B. P.; Verwer, P.; Williams, D. E. Acta Crystallogr. B 2000 56 697. 56. Gavezzotti, A.; Filippini, G. J. Am. Chem. Soc. 1995 117 12299.

PAGE 165

144 57. Motherwell, W. D. S.; Ammon, H. L. ; Dunitz, J. D.; Dzyabchenko, A.; Erk, P.; Gavezzotti, A.; Hofmann, D. W. M.; Leusen, F. J. J.; Lommerse, J. P. M.; Mooij, W. T. M.; Price, S. L.; Scheraga, H.; Schweizer, B.; Schmidt, M. U.; Van Eijck, B. P.; Verwer, P.; Williams, D. E. Acta Crystallogr. B 2002 58 647. 58. Day, G. M.; Motherwell, W. D. S.; A mmon, H. L.; Boerrigter, S. X. M.; Della Valle, R. G.; Venuti, E.; Dzyabchenko, A.; Dunitz, J. D.; Schweizer, B.; Van Eijck, B. P.; Erk, P.; Facelli, J. C.; Bazt erra, V. E.; Ferraro, M. B.; Hofmann, D. W. M.; Leusen, F. J. J.; Liang, C.; Pantelid es, C. C.; Karamertzanis, P. G.; Price, S. L.; Lewis, T. C.; Nowell, H.; Torrisi, A.; Scheraga, H. A.; Arnautova, Y. A.; Schmidt, M. U.; Verwer, P. Acta Crystallogr. B 2005 61 511. 59. Price, S. L. CrystEngComm 2004 6 344. 60. Price, S. L. Adv. Drug Deliv. Rev. 2004 56 301. 61. Vishweshwar, P.; McMahon, J. A.; Oliv eira, M.; Peterson, M. L.; Zaworotko, M. J. J. Am. Chem. Soc. 2005 127 16802. 62. Maddox, J. Nature 1988 335 201. 63. Corey E. J. Pure Appl. Chem. 1967 14 19. 64. Desiraju, G. R. Angew. Chem. Int. Ed. 1995 34 2311. 65. Allen, F. H.; Motherwell, W. D. S.; Raithby, P. R.; Shields, G. P.; Taylor, R. New J. Chem. 1999 23 25. 66. Etter, M. C. Acc. Chem. Res. 1990 23 120. 67. Nangia, A.; Desiraju, G. R. Acta Crystallogr. 1998 A54 934. 68. Gilli, P.; Bertolasi, V.; Ferretti, V.; Gilli, G. J. Am. Chem. Soc. 1994 116 909. 69. Ward, M. D. Chem. Commun. 2005 5838. 70. Bis, J. A.; Zaworotko, M. J. Cryst. Growth Des. 2005 5 1169. 71. Stanley, N.; Sethuraman, V.; Muthiah, P. T.; Luger, P.; Weber, M. Cryst. Growth Des. 2002 2 631. 72. Bertolasi, V.; Gilli, P.; Ferretti, V.; Gilli, G. Acta Crystallogr. B 2001 57 591. 73. Russell, V. A.; Etter, M. C.; Ward, M. D. Chem. Mater. 1994 6 1206. 74. Aakery, C. B.; Hughes, D. P.; Nieuwenhuyzen, M. J. Am. Chem. Soc. 1996 118 10134.

PAGE 166

145 75. Etter, M. C.; MacDonald, J. C.; Bernstein, J. Acta Crystallogr. B 1990 B46 256. 76. Etter, M. C. J. Phys. Chem. 1991 95 4601. 77. Saha, B. K.; Nangia, A.; Jaskolski, M. CrystEngComm 2005 7 355. 78. Aakery, C. B.; Beatty, A. M. Aust. J. Chem. 2001 54 409. 79. Thalladi, V. R.; Goud, B. S.; Hoy, V. J. ; Allen, F. H.; Howard, J. A. K.; Desiraju, G. R. Chem. Commun. 1996 401. 80. Walsh, R. D. B.; Bradner, M. W.; Fleischman, S.; Morales, L. A.; Moulton, B.; Rodrguez-Hornedo, N.; Zaworotko, M. J. Chem. Commun. 2003 186. 81. Leiserowitz, L. Acta Crystallogr. 1976 B32 775. 82. Etter, M. C. J. Am. Chem. Soc. 1982 104 1095. 83. Leiserowitz, L.; Schmidt, G. M. J. J. Chem. Soc. 1969 2372-&. 84. Leiserowitz, L.; Nader, F. Acta Crystallogr. 1977 33 2719. 85. Videnova-Adrabinska, V.; Etter, M. C. J. Chem. Crystallogr. 1995 25 823. 86. Fleischman, S. G.; Kuduva, S. S.; McMahon, J. A.; Moulton, B.; Walsh, R. D. B.; Rodrguez-Hornedo, N.; Zaworotko, M. J. Cryst. Growth Des. 2003 3 909. 87. McMahon, J. A.; Bis, J. A.; Vishweshwar, P.; Shattock, T. R.; McLaughlin, O. L.; Zaworotko, M. J. Z. Kristallogr. 2005 220 340. 88. Vishweshwar, P.; Nangia, A.; Lynch, V. M. CrystEngComm 2003 5 164. 89. Papaefstathiou, G. S.; MacGillivray, L. R. Org. Lett. 2001 3 3835 90. MacGillivray, L. R.; Atwood, J. L. J. Solid State Chem. 2000 152 199. 91. Aakery, C. B.; Beatty, A. M.; Helfrich, B. A. J. Am. Chem. Soc. 2002 124 14425. 92. Vishweshwar, P.; Nangia, A.; Lynch, V. M. J. Org. Chem. 2002 67 556. 93. Bhogala, B. R.; Vishweshwar, P.; Nangia, A. Cryst. Growth Des. 2002 2 325. 94. Bond, A. D. Chem. Commun. 2003 250. 95. Etter, M. C. ; Adsmond, D. A. J. Chem. Soc. 1990 589. 96. Goud, B. S.; Reddy, P. K.; Panneerselvam, K.; Desiraju, G. R. Acta Crystallogr.

PAGE 167

146 1995 C51 683. 97. Desiraju, G. R.; Sarma, J. A. R. P. J. Chem. Soc. 1983 45. 98. Aakery, C. B.; Desper, J.; Helfrich, B. A. CrystEngComm 2004 6 19. 99. Allen, F. H. ; Kennard, O. Chem. Des. Autom. News 1993 8 31. 100. Allen, F. H.; Kennard, O.; Taylor, R. Acc. Chem. Res. 1983 16 146. 101. Allen, F. H.; Taylor, R. Chem. Soc. Rev. 2004 33 463. 102. Allen, F. H.; Motherwell, W. D. S. Acta Crystallogr. 2002 B58 407. 103. Nangia, A. CrystEngComm 2002 4 93. 104. Desiraju, G. R. Acc. Chem. Res. 1991 24 290. 105. A typical example of a drastic differen ce in properties of two crystal forms of chemically identical compound is that of diamond (transparent and hard) and graphite (black and soft). Other exampl es include pharmaceuticals, e.g. ritonavir exhibits varying solubilities (and biov ailability) depending on the polymorpic form used. 106. Desiraju, G. R. CrystEngComm 2003 5 466. 107. Dunitz, J. D. CrystEngComm 2003 5 506. 108. Aakery, C. B.; Salmon, D. J. CrystEngComm 2005 7 439. 109. Ling, A. R. ; Baker, J. L. J. Chem. Soc. 1893 63 1314. 110. Anderson, J. S. Nature 1937 140 583. 111. Buck, J. S. ; Ide, W. S. J. Am. Chem. Soc. 1931 53 2784. 112. Vanniekerk, J. N.; Saunder, D. H. Acta Crystallogr. 1948 1 44. 113. Hall, B.; Devlin, J. P. J. Phys. Chem. 1967 71 465. 114. Pekker, S.; Kovts, .; Oszlnyi G.; Bnyei, G.; Klupp, G.; Bortel, G.; Jalsovszky, I.; Jakab, E.; Borondics, F.; Kamars, K.; Bokor, M.; Kriza, G.; Tompa K.; Faigel, G. Nature Mat. 2005 4 764. 115. Whler F. Annalen 1844 51 153. 116. Sakurai, T. Acta Crystallogr. 1965 19 320.

PAGE 168

147 117. Sakurai, T. Acta Crystallogr. 1968 B 24 403. 118. Watson, J. D.; Crick, F. H. C. Nature 1953 171 737. 119. Pauling, L.; Corey, R. B. Arch. Biochem. Biophys. 1956 65 164. 120. Hoogsteen, K. Acta Crystallogr. 1963 16 907-&. 121. Kim, S. H.; Rich, A. J. Mol. Biol. 1969 42 87-&. 122. Hoogsteen, K. Acta Crystallogr. 1959 12 822. 123. O'Brien, E. J. Acta Crystallogr. 1967 23 92. 124. Huang, C. M.; Leiserowitz, L.; Schmidt, G. M. J. J. Chem. Soc. Perkin Trans. 2 1973 503. 125. Zerkowski, J. A.; MacDonald, J. C.; Se to, C. T.; Wierda, D. A.; Whitesides, G. M. J. Am. Chem. Soc. 1994 116 2382. 126. Kane, J. J. ; Liao, R. F.; Lauher, J. W.; Fowler, F. W. J. Am. Chem. Soc. 1995 117 12003. 127. Shan, N.; Batchelor, E.; Jones, W. Tetrahedron Lett. 2002 43 8721. 128. Donohue, J. J. Phys. Chem. 1952 56 502. 129. Etter, M. C.; MacDonald, J. C. ref. 26 in Etter, M. C.; J. Phys. Chem. ,1991, 95 4601. 130. Etter, M. C.; Adsmond, D. ref. 28 in Etter, M. C.; J. Phys. Chem. ,1991, 95 4601. 131. Aakery, C. B.; Desper, J.; Urbina, J. F. Chem. Commun. 2005 2820. 132. Remenar, J. F.; Morissette, S. L.; Pe terson, M. L.; Moulton, B.; MacPhee, J. M.; Guzman, H. R.; Almarsson, J. Am. Chem. Soc. 2003 125 8456. 133. Vishweshwar, P.; Nangia, A.; Lynch, V. M. Cryst. Growth Des. 2003 3 783. 134. Reddy, L. S.; Nangia, A.; Lynch, V. M. Cryst. Growth Des. 2004 4 89. 135. Jetti, R. K. R.; Xue, F.; Mak, T. C. W.; Nangia, A. J. Chem. Soc. Perkin Trans. 2 2000 6 1223. 136. Melendez, R. E.; Hamilton, A. D. Top. Curr. Chem. 1998 198 97 137. Steiner, T. Acta Crystallogr. 2001 B57 103.

PAGE 169

148 138. Aakery, C. B.; Beatty, A. M.; Helfrich, B. A. Angew. Chem. Int. Ed. 2001 40 3240. 139. CSD (ConQuest 1.7, August 2005) re veals 10,709 crystal structures which contain hydroxyl moieties (one of the mo st ubiquitous funtional group with ca. 22,000 structures) in the presence of ethe rs (the most ubiqui tous functional group with ca. 36,846 entries) and 1,434 crystal stru ctures in the presence of pyridines. The three moieties are present simliltanously in 451crystal struct ues. The analysis of the 451 entries reveals that th e occurence of the hydroxyl-pyridine heterosynthon is 9 times higher (125/14) than the hydroxyl-ether heterosynthon. 140. CSD reveals that the number of entrie s that contain hydroxyl, pyridine, and amine is only 36, of which the majority additionally contain other hydrogen bond donor and acceptors. 141. Trask, A. V.; Jones, W. Top. Curr. Chem. 2005 254 41. 142. Anastas, P. ; Warner, J. C. Green Chemistry: Theory and Practice ; Oxford University Press: London, 1998. 143. Fernandez-Bertran, J. F. Pure Appl. Chem. 1999 71 581. 144. Tanaka, K.; Toda, F. Chem. Rev. 2000 100 1025. 145. Toda, F. Acc. Chem. Res. 1995 28 480. 146. Etter, M. C.; Frankenbach, G. M. Chem. Mater. 1989 1 10. 147. Etter, M. C.; Frankenbach, G. M.; Bernstein, J. Tetrahedron Lett. 1989 30 3617. 148. Etter, M. C.; Urbanczyklipkowska, Z.; Ziaebrahimi, M.; Panunto, T. W. J. Am. Chem. Soc. 1990 112 8415. 149. Etter, M. C.; Reutzel, S. M. J. Am. Chem. Soc. 1991 113 2586. 150. Etter, M. C.; Reutzel, S. M.; Choo, C. G. J. Am. Chem. Soc. 1993 115 4411. 151. Shan, N.; Toda, F.; Jones, W. Chem. Commun. 2002 2372. 152. Trask, A. V.; Haynes, D. A.; Motherwell W. D. S.; Jones W. Chem. Commun. 2006 51. 153. Shan, N.; Jones, W. Green Chemistry 2003 5 728. 154. Trask, A. V.; Motherwell, W. D. S.; Jones, W. Crys. Growth .Des. 2005 5 1013. 155. McCrone, W. C. Polymorphism. In Physics and Ch emistry of the Organic Solid-

PAGE 170

149 State ; Interscience: New York, 1965. 156. Bernstein, J. Polymorphism in Molecular Crystals. ; Claredon Press: Oxford, United Kingdom, 2002. 157. Threlfall, T. L. Analyst 1995 120 2435. 158. Davey, R. J. Chem. Commun. 2003 1463. 159. Mitscherlich, E. Ann. Chim. Phys. 1822 19 350. 160. Bauer, J.; Spanton, S.; Henry, R.; Quic k, J.; Dziki, W.; Porter, W.; Morris, J. Pharm. Res. 2001 18 859. 161. Byrn, S. R. ; Pfeiffer, R. R.; Stowell, J. G. Solid-State Chemistry of Drugs ; SSCI: West Lafayette, IN, 1999. 162. McCrone, W. C. Anal. Chem. 1950 22 954. 163. Gallagher, H. G.; Sherwood, J. N. J. Chem. Soc., Faraday Trans. 1996 92 2107. 164. Erk, P. Curr. Op. Solid State Mater. Sci. 2001 5 155. 165. Klebe, G.; Graser, F.; Hadicke, E.; Berndt, J. Acta Crystallogr. B 1989 45 69. 166. Yu, L.; Stephenson, G. A.; Mitchell, C. A.; Bunnell, C. A.; Snorek, S. V.; Bowyer, J. J.; Borchardt, T. B.; Stowell, J. G.; Byrn, S. R. J. Am. Chem. Soc. 2000 122 585. 167. Sarma, J. A. R. P.; Desiraju, G. R. Polymorphism and Pseudopolymorphism in Organic Crystals: a Cambridge Structural Database Study. Crystal Engineering: The Design and Application of Functional Solids ; Kluwer: Zaworotko, M. J.//Seddon, K. R. Dordrecht, 1999. 168. Bilton, C.; Howard, J. A. K.; Madhavi, N. N. L.; Nangia, A.; Desiraju, G. R.; Allen, F. H.; Wilson, C. C. Chem. Commun. 1999 1675. 169. Seddon, K. R. Cryst. Growth Des. 2004 4 1087. 170. Desiraju, G. R. Cryst. Growth Des. 2004, 4 1089. 171. Haleblian, J. K. J. Pharm. Sci. 1975 64 1269. 172. Kuhnert-Brandstaetter, M. Thermomicroscopy in the Analysis of Pharmaceuticals ; Pegamon Press: Oxford, 1974. 173. McMahon, J. A.; Zaworotko, M. J.; Remenar, J. F. Chem. Commun. 2004 278.

PAGE 171

150 174. Chen, S.; Guzei, I. A.; Yu, L. J. Am. Chem. Soc. 2005 127 9881. 175. ROY stands for Red, Orange, and Yellow colors of the polymorphs of 5methyl-2-[(2-nitrophenyl)-ami no]-3-thiophenecarbonitrile. 176. Admiraal, G.; Eikelenboom, J. C.; Vos, A. Acta Crystallogr. B 1982 38 2600. 177. Fabbiani, F. P. A.; Allan, D. R.; Parsons, S.; Pulham, C. R. CrystEngComm 2005 7 179. 178. Prusiner, P.; Sundaralingam, M. Acta Crystallogr. B 1976 32 419. 179. Bernstein J. Conformational Polymorphism In Organic Solid State Chemistry. ; Ed. Desiraju, G. R., Elsevier: Amsterdam, 1978. 180. Kumar, V. S. S.; Addlagatta, A.; Na ngia, A.; Robinson, W. T.; Broder, C. K.; Mondal, R.; Evans, I. R.; Howard, J. A. K.; Allen, F. H. Angew. Chem. Int. Ed. 2002 41 3848. 181. Bernstein, J.; Davey, R. J.; Henck, J. O. Angew. Chem. Int. Ed. 1999 38 3441. 182. Vishweshwar, P.; McMahon, J. A.; Zaworotko M. J. Crystal Engineering of Pharmaceutical Co-Crystals. In Frontiers in Crystal Engineering. Edward R.T. Tiekink and Jagadese J. Vittal. Wiley, Chichester, UK, 2006. 183. Vishweshwar, P.; McMahon, J. A.; Bis, J. A.; Zaworotko, M. J. J. Pharm. Sci 2006 95 1. 184. CSD search parameters: ConQuest Version 1.7, August 2005, organics, no ions, no alkali metals, number of chemical units = 1. In addition, text search "polymorph" was used and only one re fcode was counted for a polymorphic compound. 185. CSD contains 11 polymorphic co-cryst als with determined 3D coordinates: AJAJEA (AJAJEA01), EXUQUJ (EXUQUJ01), HADKUT (HADKUT01), JICTUK (JICTUK01), MACCID (MA CCID01, MACCID02), MUROXA (MUROXA01), PDTOMS10 (PDTOMS 11), PTZTCQ (PTZTCQ01), QUIDON (QUIDON02), TECCAF01 (TECCA F02), ZIGPAG (ZIGPAG01). 186. Trask, A. V.; Motherwell, W. D. S.; Jones, W. Chem. Commun. 2004 890 187. Trask, A. V.; van de Streek, J.; Haynes D. A.; Motherwell, W. D. S.; Jones W., Cryst. Growth Des 2005 5 2233. 188. Trask, A. V.; Shan, N.; Motherwell, W. D. S.; Jones, W.; Feng, S. H.; Tan, R. B. H.; Carpenter, K. J. Chem. Commun. 2005 880.

PAGE 172

151 189. Web Page, http://www.rxlist.com/top200.htm. 190. Almarsson .; Bourghol Hickey M.; Peterson M.; Zaworotko M.J.; Moulton B.; Rodrguez-Hornedo N. PCT Int. Appl. WO2004078161, 2004. 191. Stahl, P. H.; Nakano, M. Pharmaceutical Aspects of the Drug Salt Form, in Handbook of Pharmaceutical Salts: Properties, Selection, and Use ; Ed. Stahl, P. H. and Wermuth, C. G, W iley-VCH/VCHA: New York, 2002. 192. Gould, P. L. Int. J. Pharm. 1986 33 201. 193. Berge, S. M.; Bighley, L. D.; Monkhouse, D. C. J. Pharm. Sci. 1977 66 1. 194. Remenar, J. F.; MacPhee, J. M.; Larson, B. K.; Tyagi, V. A.; Ho, J. H.; McIlroy, D. A.; Hickey, M. B.; Shaw, P. B.; Almarsson, Org. Proc. Res. Dev. 2003 7 990. 195. Oswald, I. D. H.; Allan, D. R.; McGre gor, P. A.; Motherwell, W. D. S.; Parsons, S.; Pulham, C. R. Acta Crystallogr. B 2002 58 1057. 196. Ghosh, M.; Basak, A. K.; Mazumdar, S. K.; Sheldrick, B. Acta Crystallogr. 1991 C47 577. 197. Madarasz, J.; Bombicz, P.; Jarmi, K.; Ban, M.; Pokol, G.; Gal, S. J. Therm. Anal. Cal. 2002 69 281. 198. Caira, M. R. J. Crystallogr. Spectrosc. Res. 1991 21 641. 199. Caira, M. R. J. Crystallogr. Spectrosc. Res. 1992 22 193. 200. Bettinetti, G.; Caira, M. R.; Callegari A.; Merli, M.; Sorrenti, M.; Tadini, C. J. Pharm. Sci. 2000 89 478. 201. Almarsson .; Remenar J.; Peterson M. L. US Pat. Appl. US6559293, 2003. 202. Peterson M. L.; Bourghol Hickey, M. O. M.; Almarsson .; Remenar J. US Pat. Appl. US2005267209, 2005. 203. Bourghol Hickey, M.; Remenar J. PCT Int. Appl. WO 2004089313, 2004. 204. Childs S. L.; Chyall L. J.; Dunlap J. T. ; Smolenskaya V. N.; Stahly B. C.; Stahly G. P. J. Am. Chem. Soc. 2004 126 13335 205. Childs, S. PCT. Int. Appl. WO2004064762, 2004. 206. Remenar J.; MacPhee M.; Peterson M. L.; Morissette S.L.; Almarsson PCT Int. Appl. WO03101392, 2003.

PAGE 173

152 207. Amos J. G.; Indelicato J. M.; Pasini C. E.; Reutzel S. M Eur. Pat. Appl. EP0637587 A1, 1995. 208. Vishweshwar, P.; McMahon, J. A.; Pete rson, M. L.; Hickey, M. B.; Shattock, T. R.; Zaworotko, M. J. Chem. Commun. 2005 4601. 209. Monissette, S. L.; Almarsson, .; Peterson, M. L.; Remenar, J. F.; Read, M. J.; Lemmo, A. V.; Ellis, S.; Cima, M. J.; Gardner, C. R. Adv. Drug Deliv. Rev. 2004 56 275. 210. Almarsson, .; Hickey, M. B.; Peters on, M. L.; Morissette, S. L.; Soukasene, S.; McNulty, C.; Tawa, M.; MacPhee, J. M.; Remenar, J. F. Cryst. Growth Des. 2003 3 927. 211. Hickey, M. B.; Peterson, M. L. ; Almarsson private communication. 212. Aakery, C. B.; Desper, J.; Elisabeth, E. ; Helfrich, B. A.; Levin, B.; Urbina, J. F. Z. Kristallogr. 2005 220 325. 213. CSD search parameters: ConQuest Version 1.7, August 2005 release, organics only, 3D coordinates determined, and R< 7.5%. All statistics presented in this section relate to OH, Narom, and CN moieties and their engagament in forming synthons I, II, and III in the ab sence of other functional groups. 214. CSD search parameters: ConQuest Version 1.7, August 2005 release, organic compounds with 3D coordinate s determined, and R < 7.5% 215. CSD search parameters: organics, 3D coordinates determined, R < 7.5%. The histograms were generated based upon genera l searches for structures that contain the targeted interactions in a broad range of cont acts. 592 entires were for heterosynthon I, 174 entries for II, and 6588 entries for III. 216. Biradha, K.; Zaworotko, M. J. J. Am. Chem. Soc. 1998 120 6431. 217. CSD reveals 136 crystal st ructures contaning OH and Narom of which 135 entries exhibited I (99%). In one entry, REPPEH, I is absent due to a steric hindrance of the Narom moiety. The 27% of OHOH ho mosynthon occurence is due to the presence of multiple OH moieties in 37 entries. 218. CSD contains 3 crystal st ructures that possess OH, Narom, and CN moieties in the absence of other competing functiona l groups: FOMWIN, YAMMIJ, YAMMOP. 219. Huang, K. S.; Britton, D.; Etter, M. C.; Byrn, S. R. J. Mater. Chem. 1997 7 713. 220. Majerz, I.; Malarski, Z.; Sobczyk, L. Chem. Phys. Lett. 1997 274 361. 221. Brzezinski, B.; Grech, E.; Malarski, Z. ; Rospenk, M.; Schroeder, G.; Sobczyk, L.

PAGE 174

153 J. Chem. Res. 1997 151. 222. Boenigk, D.; Mootz, D. J. Am. Chem. Soc. 1988 110 2135. 223. Cowan, J. A.; Howard, J. A. K.; McIntyre, G. J.; Lo, S. M. F.; Williams, I. D. Acta Crystallogr. B 2003 59 794. 224. CSD searches for neutral C-OH and ionic C-Obonds involved specific restrictions on the oxygen atoms. For neutral C-OH bond, the presence of the hydrogen atom bonded to oxygen atom was acquired, the oxygen atom was defined to be uncharged and the number of bonded atoms was set to 2. In the case of ionic C-Obond, the charge of the oxygen was set to -1 and the number of bonded atoms was set to 1. 225. Metzger, J. O. Angew. Chem. Int. Ed. 1998 37 2975. 226. MacGillivray, L. R.; Reid, J. L.; Ripmeester, J. A. J. Am. Chem. Soc. 2000 122 7817. 227. Desiraju, G. R. a. S. T. Weak Hydrogen Bond in Structural Chemistry and Biology ; Oxford University Press: Oxford, 1999. 228. Mascal, M. Chem. Commun. 1998 303. 229. Steiner, T.; Desiraju, G. R. Chem. Commun. 1998 891. 230. Thalladi, V. R.; Gehrke, A.; Boese, R. New J. Chem. 2000 24 463. 231. Thalladi, V. R.; Smolka, T.; Boese, R.; Sustmann, R. CrystEngComm 2000 17 232. Thalladi, V. R.; Smolka, T.; Gehrke, A.; Boese, R.; Sustmann, R. New J. Chem. 2000 24 143. 233. Desiraju, G. R. Acc. Chem. Res. 1996 29 441. 234. Gu, Y. L.; Kar, T.; Scheiner, S. J. Am. Chem. Soc. 1999 121 9411. 235. Bis, J. A.; Vishweshwar, P.; Middleton, R.; Zaworotko, M. J. Cryst. Growth Des. 2006 ASAP. 236. Coupar, P. I.; Ferguson, G.; Glidewell, C. Acta Crystallogr. 1996 C52 2524. 237. Abrahamsson, S.; Larsson, G.; Vonsydow, E. Acta Crystallogr. 1960 13 770. 238. Ohms, U.; Guth, H.; Treutmann, W. Z. Kristallogr. 1983 162 299. 239. Sarma, J. A. R. P.; Desiraju, G. R. Acc. Chem. Res. 1986 19 222.

PAGE 175

154 240. Sarma, J. A. R. P.; Desiraju, G. R. J. Chem. Soc. Perkin Trans. 2 1987 1195. 241. Desiraju, G. R. J. Chem. Soc. 1990 454. 242. Reddy, D. S.; Goud, B. S.; Panneerselvam, K.; Desiraju, G. R. J. Chem. Soc. 1993 663. 243. Allen, F. H.; Hoy, V. J.; Howard, J. A. K.; Thalladi, V. R.; Desiraju, G. R.; Wilson, C. C.; McIntyre, G. J. J. Am. Chem. Soc. 1997 119 3477. 244. Thalladi, V. R.; Nusse, M.; Boese, R. J. Am. Chem. Soc. 2000 122 9227. 245. Aakery, C. B.; Hussain, I.; Desper, J. Cryst. Growth Des. 2006 6 474. 246. Aakery, C. B.; Desper, J. ; Leonard, B.; Urbina, J. F. Cryst. Growth Des. 2005 5 865. 247. Cotterill, R. M. J. J. Cryst. Growth 1980 48 582. 248. SADABS [Area-Detector Absorption Correct ion]. Siemens Industrial Automation, Inc.: Madison, WI, 1996. 249. Sheldrick, G. M. SHELXTL University of Gottingen: Germany, 1997. 250. Ahn, S.; Kariuki, B. M.; Harris, K. D. M. Cryst. Growth Des. 2001 1 107. 251. CSD search parameters: organics on ly, no alkali metals, and 3D coordinates determined. Hydrogen atoms were placed at 2,2,6,6-position to avoid steric restrains. 252. CSD contains 3 conformational is omorphs: COBLAG, CUVNAI, and NISLUW. 253. Conformational polymorphism has b een reported for EC ELON (ECELON01) 254. Brock, C. P.; Minton, R. P. J. Am. Chem. Soc. 1989 111 4586. 255. Corish, J.; Morton-Blake, D. A.; O'Donoghue, F.; Baudour, J. L.; Beniere, F.; Toudic, B. Theochem 1995 29. 256. Bowes, K. F.; Ferguson, G.; Lough, A. J.; Glidewell, C. Acta Crystallogr. B 2003 59 277. 257. Shattock, T.; Vishweshwar, P.; Wang, Z.; Zaworotko, M. J. Cryst. Growth Des. 2005 5 2046. 258. Matsuda, H.; Osaki, K.; Nitta, I. Bull. Chem. Soc. Jpn. 1958 31 611.

PAGE 176

155 259. Whler F.; Liebig, J. Annal. Pharm. 1832 3 249. 260. von Groth, P. H. R. An Introduction to Ch emical Crystallography ; London, 1906. 261. Concomitant polymorphism was reported for MACCID (MACCID01, MACCID02), JICTUK10 (JICTUK01), HADKUT01 (HADKUT), and EXUQUJ01 ( EXUQUJ) 262. Vishweshwar, P.; Thaimattam, R. ; Jaskolski, M.; Desiraju, G. R. Chem. Commun. 2002 1830. 263. Chao, M.; Schempp, E.; Rosenstein, D. Acta Crystallogr. B 1975 31 2922. 264. Lynch, D. E.; Smith, G.; Freney, D. ; Byriel, K. A.; Kennard, C. H. L. Aust. J. Chem. 1994 47 1097. 265. Hsu, I. N.; Craven, B. M. Acta Crystallogr. B 1974 B 30 994. 266. CSD search parameters: ConQuest Ve rsion 1.6, July 2004 release, organics, 3D coordinates determined, and R < 7.5% 267. Borthwick, P. W. Acta Crystallogr. B 1980 36 628. 268. Mootz, D.; Wussow, H. G. J. Chem. Phys. 1981 75 1517. 269. Mootz, D.; Hocken, J. Z. Naturforschung B J. Chem. Sci. 1989 44 1239. 270. Leiserowitz, L.; Hagler, A. T. Proc. R. Soc. London, A 1983 388 133. 271. Kuduva, S. S.; Blaser, D.; Boese, R.; Desiraju, G. R. J. Org. Chem. 2001 66 1621. 272. Odabasoglu, M.; Buyukgungor, O.; Lonnecke, P. Acta Crystallogr. 2003 C59 51. 273. Gellert, R. W.; Hsu, I. N. Acta Crystallogr. 1988 C44 311. 274. Wheeler, K. A.; Foxman, B. M. Mol. Cryst. Liq. Cryst. 1994 240 89. 275. Koshima, H.; Miyamoto, H.; Yagi, I.; Uosaki, K. Cryst. Growth Des. 2004 4 807. 276. Aakery, C. B.; Beffert, K.; Desper, J.; Elisabeth, E. Cryst. Growth Des. 2003 3 837. 277. Robert, J. J.; Raj, S. B.; Muthiah, P. T. Acta Crystallogr. 2001 E 57 O1206O1208.

PAGE 177

156 278. Kalman, A. P. L. A. G. Acta Crystallogr. 1993 B 49 1039. 279. Odabasoglu, M.; Buyukgungor, O.; Turgut, G.; Karadag, A.; Bulak, E.; Lonnecke, P. J. Mol. Struct. 2003 648 133. 280. Buyukungor, O.; Odabasoglu, M. Acta Crystallogr. 2002 C58 691. 281. Ballabh, A.; Trivedi, D. R.; Dastidar, P.; Suresh, E. CrystEngComm 2002 135. 282. CSD refcodes of 32 co-crystals of barbital: AEPDEB, AMIWUO, BARAPY10, BARBAM, BARBUR, BARHMP, BA RIMZ10, BARMPN, BIGCUP, CAFBAR20, EADBAR10, HIBJUX, HI BKEI, JICTIY, JICTOE, JICTUK, JICVAS, JICVEW, JICVIA, JICVOG, JICVUM, JUBRAZ, KEGPUH, KUFPIK, MUDSAF, MUDSEJ, MUDSIN, PIYGE J, QQQEUV, QQQFVA, WETSOD, WETTUK. 283. CSD refcodes of 14 co-crystals of sulfonamide drugs: GEYSAE, SACCAF, SANAPY, SMZTMP, SORWEB, SORW IF, STHSAM, SULTHE, VIGVOW, VUGMIT, VUGMOZ, XEXCAE, XEXCEI, YOSMOI 284. CSD refcodes of 8 co-crystal s of phenathiazine : BUNRAD, DAPXUN, LENGOA, NIWCEB, PHNSNB10, PHTNBA, PTZPMA, PTZTCQ 285. CSD refcodes of 5 co-cryst als of CBZ: UNEYOB, UNEYKH, UNEZAO, UNEZES, UNIBIC 286. CSD refcodes of 5 co-crystals of theophylline: CSATEO DUXZAX, SULTHE, THOPBA, ZEXTIF. 287. CSD refcodes of 6 co-crystals of caffeine: CAFSAL, DIJVOH, DIJVUN, SACCAF, VIGVOW, EXUQUJ 288. CSD refcodes of 2 co-crystal s of fluribuprofen: HUPPEN, HUPPIR. 289. CSD refcode of the co-crystal of ibuprofen: HUPPAJ 290. CSD refcode of the co-cry stal of itraconazole: IKEQEU 291. CSD refcode of the co-cryst al of diphenylhydantoin: DPHPZL. 292. CSD refcode of the co-crystal of trimethoprim: BIGCUP 293. Zerkowski, J. A.; MacDonald, J. C.; Whitesides, G. M. Chem. Mater. 1997 9 1933. 294. Zerkowski, J. A.; Mathias, J. P.; Whitesides, G. M. J. Am. Chem. Soc. 1994 116

PAGE 178

157 4305. 295. Zerkowski, J. A.; Whitesides, G. M. J. Am. Chem. Soc. 1994 116 4298. 296. Zerkowski, J. A.; Seto, C. T.; Whitesides, G. M. J. Am. Chem. Soc. 1992 114 5473. 297. Reboul, J. P.; Cristau, B.; Soyfer, J. C.; Astier, J. P. Acta Crystallogr. 1981 B37 1844. 298. Lisgarten, J. N.; Palmer, R. A.; Saldanha, J. W. J. Crystallogr. Spectrosc. Res. 1989 19 641. 299. Lang, M. D.; Kampf, J. W.; Matzger, A. J. J. Pharm. Sci. 2002 91 1186. 300. Grzesiak, A. L.; Lang, M. D.; Kim, K.; Matzger, A. J. J. Pharm. Sci. 2003 92 2260. 301. Louer, D.; Louer, M.; Dzyabchenko, V. A.; Agafonov, V.; Ceolin, R. Crystallogr. 1995 B 51 182. 302. Thallapally, P. K.; Nangia, A. CrystEngComm 2001 27. 303. Steiner, T. Acta Crystallogr. 1998 B 54 456. 304. Ishida, T.; In, Y.; Doi, M.; Inoue, M. Acta Crystallogr. 1989 B 45 505. 305. Caira, M. R.; Dekker, T. G.; Liebenberg, W. J. Chem. Crystallogr. 1998 28 11. 306. Ladanyi, L.; Sztruhar, I.; Budai, Z. ; Lukacs, G.; Mezei, T.; Argay, G.; Kalman, A.; Simig, G. Chirality 1999 11 689. 307. Ota, A.; Kawashima, Y.; Ohishi, H.; Ishida, T. Chem. Pharm. Bull. 1993 41 1681. 308. Marubayashi, N.; Fujii, I.; Hirayama, N. Anal. Sci. 1999 15, 813. 309. Hogberg, T.; Norinder, U.; Ramsby, S.; Stensland, B. J. Pharm. Pharmacol. 1987 39 787. 310. Furuya, T.; Fujita, S.; Fujikura, T. Anal. Sci. 1989 5 489. 311. Kolvenbag Gjcm; Blackledge, G. R. P.; Gotting-Smith, K. Prostate 1998 34 61. 312. Hickey, M. B.; Peterson, M.; Almar sson, .; Zaworotko, M. J.; Shattock, T.; McMahon J.; Bis, J. ; Remenar, J. ; Tawa, M. PCT Int. Appl. WO2005089511, 2005.

PAGE 179

158 313. Westheim, R. J. H. US Pat. Appl. US20040063782 A1, 2004. 314. CSD search for the distance distribution the NHNarom interaction (organics, 3D coordinates determined, and R < 7.5%) a fforeded 420 entries that occur within a range of 2.75 3.30 A, with the average of 3.0(1) A. 315. Galdecki, Z.; Glowka, M. L. Roczniki Chemii 1976 50 1139. 316. Cox, P. J.; Manson, P. L. Acta Crystallogr. 2003 E 59 O986-O988. 317. Chen, X. M.; Morris, K. R.; Griesser, U. J.; Byrn, S. R.; Stowell, J. G. J. Am. Chem. Soc. 2002 124 15012. 318. Stowell, J. G.; Byrn, S. R.; Zografii, G.; Yoshioka, M. 2002 private communication. 319. Cox, P. J.; Manson; P. L. Acta Crystallogr. 2003 E59 1189. 320. Umadevi, B. ; Prabakaran, P.; Muthiah, P. T. Acta Crystallogr. 2002 C58 510. 321. Bryan, R. F.; Haltiwanger, R. C.; Woode, M. K. Acta Crystallogr. 1987 C43 2412. 322. Francis, S.; Muthiah, P. T.; Bocelli, G.; Righi, L. Acta Crystallogr. 2002 E 58 O717-O719. 323. Hemamalini, M.; Muthiah, P. T.; Butcher, R. J. Acta Crystallogr. 2004 E 60 O2350-O2352. 324. Giuseppetti, G.; Tadini, C.; Bettin etti, G. P.; Giordano, F.; Lamanna, A. Acta Crystallogr. 1984 C40 650. 325. Raj, S. B.; Muthiah, P. T.; Rychlewska, U.; Warzajtis, B. CrystEngComm 2003 48. 326. Raj, S. B.; Stanley, N.; Muthiah, P. T.; Bocelli, G.; Olla, R.; Cantoni, A. Cryst. Growth Des. 2003 3 567. 327. Hemamalini, M.; Muthiah, P. T.; Bocelli, G.; Cantoni, A. Acta Crystallogr. 2003 E59 O14-O17. 328. GRAS compounds can be found at www.cfsan.fda.gov/~dms/eafus.html.

PAGE 180

159 Appendices

PAGE 181

160 Appendix 1. Experimental data for compound 1 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the sing le crystal structure (black). 56195 615.73 685.91 731.75 757.84 786.88 834.37 890.34 897.93 936.90 994.06 1008.27 1071.45 1140.20 1227.86 1251.61 1288.29 1367.00 1416.66 1479.24 1574.84 1602.15 1785.55 2232.68 2359.41 2569.80 2664.93 2791.22 2902.29 3056.55 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 %T 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 500 1000 1500 2000 2500

PAGE 182

161 Appendix 2. Experimental data for compound 2 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the sing le crystal structure (black). 679.50 777.39 822.81 888.30 938.34 1010.44 1067.91 1139.78 1249.03 1293.90 1365.26 1419.21 1480.26 1577.84 1605.53 2225.08 2363.39 2568.20 2663.22 2789.84 2893.34 3063.75 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 104 106 108 %T 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 200 400 600 800

PAGE 183

162 Appendix 3. Experimental data for compound 3 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the si ngle crystal structure (black) 615.77 677.62 785.17 826.02 888.40 937.20 1008.01 1065.04 1136.65 1216.51 1250.82 1287.39 1360.37 1421.19 1478.48 1577.66 1602.25 2229.45 2572.28 2666.11 2792.34 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 %T 1000 2000 3000 Wavenumbers (cm-1)051015202530354045 0 200 400 600

PAGE 184

163 Appendix 4. Experimental data for compound 4 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the sing le crystal structure (black). 620.46 689.21 798.97 820.94 888.87 967.80 1011.90 1067.14 1138.08 1219.75 1251.61 1292.98 1362.45 1416.96 1476.89 1573.03 1604.02 2221.07 2536.40 2773.81 2900.91 74 76 78 80 82 84 86 88 90 92 94 96 98 100 %T 1000 2000 3000 Wavenumbers (cm-1)051015202530354045 0 100 200 300 400

PAGE 185

164 Appendix 5. Experimental data for compound 5 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the si ngle crystal structure (black). 61634 690.85 728.32 754.81 828.69 847.98 1008.72 1067.82 1172.20 1216.78 1265.98 1297.83 1413.03 1450.75 1513.01 1580.46 1604.35 2214.40 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 %T 1000 2000 3000 Wavenumbers (cm-1)051015202530354045 0 200 400 600 800 1000 1200

PAGE 186

165 Appendix 6. Experimental data for compound 6 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the si ngle crystal structure (black). 60774 61847 697.53 798.11 830.62 999.21 1059.02 1168.70 1213.73 1255.33 1288.16 1408.25 1457.39 1509.50 1589.18 2218.07jb434a 88 89 90 91 92 93 94 95 96 97 98 99 100 101%T 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 -500 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500

PAGE 187

166 Appendix 7. Experimental data for compound 7 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the sing le crystal structure (black). 698.05 814.30 825.57 1007.52 1163.67 1255.09 1290.59 1456.42 1510.28 1583.25 1603.62 2215.31 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 %T 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 100 200 300

PAGE 188

167 Appendix 8. Experimental data for compound 8 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the sing le crystal structure (black). 698.94 818.42 971.42 1009.81 1061.84 1169.44 1293.64 1418.30 1463.80 1513.49 1587.87 1603.18 2220.38 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 %T 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 200 400 600 800 1000 1200 1400

PAGE 189

168 Appendix 9. Experimental data for compound 9 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the sing le crystal structure (black). 61353 637.74 649.99 699.33 781.30 810.41 825.44 1030.53 1047.82 1170.62 1198.79 1232.63 1263.27 1417.45 1483.17 1500.33 1594.37 1610.98 2237.63jb457a 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 104 106 %T 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 250 500 750 1000 1250 1500 1750 2000 2250 2500

PAGE 190

169 Appendix 10. Experimental data for compound 10 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the sing le crystal structure (black). 61125 666.43 693.24 778.46 826.70 883.18 966.71 1001.59 1065.83 1145.34 1182.27 1219.46 1276.72 1413.74 1477.94 1598.42 2237.32 2607.33 2732.36 2889.24 3037.14 3087.96 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 %T 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 100 200 300 400 500 600 700 800

PAGE 191

170 Appendix 11. Experimental data for compounds 11a and 11b DSC termograms, FT-IR spectra of Form I and Form II.

PAGE 192

171 Appendix 12. Experimental data for compound 12 DSC termogram, FT-IR spectrum and X-ray powder diffraction patterns of bulk sample (red) and calculated from the sing le crystal structure (black). 666.70 698.41 781.99 823.97 838.56 1001.73 1069.75 1153.33 1213.95 1294.72 1411.67 1499.72 1597.76 1625.70 2239.33 2637.46 3089.16 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 %T 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 100 200 300 400

PAGE 193

172 Appendix 13. Experimental data for compound 13a and 13b DSC termogram, FT-IR spectra Form I (blue) and Form II (red).

PAGE 194

173 Appendix 14. Polymorphism screen data for compound 1 FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 100 200 300 400 500 600 700 Methanol Water DMSO Ethyl acetate Chloroform Toluene Cyclohexane

PAGE 195

174 Appendix 15. Polymorphism screen data for compound 2 FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 50 100 150 200 250 Water DMSO Methanol Ethyl acetate Chloroform Toluene Cyclohexane

PAGE 196

175 Appendix 16. Polymorphism screen data for compound 3 FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 50 100 150 200 250 300 350 Water DMSO Methanol Ethyl acetate Chloroform Toluene Cyclohexane

PAGE 197

176 Appendix 17. Polymorphism screen data for compound 4 FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 50 100 150 200 250 300 350 Cyclohexane Toluene Methanol Ethyl acetate Chloroform DMSO Water

PAGE 198

177 Appendix 18. Polymorphism screen data for compound 5 FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 100 200 300 400 500 600 700 Water DMSO Methanol Ethyl acetate Chloroform Toluene Cyclohexane

PAGE 199

178 Appendix 19. Polymorphism screen data for compound 6 FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 50 100 150 200 250 Cyclohexane Toluene Chloroform Ethyl acetate Methanol DMSO Water

PAGE 200

179 Appendix 20. Polymorphism screen data for compound 7 FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 50 100 150 200 Cyclohexane Toluene Chloroform Ethyl acetate Methanol DMSO Water

PAGE 201

180 Appendix 21. Polymorphism screen data for compound 8 X-ray powder diffraction patte rns of powders obtained based upon solvent-drop grinding with: cyclohexane, toluene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 -50 0 50 100 150 200 250 300 350 Water DMSO Methanol Ehtyl actetate Chloroform Toluene Cyclohexane

PAGE 202

181 Appendix 22. Polymorphism screen data for compound 9 FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 50 100 150 200 250 300 Cyclohexane Chloroform Toluene Ethyl acetate Methanol DMSO Water

PAGE 203

182 Appendix 23. Polymorphism screen data for compound 10 FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 50 100 150 200 250 300 350 400 450 500 550 600 Water DMSO Methanol Ethyl acetate Chloroform Toluene Cyclohexane

PAGE 204

183 Appendix 24. Polymorphism screen data for compound 11a and 11b FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 250 500 750 1000 1250 1500 Water DMSO Methanol Ethyl acetate Chloroform Toluene Cyclohexane

PAGE 205

184 Appendix 24. Polymorphism screen data for compound 11a and 11b (continued) Comparison of experimental (pink) and calculated X-ray powder diffraction patterns of 4,4-biphenol DMSO solvate, ECELON01 (black). 051015202530354045 0 200 400 600 800 1000 1200 DMSO-drop grinding simulated (ECELON01)

PAGE 206

185 Appendix 25. Polymorphism screen data for compound 12 FT-IR spectra and X-ray powder diffracti on patterns of powders obtained based upon solvent-drop grinding with: cyclohexane, tolu ene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 0 100 200 300 400 500 Ethyl acetate Water DMSO Methanol Chloroform Toluene Cyclohexane

PAGE 207

186 Appendix 26. Polymorphism screen data for compound 13a and 13b X-ray powder diffraction patte rns of powders obtained based upon solvent-drop grinding with: cyclohexane, toluene, chloroform, ethyl acetate, methanol, DMSO, and water. 051015202530354045 -50 0 50 100 150 200 250 300 Water Cyclohexane Toluene Chloroform Ethyl acetate Methanol DMSO

PAGE 208

187 Appendix 27. Experimental data for compound 14 FT-IR spectrum and X-ray powder diffracti on patterns of bulk sample (red) and .calculated from the single crystal structure (black). 688.24 704.44 733.86 767.18 789.80 853.70 924.82 1138.43 1156.98 1169.12 1292.30 1370.79 1496.02 1551.67 1604.63 1619.19 1676.08 2927.33 3341.78 3470.40 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 %Transmittance 1000 2000 3000 4000 Wavenumbers (cm-1) 051015202530354045 0 500 1000 1500 2000 2500 3000

PAGE 209

188 Appendix 28. Experimental data for compound 15 FT-IR spectrum and X-ray powder diffracti on patterns of bulk sample (red) and calculated from the single crystal structure (black). 688.00 730.65 773.19 934.70 983.35 1076.19 1151.91 1257.69 1380.41 1430.08 1484.51 1606.31 1638.50 1668.33 1708.22 74 76 78 80 82 84 86 88 90 92 94 96 98 %Transmittance 1000 2000 3000 4000 Wavenumbers (cm-1) 051015202530354045 0 200 400 600 800 1000 1200 1400 1600 1800

PAGE 210

189 Appendix 29. Experimental data for compound 16 FT-IR spectrum and X-ray powder diffracti on patterns of bulk sample (red) and calculated from the single crystal structure (black). 741.48 765.57 816.68 852.39 880.34 965.48 1017.83 1142.76 1155.57 1255.90 1350.55 1378.14 1395.41 1436.16 1498.50 1633.41 1680.43 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 %Transmittance 1000 2000 3000 4000 Wavenumbers (cm-1) 051015202530354045 0 200 400 600 800 1000 1200 1400 1600

PAGE 211

190 Appendix 30. Experimental data for compound 17 FT-IR spectrum and X-ray powder diffracti on patterns of bulk sample (red) and calculated from the single crystal structure (black). 60741 674.27 714.25 757.13 833.19 980.05 1042.74 1065.08 1149.84 1170.66 1258.62 1366.90 1486.18 1555.70 1596.89 1644.59 1680.45 78 80 82 84 86 88 90 92 94 96 98 100 102%Transmittance 1000 2000 3000 4000 Wavenumbers (cm-1) 051015202530354045 0 200 400 600 800 1000

PAGE 212

191 Appendix 31. Experimental data for compound 18 FT-IR spectrum and X-ray powder diffracti on patterns of bulk sample (red) and calculated from the single crystal structure (black). 711.58 742.88 782.00 1004.31 1098.63 1152.82 1266.24 1353.75 1407.30 1500.24 1555.47 1666.31 2956.97 78 80 82 84 86 88 90 92 94 96 98 %Transmittance 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 200 400 600 800 1000 1200 1400 1600

PAGE 213

192 Appendix 32. Experimental data for compound 19 FT-IR spectrum and X-ray powder diffracti on patterns of bulk sample (red) and calculated from the single crystal structure (black). 639.81 718.01 753.33 816.39 883.16 908.24 987.64 1016.64 1045.29 1142.94 1158.95 1259.73 1335.43 1357.20 1376.85 1394.76 1449.63 1497.43 1549.40 1684.42 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 %Transmittance 1000 2000 3000 4000 Wavenumbers (cm-1) 051015202530354045 0 200 400 600 800 1000 1200 1400 1600

PAGE 214

193 Appendix 33. Experimental data for compound 20 FT-IR spectrum and X-ray powder diffracti on patterns of bulk sample (red) and calculated from the single crystal structure (black). 636.20 660.61 751.20 793.01 833.10 854.31 914.37 983.11 1096.36 1136.30 1156.56 1187.52 1256.41 1354.49 1381.99 1439.07 1494.27 1658.81 2932.14 3469.36 10 20 30 40 50 60 70 80 90 100 %Transmittance 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 1000 2000 3000 4000

PAGE 215

194 Appendix 34. Experimental data for compound 21 FT-IR spectrum and X-ray powder diffracti on patterns of bulk sample (red) and calculated from the single crystal structure (black). 642.51 758.05 825.07 857.17 898.21 914.34 979.52 1024.64 1092.50 1190.50 1265.39 1311.14 1367.01 1395.70 1458.87 1489.75 1549.40 1644.23 1681.47 2870.21 2942.61 3314.90 76 78 80 82 84 86 88 90 92 94 96 98 %Transmittance 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000

PAGE 216

195 Appendix 35. Experimental data for compound 22 FT-IR spectrum and X-ray powder diffracti on patterns of bulk sample (red) and calculated from the single crystal structure (black). 60368 645.80 753.82 800.13 819.30 884.85 989.13 1105.60 1154.37 1206.84 1237.27 1317.94 1350.16 1534.56 1627.50 1673.26 3148.64 3419.98 78 80 82 84 86 88 90 92 94 96 98 %Transmittance 1000 2000 3000 Wavenumbers (cm-1) 051015202530354045 0 500 1000 1500 2000 2500

PAGE 217

196 Appendix 36. Experimental data for compound 23 X-ray powder diffraction pattern s of bulk sample (black) a nd calculated from the single crystal structure (red) of compound 14 051015202530354045 0 200 400 600 800 1000

PAGE 218

197 Appendix 37. Experimental data for compound 24 X-ray powder diffraction pattern s of bulk sample (red) and calculated from the single crystal structure (black). 051015202530354045 200 250 300 350 400 450 500 550 600 650

PAGE 219

198 Appendix 38. Experimental data for compound 25 X-ray powder diffraction pattern s of bulk sample (red) and calculated from the single crystal structure (black). 051015202530354045 -100 0 100 200 300 400 500 600 700 800 900 1 000 1 100

PAGE 220

199 Appendix 39. Experimental data for compound 26 FT-IR spectrum and X-ray powder diffracti on patterns of bulk sample (red) and calculated from the single crystal structure (black). 051015202530354045 0 100 200 300 400 500 600 700

PAGE 221

About the Author Joanna A. Bis received M. Sc. degree in analytical chemistry from Jagiellonian University, Cracow, Poland, in 2002, where she worked in the area of analysis of inks extracted from documents in the context of criminalistics. While studying there she was honored with the Socrates Erasmus Programme scholarship for academic achievements. This gave her the opportunity to study and conduct a research proj ect at Strathclyde University, Glasgow, Scotland. In 2002, Joanna joined the University of South Florida a nd started working toward her Ph. D. in Dr. Michael Zaworo tkos research group. While in the Ph.D. program, she obtained a Research Assistants hip from TransForm Pharmaceuticals Inc. Joanna is a co-inventor on a patent app lication, has co-authore d four scientific publications, and has presented her research at regional, national, and international scientific meetings of the American Chemical Society, Scientific Advisory Board of TransForm Pharmaceuticals Inc., and Intern ational Quality & Productivity Center.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001787550
003 fts
005 20070508130535.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 070508s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001424
040
FHM
c FHM
035
(OCoLC)124066589
049
FHMM
090
QD31.2 (ONLINE)
1 100
Bis, Joanna A.
0 245
Crystal engineering of organic compounds including pharmaceuticals
h [electronic resource] /
by Joanna A. Bis.
260
[Tampa, Fla] :
b University of South Florida,
2006.
3 520
Neutral or charge-assisted hydrogen bonds occurring between organic molecules represent strong and directional forces that mediate the molecular self-assembly into well defined supramolecular architectures. A proper understanding of hydrogen bonding interactions, their types, geometries, and occurrence in supramolecular motifs, is a prerequisite to crystal engineering, i.e. to the rational design of functional solid materials.Multiple-component organic crystals represent ideal systems to study the intermolecular interactions between the constituent molecules that can be pre-selected for their hydrogen bonding sites and geometrical capabilities. In particular, the systematic structural analysis of supramolecular systems that are comprised of simple molecules facilitates the development of strategies for the rational design of new multiple-component compounds involving more complex components such as drug molecules.The work presented herein shows a combination of systematic database and experimental studies in the context of reliability and hierarchy of several hydrogen bonded supramolecular synthons that exist in a series of model co-crystals and organic salts. The acquired paradigms are ultimately utilized in crystal engineering of pharmaceuticals. In addition, the viability of a mechanichemical approach toward supramolecular synthesis in the context of its efficacy and the effect on polymorphism in multiple-component compounds is also addressed.
502
Dissertation (Ph.D.)--University of South Florida, 2006.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 199 pages.
590
Adviser: Michael J. Zaworotko, Ph.D.
653
Supramolecular chemistry.
Supramolecular synthon.
Hydrogen bond.
Co-crystal.
Polymorphism.
690
Dissertations, Academic
z USF
x Chemistry
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1424