USF Libraries
USF Digital Collections

Crystal engineering of binary compounds containing pharmaceutical molecules

MISSING IMAGE

Material Information

Title:
Crystal engineering of binary compounds containing pharmaceutical molecules
Physical Description:
Book
Language:
English
Creator:
Morales, Leslie Ann
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
pharmaceuticals
supramolecular synthons
supramolecular chemistry
crystal engineering
hydrogen bonding
Dissertations, Academic -- Chemistry -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: The synthesis or the interaction between two or more molecules is known as supramolecular chemistry. The concept of supramolecular chemistry can be applied to the design of new pharmaceutical materials affording new compositions of matter with desirable composition, structure and properties. The design of a two-molecule, or binary, compound using complementary molecules represents an example of an application of crystal engineering. Crystal engineering is the understanding of intermolecular interactions, in the context of crystal packing, in the design of new solid materials. By identifying reliable connectors through molecular recognition or self-assembly, one can build predictable architectures. The study of supramolecular synthesis was accomplished using known pharmaceutical molecules such as Nifedipine (calcium channel blocker used for cardiovascular diseases) and Phenytoin (used as an anticonvulsant drug) and model compounds containing synthons common in pharmaceutical drugs (Crown ethers and Trimesic acid with ether linkages and carboxylic acid dimers, respectively) with complementary molecular additives. The co-crystals formed were characterized by various techniques (IR, m.p., XPD, single X-ray diffraction) and preliminary results were found to exhibit characteristics different from the parent compounds as a direct result of hydrogen bonding and self-assembly interactions. These crystalline assemblies could afford improved solubility, dissolution rate, stability and bioavailability.
Thesis:
Thesis (M.S.)--University of South Florida, 2003.
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 Leslie Ann Morales.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 80 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 - 001441493
oclc - 53994306
notis - AJM5933
usfldc doi - E14-SFE0000166
usfldc handle - e14.166
System ID:
SFS0024862:00001


This item is only available as the following downloads:


Full Text
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 001441493
003 fts
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 031203s2003 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000166
035
(OCoLC)53994306
9
AJM5933
b SE
SFE0000166
040
FHM
c FHM
090
QD31.2
1 100
Morales, Leslie Ann.
0 245
Crystal engineering of binary compounds containing pharmaceutical molecules
h [electronic resource] /
by Leslie Ann Morales.
260
[Tampa, Fla.] :
University of South Florida,
2003.
502
Thesis (M.S.)--University of South Florida, 2003.
504
Includes bibliographical references.
516
Text (Electronic thesis) 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 80 pages.
520
ABSTRACT: The synthesis or the interaction between two or more molecules is known as supramolecular chemistry. The concept of supramolecular chemistry can be applied to the design of new pharmaceutical materials affording new compositions of matter with desirable composition, structure and properties. The design of a two-molecule, or binary, compound using complementary molecules represents an example of an application of crystal engineering. Crystal engineering is the understanding of intermolecular interactions, in the context of crystal packing, in the design of new solid materials. By identifying reliable connectors through molecular recognition or self-assembly, one can build predictable architectures. The study of supramolecular synthesis was accomplished using known pharmaceutical molecules such as Nifedipine (calcium channel blocker used for cardiovascular diseases) and Phenytoin (used as an anticonvulsant drug) and model compounds containing synthons common in pharmaceutical drugs (Crown ethers and Trimesic acid with ether linkages and carboxylic acid dimers, respectively) with complementary molecular additives. The co-crystals formed were characterized by various techniques (IR, m.p., XPD, single X-ray diffraction) and preliminary results were found to exhibit characteristics different from the parent compounds as a direct result of hydrogen bonding and self-assembly interactions. These crystalline assemblies could afford improved solubility, dissolution rate, stability and bioavailability.
590
Adviser: Zaworotko, Michael
653
pharmaceuticals.
supramolecular synthons.
supramolecular chemistry.
crystal engineering.
hydrogen bonding.
690
Dissertations, Academic
z USF
x Chemistry
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.166



PAGE 1

Crystal Engineering of Binary Compounds Containing Pharmaceutical Molecules By Leslie Ann Morales A thesis submitted in partial fulfillment Of the requirements for the degree of Master of Science Department of Chemistry College of Arts and Sciences University of South Florida Major Professor: Michael J. Zaworotko, Ph.D. Gregory Baker, Ph.D. Edward Turos, Ph.D. Date of Approval: October 29, 2003 Keywords: crystal engineering, supramol ecular chemistry, supramolecular synthons, Pharmaceuticals, hydrogen bonding Copyright 2003, Leslie Ann Morales

PAGE 2

Acknowledgements The author would like to express her sinc ere appreciation to Dr. Michael Zaworotko for his guidance and support. Also to Dr. Rosa Walsh, Dr. Brian Moulton, Dr. Victor Kravtsov and Dr. Srinivassan Kuduva for thei r contributions with the crystallographic data in this work. The aut hor is also grateful to Dr. Rosa Walsh and Dr. John Osegovic for their advice and input duri ng her graduate career as well as in the writing of this thesis. Most importantly, the author would like to thank her parents, Hector and Belkys Morales, her brother Hector, sister-inlaw Edna and niece Alyssa for being my inspiration, my constant support and my mo tivation. A special thank you goes to her grandmother Ana Luisa Navarro for always be ing there, her loving friends Diana Acosta, Lena Lago, Dahianna Ramirez and Christopher Brazel. Chris, even though you came at the tail end of this journey, your support a nd advice proved to be immeasurable. You came at the perfect time, thank you very much. Lastly, the author extends he r sincere appreciation to her laboratory friends, Jennifer McMahon, Heba Abourahma, Joanna Bis, Tani se Shattock, Kadine Mohomed, LaNetra Clayton, Vicky Likourinou, Jarrod Eubank, John Perry and Gregory McManus for their help and for making the last couple of year s more enjoyable and interesting. Thank you once again and God bless you all.

PAGE 3

Table of Contents List of Tables iv List of Figures v List of Abbreviations vii Abstract viii Chapter 1---Introduction 1 1.1 Supramolecular Chemistry and Crystal Engineering 1 1.2 Supramolecular Synthons 2 1.3 Polymorphism 3 1.4 Cambridge Structural Database 4 1.4.1 Nifedipine Synthons 5 1.4.2 Phenytoin Synthons 6 1.4.3 Crown Ether Synthons 7 1.4.4 Trimesic acid Synthons 8 1.4.5 Secondary Hydrogen Bonds 9 Chapter 2---Nifedipine 11 2.1 Description 11 2.2 Strategy 12 2.3 Structures 13 2.3.1 Nifedipine/formamide (A) 13 2.3.2 Nifedipine/isonicotinamide (B) 14 2.3.3 Nifedipine/4,4-dipyridyl (C) 15 2.4 Synthesis and Results 16 2.4.1 Nifedipine/formamide (A) 16 2.4.2 Nifedipine/isonicotinamide (B) 17 2.4.3 Nifedipine/4,4-dipyridyl (C) 17 2.5 Discussion 18 Chapter 3---Phenytoin 19 3.1 Description 19 3.2 Strategy 21 3.3 Structures 22 3.3.1 Phenytoin/4(1H)pyridone (A) 22 3.3.2 Phenytoin/4,4-dipyridyl (B) 23 3.3.3 Phenytoin/Trans-1,2-(4-pyridyl) ethylene 24 i

PAGE 4

3.4 Synthesis and Results 25 3.4.1 Phenytoin/4(1H)pyridone (A) 25 3.4.2 Phenytoin/4,4-dipyridyl (B) 26 3.4.3 Phenytoin/trans-1,2-(4-pyridyl) ethylene 26 3.5 Discussion 27 Chapter 4---Crown Ethers 29 4.1 Description 29 4.2 Polymorphism 30 4.3 Strategy 30 4.4 Structures 32 4.4.1 DB18C6/4-nitroaniline (A) 32 4.4.2 DB18C6/2-methyl-4-nitroaniline (B) 34 4.4.3 DB18C6/nicotinamide Polymorph A (C) 35 4.4.4 DB18C6/nicotinamide Polymorph B (D) 36 4.4.5 18C6/2,6-diaminopurine (E) 36 4.4.6 18C6/dicyandiamide (F) 37 4.4.7 18C6/5-aminoisophthalic acid (G) 38 4.4.8 18C6/tetrafluorisophthalic acid (H) 39 4.5 Synthesis and Results 39 4.5.1 DB18C6/4-nitroaniline (A) 39 4.5.2 DB18C6/2-methyl-4-nitroaniline (B) 40 4.5.3 DB18C6/nicotinamide Polymorph A (C) 40 4.5.4 DB18C6/nicotinamide Polymorph B (D) 41 4.5.5 18C6/2,6-diaminopurine (E) 42 4.5.6 18C6/dicyandiamide (F) 42 4.5.7 18C6/5-aminoisophthalic acid (G) 43 4.5.8 18C6/tetrafluorisophthalic acid (H) 43 4.6 Discussion 44 Chapter 5---Trimesic Acid 45 5.1 Porous Materials 45 5.2 TMA Polymorphic Forms 47 5.3 Strategy 48 5.4 Structures 49 5.4.1 TMA/biphenyl (A) 49 5.4.2 TMA/anthracene (B) 50 5.4.3 TMA/acetic acid (C) 51 5.4.4 TMA/naphthalene (D) 52 5.4.5 TMA/toluene (E) 53 5.5 Synthesis and Results 53 5.5.1 TMA/biphenyl (A) 53 5.5.2 TMA/anthracene (B) 54 5.5.3 TMA/acetic acid (C) 54 5.5.4 TMA/naphthalene (D) 55 5.5.5 TMA/toluene (E) 56 ii

PAGE 5

5.6 Discussion 57 Chapter 6---Conclusion 59 References 62 Appendices 65 Appendix A: Crystallographic data for nifedipine structures 66 Appendix B: Crystallographic data for phenytoin structures 67 Appendix C: Crystallographic data for crown ether structures 68 Appendix D: Crystallographic data for trimesic acid structures 69 iii

PAGE 6

List of Tables Table 1. Nifedipine synthon hydrogen bond distances 5 Table 2. Phenytoin synthon hydrogen bond distances 6 Table 3. Crown ether synthon hydrogen bond distances 7 Table 4. Trimesic acid synthon hydrogen bond distances 8 Table 5. Secondary hydrogen bond synthon distances 9 Table 6. Nifedipine supramolecular complexes 13 Table 7. Phenytoin supramolecular complexes 22 Table 8. Crown ether supramolecular complexes 31 Table 9. Trimesic acid COOH COOH synthon distances 46 Table 10. Trimesic acid supramolecular complexes 49 Table 11. TMA/toluene before and after TGA analysis 57 Table 12. Crystallographic data for nifedipine structures 65 Table 13. Crystallographic data for phenytoin structures 66 Table 14. Crystallographic data for crown ether structures 67 Table 15 Crystallographic data for trimesic acid structures 68 iv

PAGE 7

List of Figures Figure 1. Supramolecular homo & hetero synthons 2 Figure 2. Nifedipine supramolecular synthons 5 Figure 3. Phenytoin supramolecular synthons 6 Figure 4. Crown ether supramolecular synthons 7 Figure 5. Trimesic acid supramolecular synthons 8 Figure 6. Secondary hydrogen bond synthons 9 Figure 7. Pure nifedipine model 11 Figure 8. Nifedipine co-crystal formers 13 Figure 9. Nifedipine/formamide 14 Figure 10. Nifedipine/isonicotinamide 15 Figure 11. Nifedipine/4,4-dipyridyl 16 Figure 12. Pure nifedipine model 19 Figure 13. Phenytoin hydrogen-bonded motif 20 Figure 14. Phenytoin co-crystal formers 22 Figure 15. Phenytoin/4(1H)pyridone 23 Figure 16. Phenytoin/4,4-dipyridyl 24 Figure 17. Phenytoin/trans-1,2-(4-pyridyl) ethylene 25 Figure 18. Pure Dibenzo-18-crown-6 29 Figure 19. Pure 18-crown-6 30 v

PAGE 8

Figure 20. Crown ether co-crystal formers 32 Figure 21. DB18C6/4-nitroaniline 33 Figure 22. DB18C6/4-nitroaniline polar nanotubes 33 Figure 23. Polar axis views of DB18C6/4-nitroaniline polar nanotubes 34 Figure 24. DB18C6/2-methyl-4-nitroaniline 34 Figure 25. DB18C6/nicotinamide Polymorph A 35 Figure 26. DB18C6/nicotinamide Polymorph B 36 Figure 27. DB18C6/2,6-diaminopurine 37 Figure 28. 18C6/dicyandiamide 38 Figure 29. 18C6/5-aminoisophthalic acid 38 Figure 30. 18C6/tetrafluoroisophthalic acid 39 Figure 31. Pure trimesic acid model 45 Figure 32. Trimesic acid carboxylic acid synthon 46 Figure 33. Trimesic acid polymorphic forms 48 Figure 34. Trimesic acid Co-crystal formers 49 Figure 35. Trimesic Acid /biphenyl 50 Figure 36. Trimesic Acid/anthracene 50 Figure 37. Trimesic Acid/acetic acid 51 Figure 38. Trimesic Acid/naphthalene 52 Figure 39. Trimesic Acid/toluene 53 vi

PAGE 9

List of Abbreviations CSD= Cambridge Structural Database DB18C6= Dibenzo-18-crown-6 18C6= 18-crown-6 TMA= Trimesic acid or 1,3,5-benzenetricarboxylic acid FTIR=Fourier Transform Infra-red Bipy= 4,4-dipyridyl THF= tetrahydorfuran AcOH= Acetic acid TGA= Thermal Gravimetric Analysis vii

PAGE 10

Crystal Engineering of Binary Compounds Containing Pharmaceutical Molecules Leslie Ann Morales ABSTRACT The synthesis or the interaction between two or more molecules is known as supramolecular chemistry. The concept of supramolecular chemistry can be applied to the design of new pharmaceutical materials affording new compositions of matter with desirable composition, structure and properties. The design of a two-molecule, or binary, compound using complementary molecules represents an example of an application of crystal engineering. Crystal engineering is the understanding of intermolecular interactions, in the context of crystal packing, in the design of new solid materials. By identifying reliable connectors through molecular recognition or self-assembly, one can build predictable architectures. The study of supramolecular synthesis was accomplished using known pharmaceutical molecules such as Nifedipine (calcium channel blocker used for cardiovascular diseases) and Phenytoin (used as an anticonvulsant drug) and model compounds containing synthons common in pharmaceutical drugs (Crown ethers and viii

PAGE 11

Trimesic acid with ether linkages and carboxylic acid dimers, respectively) with complementary molecular additives. The co-crystals formed were characterized by various techniques (IR, m.p., XPD, single X-ray diffraction) and preliminary results were found to exhibit characteristics different from the parent compounds as a direct result of hydrogen bonding and self-assembly interactions. These crystalline assemblies could afford improved solubility, dissolution rate, stability and bioavailability. ix

PAGE 12

Chapter 1 Introduction 1.1 Supramolecular Chemistry and Crystal Engineering The concept of crystal engineering was introduced by Pepinsky in 1955 1 and was first applied by Schmidt in the context of covalent bond formation in the solid state, i.e. topochemical reactions. 2 More recent developments have expanded this concept into areas as diverse as supramolecular synthesis, crystal structure analysis and prediction. 3,4 Crystal engineering is a field that offers varying paths for the rational design of functional solids. Supramolecular synthesis offers a paradigm for the facile design and isolation of supramolecular complexes using self-assembly of different, but complementary supramolecular synthons. The subsequent ability to generate a diverse range of multiple component crystalline solids affords new compositions of matter and an ability to engage in systematic analysis of the factors that control structure/function relationships in the solid state. Self-assembly of pre-selected molecular components, 5 or supramolecular synthesis, 4,6 represents a paradigm for the generation of chemical structures with nanoscale features. Supramolecular synthetic approaches offer a number of attractive features: the ability to exploit readily available molecules or ions in novel ways, their inherent modularity 1

PAGE 13

affords compositional and structural diversity, and the use of non-covalent interactions (such as ion-ion, ion-dipole, dipole-dipole, stacking, Van der Waals forces and hydrogen bonding) makes for facile synthesis. Supramolecular synthesis offers a combinatorial type of approach to synthesis of new structures and can afford a wide range of structural diversity without the need to break and form covalent bonds. The paradigm is exemplified by the use of geometric considerations to guide the preparation of nanoscale hydrogen bonded polyhedra 7 and by development of rational approaches to the design of ternary structures.The fact that hydrogen bonds are formed in a hierarchical fashion (strongest H-bond acceptor to strongest H-bond donor, and so on) and that crystal stability depends on specific intermolecular interactions 8 is the basis for this arrangement. (a)(a)(b)(b) ( c ) Figure 1. Schematic representation of (a) supramolecular homosynthon (i.e. carboxylic acid dimer), (b) supramolecular heterosynthon (i.e. a carboxylic acidpyridyl dimer), (c) supramolecular heterosynthon (i.e. ether linkage-amine). 1.2 Supramolecular Synthons Supramolecular synthons 6 depict the possible ways in which complementary functionalities of molecules interact by non-covalent interactions. One of such 2

PAGE 14

interactions is the hydrogen bond. A hydrogen bond is a strong, directional, selective and stabilizing interaction between a hydrogen atom and an electronegative atom. As revealed by Figure 1, supramolecular synthons can either be self-complementary (a supramolecular homosynthon such as a carboxylic acid dimer 9 ) or the result of complementary interactions between different moieties (a supramolecular heterosynthon such as a DNA base pair or the well-known carboxylic acid-pyridyl interaction 10 ). The latter brings with it the opportunity to build multiple component superstructures and is the main focus of this study. 1.3 Polymorphism The occurrence of polymorphism in crystal engineering cannot be ignored. Polymorphism is defined as the phenomenon where the same chemical substance exists in different crystalline forms or different crystalline patterns. 11 This trend is more common among molecules with flexible conformations capable of hydrogen bonding. 6 The internal arrangement of molecules in the solid state has a pronounced influence on chemical and physical properties. In this context, the existence of more than one crystalline form of a given compound, typically in the form of polymorphs or solvates, represents both a problem and an opportunity 12 in organic synthesis and early stage development of pharmaceuticals. Supramolecular synthons are the structural units within molecules, which can be formed and/or assembled by synthetic operations 6 The supramolecular synthons present in polymorphic forms may be intact from form to form, therefore it can be stated that the supramolecular equivalents of structural isomers are crystalline polymorphs 6 The criteria 3

PAGE 15

for assessing the existence of polymorphs are different unit cell parameters, crystal packing arrangements and physical properties. 11 The study of the phenomenon of polymorphism has many implications. It has increased due to the great importance of legal and regulatory issues that faces the pharmaceutical industry to date. 13 The US Patent Office permits a new phase (i.e. polymorph, solvate or binary compound) of a drug to be patented if there is any processing or bioavailability benefit to the new solid phase. This may be problematic since many drug molecules are prone to polymorphism and crystal size and morphology can vary for any given phase. This can have significant commercial relevance because patents can be extended or effectively voided, translating into monetary gain or loss. 1.4 Cambridge Structural Database As previously stated, crystal engineering and supramolecular chemistry are intricately intertwined and the common theme between the two is the connection between structure and architecture control via intermolecular interactions, namely hydrogen bonding between supramolecular synthons. Self-assembly of the targeted synthons through hydrogen bonding of complementary functional groups contributes to the control and possible prediction of the supramolecular structure, thus changing the physical properties of the material. Rationalization and design of these supramolecular solids is the basis of crystal engineering. A survey of existing supramolecular synthons and structures can be carried out through the Cambridge Structural Database (CSD). 14 By analyzing the existing structures and/or synthons in the CSD, new compositions of matter can be obtained, achieving new phases with new properties. An extensive analysis of the 4

PAGE 16

supramolecular synthons involved in the compounds described in this thesis was carried out and a library with average hydrogen bond distances was formulated. The CSD has a total of 296,427 hits as of July 2003. Out of the 296,427 hits, 127,003 hits are considered organic, and out of that subset, 118,322 are organics with no metals, which accounts for ~40% of the total hits in the CSD. 14 All searches had as search parameters R factor <0.075, no ions and organics only. All searches were done with specific hydrogen bond distances, which were narrowed down through statistics and histograms. The distances ranged from the two non-hydrogen atoms sharing the hydrogen atom since the hydrogen atom cannot be given a specific position due to its movement. Nifedipine Synthons 1.4.1 NO O HN N HN O HN O O HN 1 2 3 4 Figure 2. Supramolecular synthons in nifedipine Table 1 Nifedipine Hydrogen bonded synthons and distances. # hits distance min Max ave std Single* Multi* 1 147 2.8-3.25 2.801 3.247 3.074 0.1186 122 25 2 55 2.8-3.4 2.847 3.391 3.148 0.1313 48 7 3 244 2.8-3.3 2.806 3.298 3.006 0.1172 224 20 4 118 2.8-3.1 2.801 3.097 2.917 0.0752 106 15 *Single and multi-component corresponds to the homo/heterosynthons being formed by the synthons on the same molecules (single) or the formation between two molecules (multi) with complementary synthons. 5

PAGE 17

Nifedipine has three possible hydrogen-bonding sites, the secondary amine (H-bond donor), the ester carbonyl (H-bond acceptor) and the nitro group (H-bond acceptor). Out of these three possible sites, the secondary amine is involved in hydrogen bonding most often and there are many more structures or hits of this interaction in the CSD as can be seen above. A secondary amide is involved in all of the CSD searches that were carried out in the nifedipine analysis. The ester carbonyl and the nitro group form secondary hydrogen bonds. nifedipine will always be involved in a heterosupramolecular synthon. The use of the secondary amine as a designing tool seems promising since interactions with this functional group is very prevalent in the CSD. Phenytoin Synthons 1.4.2 N HN O HN ON H N H O O X X HN X X 1 2 3 4 Figure 3. Supramolecular synthons in phenytoin Table 2. Phenytoin Hydrogen bonded synthons and distances. # hits Distance min max mean Single* Multi* 1 970 2.4-3.3 2.707 3.297 2.908 0.1015 804 166 2 340 2.7-3.2 2.701 3.2 2.939 0.1195 283 57 3 1441 2.7-3.2 2.7 3.198 2.924 0.1014 1224 217 4 147 2.8-3.25 2.801 3.247 3.074 0.1186 122 25 *Single and multi-component corresponds to the homo/heterosynthons being formed by the synthons on the same molecules (single) or the formation between two molecules (multi) with complementary synthons. 6

PAGE 18

Phenytoin has two hydrogen bond donors (secondary amine) and acceptors (carbonyls) and they are complementary with one another as seen from the CSD searches. Search number 2 is a subset of search 3 and makes the results more relevant toward phenytoin. Phenytoin can hydrogen bond as amide dimers or as heterosynthons between complementary functional groups or synthons present on different molecules. As seen in the table above, the amide dimer is prevalent in structures containing the same molecule and the average distance is very similar when considering other ways of forming synthons. Crown Ethers Synthons 1.4.3 O HN X O HN H O HN N 1 2 3 O HN O HN X + 18C6 4 5 Figure 4. Supramolecular synthons in crown ethers Table 3. Possible ether hydrogen bonded synthons and distances. # hits distance min max ave std Single* Multi* 1 473 2.8-3.35 2.801 3.349 3.112 0.1319 328 141 2 225 2.8-3.35 2.801 3.348 3.103 0.1256 133 92 3 36 2.8-3.35 2.801 3.345 3.084 0.1226 26 10 4 446 2.8-3.35 2.801 3.349 3.114 0.1323 319 127 5 21 2.8-3.35 2.801 3.314 3.062 0.1246 21 0 *Single and multi-component corresponds to the homo/heterosynthons being formed by the synthons on the same molecules (single) or the formation between two molecules (multi) with complementary synthons. 7

PAGE 19

Crowns ethers are deficient in hydrogen bond donors; therefore they are very susceptible to forming hydrogen bonds when molecules with amines or any other complementary hydrogen bonding functional group are present. Searches 2, 3 and 4 are all subsets of search 1. Search 1 encompasses all possibilities, from primary amines, secondary amides, pyridines, etc. Most of the structures or hydrogen-bonded synthons found in the CSD regarding ethers involved the same molecule. On the contrary, when 18-crown-6 (18C6) was specified as the ether to be analyzed, none of the 21 hits involved single molecules since 18C6 only has hydrogen bond donors. In this case, molecules with complementary functional groups for hydrogen bonding were needed. This is an example of how new phases and compositions of matter can be manipulated or designed in order to achieve different physical properties. Trimesic acid Synthons 1.4.4 O OH O OH 1 Figure 5. Supramolecular synthons in trimesic acid Table 4. Trimesic acid carboxylic acid dimer hydrogen bond distances # hits distance min max ave std Single* Multi* 1 1193 2.4-3.2 2.554 3.196 2.657 0.0644 1059 134 *Single and multi-component corresponds to the homo/heterosynthons being formed by the synthons on the same molecules (single) or the formation between two molecules (multi) with complementary synthons. Trimesic acid forms honeycomb networks by means of carboxylic acid dimers. The carboxylic acid dimer is the basis for all of the TMA structures that will be discussed in 8

PAGE 20

Chapter 5. There are many forms of TMA and all involve the carboxylic acid dimer. The new structures incorporate a neutral guest in the cavities formed by six trimesic acid molecules bonding in a honeycomb fashion (~14 pore diameter). The distances of the hydrogen bonds between the carboxylic acid functional groups on separate molecules are very consistent when it comes to the network. 1.4.5 Secondary Hydrogen Bonds H H H X N H O X X 1 2 Figure 6. Secondary hydrogen bonds involving methyl hydrogens and hydrogen bond acceptors nitrogen and oxygen. There are two secondary hydrogen bonds which need to be specified since they are used to further stabilize the crystal packing of the supramolecular structures discussed in this work. Synthon 1 is seen in the nifedipine/4,4-dipyridyl structure while synthon 2 is seen in the phenytoin/trans-1,2-(4-pyridyl) ethylene structure. Further discussion on all of these structures will be available in the following chapters. Table 5. Possible secondary hydrogen bond synthons and distances # hits distance min max ave std Single* Multi* 1 10317 3.3-4.2 3.4 3.199 3.8872 0.1038 8606 1711 2 13921 3.35-4.2 3.349 4.199 3.8976 0.1995 11766 2155 *Single and multi-component corresponds to the homo/heterosynthons being formed by the synthons on the same molecules (single) or the formation between two molecules (multi) with complementary synthons. 9

PAGE 21

This thesis is based on the principles of crystal engineering in order to generate new compositions of matter through supramolecular synthesis and self-assembly of complementary synthons. This presents an increase in the potential need to control solid-state structures, changing their physical properties and leading to improved solubility and bioavailability. The study chose pharmaceutical molecules as the target system since they contain a wide range of synthons and understanding of the crystal structure physical property relationship is essential for the continued application of these essential chemicals toward improving the overall health of society. 10

PAGE 22

Chapter 2 Nifedipine 2.1 Description N H NO2 COOCH3H3COOCCH3 CH3 Figure 7. Pure nifedipine stick models. Oxygen atoms are in red (hydrogen bond acceptor); nitrogen atoms are in blue (hydrogen bond donor). All hydrogen atoms on the figure to the right have been omitted for clarity. Nifedipine, 15 with MW of 246.34g/mol and m.p. of 172-174C, is a calcium antagonist used in the treatment of cardiovascular disorders such as hypertension and angina pectoris. Nifedipine exhibits low water solubility (less than 10g/mL) and erratic bioavailability. There is a direct correlation between structure and function of new pharmaceutical phases of nifedipine. Modifying the structure of the compound, thus creating new and different phases, may affect different properties. Differing the phases or crystalline self-assemblies of nifedipine may improve the drug's solubility, dissolution rate, stability and bioavailability. 11

PAGE 23

Dihydropyridine chemistry 16 began in 1882 when Hantzsch first reported dihydropyridines as stable intermediates in the pyridine synthesis that bears his name. The recent interest in dihydropyridines can be traced to the coenzyme reduced nicotinamide adenine dinucleotide (NADH) and the unique ability of this compound in biological systems to reduce unsaturated functional groups. 15 In 1949 A.P. Phillips reported on the weak analgesic and curare-like activities of a few dihydropyridine derivatives. In the search for orally active drugs for the treatment of coronary insufficiency, F. Bosset synthesized in 1966 a compound designated Bay a 1040, which following its introduction as nifedipine in 1975 has since become one of the major cardiovascular drugs. Nifedipine 16 is sensitive to light in solid form and extremely sensitive to light in dissolved state in solution. In the crystal lattice, the almost flat dihydropyridine ring lies at a practically perpendicular angle to the nitrophenyl group, the ortho-nitro group facing away from the dihydropyridine ring. Under the influence of visible and ultra-violet light nifedipine in solutions is converted into a nitroso compound. 2.2 Strategy In view of the fact that nifedipine contains hydrogen bond acceptors and donors, complementary molecules can be compounds with both acceptor and donor synthons. The main hydrogen-bonding site is the secondary amine, which can bond to carbonyl as well as nitrogen (e.g. pyridines). Crystals of A, B and C were obtained via slow evaporation of stoichiometric amounts of nifedipine and cocrystal formers in the appropriate solvents. All crystallization 12

PAGE 24

experiments were conducted in an unmodified atmosphere and the solvents were dried by standard methods prior to use. All chemicals used are commercially available and were purchased from Aldrich Table 6. Nifedipine Supramolecular Complexes N ifedipine/formamide A N ifedipine/isonicotinamide B N ifedipine/4,4'-dipyridyl C NH2 O H N NH2 O N N FORMAMIDE ISONICOTINAMIDE 4,4'-DIPYRIDYL Figure 8. Structures of the co-crystal formers used in the supramolecular complexes involving nifedipine. 2.3 Structures 2.3.1 Nifedipine/formamide (A) Crystallization of nifedipine from formamide results in a 1:1 supramolecular complex that is sustained by amine-carbonyl H-bonds. In A, as revealed by Figure 3 (synthons 3), one of the hydrogen atoms in the amino moiety of the formamide hydrogen bonds to the carbonyl oxygen of the ester on one nifedipine molecule with bond distance of 3.024, which falls well within the distance range found in the CSD and stated in chapter 1 (nifedipine search 3) of 2.8-3.3 The other amine hydrogen atom found in formamide is engaged in a hydrogen bond with the nitro oxygen on another nifedipine 13

PAGE 25

molecule. The distance for this bond is 3.155 and falls within the range obtained from the CSD search (nifedipine search 2) for this synthon (2.8-3.4 ). Formamide has two hydrogen bond donors (the amine hydrogens) but also has a hydrogen bond acceptor (the aldehyde carbonyl). The carbonyl is also involved in a hydrogen bond with a third nifedipine molecule. The carbonyl is bonded to the secondary amine of the nifedipine molecule and has a distance of 2.906 and falls within the range set from the CSD search (nifedipine search 4) of 2.8-3.1 Figure 9. The 1:1 adduct formed between nifedipine and formamide. The hydrogen-bonded motif is also shown. 2.3.2 Nifedipine/isonicotinamide (B) Crystallization of nifedipine and isonicotinamide from methanol results in a 1:1 supramolecular complex that is sustained by amine-carbonyl and amine-pyridine H-bonds. In B, as revealed by Figure 4, one of the hydrogen atoms in the amino moiety of the Isonicotinamide hydrogen bonds to the carbonyl oxygen of the ester on one nifedipine molecule with bond distance of 3.097 which falls well within the distance range found in the CSD and stated in chapter 1 (nifedipine search 3) of 2.8-3.3 The other amine 14

PAGE 26

hydrogen atom found in isonicotinamide is engaged in a hydrogen bond with another isonicotinamide molecule. The distance for the amine hydrogen and the pyridine nitrogen bond on separate isonicotinamide molecules is 2.911 and falls within the range obtained from the CSD search (nifedipine search 1) for this synthon (2.8-3.25 ). Isonicotinamide has two hydrogen bond donors (the amine hydrogens) but also has a hydrogen bond acceptor (the amide carbonyl). This carbonyl oxygen is involved in a hydrogen bond to a second nifedipine molecule. The carbonyl is bonded to the secondary amine of the nifedipine molecule and has a distance of 2.878 and falls within the range set from the CSD search (nifedipine search 4) of 2.8-3.1 B differs from A in that a new synthon is involved, the amine-pyridine hydrogen bond is seen in B which facilitates many more arrangements and new phases for nifedipine. Figure 10. The 1:1 adduct formed between nifedipine and isonicotinamide. 2.3.3 Nifedipine/4,4-dipyridyl (C) Crystallization of nifedipine and 4,4-dipyridyl from methanol results in a 1:1 supramolecular complex which is sustained by amine-pyridine H-bonds. 4,4-dipyridyl is only a hydrogen bond acceptor; therefore the secondary amine in the nifedipine is the 15

PAGE 27

most likely hydrogen bonding site. In C, as revealed by Figure 5, the amine-pyridine hydrogen bond is evident with a bond distance of 3.134. This distance does fall within the nifedipine CSD search from chapter 1 (search 1, 2.8-3.25 ). A secondary hydrogen bond is formed between one methyl hydrogen (from the ester group) and the other pyridine nitrogen available. This bonds purpose is for stability in the crystal packing and the distance is 3.295 The range distance for this secondary hydrogen bond is 3.3-4.2 ; therefore, it is on the lower limit of this type of bond (Search 1, section 1.25). Figure 11. The 1:1 adduct formed between nifedipine and 4,4-dipyridyl. 2.4 2.4.1 Synthesis and Results Nifedipine/formamide (A) Synthesis: In a typical reaction, 32 mg (1.29x10 -4 mol) of nifedipine was dissolved in 2 mL of formamide. The reaction vial was kept tightly capped to avoid any solvent evaporation. The vial was covered in aluminum foil to protect from light. In approximately two days yellow crystals appeared. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix A Crystal packing: Crystallization of nifedipine from formamide results in a 1:1 supramolecular complex that is sustained by amine-carbonyl H-bonds. Distances range between 3.024-3.155. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3443, 3303 cm -1 (N-H stretch, 1 amine, formamide); 1684 cm -1 (C=O stretch, formamide) 1310, 1248, 1217 cm -1 (C-O stretching). 16

PAGE 28

Melt-temp: 136-138C X-ray powder diffraction: in 2, simulated derived from the single crystal data; experimental (simulated): 8.461(8.850), 9.955(9.923), 12.682(12.893), 13.080(13.420), 14.954(15.307), 16.966(17.022), 17.082 (17.034), 21.036(20.827), 23.539(23.437), 24.658(24.682), 26.725(26.715), 27.039(27.028). 2.4.2 2.4.3 Nifedipine/isonicotinamide (B) Synthesis: In a typical reaction, 31 mg (1.26x10 -4 mol) of nifedipine and 28 mg (2.29x10 -4 mol) of isonicotinamide were dissolved in approximately 1 mL of methanol. Slow evaporation of the solvent for approximately a week yielded 1:1 crystals. The reaction vial was covered in aluminum foil to prevent light degradation. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix A Crystal packing: Crystallization of nifedipine and isonicotinamide from methanol results in a 1:1 supramolecular complex that is sustained by amine-carbonyl and amine-pyridine H-bonds. Distances range between 2.878-3.097. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3408 cm -1 (N-H stretch, 2 amine); 3286 cm -1 (-OH stretch); 1669 cm -1 (C=O stretch); 1209, 1118 cm -1 (C-O stretching). Melt-temp: 148-153C X-ray powder diffraction: in 2, simulated derived from the single crystal data; experimental (simulated): 10.418(10.636), 11.561(11.823), 11.880(11.849), 13.003(13.212), 19.404(19.739), 19.660(19.771), 21.142(21.167), 22.739(22.779), 23.739(23.774), 26.399(26.407), 31.820(31.806). Nifedipine/4,4-dipyridyl (C) Synthesis: In a typical reaction, 30 mg (1.22x10 -4 mol) of nifedipine and 43 mg (2.75x10 -4 mol) of 4,4-dipyridyl were dissolved in approximately 1 mL of methanol. Slow evaporation of the solvent for approximately two week yielded 1:1 nifedipine/4,4-dipyridyl crystals. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix A Crystal packing: Crystallization of nifedipine and 4,4-dipyridyl from methanol results in a 1:1 supramolecular complex that is sustained by amine-pyridine H-bonds. 4,4-Dipyridyl is only a hydrogen bond acceptor; therefore the secondary amine in the 17

PAGE 29

nifedipine is the most likely hydrogen bonding site. Distances range between 3.134-3.295. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3275 cm -1 (N-H stretch, 2 amine); 3213 (-OH stretch); 1687 cm -1 (C=O stretch); 1263, 1208 cm -1 (C-O stretching). Melt-temp: 137-139C X-ray powder diffraction: experimental done, calculated could not be done. 4.039, 11.459, 12.281, 19.362, 20.200, 21.601, 24.802, 24.958, 25.340, 25.419. 2.5 Discussion Presented above were three examples of the use of supramolecular synthesis to generate new compositions of matter containing molecules of interest to materials scientists. These supramolecular complexes show how secondary amines, nitro groups and carbonyls present in pharmaceutical drugs can be exploited via crystal engineering. This has increased the potential to control solid-state structures, changing their physical properties and leading to improved solubility and bioavailability. Pure nifedipine H-bonds forming a chain, secondary amine to ether carbonyl, but as seen in the previous co-crystals, other synthons are exploited and the crystal structure therefore has to adjust. Careful analysis of the existing supramolecular synthons on nifedipine provides an educated guess that allows scientists to choose other molecules with complementary components, designing supramolecules. The exact connection sites for the synthons cannot always be predicted since other factors such as solvent, sterics and number of possible hydrogen bonding sites has a powerful effect on the outcome. 18

PAGE 30

Chapter 3 Phenytoin 3.1. Description N N O H O H Figure 12. Pure phenytoin stick models. Oxygen atoms (in red) are the hydrogen bond acceptors and the nitrogen atoms (in blue) are the hydrogen bond donors. Hydrogen atoms have been deleted for clarity. Phenytoin, 17 or 5,5-diphenylhydantoin, is a pharmaceutical drug used for the treatment of seizures. It is sparingly soluble in water (20-25mcg/mL). Compounds with aliphatic side groups at position C5 are commonly used as sedatives, and phenyl rings at this position lead to anticonvulsant properties without the need to impair consciousness. It is also sparingly soluble in ethanol and acetone and has a M.W. of 252.272g/mol and m.p. of 295C. According to Chakrabati, 18 phenytoin exists in two forms. Form 1: rodlike plates or regular crystalline shape, and form 2: needle shaped. The existence of these two forms is due to differences in the crystalline state. 19

PAGE 31

In 1938 Merrit and Putnam 19 found that 5,5-diphenylhydantoin (phenytoin) showed anticonvulsant properties. Twelve years later it was stated that apart from antiepileptic activity, phenytoin possessed antiarrhythmic activity. It belongs to class 1b according to Vaughan Williams classification of antiarrhythmic agents modulating voltage-gated sodium channels conductance. Limited clinical effectiveness and undesired side effects are the driving force to using supramolecular chemistry in order to change the structure, therefore changing the physical properties since there is a direct correlation between the two. Diphenylhydantoin and phenobarbital are the most widely used drugs in the treatment of grand mal, psychomotor, and focal epilepsies. The similarity of barbiturates to uracil has prompted investigations into their hydrogen-bonding capabilities with adenine molecules. Some of the biological effects of these drugs may derive from an ability to specifically bind and inactivate adenine-containing nucleic acids or coenzymes. 20 Comparisons of the complexes formed by these drugs with adenine containing molecules may prove helpful in understanding the modes of action of these drugs. 20 Figure 13. Hydrogen bonded motif of pure phenytoin. The crystals are orthorhombic and systematic absences indicate space groups Pnma or Pn2 1 a. Pn2 1 a (centrosymmetric space group), indicated by structure analysis, was yielded by the hydantoin group atomic positions even though the true space group for phenytoin 20

PAGE 32

is non-centrosymmetric, but the proximity of the hydantoin group plane was very close to 0 or (approximating a centrosymmetric arrangement in the unit cell). The hydrogen-bonded motif, figure 2, shows the hydrogen bonds between amine hydrogens and carbonyl oxygens between phenytoin molecules with distances ranging from 2.784-2.844 These distances fall within the range obtained from the CSD search in Chapter 1 (phenytoin search 2) which is 2.7-3.2 3.2 Strategy In view of the fact that phenytoin has both hydrogen bond acceptors (carbonyl oxygens) as well as hydrogen bond donors (secondary amines), a complementary molecule can be one that has at least one hydrogen bond acceptor moiety (example amines, alcohols, carboxylic acids, etc.) or at least one hydrogen bond donor moiety (amines, amides, alcohols, etc). Phenytoin has the added characteristic that it can form amide dimers between two phenytoin molecules since the synthon is complementary with itself. Crystals of A, B, and C were obtained via slow evaporation of stoichiometric amounts of phenytoin and cocrystal formers in the appropriate solvents. All crystallization experiments were conducted in an unmodified atmosphere and the solvents were dried by standard methods prior to use. All chemicals used are commercially available and were purchased from Aldrich 21

PAGE 33

Table 7. Phenytoin Supramolecular Complexes Phenytoin/4(1H) pyridone A Phenytoin/4,4'-dipyridyl B Phenytoin/trans-1,2-bis(4-pyridyl) ethylene C N H H H H H O N N N N 4(1H)-PYRIDONE 4,4'-DIPYRIDYL TRANS-1,2-BIS(4-PYRIDYL)ETHYLENE Figure 14. Structures of the co-crystal formers used in the supramolecular complexes involving phenytoin. 3.3 Structures 3.3.1 Phenytoin/4(1H)-pyridone (A) Crystallization of phenytoin and 4-hydroxypyridine (which later tautomerized to 4(1H)-pyridone) from ethanol results in a 1:1 supramolecular complex that is sustained by amine-carbonyl H-bonds between phenytoin molecules and pyridone molecules. In A, as revealed by Figure 4, there is a chain formed of phenytoin molecules. The pyridones also arrange consecutively with one another forming another chain by means of N-H O=C bonds. These hydrogen bonds average 2.665 well within the range obtained from the CSD (2.7-3.2 ). These two independent chains, phenytoin and pyridone, are connected to each other by means of another N-H O=C bond. The carbonyl present in the pyridone molecule is bifurcated, H-bonding to both the secondary amine in phenytoin as well as the secondary amine in the pyridone molecule. In the crystal packing of A one 22

PAGE 34

can observe that two pyridone chains are sandwiched between two phenytoin chains. If one examines figure 2 one can see how that motif was simply expanded enough to fit the two H-bonded pyridone chains. Figure 15. The 1:1 adduct formed between phenytoin and 4(1H) pyridone. The hydrogen bonded crystal packing is shown to the left. 3.3.2 Phenytoin/4,4-dipyridyl (B) Crystallization of phenytoin and 4,4-dipyridyl from 2,4-pentanedione results in a 1:1 supramolecular complex that is sustained by amine-pyridine H-bonds between the phenytoin and 4,4-dipyridyl (Bipy) molecules as well as phenytoin amide dimers. In B, (Figure 5) there is a chain of phenytoin and bipy molecules formed. The N-H N distance is 2.861 and falls within the distance range obtained from the CSD search in Chapter 1 (phenytoin search 4), which is 2.8-3.25 The phenytoin molecules also H-bond and form amide dimers. The hydrogen bond distance between the N-H O=C dimers on separate phenytoin molecules is 2.791 and according to the information obtained from the CSD searches on Chapter 1 (phenytoin search 1) fall within the distance range of 2.4-3.3. The phenytoin molecules arrange in dimers and the secondary amine not involved in the dimer is available to H-bond to the pyridine nitrogen 23

PAGE 35

of the 4,4-dipyridyl, forming a zigzag chain. The other carbonyl is not involved in any other hydrogen bonds. The crystal packing of this supramolecular structure orients the hydrophobic phenyl rings on the inside of the chains, keeping the hydrophilic parts of the structure on the outside and susceptible to forming bonds with complementary synthons (figure 5). Figure 16. The 1:1 adduct formed between phenytoin and 4,4-dipyridyl. The hydrogen bonded crystal packing is shown. 3.3.3 Phenytoin/trans-1,2-bis(4-pyridyl) ethylene (C) Crystallization of phenytoin and trans-1,2-bis(4-pyridyl) ethylene from 2,4-pentanedione results in a 1:5 supramolecular complex that is sustained by amine-pyridine H-bonds between the phenytoin and trans-1,2-bis(4-pyridyl) ethylene molecules as well as dimers between the phenytoin amides. In C (Figure 6), there is a chain formed of phenytoin and trans-1,2-bis(4-pyridyl) ethylene molecules. The N-H N distance is 2.844 and falls within the distance range obtained from the CSD search on Chapter 1 (phenytoin search 4) which is 2.8-3.25 The phenytoin molecules also H-bond and form amide dimers. The hydrogen bond distance between the N-H O=C on separate phenytoin molecules is 2.891 and according to the information obtained from the CSD searches on Chapter 1 (phenytoin search 1) falls within the distance range of 2.4-3.3. Structure C is very similar to structure B in its hydrogen bonding, the only difference is 24

PAGE 36

that in addition to forming dimers between the phenytoin molecules and N-H N bond to the additive, there are also additives interacting in a secondary hydrogen bond with the available phenytoin carbonyl. The bond distance is 3.386 and falls within the range for the CH O carbonyl synthon (range 3.1-3.9 ). This hydrogen bond permits the trans-1,2-bis(4-pyridyl) ethylene to stay in the cavity. The phenytoin molecules arrange in dimers and the secondary amine not involved in the dimer is available to H-bond to the pyridine nitrogen of the trans-1,2-bis(4-pyridyl) ethylene, forming a zigzag chain. The crystal packing of this supramolecular structure arranges in such a way that allows for the accommodation of an additive molecule in the hydrophobic cavity. Figure 17. The 1:5 adduct formed between phenytoin and trans-1,2-bis(4-pyridyl) ethylene. The hydrogen bonded crystal packing is shown. 3.4 Synthesis and Results 3.4.1 Phenytoin/4(1H)-pyridone (A) Synthesis: In a typical reaction, 28 mg (1.11x10 -4 mol) of phenytoin and 11 mg (1.16x10 -4 mol) of 4-hydorxypyridine (which later tautomerizes to 4(1H)pyridone) were dissolved in approximately 2 mL ethanol and 1 mL acetone. The reaction vial was kept tightly capped to avoid solvent evaporation. After a few weeks clear 1:1 crystals formed. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix B Crystal packing: Crystallization of phenytoin and 4-hydroxypyridine (which later tautomerized to 4(1H)-pyridone) from ethanol results in a 1:1 supramolecular complex 25

PAGE 37

that is sustained by amine-carbonyl H-bonds between phenytoin molecules and pyridone molecules. Distances averaged 2.665. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3311 cm -1 (N-H stretch, 2 amine, phenytoin); 1758, 1711, 1497cm -1 (C-O stretching). Melt-temp: 234-239C X-ray powder diffraction: in 2, simulated derived from the single crystal data; experimental (simulated): 5.336(5.212), 11.332(11.137), 15.250(15.097), 16.419(16.228), 17.000(17.763), 17.977(18.019), 19.364(19.362), 19.679(19.821), 20.055(20.341), 21.417(21.160), 23.724(23.341), 26.348(26.054), 26.628(26.418), 27.643(27.260), 29.868(29.479). 3.4.2 Phenytoin/4,4-dipyridyl (B) Synthesis: In a typical reaction, 28 mg (1.11x10 -4 mol) of phenytoin and 19 mg (1.22x10 -4 mol) of 4,4-dipyridyl were dissolved in approximately 1 mL of 2,4-pentanedione. Slow evaporation of the solvent for a few weeks yielded 1:1 crystals. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix B Crystal packing: Crystallization of phenytoin and 4,4-dipyridyl from 2,4-pentanedione results in a 1:1 supramolecular complex that is sustained by amine-pyridine H-bonds between the phenytoin and 4,4-dipyridyl (bipy) molecules as well as phenytoin amide dimers. Distances range between 2.791-2.861. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). N/A Melt-temp: 225-227C X-ray powder diffraction: experimental done, calculated could not be done. 12.974, 17.461, 18.798, 20.613, 22.180, 24.902, 26.717, 28.599, 31.931. 3.4.3 Phenytoin/trans-1,2-bis(4-pyridyl) ethylene (C) Synthesis: In a typical reaction, 29 mg (1.15x10 -4 mol) of phenytoin and 108.35 mg (5.93x10 -4 mol) of trans-1,2-bis(4-pyridyl) ethane were dissolved in approximately 1 mL 2,4-pentanedione. Slow evaporation of the solvent for approximately a week yielded yellow 1:5 crystals. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix B 26

PAGE 38

Crystal packing: Crystallization of phenytoin and trans-1,2-bis(4-pyridyl) ethane from 2,4-pentanedione results in a 1:5 supramolecular complex that is sustained by amine-pyridine H-bonds between the phenytoin and trans-1,2-bis(4-pyridyl) ethane molecules as well as dimers between the phenytoin. Distances range between 2.844-3.273. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3264, 3194 cm -1 (N-H stretch, 1 amine); 3066, 3029 cm -1 alkenes; 1713 cm -1 (C=O stretch); 1310, 1248, 1209 cm -1 (C-O stretching), 985 cm -1 (aromatic hydrogens). Melt-temp: 254-257C X-ray powder diffraction: experimental done, calculated could not be done. 15.919, 19.370, 19.866, 2.159, 22.089, 22.215, 23.815, 34.826, 36.701. 3.5 Discussion Presented above were three examples of the use of supramolecular synthesis to generate new compositions of matter that contains molecules of interest to materials scientists. These supramolecular complexes show how carbonyl and secondary amines if viewed separately, or known as amides if both synthons (functional groups) are viewed together, in pharmaceutical drugs can be exploited via crystal engineering. This has increased the potential to control solid-state structures, changing their physical properties and leading to improved solubility and bioavailability. Pure phenytoin makes full use of all available synthons when hydrogen bonding, forming a tape. In the co-crystals, modifications to this hydrogen bonded motif are necessary when accommodating other complementary compounds. Not all available synthons will be exploited, as see in structure A, where one of the two carbonyls available in each phenytoin molecule is not involved in a hydrogen bond with another phenytoin molecule, additive molecule or solvent molecule. In structure B, the supramolecular tape found in pure phenytoin has been expanded to accommodate the bipy molecules. Phenytoin molecules still hydrogen 27

PAGE 39

bond to each other forming dimers. One secondary amine is involved in the hydrogen bond with the bipy molecule, but the other carbonyl is not involved in a synthon (as seen with structure A). Structure C is very similar to structure B, having the same hydrogen bonding with the additive and also an available carbonyl not involved in a hydrogen bond, but differs in that it fits another additive molecule in the cavities formed from the arrangements of the Phenytoin/trans-1,2-(4-pyridyl) ethylene chains. This elucidates the fact that just because the same synthons are being used, does not mean that the same arrangements will be seen from structure to structure. 28

PAGE 40

Chapter 4 Crown Ethers 4.1 Description Crown ethers are toxic compounds and serve as model compounds for pharmaceutical molecules that contain ether linkages. Ether moieties are deficient in H-bond donors and contain excess H-bond acceptors. They are therefore predisposed to become involved in supramolecular heterosynthons and it should not be surprising that well known ethers such as 18-Crown-6 21 (18C6) and Dibenzo-18-Crown-6 22 (DB18C6) form co-crystals and solvates with complementary molecules. Since the crown ether has only hydrogen bond acceptors (oxygen atoms), a complementary molecule should be one that has at least one hydrogen bond donor moiety (amines, alcohols, carboxylic acids, etc.). The pure phase for 18C6 21 has been isolated, and the pure phase for DB18C6 has been solved, but is pending publication. OOOOOO Figure 18. Pure Dibenzo-18-crown-6 stick models. Oxygens (in red) are the hydrogen bond acceptors. 29

PAGE 41

OOOOOO Figure 19. Pure 18-crown-6 stick models. Oxygens (in red) are the hydrogen bond acceptors. 4.2 Polymorphism The phenomenon of polymorphism in crystal engineering cannot be ignored. Polymorphism is defined as the phenomenon where the same chemical substance exists in different crystalline forms or different crystalline patterns. 11 This phenomenon is more common among molecules with flexible conformations capable of hydrogen bonding. 6 Supramolecular synthons are the structural units within molecules, which can be formed and/or assembled by synthetic operations. 6 The supramolecular synthons present in polymorphic forms may be intact from form to form, therefore it can be stated that the supramolecular equivalents of structural isomers are crystalline polymorphs. 6 The criteria for assessing the existence of polymorphs are different unit cell parameters, crystal packing arrangements and physical properties. 6 4.3 Strategy In view of the fact that the crown ethers only have hydrogen bond acceptors (oxygen atoms), a complementary molecule should be one that has at least one hydrogen bond acceptor moiety (example amines, alcohols, carboxylic acids, etc.). A survey of the 30

PAGE 42

Cambridge Structural Database (CSD) 14 shows that out of 362 hits (146 single-component, 208 multi-component) for 18C6 foundation, 177 included metals, 39 had hydrogen bonding to amine groups, and the remaining 88 hits were of crown derivatives. The same trend was observed for DB18C6. Out of the 26 hits (15 single-component, 11 multi-component), 16 included metals, and only one hit had a hydrogen bond to an amine group. The remaining three hits were DB18C6 derivatives. No hits showed hydrogen bonding with amine groups and the remaining 13 hits were of crown derivatives. Crystals of A, B, C, D, E, F, G, and H were obtained via slow evaporation of stoichiometric amounts of crown ether and cocrystal formers in the appropriate solvents. All crystallization experiments were conducted in an unmodified atmosphere and the solvents were dried by standard methods prior to use. All chemicals used are commercially available and were purchased from Aldrich Table 8. Crown ether Supramolecular Complexes Dibenzo-18-crown-6/4-nitroaniline A Dibenzo-18-crown-6/2-methyl-4-nitroaniline B Dibenzo-18-C-6/nicotinamide Polymorph A C Dibenzo-18-C-6/nicotinamide Polymorph B D 18-crown-6/2,6-diaminopurine E 18-crown-6/dicyandiamide F 18-crown-6/5-aminoisophthalic acid G 18-crown-6/tetrafluoroisophthalic acid/H2O H 31

PAGE 43

N NH2 O NO2CH3 NH2 NO2NH2 N N NHN NH2 NH2 NH NH2 N NH 4-NITROANILINE 2-METHYL-4-NITROANILINE NICOTINAMIDE 2,6-DIAMINOPURINE DICYANDIAMIDE COOH HOOCNH2 COOH HOOCF F F F 5 -AMINOISOPHTHALIC ACID TETRAFLUOROISOPHTHALIC ACID Figure 20. Structures of the co-crystal formers used in the supramolecular complexes. 4.4 Structures 4.4.1 Dibenzo-18-crown-6/4-nitroaniline (A) Crystallization of DB18C6 and 4-nitroaniline from benzene affords a 1:1 co-crystal, A. Grinding a 1:1 mole ratio of DB18C6 and 4-nitroaniline also forms the co-crystal. As revealed by Figure 4, the amine moiety forms H-bonds with four of the six ether oxygens and generates a supramolecular complex. The hydrogen bond distances between the oxygen atoms in the ether linkages closest to the aromatic rings and the bifurcated 32

PAGE 44

hydrogens in the amino group range from 3.050-3.145, with the 4-nitroaniline bonded to the outside of the crown ether bowl. Figure 21. The 1:1 adduct formed between DB-18-C-6 and 4-nitroaniline. These supramolecular complexes pack into tubular arrangements with solvent channels on the interior of the nanotubes (Figure 5). The tubular structure is not sustained by stacking or C-H OC hydrogen bonding. The tubular arrangements are held together through steric effects. Figure 22. Stick and space-filling models of one of the nanotubes formed by self-assembly of DB18C6/4-nitroaniline. The nanotubes have an exterior diameter of ca. 2.3nm and the solvent channels have an effective cross-section of 0.35nm. The nature of the crystal packing is of particular note since the nanotubes are inherently polar and they crystallize parallel to one another. Furthermore, as revealed by Figure 6, the 4-nitroaniline molecules are oriented at ca. 60 o to the polar axis. This 33

PAGE 45

represents a close to ideal alignment of dipoles that in most systems is usually frustrated because of the tendency of molecules with high polarizability to pack in anti-parallel arrangements. 23 (a)(b) Figure 23. Views of the orientation of the p-nitroaniline molecules in 1: (a) parallel to the polar axis; (b) looking down the polar axis. 4.4.2 Dibenzo-18-crown-6/2-methyl-4-nitroaniline (B) Figure 24. The 1:1 adduct formed between DB-18-C-6 and 2-methyl-4-nitroaniline. Crystallization of DB18C6 and 2-methyl-4-nitroaniline also results in a 1:1 supramolecular complex that is sustained by amine-ether H-bonds. In B, as revealed by Figure 7, the amino moiety hydrogen bonds through the outside of the bowl to three oxygen atoms of the crown. One amino hydrogen is bifurcated, the other is not. The distances between the hydrogen atoms of the 2-methyl-4-nitroaniline and the oxygen atoms of the ether linkages range from 2.695-3.465. However, in this case, non34

PAGE 46

centrosymmetric crystals are obtained. The molecule contains a donor amine and an acceptor nitro group at the para position. This positioning allows for maximum acentricity of the molecule. A methyl group is substituted at the ortho position, which affects the non-centrosymmetric structure. The methyl group may also act as weak H-bond donor, but it is not evident in this case. 24 4.4.3 Dibenzo-18-crown-6/nicotinamide Polymorph A (C) Nicotinamide, 25 vitamin B 3 and a well known hydrotropic agent, was chosen as a complementary additive to DB18C6 due to the presence of the amino group. Crystallization of DB18C6 and nicotinamide from THF results in a 1:1 supramolecular complex that is sustained by amine-ether H-bonds. Nevertheless, this supramolecular organic solid is distinguished by the fact that it can form two crystalline phases that differ in how the amine moieties H-bond to the DB18C6 molecules. In C, the amine group of the nicotinamide molecule is hydrogen bonded to the ether oxygens through the outside part of the crown bowl. The hydrogen bond distances between the hydrogen to the oxygen range from 3.134 to 3.362 Figure 25. Outside bowl polymorphic form of the 1:1 adduct formed between DB18C6 and nicotinamide. Crystal Packing of the 1:1 polymorphic co-crystals of DB18C6 and nicotinamide. 35

PAGE 47

4.4.4 Dibenzo-18-crown-6/nicotinamide Polymorph B (D) In D the nicotinamide molecule is hydrogen bonded to the inside of the crown bowl with hydrogen bond distances ranging from 2.995 to 3.043. A notable difference between the hydrogen bonding of the two polymorphic forms is the fact that in C, one hydrogen from the amino group bifurcates and bonds to two oxygen atoms closets to one benzene ring. Figure 26. Inside bowl polymorphic form of the 1:1 adduct formed between DB18C6 and nicotinamide Crystal Packing of the 1:1 polymorphic co-crystals of DB18C6 and nicotinamide. In D, each hydrogen atom from the amine is involved in a hydrogen bond with one of the oxygen atoms, not being the same two oxygen atoms as in C. In the CSD, 5 seven structures have been reported of co-crystals involving crown ethers [DB18C6 (1 structure) and 18C6 (6 structures)] hydrogen bonded to amino groups. 4.4.5 18-crown-6/2,6-diaminopurine (E) Crystallization of 18C6 and 2,6-diaminopurine from ethanol affords a 1:1 co-crystal, E. As revealed by Figure 11, co-crystal E is sustained by hydrogen bonds between the secondary amine hydrogen from the 2,6-diaminopurine molecule and an ether linkages 36

PAGE 48

from the crown. Two 2,6-diaminopurine molecules are involved in a dimer (NH 2 N). The secondary amine is then H-bonded to a crown oxygen. Distances range between 2.836-3.072. Figure 27. The 1:1 adduct of 18C6 and 2,6-diaminopurine 4.4.6 18-crown-6/dicyandiamide (F) Crystallization of 18C6 and dicyandiamide affords a 1:1 supramolecular complex. The co-crystal, F, as shown in Figure 12, is sustained by hydrogen bonds from the two amino group hydrogens from the dicyandiamide and two ether linkages from the crown. Distances range between 2.901-2.998. Only one hydrogen atom from each amine group is involved in a hydrogen bond with the ether linkages. The other hydrogen atoms are hydrogen bonded to other dicyandiamides, as in the 2,6-diaminopurine dimers of co-crystal E. Crown ether sheets are interlinked to each other by means of the dicyandiamide. 37

PAGE 49

Figure 28. The 1:1 adduct of 18C6 and dicyandiamide. 4.4.7 18-crown-6/5-aminoisophthalic acid (G) Crystallization of 18C6 and 5-aminoisophthalic acid affords a 1:1 supramolecular complex. The co-crystal, G, as shown in Figure 13, is sustained in a honeycomb network by carboxylic acid dimers between the 5-aminoisophthalic acids and between amino hydrogens from the 5-aminoisophthalic acid and the crown ether linkages. Distances range between 2.545-2.748 for the carboxylic acid dimer and 3.008-3.179 for the two amino hydrogens and two ether linkages. Two zigzag 5-aminoisophthalic acid chains are H-bonded to each other through the amino-ether H-bonds, forming the honeycomb network. Each crown ether is bonded to two amine groups of the 5-aminoisophthalic acid molecule, one above and one below. Figure 29. The 1:1 adduct of 18C6 and 5-aminoisophthalic acid. 38

PAGE 50

4.4.8 18-crown-6/tetrafluoroisophthalic acid/H 2 O (H) Crystallization of 18C6 and tetrafluoroisophthalic acid/H 2 O affords a 1:1 supramolecular complex. The co-crystal, H, as shown in Figure 14, is sustained by hydrogen bonds through a bridging water molecule. The hydroxy group of the carboxylic acid is H-bonded to the oxygen in the water molecule and the distances range from 2.856-2.891. The water hydrogens are then H-bonded to two ether linkages at a distance of 2.553 The tetrafluoroisophthalic acid bonded to the water molecule, which is then H-bonded to the crown ether forming a screw. Figure 30. The 1:1 adduct of 18C6 and tetrafluoroisphthalic acid and water. 4.5 Synthesis and Results 4.5.1 Dibenzo-18-crown-6/4-nitroaniline 1:1 (A) Synthesis: In a typical reaction, 30 mg (8.32x10 -5 mol) Dibenzo-18-crown-6 and 17 mg (1.23x10 -4 mol) 4-nitroaniline were dissolved in approximately 5 mL benzene. Slow evaporation of the solvent for approximately a week yielded yellow 1:1 dibenzo-18-crown-6/4-nitroaniline crystals. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix C 39

PAGE 51

Crystal packing: The co-crystals are sustained by hydrogen bonds between the amine hydrogens and four ether linkages from the crown. The two amino hydrogens involved in the H-bond are bifurcated. Distances range between 3.050-3.115. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3459 cm -1 (N-H stretch, 1 amine, 4-nitroaniline); 1310, 1248, 1217 cm -1 (C-O stretching, crown). Melt-temp: 120-122C X-ray powder diffraction: experimental done, calculated could not be done. 16.451, 19.422, 20.340, 21.284, 21.422, 21.601, 21.749, 24.416. 4.5.2 Dibenzo-18-crown-6/2-methyl-4-nitroaniline 1:1 (B) Synthesis: In a typical reaction, 30 mg (8.32x10 -5 mol) dibenzo-18-crown-6 and 14 mg (9.20x10 -5 mol) of 2-methyl-4-nitroaniline were dissolved in approximately 5 mL benzene. Slow evaporation of the solvent for approximately a week yielded yellow 1:1 dibenzo-18-crown-6/2-methyl-4-nitroaniline crystals. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix C Crystal packing: The co-crystals are sustained by hydrogen bonds between the amine hydrogens and three oxygens from the crown. One of the amino hydrogens involved in the H-bond are bifurcated, the other is involved in a H-bond with just one oxygen. The H-bond distances range between 2.484-3.465. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3369 cm -1 (N-H stretch, 1 amine, 2-methyl-4-nitroaniline); 1306, 1248cm -1 (C-O stretching, crown). Melt-temp: 137-139C X-ray powder diffraction: N/A 4.5.3 Dibenzo-18-crown-6/Nicotinamide 1:1 Polymorph A (C) Synthesis: In a typical reaction, 29 mg (8.05x10 -5 mol) Dibenzo-18-crown-6 and 28 mg (1.53x10 -4 mol) of nicotinamide were dissolved in approximately 4 mL THF. Slow evaporation of the solvent for approximately a week yielded white 1:1 dibenzo-18-crown-6/nicotinamide crystals. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix C 40

PAGE 52

Crystal packing: The co-crystals are sustained by hydrogen bonds between one of the amino hydrogens and two ether linkages from the crown. The amino hydrogen involved in the H-bond is bifurcated. Distances range between 3.134-3.161. The other hydrogen is at a distance of 3.330-3.362 away from two ether linkages as well. This co-crystal has the nicotinamide molecule bonded through the outside of the crown ether bowl and will be referred to as Polymorph A. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3334 cm -1 (N-H stretch, 1 amine, nicotinamide); 1326, 1252, 1227 cm -1 (C-O stretching, crown). Melt-temp: N/A X-ray powder diffraction: in 2, simulated derived from the single the single crystal data; experimental (simulated): 13.239(13.092), 17.761(17.853), 19.679(19.482), 22.006(22.134), 22.358(22.383), 24.142(24.170), 26.479(26.360), 30.239(30.256). 4.5.4 Dibenzo-18-crown-6/nicotinamide 1:1 Polymorph B (D) Synthesis: In a typical reaction, 290 mg (8.05x10 -4 mol) Dibenzo-18-crown-6 and 123 mg (1.01x10 -3 mol) of nicotinamide were dissolved in approximately 24 mL THF. Slow evaporation of the solvent for approximately a week yielded white 1:1 dibenzo-18-crown-6/nicotinamide crystals. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix C Crystal packing: The co-crystals are sustained by hydrogen bonds between the two amino hydrogens and two ether linkages from the crown. Distances range between 2.995-3.043. This co-crystal has the nicotinamide molecule bonded through the inside of the crown ether bowl and will be referred to as Polymorph B. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3349 cm -1 (N-H stretch, 1 amine, Nicotinamide); 1256, 1217 cm -1 (C-O stretching, crown). Melt-temp: 158-160C X-ray powder diffraction: in 2, simulated derived from the single the single crystal data; experimental (simulated): 13.038(13.081), 17.736(17.842), 20.100(20.133), 20.985(21.125), 22.225(22.201), 22.432(22.418), 23.716(23.751), 26.219(26.337), 27.753(27.929), 30.052(30.191). 41

PAGE 53

4.5.5 18-crown-6/2,6-diaminopurine 1:1 (E) Synthesis: In a typical reaction, 31 mg (1.17x10 -4 mol) 18-crown-6 and 30 mg (2x10 -4 mol) 2,6-diaminopurine were dissolved in approximately 4 mL ethanol. Slow evaporation of the solvent for approximately a week yielded white 1:1 18-crown-6/2,6-diaminopurine crystals. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix C Crystal packing: The co-crystals are sustained by hydrogen bonds between the secondary amine hydrogen and ether linkages from the crown. Two 2,6-diaminopurine molecules are involved in a dimer (NH 2 N). The secondary amine is then H-bonded to a crown oxygen. Distances range between 2.836-3.072. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3319 cm -1 (N-H stretch, 1 amine, 2,6-diaminopurine); 3443 and 3478 cm -1 (secondary amine); 1391 cm -1 (C-O stretching, crown). Melt-temp: 223-227C X-ray powder diffraction: in 2, simulated derived from the single the single crystal data; experimental (simulated): 9.382(9.459), 12.641(12.767), 18.441(18.347), 18.899(18.982), 21.077(21.468), 22.960(22.640), 24.669(24.695), 24.801(24.877), 28.204(28.221). 4.5.6 18-crown-6/dicyandiamide 1:1 (F) Synthesis: In a typical reaction, 30 mg (1.13x10 -4 mol) 18-crown-6 and 9 mg (1.07x10 -4 mol) of dicyandiamide were dissolved in a solvent system of 2mL of ethyl ether and 2mL of methanol. Heated the solution, then slow evaporation of the solvent system for approximately a week. Yielded colorless 1:1 dibenzo-18-crown-6/nicotinamide crystals. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix C Crystal packing: The co-crystals are sustained by hydrogen bonds between the two amino hydrogens from the dicyandiamide and two ether linkages from the crown. Distances range between 2.901-2.998 Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3341, 3147 cm -1 (N-H stretch, 2 amine); 2878 cm -1 (2 amide); 1640 cm -1 (C=O),1096 cm -1 (C-O stretching). Melt-temp: 198-201 X-ray Powder Diffraction: in 2, simulated derived from the single the single crystal data; experimental (simulated): 9.680(9.837), 14.897(15.124), 17.239(17.181), 42

PAGE 54

19.521(19.748), 20.403(20.389), 23.359(23.325), 23.519(23.802), 23.779(23.849), 26.281(26.441), 27.567(27.519), 27.856(27.844), 29.143(29.483), 29.998(29.810). 4.5.7 18-crown-6/5-aminoisophthalic acid 1:1 (G) Synthesis: In a typical reaction, 30 mg (1.134x10 -4 mol) of 18-crown-6 and 21 mg (1.23x10 -4 mol) of 5-aminoisophthalic acid were dissolved separately in 2mL toluene and 2mL THF respectively. The two solvent mixtures were layered on top of one another according to their densities, with a 1mL layer of blank THF separating the two layers to make diffusion slower. The reaction vial remained tightly capped in order to avoid solvent evaporation. A few days later 1:1 crystals emerged. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix C Crystal packing: Crystallization of 18-C-6 and 5-aminoisophthalic acid affords a 1:1 supramolecular complex sustained in a honeycomb network by carboxylic acid dimers and between amino hydrogens and the crown ether linkages. Distances range between 2.545-2.748 for the carboxylic acid dimer and 3.008-3.179 for the two amino hydrogens and two ether linkages. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). 3437, 3356 cm -1 (N-H stretch, 2 amine); 2878 cm -1 (2 amide); 1342, 1261, 1110 cm -1 (C-O stretching). Melt-temp: 276-280C X-ray powder diffraction: in 2, simulated derived from the single the single crystal data; experimental (simulated): 14.528(14.761), 18.144(18.407), 19.100(19.521), 20.502(20.953), 22.558(22.317), 24.021(24.425), 24.898(25.098), 25.880(25.954). 4.5.8 18-crown-6/tetrafluoroisophthalic acid 1:1 (H) Synthesis: In a typical reaction, 30 mg (1.134x10 -4 mol) of 18-crown-6 and 25 mg (1.05x10 -4 mol) tetrafluoroisphthalic acid were dissolved in approximately 1 mL of methanol. Slow evaporation of the solvent for approximately a month yielded yellow 1:1 crystals. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix C Crystal packing: Crystallization of 18-C-6 and tetrafluoroisophthalic acid/H 2 O affords a 1:1 supramolecular complex sustained by hydrogen bonds of OH groups and ether linkages through a bridging water molecule. The HO O distance is 2.553. 43

PAGE 55

Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). N/A, not enough crystal sample and could not reproduce results. Melt-temp: N/A X-ray powder diffraction: N/A 4.6 Discussion Presented above were eight examples of the use of supramolecular synthesis to generate new compositions of matter that contains molecules of interest to materials scientists. These results also indicate that hitherto inaccessible alignments of dipoles can be generated and that polymorphism in multiple component phases further adds to the structural diversity of such compounds. Although all structures presented involved hydrogen bond donors (amines, carboxylic acids, etc), the crystal structure was not uniform from co-crystal to co-crystal. Geometry, sterics, solvent systems, and synthons in the additives accounted for many differences when compared to the pure crystal structures of the pure crown ethers. It is important to point out that not all available hydrogen atoms were involved, in the hydrogen bonding. There are distance ranges (obtained from statistical data) that must be taken into account. These supramolecular complexes show how ether linkages in pharmaceutical drugs can be exploited via crystal engineering. This has increased the potential to control solid-state structures, changing their physical properties and leading to improved solubility and bioavailability. 44

PAGE 56

Chapter 5 Trimesic Acid 5.1 Porous Materials Trimesic acid (benzene-1,3,5-tricarboxylic acid, TMA) predictably generates honeycomb networks with interesting supramolecular properties because of its molecular symmetry and complementary hydrogen-bonding capabilities. 26 OH O O O OH OH Figure 31. Trimesic acid stick models. The hydrogens have been omitted for clarity. Hydrogen-bonded supramolecular synthons represent a primary tool for crystal engineering. 27 The rational design of host lattices using the principles of crystal engineering 9,28,29 and self-assembly through supramolecular synthons has increased in the last few years. Of particular interest are the efforts to create new clathrates or nanoporous materials, in which the target network is porous, contrary to Kitaigorodskis principle of close crystal packing. 30 Open framework structures of Trimesic acid can be designed via crystal engineering to build porous hydrogen bonded honeycomb networks. 45

PAGE 57

The directional motif formed by the acid acid dimer (Figure 1), along with the symmetric placement of acid groups around the benzene ring, derives the desired honeycomb framework with 14 pores or cavities. 31 These pores encapsulate aromatic guests. Figure 32. AcidAcid Dimer motif representing a hydrogen-bonded supramolecular homosynthon. O OH OHO Table 9. Table shows the results of a CSD analysis of 1116 entries with the COOH COOH synthons. 0.06443 2.554-3.196 2.657 Range Average Porous materials are crystalline or amorphous solids that permit the inclusion of small molecules through holes in their structures. Structural chemists and materials scientists have been interested in building these frameworks not only to understand their design principles, but also because of their diverse commercial applications such as chemical separation, asymmetric synthesis, catalysis, storage, etc. Interpenetration of the host lattice is undesirable for porous networks because it fills space and usually produces global structural changes. 29 Very few organic hydrogen bonded networks have been analyzed to date. A survey of the Cambridge Structural Database 14 (CSD) showed the crystal structure of tetrakis(4-ethynylphenyl)methane as the first example of a diamondoid packing motif generated by weak hydrogen bonding. A three-dimensional network is formed by hydrogen bonds 46

PAGE 58

between the alkynyl groups of four separate molecules, which meet to form the nodes of the network. This network is absent of any guest molecules. 32 Another example of an aromatic hydrogen bonded network is the orthogonal anthracene-bis(resorcinol) tetraol, a host structure that builds an extensive hydrogen-bonded network resulting in a molecular sheet. This sheet contains large supramolecular cavities, which incorporate two molecules of solvent guest. 33 The crystal packing of this structure is comparable to the 3-connected honeycomb network coordination polymer, topologically equivalent to the 2-D brick wall. 34 This organic network is held together by H-bonds between the OH groups of the host molecules creating cavities. Each supramolecular cavity can bind a guest molecule (i.e. benzophenone) via hydrogen bonded OH groups of the host and the carbonyl group of the guest. 32 5.2 TMA Polymorphic Forms Trimesic acid has been identified in four polymorphic forms, 35 36 36 and 37 (Figure 2). All of these polymorphs contain the carboxylic acid homosynthon, which generates the expected honeycomb network with an approximate 14 diameter void. The -form was developed by a solvent free host with non-planar two-dimensional interpenetrated networks with large rings created by six molecules of Trimesic acid. 11 The -form is enantiotropically transformed from heating the -form at high temperature (~543K). 12 Since the -form is a polycrystalline phase, no known crystal structure exists for the polymorph. 12 The -form is the flat triple catenated honeycomb network with inclusion of small guest molecules and it is less closely packed than the -form. 12 The 47

PAGE 59

form possesses large channels which do not interpenetrate, but rather include guest molecules of bigger size. 13 1) 2) 3) Figure 33. Schematic representations of TMA forms: 1) -form; non-planar interpenetrated honeycombs, 2)-form; planar triply interpenetrated honeycombs, and 3) -form; non-interpenetrated honeycombs. There is no known form of the -form at room temperature. 5.3 Strategy Attempts to prepare the pure trimesic acid honeycomb network (with COOH dimer distances ranging from 2.609-2.657 ) phase by breaking the interpenetration was carried out with different guest molecules, such as toluene, naphthalene, anthracene, acetic acid and biphenyl. During this study, a new TMA honeycomb network truncated by acetic acid molecules was observed. Additionally, this new form of TMA, b-form, was observed when toluene and naphthalene are the aromatic guests held in the TMA honeycomb network cavities. Having studied each of these phases structurally, we then studied the thermal stability of these crystals by Thermal Gravimetric Analysis (TGA) experiments since the honeycomb network remained in single crystal form even after guest removal. 48

PAGE 60

Table 10. Trimesic Acid Supramolecular Complexes Trimesic Acid/Biphenyl A Trimesic acid/Anthracene B Trimesic acid/Acetic acid C Trimesic acid/Naphthalene D Trimesic acid/Toluene E CH3 O CH3 OH BIPHENYL ANTHRACENE ACETIC ACID NAPHTHALENE TOLUENE Figure 34. Structures of the co-crystal formers used in the supramolecular complexes with TMA. 5.4 Structures 5.4.1 Trimesic acid/biphenyl (A) Laboratory results similar to the TMA -form include the TMA honeycomb networks with biphenyl and anthracene as the aromatic guests occupying the cavities. Crystals of TMA/biphenyl A were grown from CS 2 /AcOH/Ethanol (EtOH); crystallized in the monoclinic P2 1 noncentric space group and the asymmetric unit cell consists of four trimesic acid molecules and three biphenyl molecules. The TMA/biphenyl structure differs from other TMA honeycomb networks because it does not include any solvent 49

PAGE 61

molecules. The honeycomb is held together by carboxylic acid dimers ranging in distance from 2.605 to 2.627 falling within the acceptable range reported in Figure 1. Figure 35. TMA honeycomb network with biphenyl guests in the cavity. 5.4.2 Trimesic acid/anthracene (B) Upon crystallization of TMA/anthracene B from CS 2 /AcOH/EtOH solvent system, crystals were obtained in the 2:1:1 ratio of TMA/anthracene/AcOH in the monoclinic P2 1 /c space group. The structure is not interpenetrated and each honeycomb framework (TMA dimer distance of 2.590-2.614 ) encloses one anthracene molecule by creating C-H O (anthracene TMA) interactions and an acetic acid dimer orthogonal to it (acetic acid dimer distance of 2.643 ). Again, the C-H O hydrogen bond is due to guest inclusion rather than the stabilization of the guest by the C-H O hydrogen bonds. Figure 36. TMA honeycomb with anthracene and acetic acid dimer guests in the cavity. 50

PAGE 62

5.4.3 Trimesic acid/acetic acid (C) Crystals of TMA/acetic acid (AcOH) C were obtained when trimesic acid was crystallized from CS 2 /AcOH/EtOH. It crystallizes in the P space group with six molecules of trimesic acid and three molecules of acetic acid in the asymmetric unit. Of the three acetic acid molecules, two produce the acetic acid dimer (Dimer distance of 2.647) and the third one is used in constructing the truncated TMA honeycomb network (dimer distance of 2.717 ). Analysis of the packing of TMA/AcOH reveals that they are closely related to the -form, but differ by the incorporation of the acetic acid in the network. The acetic acid forms the truncated honeycomb by terminating the normal two-dimensional honeycomb framework. It is unclear why the dimer between the acetic acid (strong acid) trimesic acid (weak acid) COOH COOH synthon is formed. It may be rationalized that in order to include the guest, the void space must be created by terminating the two-dimensional honeycomb framework with acetic acids. Figure 37. Truncated acetic acid TMA honeycomb network. 51

PAGE 63

Results showing the new TMA polymorph with acetic acid truncated honeycomb networks, b-form, were obtained with naphthalene and toluene as the guest molecules (Figure 6 and 7). 5.4.4 Trimesic acid/Naphthalene (D) Crystallization from CS 2 /AcOH/EtOH resulted in a 6:1:1 ratio of TMA/naphthalene/acetic acid. Due to the smaller size of the guest, the framework is triply interpenetrated (as in TMA polymorphic forms and ) but differs from these forms because the acetic acid instead of creating dimers with itself, generates the acid acid (distance of 2.624-2.704 ) dimer with trimesic acid, truncating the TMA honeycomb network. Trimesic acid forms an infinite two-dimensional honeycomb network (TMA dimer distance of 2.582-2.623 ), which interpenetrates with three truncated honeycomb networks. The naphthalene guest molecules are disordered, occupying the void created between the interpenetration rather than fitting into the honeycomb framework. Figure 38. TMA honeycomb networks with acetic acid truncated networks with naphthalene. 52

PAGE 64

5.4.5 Trimesic acid/Toluene (E) Crystallization from CS2/AcOH/EtOH resulted in a 6:1:1 ratio of TMA/toluene/acetic acid. Due to the smaller size of toluene, the framework is triply interpenetrated (as in TMA polymorphic forms and ) but differs from these forms because the acetic acid instead of creating dimers with itself, generates the acid acid (as in D) (TMA dimer distance of 2.605-2.632 TMA-Acetic acid dimer distance of 2.6209-2.710 ). Figure 39. TMA honeycomb networks with acetic acid truncated networks with toluene guests. 5.5 Synthesis and Results 5.5.1 Trimesic Acid/biphenyl (A) Synthesis: In a typical reaction, 58 mg (2.76x10 -4 mol) of TMA and 47 mg (3.05x10 -4 mol) of biphenyl were layered in 6 mL ethanol and 6 mL CS 2 respectively. A 4 mL layer of acetic acid separated the two layers. Slow evaporation and in approximately two weeks crystals appeared. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix D 53

PAGE 65

Crystal Packing: TMA honeycomb networks with biphenyl as the aromatic guests occupying the cavities. The honeycomb is held together by carboxylic acid dimers ranging in distance from 2.605 to 2.627. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). Not much changes from pure TMA since the honeycomb network remained the same. The peaks did shift slightly downfield. 3127 cm -1 (-OH stretch), 1691 cm -1 (C=O stretch), 1261 cm -1 (C-O stretch). Melt-temp: 362-370C X-ray powder diffraction: in 2, simulated derived from the single the single crystal data; experimental (simulated): 9.296(9.297), 12.240(12.112), 16.081(16.374), 17.243(17.480), 22.206(22.330), 26.721(26.769), 33.499(33.370). 5.5.2 Trimesic Acid/anthracene (B) Synthesis: In a typical reaction, 20 mg (9.51x10 -4 mol) of TMA and 17 mg (9.54x10 -5 mol) of anthracene were layered in 1 mL ethanol and 1 mL CS 2 respectively. A 1 mL layer of acetic acid separated the two layers. Slow evaporation and in approximately two to three weeks crystals appeared. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix D Crystal packing: Crystallization of TMA with anthracene resulted in a structure that is not interpenetrated and each honeycomb framework (TMA dimer distance of 2.590-2.614 ) encloses one anthracene molecule by creating C-HO (anthraceneTMA) interactions and an acetic acid dimer orthogonal to it (acetic acid dimer distance of 2.643 ). Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). Not much changes from pure TMA since the honeycomb network remained the same. The peaks did shift slightly downfield. 3209 cm -1 (-OH stretch), 1698 cm -1 (C=O stretch), 1276 cm -1 (C-O stretch). Melt-temp: started melting 278C, decomposed before melting completely. X-ray powder diffraction: in 2, simulated derived from the single the single crystal data; experimental (simulated): 11.035(11.056), 12.526(12.480), 13.010(13.070), 14.198(14.121), 17.324(17.153), 22.979(22.905), 23.080(23.035), 25.119(25.111), 26.300(26.314), 27.121(27.188), 30.011(30.002). 54

PAGE 66

5.5.3 Trimesic Acid/acetic Acid (C) Synthesis: In a typical reaction, 24 mg (1.14x10 -4 mol) of TMA was dissolved in 1 mL of ethanol. Separately, 4 drops of benzene were dissolved in 1 mL of CS 2 The two layers were put in a reaction vial with a 1 mL layer of acetic acid between then to slow down diffusion. Slow evaporation and in approximately two weeks crystals appeared. The benzene served only as a template and did not appear in the crystal structure. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix D Crystal Packing: Crystallization of TMA and acetic acid results in a 1:1 supramolecular complex that is sustained by carboxylic acid H-bonds. The acetic acid forms the truncated honeycomb by terminating the normal two-dimensional honeycomb framework. Distances range between 2.647-2.717 Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). Not much changes from pure TMA since the honeycomb network remained the same. The peaks did shift slightly downfield. 3117 cm -1 (-OH stretch), 1684 cm -1 (C=O stretch), 1397 and 1275 cm -1 (C-O stretch). Melt-temp: 363-364C X-ray powder diffraction: in 2, simulated derived from the single the single crystal data; experimental (simulated): 12.887(12.906), 19.962(19.669), 20.037(20.039), 22.463(22.461), 24.442(368), 26.841(26.974), 29.183(29.173), 30.319(30.333), 30.880(30.879), 33.475(33.478), 35.127(35.127), 36.158 (36.154), 37.638(37.674), 39.562(39.561). 5.5.4 Trimesic Acid/naphthalene (D) Synthesis: In a typical reaction, 21 mg (9.99x10 -5 mol) of TMA and 11 mg (8.58x10 -5 mol) of naphthalene were dissolved separately in 1 mL ethanol and 1 mL CS 2 respectively. A 1 mL layer of acetic acid separated the two layers. Slow evaporation and in approximately two weeks crystals appeared. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix D Crystal Packing: Crystallization from CS2/AcOH/EtOH resulted in a 6:1:1 ratio of TMA/naphthalene/acetic acid. Due to the smaller size of the guest, the framework is triply interpenetrated (as in TMA polymorphic forms and ), but differs from these forms because acetic acid instead of creating dimers with itself, generates the acid acid 55

PAGE 67

dimer (distance of 2.624-2.704) with TMA, truncating the TMA honeycomb network. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). Not much changes from pure TMA since the honeycomb network remained the same. The peaks did shift slightly downfield. 3138 cm -1 (-OH stretch), 1684 cm -1 (C=O stretch), 1268 cm -1 (C-O stretch). Melt-temp: 360-363C X-ray powder diffraction: in 2, simulated derived from the single the single crystal data; experimental (simulated): 11.858(11.473), 12.701(12.742), 16.457(16.469), 22.960(22.900), 24.459(24.550), 26.658(26.656), 27.824(27.806), 30.897(30.877), 35.265(35.260). 5.5.5 Trimesic Acid/toluene (E) Synthesis: In a typical reaction, 25 mg (1.19x10 -4 mol) of TMA was dissolved separately in 1 mL ethanol and ~5 drops of toluene was dissolved in 1 mL CS 2 respectively. A 1 mL layer of acetic acid separated the two layers. Slow evaporation and in approximately two days crystals appeared. Crystal data: (Bruker SMART-APEX CCD Diffractometer). Appendix D Crystal Packing: Crystallization from CS 2 /AcOH/EtOH resulted in a 6:1:1 ratio of TMA/toluene/acetic acid. Due to the smaller size of the guest, the framework is triply interpenetrated, but differs from these forms because acetic acid instead of creating dimers with itself, generates the acid acid dimer (distance of 2.6209-2.710) with TMA, truncating the TMA honeycomb network. Infrared Spectroscopy: (Nicolet Avatar 320 FTIR). Not much changes from pure TMA since the honeycomb network remained the same. The peaks did shift slightly downfield. 3127 cm -1 (-OH stretch), 1695 cm -1 (C=O stretch), 1397 and 1265 cm -1 (C-O stretch). Melt-temp: 366-367C X-ray powder diffraction: in 2, simulated derived from the single the single crystal data; experimental (simulated): 16.344(16.345), 20.080(20.205), 22.335(22.283), 24.522(24.567), 25.487(25.508), 27.001(27.017), 28.061(28.055), 29.301(29.351), 30.582(30.503), 30.841(30.822), 33.028(33.036), 36.099(36.096), 37.703(37.700), 39.079(39.062). 56

PAGE 68

5.6 Discussion Is it really possible to maintain the crystallinity and pores present in the TMA honeycomb networks even in the absence of guests? Thermal Gravimetric Analysis (TGA) experiments carried out with the b-form crystal of TMA/toluene and the single crystal were performed. TGA measures the weight change in materials as a function of time and temperature. The measurement provides basic information about the thermal stability of a chemical and its composition (change in mass of a sample vs. temperature). 38,39 The samples were heated to 180C to ensure the removal of all guest molecules (a weight % loss of ~12% was evident in both TGA spectra). Single crystal X-ray data revealed that cell parameters before TGA and after guest removal by TGA remained the same, confirming retention of crystallinity and the hydrogen-bonded network. Table 11. TMA/toluene crystal data before and after TGA analysis. Before TGA After TGA a () 15.4991 15.2245 b () 16.2684 16.345 c () 16.4957 16.54 () 119.309 119.556 () 90.907 90.477 () 117.171 116.276 Volume ( 3 ) N/A 3072.9 Space grp P-1 P-1 57

PAGE 69

The large open honeycomb network of TMA is built by strong, yet, flexible supramolecular synthons (acid acid) and by accommodating aromatic guests in the cavities. These guest molecules inhibit the honeycomb networks from interpenetrating and assembling into a close packed conformation. Removal of the guest can be accomplished while retaining the hydrogen bonded aromatic framework. 58

PAGE 70

Chapter 6 Conclusion The concept of crystal engineering was first coined by Schmidt in the context of covalent synthesis in the solid state, i.e. topochemical reactions. 2 However, recent development of this subject has expanded it into areas as diverse as supramolecular synthesis, 3,4,9,40 crystal structure analysis and prediction 41-45 and functional materials. 46-48 As previously stated, the concept of crystal engineering was originally introduced in the context of stereochemical control of photochemical reactions, 2 but it has subsequently been shown to have wider implications for materials science. In the context of pharmaceuticals, there are important process and intellectual property implications related to control and reproducibility of composition and polymorphism. 12,49 This work focused upon how previously known supramolecular synthons could be exploited to rationally generate binary crystals with engineered compositions. The study chose pharmaceutical molecules as the target system since they contain a wide range of synthons and understanding of the crystal structure physical property relationship is essential for the continued application of these essential chemicals toward improving the overall health of society. Hydrogen bonded supramolecular synthons represent a prototypal tool for crystal engineering 9 and can be categorized as follows: 1) self-complementary synthons, in 59

PAGE 71

which a single molecule self-assembles to form a homosupramolecular synthon (e.g. carboxylic acid dimer) or 2) two or more complementary components, thereby creating a multiple component or heterosupramolecular synthon (e.g. the hetero-dimer formed by pyridines and carboxylic acids). Supramolecular synthetic approaches offer a number of attractive features: the ability to exploit readily available molecules or ions in novel ways; their inherent modularity affords compositional and structural diversity; the use of non-covalent interactions makes for facile synthesis. In short, supramolecular synthesis offers a combinatorial type of approach to synthesis of new structures and can afford a wide range of structural diversity without the need to break and form covalent bonds. The phenomenon of polymorphism in crystal engineering cannot be ignored. Polymorphism is defined as the phenomenon where the same chemical substance exists in different crystalline forms or different crystalline patterns. 11 This phenomenon is more common among molecules with flexible conformations capable of hydrogen bonding. 6 Supramolecular synthons are the structural units within molecules, which can be formed and/or assembled by synthetic operations 6 The supramolecular synthons present in polymorphic forms may be intact from form to form, therefore it can be stated that the supramolecular equivalents of structural isomers are crystalline polymorphs. 6 The criteria for assessing the existence of polymorphs are different unit cell parameters, crystal packing arrangements and physical properties. 11 Rationally generating organic binary crystals that contain pharmaceutical components by exploiting supramolecular chemistry concepts should further the development of other novel pharmaceutical phases. Indeed, the creation of model binary crystals provides 60

PAGE 72

supramolecular insight that could lead to structures that have a pharmaceutical advantage 50 over the pure phase pharmaceuticals. The necessary information to create and develop binary crystals should be stored already in the known pharmaceutical molecule. It should be intuitive that modular structures, e.g. binary crystals based upon more than one component, are often more diverse and controllable than single component phases. It is now evident that supramolecular chemistry, defined as chemistry beyond the molecule, 4 and supramolecular assemblies are inherently linked to the concepts of crystal engineering. 61

PAGE 73

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.13.14. 15. Pepinsky, R. Phys. Rev. 1955, 100, 971 Schmidt, G.M. J. Pure Appl. Chem. 1971, 27, 647. Desiraju, G.R. Crystal Engineering--The Design of Organic solids; Elsevier: Amsterdam, 1989 Lehn, J.M. Supramolecular Chemistry: Concepts and Perspectives; VCH: Weinheim, Germany, 1995 Mathias, J.P.; Stoddard, J.F. Chem. Soc. Rev. 1992, 21(4), 215-225 Desiraju, G.R. Angew. Chem. Int. Ed. Engl, 1995, 34(21), 2311-2327 MacGillivray, L.R.; Atwood, J.L Nature 1997, 389(6650), 469-472. Aakeroy, C.B; Beatty, A.M; Helfrich, B.A. Angew. Chem. Int. Ed. Engl, 2001, 40(7), 324 Etter, M.C. Acc. Chem. Res, 1990, 23(4), 120-126 Chatterjee, S; Pedireddi, V.R.; Rao, C.N.R. Tetrahedron Letters, 1998, 39, 2843-2846 Nangia, A; Desiraju, G.R. Topics in Current Chemistry; Springer Verlag Berlin Heidelberg, 1998, 198, 57-95 Bernstein, J; Davey, R.J. and Henck, J.O. Agnew.Chem.,Int. Ed.Engl., 1999, 38, 3441. Etter, M.C.; Frankenbach, G.M. Chem. Mater. 1989, 1, 10. CCDC, The Cambridge Structural Database, ConQuest 1.3, Cambridge, UK, 2003 Gaudio, A.C.; Korolkovas, A.J. Pharm. SCi., 1994, 83, 1110-1115, 126 62

PAGE 74

16.17. 18.19.20.21.22.23.24.25. 26.27.28.29. Ali, S.L. Analytical profiles of drug solubilities pp. 224-265. Camerman, A; Camerman, N. Acta Cryst. (1971) B27, 2205. Chakrabati, S; van Severen, R. Studies on the Crystalline Form of Phenytoin. Farmaceutisch Instituut, Gent, Belgium. Kiec-Kononowicz, K.; Stadnicka, K; Mitka, A. European Journal of MedicinalCchemistry 38 (2003) 555-566. Mastropaolo, D; Camerman, A; Camerman, N. Molecular Pharmacology, 23:273-277. Maverick, E.; Seiler, P; Schweizer, W.B. Acta Cryst. 1980; B36, 615-620. Rogers, R.D.; Kurihara, L.K.; Richards, P.D. J. Chem. Soc, Chem Commun., 1987; 604-606. Borecka-Bednarz, B; Bree, A.V; Patrick, B.O. Can J. Chem, 1998, 76, 1616-1632. Lipscomb, G.F; Garito, A.F.; Narang, R.S., J. Chem. Phys. 1981, 75(3), 1509-1516. Coffman, R.E.; Kildsig, D.O. J. of Pharmaceutical Sciences, 1996, 85(8), 848-853. Sharma, C.V.K. and Zaworotko, M.J.; Chem Commun., 1996, 2655-2656. Desiraju, G.R. and Sharma, C.V.K.; Perspectives in Supramolecular Chemistry, The Crystal as a Supramolecular Entity; ed. G.R. Desiraju; Wiley, Chichester, 1996. Aakeroy, C.B and Seddon, K.R. Chem. Soc. Rev, 1993, 22, 397. Nangia, A. Current Opinion in Solid State and Materials Sciences, 2001, 5, 115-122 30. Kolotuchin, S.; Thiessen, P.; Felon, E. Chem. Eur. J., 1999, 5, 2537-2547 31. Herbstein, F.H.; Kapon, M. and Reisner, G.M.Journal of Inclusion Phenomena, 1987, 5, 211-214. 32. Galoppini, E. and Gilardi, R. Chem Commun. 1999, 173-174. 63

PAGE 75

33. Endo, K. Sawaki, T.; Koyanagi, M.; Kobayashi, K.; Masuda, H. and Aoyama, Y. J. Am. Chem. Soc, 1995, 117, 8341-8352 34. Aitipamula, S.; Thallapally, P.; Thaimattam, R.; Jasklski, M. and Desiraju, G.R. Organic Letters, 2002, 4, 921-924 35. Duchamp, D.J. and Marsh, R.E. Acta Cryst, 1969, B25, 5. 36. Herbstein, F.H.; Kapon, M. and Reisner, G.M. Acta Cryst, 1985, B41, 348-354. 37. Ermer, O and Neudrfl, J. Helvetica Chimica Acta, 2001, 84, 1268-1313. 38. Applications of Thermogravimetric Analysis. www.impactanalytical.com 39. Thermal Analysis. http://curie.umd.umich.edu/srsmith/TGA.html 40. Desiraju, G.R., Stimulating Concepts in Chemistry, 2000, 293-306. 41. Aakeroy, C.B. Acta Crystallogr., Sect B, 197, 53, 569. 42. Desiraju, G.R. J. Am. Chem. Soc, 1999, 121, 1936. 43. Zaworotko, M.J. Chem. Commun. 2001, 1 44. Moulton, B.; Zaworotko, M.J. Chem Rev. 2001, 101, 1629. 45. Aakeroy, C.B. ; Beatty, A.M. Aust. J. Chem, 2001, 54, 409. 46. G. A. Jeffrey, W. Saenger, Hydrogen bonding in biological structures; Springer-Verlag: Berlin, 1991 47. G. A. Jeffrey, An introduction to hydrogen bonding; Oxford University Press: New York, 1997 48. G. R. Desiraju, T. Steiner, The Weak hydrogen bond in structural chemistry and biology; Oxford University Press: Oxford, 1999 49. M. Amai, T. Endo, H. Nagase, H. Ueda and M. Nakagaki, 1998 Acta Crystallogr., Sect C, 54, 1367. 50. Dunitz, J.D. and Bernstein, J. Acc. Chem. Res., 1995, 28, 193. 64

PAGE 76

Appendices 65

PAGE 77

Appendix A: Crystallographic Data for Nifedipine Structures A B C Formula C 18 H 21 N 3 O 7 C 92 H 96 N 12 O 28 C 27 H 26 N 4 O 6 m.p. (C) 136-138 148-153 137-139 M ol. Wt. 391.38 1817.81 502.52 Crystal System Triclinic Monoclinic Triclinic S pace Group P-1 P2(1)n P-1 a () 7.8268(11) 7.9310(14) 7.6332(7) b () 10.667(15) 12.4945(22) 9.8375(9) c () 11.8401(16) 22.2611(37) 16.6567(15) () 70.065(3) 90 101.689(2) () 76.766(3) 90.198(8) 93.132(2) () 80.56(3) 90 95.396(2) Volume ( 3 ) 900.7(2) 2205.93(4) 1215.92(19) calc density (mg/cm 3 ) 1.443 1.373 1.373 S olven t Formamide Methanol Methanol 66

PAGE 78

Appendix B: Crystallographic data for Phenytoin Structures A B C Formula C 80 H 22 N 4 O 2 C 80 H 64 N 12 O 8 C 27 H 22 N 4 O 2 m. p ( C ) 234-239 225-227 254-257 M ol. Wt. 347.37 440.48 434.49 Crystal System Monoclinic Monoclinic Triclinic S pace Group P21/c P21/c P-1 a () 16.6583(19) 8.0106(10) 8.620(14) b () 8.8478(10) 26.2398(33) 9.929(14) c () 11.9546(14) 7.981(10) 13.29(2) () 90 90 87.47(4) () 96.618(2) 91.069(2) 84.03(3) () 90 90 86.13(3) Volume ( 3 ) 1750.2(3) 1677.4(4) 1128(3) calc density (mg/cm 3 ) 1.318 1.308 1.28 S olven t Ethanol/acetone 2,4-pentanedione 2,4-pentanedione 67

PAGE 79

Appendix C: Crystallographic Data for Crown Ether Structures A B C D E F G H Formula C 58 H 66 N 6 O 18 C 27 H 30 N 2 O 7 C 26 H 30 N 2 O 7 C 26 H 30 N 2 O 7 C 51 H 87 N 18 O 18 C 16 H 32 N 8 O 6 C 60 H 60 N 2 O 20 C 78 H 132 O 39 F 16 m. p ( C ) 120-122 137-139 N/A 158-160 284-286 194-198 276-280 1997.84 M ol. Wt. 1135.17 512.55 482.52 482.52 1240.206 432.5 1129.14 1997.84 Crystal System Orthorhombic Monoclinic Monoclinic Orthorhombic Monoclinic Monoclinic Orthorhombic Monoclinic S pace Group Fdd2 P2 1 /n P2 1 /c P2 1 2 1 2 1 P2 1 /c I2/a C2 P2 1 /n a () 42.769(3) 13.5993(14) 8.4015(13) 7.9634(28) 15.452(2) 7.6625(48) 8.8849(28) b () 52.193(3) 9.3950(10) 9.8132(16) 13.650(2) 13.8565(51) 8.1949(10) 16.929(99) 20.584(71) c () 11.1128(7) 20.885(2) 17.266(3) 20.705(3) 12.6933(40) 18.215(3) 11.713(73) 14.102(42) () 90 103.445(2) 111.615(3) 90 94.661(12) 99.445(2) 90 105.952(8) Volume ( 3 ) 24806(3) 2595.2(5) 2513.3(7) 2374.5(7) 1396.008 2275.2(6) 1579.4(16) 2479.67(4) calc density mg /cm 3 ) 1.216 1.312 1.275 1.35 1.475 1.263 1.317 1.338 Ether/ S olven t Benzene Benzene THF THF Ethanol Methanol Toluene/ THF Methanol 15.956(3) 68

PAGE 80

69 Appendix D: Crystallographic Data for Trimesic Acid Structures. A B C D E Formula C 72 H 54 O 24 C 34 H 24 O 24 C 120 H96O8 4 C 133 H 98 O 42 C 126 H 112 O4 8 m.p. (C) 362-370 decomposed 363-364 360-363 366-367 M ol. Wt. 186.16 656.53 2881.97 2368.11 2418.18 Crystal System Monoclinic Monoclinic Triclinic Triclinic Triclinic S pace Group P21 P-1 P-1 P-1 P-1 a () 9.9791(10) 10.4027(8) 15.4543(14) 15.4148(21) 15.4991(34) b () 28.424(3) 28.560(2) 16.3494(15) 16.3203(22) 16.2684(35) c () 11.0164(11) 10.4430(8) 16.5130(14) 16.3669(23) 16.4957(36) () 90 90 119.409(2) 119.516(2) 119.309(6) () 104.243(3) 105.226(2) 90.449(2) 90.349(3) 90.907(9) () 90 90 117.481(1) 117.499(2) 119.171(5) Volume ( 3 ) 3028.7(5) 2993.7(4) 3072.21(2) 3029.29(7) 3069.21(5) calc density (mg/cm 3 ) 1.429 1.312 1.558 1.298 1.308 S olven t CS 2 /AcOH/EtOH CS 2 /AcOH/EtOH CS 2 /AcOH/EtOH CS 2 /AcOH/EtOH CS 2 /AcOH/EtOH