USF Libraries
USF Digital Collections

Crystal engineering of multiple component crystal forms of active pharmaceutical ingredients

MISSING IMAGE

Material Information

Title:
Crystal engineering of multiple component crystal forms of active pharmaceutical ingredients
Physical Description:
Book
Language:
English
Creator:
Weyna, David Rudy
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Crystal Engineering
Pharmaceutical Cocrystal
Pharmacokinetics
Physicochemical Properties
Zwitterion
Dissertations, Academic -- Chemistry, Pharmaceutical -- Doctoral -- USF   ( lcsh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Enhancing the physicochemical properties of solid-state materials through crystal engineering enables optimization of these materials without covalent modification. Cocrystals have become a reliable means to generate novel crystalline forms with multiple components and they exhibit different physicochemical properties compared to the individual components. This dissertation exemplifies methodologies to generate cocrystals of active pharmaceutical ingredients (API's) based upon knowledge of supramolecular interactions (supramolecular synthons), while focusing on enhanced delivery through in vitro and in vivo processes with both salts and cocrystals respectively. The utility of mechanochemistry involving small amounts of an appropriate solvent, or solvent drop grinding (SDG), has been shown to reliably reproduce cocrystals with the anti-convulsant carbamazepine that were originally obtained by solution crystallization. This technique has been confirmed as a reliable screening method using solvents in which both components exhibit some solubility. The benefits of this technique lie in the time and cost efficiency associated with it as well as its inherently small environmental impact making it a "Green" method. SDG was also used as an efficient way to discover cocrystals of the anti-inflammatory meloxicam with carboxylic acids after analysis of existing reports and the analysis of structural data from the Cambridge Structural Database (CSD) to guide the choice of coformer. It has been shown that SDG can be used to screen for cocrystalline forms that are also obtainable by solution crystallization which is important in later stage development and manufacturing including but not limited to large scale up processes. Single crystals suitable for single crystal X-ray diffraction were obtained with meloxicam and two of the coformers, fumaric and succinic acid. Some of the meloxicam cocrystals exhibited enhanced pharmacokinetic (PK) profiles in rats exemplifying significantly higher serum concentrations after only fifteen minutes and consistently higher exposure over the time studied while others maintained lower exposure. This reveals that cocrystals can fine tune the PK profile of meloxicam in order to reduce or enhance exposure. Two different sulfonate salts, 4-hydroxybenzenesulfonate (p-phenolsulfonate) and 4-chlorobenzenesulfonate, of the anti-spastic agent (R,S) baclofen were developed by strategically interrupting the intramolecularly stabilized zwitterionic structure of baclofen. This zwitterionic structure results in low solubility associated with physiological pH required for intrathecal administration. Structural data for both salts in the form of single crystal X-ray diffraction data was successfully obtained. Solubility based on baclofen was assessed and shown to increase in pure water and at pH's 1 and 7. Only the 4-chlorobenzenesulonate salt maintained an increased solubility over two days at pH 7 making it a viable candidate for further study in terms of intrathecal administration. During crystallization experiments with (R,S) baclofen two polymorphic forms of the baclofen lactam were generated, Forms II and III. Both forms are conformational polymorphs confirmed by single crystal X-ray diffraction and Form II has a Z' of 4 with an unusual arrangement of enantiomers.
Thesis:
Disseration (Ph.D.)--University of South Florida, 2011.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by David Rudy Weyna.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 204 pages.
General Note:
Includes vita.

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:
usfldc doi - E14-SFE0004855
usfldc handle - e14.4855
System ID:
SFS0028121:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Crystal Engineering of Multiple Component Crystal Forms of Active Pharmaceutical Ingredients by David R. Weyna A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemistry College of Arts and Sciences University of South Florida Major Professor: Michael J. Zaworotko, Ph.D. Roman Manetsch, Ph.D. John Koomen, Ph.D. Peter Karpinski, Ph.D. Mazen Hanna, Ph. D. Date of Approval: March 23, 2011 Keywords: Pharmaceutical Cocrystal, Supramolecular Chemistry, Zwitterion, Physicochemical Propert ies, and Pharmacokinetics Copyright 2011, David R. Weyna

PAGE 2

Dedication For Bryana and Izabelle

PAGE 3

Acknowledgements I would like to thank Dr. Mi chael Zaworotko first and foremost for the chance to work in his group and his advice as a mentor throughout the doctoral program. I would like to thank the various Zaworotko group members for their helpful input through discussions and hands on training, and th e members of my committee, Dr. Roman Manetsch, Dr. John Koomen, Dr. Peter Karpinski, and Dr. Mazen Hanna. I would also like to thank Thar Ph armaceuticals Inc. for the opportunity to conduct industrial research while concurren tly finishing the doctoral program at the University of South Florida including Dr Ning Shan, Dr. Mazen Hanna, Dr. Miranda Cheney, Ray Houck, and Brian Moyer.

PAGE 4

Table of Contents List of Tables v List of Figures vii Abstract xiii Chapter 1: Introduction 1 1.1 Supramolecular Chemistry 1 1.1.1 van der Waals Forces 1 1.1.2 Coordination Chemistry 2 1.1.3 Hydrogen Bonding 3 1.1.4 Supramolecular Chemistry and its Biological Significance 3 1.2 Crystal Forms 5 1.2.1 Single Component Molecular Crystals 5 1.2.2 Salts 6 1.2.3 Other Multiple Component Crystals: Cocrystals, Solvates, and Hydrates 7 1.2.4 Polymorphism 8 1.3 Solid-State Characterization 10 1.4 Crystal Engineering 12 1.4.1 Background 12 i

PAGE 5

1.4.2 Supramolecular Synthons 14 1.4.3 The Cambridge Structural Database 17 1.4.4 Crystal Structure Prediction 19 1.5 Pharmaceutical Cocrystals 20 1.5.1 Synthesis 20 1.5.2 History 22 1.5.3 Crystal Form Impact and Physicochemical Property Manipulation 24 1.5.4 Intellectual Property 33 1.6 References Cited 35 Chapter 2. Mechanochemistry: Solvent Dr op Grinding vs. Solution Evaporation for the Synthesis of Pharm aceutical Cocrystals Involving Carbamazepine 54 2.1 Background 54 2.2 Experimental Details 57 2.2.1 Materials 57 2.2.2 Methods 57 2.3 Results and Discussion 59 2.3.1 Reproducibility 59 2.3.2 Solvent Choice 60 2.3.3 Utility as a Screening Technique 61 2.4 Conclusions 61 2.5 References Cited 62 ii

PAGE 6

Chapter 3. Cocrystallization and Pharmac okinetic Enhancement with Meloxicam 67 3.1 Background 67 3.2 Experimental Details 71 3.2.1 Materials 71 3.2.2 Methods 71 3.3 Results and Discussion 78 3.3.1 Cambridge Structural Database Analysis 78 3.3.2 Crystal Structure Descriptions 83 3.3.3 Meloxicam Crystal Forms: Cocrystals or Salts? 86 3.3.4 Cocrystal Stoichiometries 88 3.3.5 In vivo Performance via Rat Pharmacokinetic Studies 91 3.4 Conclusions 95 3.5 References Cited 96 Chapter 4. Crystalline Forms of (R,S ) Baclofen: A Zwitterionic Active Pharmaceutical Ingredient 103 4.1 Background 103 4.2 Experimental Details 107 4.2.1 Materials 107 4.2.2 Methods 107 4.3 Results and Discussion 111 4.3.1 Pure Baclofen 111 4.3.2 Sulfonate Salts of (R,S) Bacl ofen and their Crystal Structure Descriptions 113 iii

PAGE 7

4.3.3 Pure Water Dissolution 119 4.3.4 pH 1 (0.1 N HCl) Dissolution 122 4.3.5 pH 7 Sodium Phosphate Buffer Dissolution 124 4.4 Conclusions for Baclofen Salts 128 4.5 Baclofen Lactam Polymo rphism: High and Low Z Structures 129 4.5.1 Background 129 4.5.2 Synthesis 132 4.5.3 Cambridge Structural Database Statistics 132 4.5.4 Crystal Structure Descriptions 134 4.5.5 Conclusions for Baclofen Lactam Polymorphs 140 4.6 References Cited 140 Chapter 5. Summary and Future Directions 146 5.1 Summary 146 5.2 Future Directions 150 Appendices 153 Appendix 1. Experimental Data for Carbamazepine 153 Appendix 2. Experimental Data for Meloxicam 162 Appendix 3. Experimental Data for Baclofen and Baclofen Lactam 180 About the Author 187 iv

PAGE 8

List of Tables Table 1.1: Selected bo nd energies for comparison. 4 Table 1.2: Occurrence of H-Bonding functional groups in the top 100 prescription APIs. 21 Table 2.1: Summary of SDG results for Cbz: Red = starting materials, Green = cocrystal formed, and Hyd = hydrate. Indicates Unconverted Starti ng Materials by Powder X-Ray Diffraction. 60 Table 3.1: CSD statistics for th iazole supramolecular synthons. 80 Table 3.2: Molecular diagrams, p Ka information, and melting points for meloxicam and cocrystals 1 10. 82 Table 3.3: Crystal struct ure parameters for 1 and 2. 86 Table 3.4: Pharmacokinetic data fo r meloxicam and meloxicam cocrystals. 93 Table 4.1: pKa and pKa values. 113 Table 4.2: Crystal structur e parameters for salts 2-3. 118 Table 4.3: Melting Points as determined by DSC (See Appendix 10). 119 Table 4.4: pH values for water dissolution. 120 Table 4.5: pH values for 0.1 N HCl dissolution. 123 Table 4.6: CSD statistics for Z > 1. 133 Table 4.7: CSD Refi ned statistics for Z 4. 134 Table 4.8: Selected bond distances and angles for Forms II (4) and III (5) [ and ]. 138 v

PAGE 9

Table 4.9: Selected torsion angles (o) for Forms I, II (4), and III (5). 138 Table 4.10: Crystallographic data for forms II (4) and III (5). 139 vi

PAGE 10

List of Figures Figure 1.1: Single-crysta l to single-crystal photodi merization within a cocrystal. 12 Figure 1.2: a) AmideAmide supramolecu lar homosynthon between CBZ molecules. b) AcidAmide supramolecular heterosynthon between aspirin and CBZ. 14 Figure 1.3: CarboxylateWeakly Acidic Hydroxyl supramolecular heterosynthon, CSD Refcode SERASC10 (SerineAscorbic Acid). 16 Figure 1.4: Halogen bonded chain between tetrafluoro-1,4-diiodobenzene and piperazine, CSD Re fcode DIVCUH. 17 Figure 1.5: D-HA (DA distance) in ngstroms for the carboxylic acidNarom supramolecular heterosy nthon. CSD parameters; Aug. 2010 Update, Only Organics, R 0.075, 3D Coordinates Determined, and as drawn in white box. 18 Figure 1.6: Sodium hydroge n divalproate oligomer. 23 Figure 1.7: Dissolution profiles into 0.1 N HCl at 25 C for itraconazole pharmaceutical corystals. Squares = Marketed Form Diamonds = Pure API, Upside down triangles = L-Malic Acid Cocrystal, Right side up triangles = Succinic acid Cocrystal, and Circles = LTartaric Acid Cocrystal. 27 Figure 1.8: a) Intrinsic dissolution profiles for fluoxetine HCl pharmaceutical cocrystals in water at 10o C. b) Spring and parachute of fluoxetine HCl:succinic acid during powde r dissolution in water at 20o C. 28 Figure 1.9: Plasma concentration over time for the modafinil:malonic acid pharmaceutical cocrystal over 24 hrs compared to pure API. 29 Figure 1.10: Plasma concentration of the glutaric acid pharm aceutical cocrystal over time in dogs at 50mg/kg oral dosing compared to parent API, Open Circles = Cocrystal, Filled Circles = Pure API. 30 vii

PAGE 11

Figure 1.11: Plasma concentrati on in dogs for the CBZ:saccharin pharmaceutical cocrystal in a capsule compared to the marketed form immediate release tablet at a 200 mg dose equivalent for CBZ. 31 Figure 1.12: Powder dissolution profiles over 4 hrs for AMG 517 pharmaceutical cocrystals. AMG 517= triangles, benzoic acid = open squares, benzamide = filled squares, cinnamic acid = open circles, cinnamide = filled circles. 32 Figure 2.1: Chemical diagram of CBZ. 56 Figure 2.2: Chemical diagrams of the cocrystal formers employed. 57 Figure 3.1: Meloxicam and its ionization states. 68 Figure 3.2: Meloxicam supramolecular chains sustained by sulfonyl:amide imers and thiazole-alcohol supr amolecular synthons, CSD Refcode SEDZOQ. 78 Figure 3.3: Supramolecular synthons obs erved in Meloxicam:Fumaric Acid (2:1), Cocrystal 1. 83 Figure 3.4: Supramolecular layers stacking in Meloxicam:Fumaric Acid (2:1), Cocrystal 1. 84 Figure 3.5: Supramolecular synt hons observed in Meloxicam:Succinic Acid (2:1), Cocrystal 2. 85 Figure 3.6: Supramolecular layers st acking in Meloxicam:Succinic Acid (2:1), Cocrystal 2. 85 Figure 3.7: PXRD patterns for Meloxicam:Maleic Acid slurries. 89 Figure 3.8: DSC of Meloxi cam:Maleic Acid slurries. 90 Figure 3.9: Serum concentration for me loxicam and meloxicam cocrystals in rats over 4 hours. 92 Figure 4.1: Baclofen s zwitterionic structure. 105 Figure 4.2: Sulfonic ac ids paired with baclofen. 107 Figure 4.3: Supramolecular arrangemen t of (R) Baclofen HCl from the CSD Refcode CRBMZC10. 112 viii

PAGE 12

Figure 4.4: Single Crystals of (R ,S) Baclofen monohydrate, 1, 20x magnification. 112 Figure 4.5: a) Single Crystals of Baclofen:p-Phenolsulf onate monohydrate, 2. b) Asymmetric unit for 2. 114 Figure 4.6: Overall supramolecular packing motif for Baclofen:pPhenolsulfonate monohydrate, 2. 115 Figure 4.7: Single Crystals of baclofen:4-chlorobenzenesulfonate monohydrate, 3. 116 Figure 4.8: Supramolecular interactions within the unit cell for baclofen:4chlorobenzenesulfonate monohydrate, 3. 117 Figure 4.9: Overall suprmolecular packing motif for baclofen:4chlorobenzenesulfonate monohydrate, 3, looking down the b axis. 117 Figure 4.10: 24 hour dissolution profiles in water at 37o C. 121 Figure 4.11: 1 hour dissolution profiles in water at 37o C. 121 Figure 4.12: 24 hour dissolution profiles in 0.1 N HCl at 37o C. 123 Figure 4.13: 1 hour dissolu tion profiles in 0.1 N HCl at 37o C. 124 Figure 4.14: 24 hour dissolution profiles in pH 7 buffer at 37o C. 126 Figure 4.15: 1 hour dissolution profiles in pH 7 buffer at 37o C. 126 Figure 4.16: Baclofen lactam chemical diagram. 132 Figure 4.17: 50% probability ORTEP diagra m for the asymmetric unit of Form II. 134 Figure 4.18: Pairing between R and S b aclofen lactam molecules in Form II. 136 Figure 4.19: Lactam dimer in Form III. 137 Figure 1.1A: FT-IR data for SDG with Carbamazepine and 4,4-Bipyridine. 154 Figure 1.2A: PXRD data for SDG with Carbamazepine and 4,4-Bipyridine. 155 Figure 1.3A: FT-IR data for SDG with Carbamazepine and 4-Aminobenzoic Acid. 156 ix

PAGE 13

Figure 1.4A: PXRD data for SDG with Carbamazepine and 4-Aminobenzoic Acid. 156 Figure 1.5A: FT-IR data for SDG with Carbamazepine and 2,6Pyridinedicarboxylic Acid. 157 Figure 1.6A: PXRD data for S DG with Carbamazepine and 2,6Pyridinedicarboxylic Acid. 157 Figure 1.7A: FT-IR data for SDG with Carbamazepine and Benzoquinone. 158 Figure 1.8A: PXRD data for SDG with Carbamazepine and Benzoquinone. 158 Figure 1.9A: FT-IR data for SDG with Carbamazepine and Terephthalaldehyde. 159 Figure 1.10A: PXRD data for SDG with Carbamazepine and Terephthalaldehyde. 159 Figure 1.11A: IR data for SDG with Carbamazepine and Saccharin. 160 Figure 1.12A: PXRD data for SDG with Carbamazepine and Saccharin. 160 Figure 1.13A: Ft-IR data for SDG with Carbamazepine and Nicotinamide. 161 Figure 1.14A: PXRD data for SDG with Carbamazepine and Nicotinamide. 161 Figure 1.15A: IR data for SDG with Carbamazepine and Aspirin. 162 Figure 1.16A: PXRD data for SDG with Carbamazepine and Aspirin. 162 Figure 2.1A: Melting point correlation between meloxicam cocrystals and coformers. 163 Figure 2.2A: PXRD data for 1. 163 Figure 2.3A: FT-IR data for 1. 164 Figure 2.4A: DSC data for 1. 164 Figure 2.5A: PXRD data for 2. 165 Figure 2.6A: FT-IR data for 2. 165 Figure 2.7A: DSC data for 2. 166 Figure 2.8A: FT-IR data for 3. 166 x

PAGE 14

Figure 2.9A: [1H] NMR Spectrum For 1:1 Meloxicam:Maleic Acid Slurry in EtOAc. 167 Figure 2.10A: [1H] NMR Spectrum For 1:2 Meloxicam:Maleic Acid Slurry in EtOAc. 168 Figure 2.11A: [1H] NMR Spectrum For 2:1 Meloxicam:Maleic Acid Slurry in EtOAc. 169 Figure 2.12A: PXRD data for 4. 170 Figure 2.13A: FT-IR data for 4. 170 Figure 2.14A: DSC data for 4. 171 Figure 2.15A: PXRD data for 5. 171 Figure 2.16A: FT-IR data for 5. 172 Figure 2.17A: DSC data for 5. 172 Figure 2.18A: PXRD data for 6. 173 Figure 2.19A: FT-IR data for 6. 173 Figure 2.20A: DSC data for 6. 174 Figure 2.21A: PXRD data for 7. 174 Figure 2.22A: FT-IR data for 7. 175 Figure 2.23A: DSC data for 7. 175 Figure 2.24A: PXRD data for 8. 176 Figure 2.25A: FT-IR data for 8. 176 Figure 2.26A: DSC data for 8. 177 Figure 2.27A: PXRD data for 9. 177 Figure 2.28A: FT-IR data for 9. 178 Figure 2.29A: DSC data for 9. 178 xi

PAGE 15

Figure 2.30A: PXRD data for 10. 179 Figure 2.31A: FT-IR data for 10. 179 Figure 2.32A: DSC data for 10. 180 Figure 3.1A: TGA for (R,S) Baclofen Monohydrate. 181 Figure 3.2A: DSC for (R,S ) Baclofen Monohydrate. 181 Figure 3.3A: TGA for (R,S) B aclofen:p-Phenolsulfonate. 182 Figure 3.4A: DSC for (R,S) B aclofen:p-Phenolsulfonate. 182 Figure 3.5A: TGA for (R,S) Bacl ofen:4-Chlorobenzenesulfonate. 183 Figure 3.6A: DSC for (R,S) Bacl ofen:4-Chlorobenzenesulfonate. 183 Figure 3.7A: PXRD data for baclofen dissolution. 184 Figure 3.8A: PXRD data for Baclofen :p-Phenolsulfonate dissolution. 184 Figure 3.9A: PXRD data for Baclofen :4-Chlorobenzenesulfonate in pH 7 sodium phosphate buffer. 185 Figure 3.10A: Histogram for amide-amide dimer, 1st contact. CSD version 5.31 Aug. 2010 update with search pa rameters; 3D coordinates, organics, and R 0.05. Contact between donor and acceptor defined as 2.7 3.3 185 Figure 3.11A: Histogram for amide-amide dimer, 2nd contact. CSD version 5.31 Aug. 2010 update with search pa rameters; 3D coordinates, organics, and R 0.05. Contact between donor and acceptor defined as 2.7 3.3 186 Figure 3.12A: Histogram for secondary amidesecondary amide dimer (includes lactams), 1st contact. CSD version 5.31 Aug. 2010 update with search parameters; 3D coordinates, organics, and R 0.05. Contact between donor and acceptor defined as 2.5 3.1 186 Figure 3.13A: Histogram for secondary amidesecondary amide dimer (includes lactams), 2nd contact. CSD version 5.31 Aug. 2010 update with search parameters; 3D coordinates, organics, and R 0.05. Contact between donor and acceptor defined as 2.5 3.1 187 xii

PAGE 16

Abstract Enhancing the physicochemical properties of solid-state materials through crystal engineering enables optimization of these materials without covalent modification. Cocrystals have become a reliable means to generate novel crystalline forms with multiple components and they exhibit different physicochemical properties compared to the individual components. This disserta tion exemplifies methodologies to generate cocrystals of active pharmaceutical ingredients (APIs) based upon knowledge of supramolecular interactions (supramolecula r synthons), while focusing on enhanced delivery through in vitro and in vivo processes with both salts a nd cocrystals respectively. The utility of mechanochemistry invo lving small amounts of an appropriate solvent, or solvent drop grinding (SDG), has b een shown to reliably reproduce cocrystals with the anti-convulsant carbamazepine that were originally obtained by solution crystallization. This technique has been conf irmed as a reliable screening method using solvents in which both components exhibit some solubility. The benefits of this technique lie in the time and cost efficiency associat ed with it as well as its inherently small environmental impact making it a Green me thod. SDG was also used as an efficient way to discover cocrystals of the anti-in flammatory meloxicam with carboxylic acids after analysis of existing reports and the anal ysis of structural da ta from the Cambridge Structural Database (CSD) to guide the choi ce of coformer. It has been shown that SDG can be used to screen for cocrystalline forms that are also obtainable by solution xiii

PAGE 17

xiv crystallization which is important in later stage development and manufacturing including but not limited to large scale up processes. Si ngle crystals suitable for single crystal Xray diffraction were obtained with meloxica m and two of the coformers, fumaric and succinic acid. Some of the meloxicam cocr ystals exhibited enha nced pharmacokinetic (PK) profiles in rats exemp lifying significantly higher seru m concentrations after only fifteen minutes and consistently higher exposure over the time studied while others maintained lower exposure. This reveals that cocrystals can fine t une the PK profile of meloxicam in order to reduce or enhance exposure. Two different sulfonate salts, 4-hydroxyben zenesulfonate (p-phenolsulfonate) and 4-chlorobenzenesulfonate, of the anti-spast ic agent (R,S) baclofen were developed by strategically interrupting the intramolecula rly stabilized zwitte rionic structure of baclofen. This zwitterionic structure resu lts in low solubility associated with physiological pH required for intrathecal admini stration. Structural data for both salts in the form of single crystal X -ray diffraction data was successfully obtained. Solubility based on baclofen was assessed and shown to increase in pure water and at pHs 1 and 7. Only the 4-chlorobenzenesulonate salt maintain ed an increased sol ubility over two days at pH 7 making it a viable candidate fo r further study in terms of intrathecal administration. During crystallization experime nts with (R,S) baclofen two polymorphic forms of the baclofen lactam were generated, Forms II and III. Both forms are conformational polymorphs confirmed by single crystal X-ray diffraction and Form II has a Z of 4 with an unusual arrangement of enantiomers.

PAGE 18

Chapter 1: Introduction 1.1 Supramolecular Chemistry 1.1.1 van der Waals Forces When two atoms come together to form a molecule (e.g. N2) they are held together by a covalent bond, which can be descri bed as sharing of electron density that is delocalized over the entire molecule resulting in a strong chemical bond.1 The interaction of atoms or molecules with one another is th e fundamental basis for the physical state in which they exist under ambient conditions a nd ultimately affects their physicochemical properties. Johannes Diderik van der Waal s was born in 1837 in the Netherlands and was one of the first professors of physics at the University of Amsterdam, then called Athenaeum Illustre of Amsterdam. During van der Waalss work towards understanding the different phases of physical chemistry he published the van der Waals equation of state that contained a consta nt related to the strength of attraction between species.2 The idea of intermolecular attractiv e forces were considered ear lier by Borelli and Jurin with respect to capillary action where they explain their results by attractive forces between the molecules of the tube and the liquid.3 Once the existence of these forces was accepted, explanations were sought after. Debye suggested molecules have a deformable distribution of charge and are therefore this distribution is not rigid and can be polarized in an external electromagnetic field resulti ng in attractive forces if the field is nonuniform.4 For molecules with dipoles and quadr upoles in the gaseous state at low 1

PAGE 19

temperature Keesoms alignment effect was developed, which says that as a molecule rotates and moves in space it will frequently co llide with another molecule resulting in the molecules positioning in such a way that they are caught in an attractive position.5 Debyes theory fills in the gaps for the persistence of van der Waals forces at higher temperatures not accounted for by Keesom For nonpolar molecules the very weak attractive van der Waals forces can be e xplained by London Dispersi on forces which are related to the deformable distribution of charge mentione d earlier, whereby temporary very weak dipoles are formed re sulting in electrostatic attraction.3 This work led to the measurement and reporting of van der Waals volumes and radii in 1964 once X-Ray diffraction data of crystalline materials became sufficiently available and as of 2010 the manuscript has been cited 10,767 times.6 X-ray diffraction is now considered the gold standard for the characterization of solid-state materials. van der Waals equation of state has been updated several times to account fo r progress in the field of physical chemistry to its current status as the Ellio tt, Suresh, Donohue equation of state.7 These forces can be summed up as weak non-directiona l electrostatic interactions usually less than 5 kJ/mol.8 1.1.2 Coordination Chemistry The stronger side of electr ostatic interaction involves charged species i.e. ionic bonds and coordination bonds also known as coordinate covalent. Coordination chemistry involves a metal ion coordinating with a donating ligand. The ligand in this case can be organic and can be negatively charged, as in the case of a carboxylate, or contain a lone pair of electr ons, as in the case of the nitr ogen atom of 4, 4-bipyridine. For the neutral pyridine molecule there is a dipole ion interac tion with the positive charge on the metal, which may be consid ered a coordinate covalent bond. These 2

PAGE 20

interactions can lead to discrete structures, polymers, and three-dimensional (3D) networks depending on the diffe rent coordination geometries of the metal employed but are not considered completely covalent in nature because they are reversible.9 These coordination polymers can have porosity or cag es similar to zeolites and when organic ligands are used they are termed metal organic frameworks (MOFs).10 1.1.3 Hydrogen Bonding A hydrogen bond (H-Bond) can be described as an exaggerated dipole dipole interaction in which a hydrogen atom is atta ched to an electronegative atom and is attracted to another dipole in close proximity, in what is termed a donor (D) acceptor (A) relationship, where the atom covalently linked to the proton (H) is the donor. The relationship can be generally described by, D-HA. Linus Pauling described the H-bond as electrostatic in nature early on and he believed the donor atom and the acceptor atom needed to be sufficiently electronegative in order to have enough el ectrostatic attraction to result in a bond.11 H-bonds are strong and directional by nature compared to van der Waals forces with bond energies ranging from 4 120 kJ/mol.8 The weaker end of the Hbond energy spectrum merges into van der Waals forces where there lies a grey area between the two.12 1.1.4 Supramolecular Chemistry and its Biological Significance Other intermolecular interactions include C-H stacking, and cation interactions.8 All of these intermolecular interacti ons including those mentioned in the two previous sections can lead to self asse mbled aggregates called supermolecules, where there lies a host guest type of interaction.13 The term supermolecule dates back to 1949 3

PAGE 21

when the term bermolekeln was used for an intermolecular complex.14 This self assembly based on intermolecular interactions is termed supramolecular chemistry by Jean-Marie Lehn which he defined as follows, Supramolecular chemistry may be defined as chemistry beyond the molecule, bearing on the organized entities of higher complexity that result from the association of two or more chemical species held together by intermolecular forces.15 Supramolecular chemistry could be traced back to investigations into receptor binding in biol ogical systems when a lock and key model was described by Emil Fischer.16 H-bonding is the lead intermol ecular force directing self assembly involving organic compounds and has been described as the masterkey interaction in supramolecular chemistry.8 Each of the intermolecular interactions described can affect supramolecular self asse mbly and their bond energies are compared in Table 1.1. Although H-bonds are usually the primary directing interaction for self assembly, other weaker intermolecular interac tions can serve to stabilize the resulting structure. Table 1.1: Selected bond energies for comparison. Bond Type Bond Energy (kJ/mol) Covalent 150-450 Ionic 100-350 Ion Dipole 50-200 H-Bond 4-120 Cation 5-80 Dipole-Dipole 5-50 Stacking <50 van der Waals <5 4

PAGE 22

Using nature as a template for instru ction and molecular recognition, the field of supramolecular chemistry has includ ed research involving crown ethers17-18, host guest systems19-20, cryptands21-22, and cavitands.23 One of the most important discoveries in biological sciences was the determination of the structure of deoxyribonucleic acid (DNA),24 which consists of nucleotide polymers connected via phosphodiester backbones that H-bond between base pairs creating a double helix. This genetic material is necessary for cellular replication and the polymeric strands must be unwound in order to be decoded and translated into instructions fo r cellular components, th erefore a reversible supramolecular process is necessary to read the DNA. H-bonding and supramolecular recognition are necessary for many biological functions including but not limited to receptor binding, immunological respons es, protein folding and function, and coordination bonds responsible for hem oglobins ability to transport oxygen.8 1.2 Crystal Forms 1.2.1 Single Component Molecular Crystals Molecules tend to arrange themselves into regular repeating units in the solid state creating a 3D molecular array which is defi ned as a crystal. If there is no regular repeating unit the solid is termed amorphous for example glass. Amorphous compounds are known to be unstable and reactive compared to their crystalline counterparts.25 An example of a molecular crysta lline substance is table suga r which consists of sucrose crystals. The regular repeating 3D unit of a crystal is termed the unit cell and the way molecules arrange themselves inside the uni t cell is referred to as crystal packing. Crystals can vary in size from a few nanometers to meters26 and exhibit a particular shape 5

PAGE 23

or morphology. The points in the unit cell that define where molecules reside can be referred to as the crystal lat tice. The crystal latt ice can be simplified to a 3D box with points inside representing lat tice points and if those points were spheres then different centering within the box will de termine where other points can reside. There are 7 ways in which a lattice can arrange in the unit cell, called a crystal system, and when combined with different lattice centerings the result is 14 Bravais Lattices (infinite set of points generated by discrete translations) and within these there are only 230 possible arrangements called space groups, which are based on mathematically generated symmetry operations.25 The study and determination of cr ystal structures comprises the field of crystallography and Z is defined as th e number of formula units in the unit cell, while Z is the number of formula units in th e crystallographic asymmetric unit. Z can be strictly defined as the number of formula un its in the unit cell divi ded by the number of independent general positions.27 Covalent network solids co ntaining only one type of atom, diamond for example, should also be me ntioned, however, due to the scope of this manuscript they are no t discussed further. 1.2.2 Salts A crystalline salt is an ionic solid in volving a charge-charge interaction between ions of opposite charge. Organi c and inorganic salts, like sodium chloride composed of alternating sodium and chloride ions, are held together solely by very strong electrostatic interactions. An organic sa lt can result from ionizable functional groups or in combination with permanent charge states as in a quarternary nitr ogen atom carrying a positive charge. Primary, secondary, or tertiary ammonium salts are examples of charge 6

PAGE 24

assisted H-bonds28, which typically contain ion-dipole electrostatic interactions with bond energies between those of ionic bonds and H-bonds as exemplified in Table 1. Inorganic ions can also readily pair with organic ions as is the case of pharmaceutical hydrochloride salts. 1.2.3 Other Multiple Component Crystals: Solvates, Hydrates, and Cocrystals When more than one entity or molecule crystallizes together in a stoichiometric ratio it is referred to as a multiple com ponent crystal. When a single component crystalline material has solvent as part of th e crystalline arrangement in a stoichiometric ratio the binary crystal is termed a solvate and when water is the solvent involved the crystalline form is referred to as a hydrate. In these instances the solvent helps stabilize the overall packing of the unit cell. The term cocrystal is used very loosely depending on the scientific field it applies to and in terms of solid-state chemistry there is current debate about how narrowly or br oadly one defines a cocrystal.29-34 For this work a narrow definition of cocrystal is applied: a st oichiometric multiple component crystal that is formed between two compounds that are so lids under ambient conditions in which at least one cocrystal former (s econd molecule) is molecular and forms a supramolecular synthon (specific intermolecular interaction to be discussed further in section 1.4.2) with the remaining cocrystal former(s).35-36 One of the first uses of the term cocrystal in the context of solid-state chemistry dates back to 1967 describing the H-bonded complex between 9-methyladenine and 1methylthymine, first reported by Hoogsteen.37-38 Cocrystals as defined here began to appear in the literature in 1844 via a gr inding experiment by Friedrich Whler who 7

PAGE 25

prepared the 1:1 quinone:hydroquinone cocrys tal by kugelchen or little ball likely referring to a ball mill type of apparatus.39 Cocrystals were not named as such early on and over the years many cocrystals were discovered and report ed but labeled under different nomenclature including molecular complexes40, addition compounds41, organic molecular compounds42, heteromolecular crystals43, solid-state complexes44, and molecular compounds.45 Both single component and mu ltiple component crystalline materials can have the same building blocks arranged in a different manner considered a separate crystalline entity of th e same molecules or a polymorph. 1.2.4 Polymorphism The generally accepted definition of pol ymorphism is the ability of a compound to exist in more than one crystalline stat e. Bernstein, however, provides a more narrow definition limited by what he considers as safe criterion for classifying polymorphic systems which states, classification of a syst em as polymorphic would be if the crystal structures are different but lead to identical liquid and vapor states.46 Pseudopolymorphism is a term McCrone proposed which he defines as, a convenient term to use to describe a variety of phe nomena sometimes confused with polymorphism. They include desolvation, sec ond-order transitions (some of which are polymorphism), dynamic isomerism, mesomorphism, grain growth, boundary migration, re-crystallization in the solid state and la ttice strain effects.47A prototypical example of a polymorphic system would be 5-methyl-2-[(2-nitrophenyl)-amino]-3-thiophenecarbonitrile, also known as ROY, which exhibits re d, orange, and yellow polymorphs.48-49 For organic crystals there are two main types of polymorphism; packing polymorphism, which refers 8

PAGE 26

to the 3D arrangement of individual molecules with respect to one another, and conformational polymorphism, which involve s molecules that can adopt different geometries through rotation about single bonds, bond stretching/compressing, and bending of bond angles.50 Conformational polymorphism bega n to appear in the literature in the 1970s.51-52 Different molecular geometries that lead to conformational polymorphism in order to satisfy a particular solid-state arrangement under a particular set of physical conditions can often lead to structures with more than one symmetry independent molecule, Z > 1. It should also be noted that intramolecular and intermolecular interactions can affect both conformation and packing characteristics and therefore also play a role in polymorph c ontrol. There are two types of polymorphic systems with respect to polymorphic tran sformations, monotropic and enantiotropic.46 Considering for the sake of discussion that there are two polymorphs, A and B, monotropic systems represen t a situation where once A has converted to B the transformation is irreversible, while enantiotropic systems are reversible. According to McCrones statement, the number of forms known for a given compound is proportional to the time and money spent in research on that compound, there are virtually endless possibilities and likely there is great validity in this statement, however, existing literature and databases do not represent the full sc ope of polymorphs discovered since many structures may not be of inte rest and not reported, therefor e it is difficult to confirm. Polymorphism as a whole is an importan t phenomenon with resp ect to intellectual property and physicochemical properties especially with respect to pharmaceutical science.25, 46, 53-54 9

PAGE 27

1.3 Solid-State Characterization X-ray diffraction is the gold standard for solid-state characterization but multiple other complimentary techniques such as attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTI R), Raman spectroscopy, solid-state nuclear magnetic resonance (SS NMR), microscopy, di fferential scanning calorimetry (DSC), and thermal gravimetric analysis (TGA) are used in conjunction.25 X-ray diffraction is well suited for solid phase characterization due to the X-ray region of electromagnetic radiation containing wavelengths smaller th an the diameter of an atom since the principles of optics dictate that the wavelength of light used to observe an object must be smaller than the object itself.55 This is only part of the story however because no lens exists that can focus X-rays and therefore cr eating a direct image is not possible. Since Xrays have the ideal wavelength fo r reflecting off of atoms they can be directed at a crystal resulting in scattering of the radiation in a specific manner called a diffraction pattern. This was first interpreted by William H. Bragg in 1913 and he developed the Bragg equation which remains the basis for this technique, Eq. 1.1.25 Eq.(1.1) This equation states that n, an integer multiple of the wavelength, equals twice the distance, d, between planes of atoms multiplied by the sine of the angle of incidence, while the intensity of diffraction is determ ined by the electron density of the atom. 10

PAGE 28

This technique is used for single crystal anal ysis resulting in a single crystal structure with the aid of mathematical analysis and is also amenable to powders in which case it is known as powder X-ray diffraction (PXRD). PXRD usually only provides a diffraction pattern and not a single crysta l structure but in some cases where the diffraction pattern intensity is large enough th e structure can be delineated mathematically from the pattern.56 The other techniques mentioned each have their own advantages in diagnosing a solid phase. ATR-FTIR utilizes absorption of infrared light based on vibrational intramolecular and intermolecular movement which results in a spectral output. New intermolecular interactions between molecules results in shifting of peaks or even new peaks. The other spectroscopic techniques, SS NMR and Raman spectroscopy (measurement of scattering of infrared light), will also result in form specific spectra and similarly, significant new intermolecular interac tions in a new solid-state phase will cause shifts in the spectrum. DSC is a valuable tool for determining heats of fusion and resultant enthalpies associated with a specifi c crystal form and can also be applied to study polymorphic behavior such as reversible phase transi tions and the presence of impurities. TGA is ideal for studying solvates and hydrates as it determines weight loss over time upon heating which can delineate th e stoichiometry of solvent present and the qualitative strength of the interaction. Mi croscopy is useful in determining crystal habit/morphology, twinning in some cases, and with the use of a polarizer, whether the crystal is polar. Hot-stage microscopy coupl ed with other techniques is useful for determining polymorphic transformations and single crystal to single crystal transitions.46 11

PAGE 29

1.4 Crystal Engineering, Supramolecular Synt hons, and the Cambridge Structural Database 1.4.1 Background Designing crystalline materials for specified applications vi a manipulation of intermolecular interactions has been estab lished as the field of crystal engineering.9, 57-58 The term crystal engineering was introdu ced by Pepinsky in 1955 when he revealed controllable unit cell dimensions and symmetries with metal organic complexes.59 This nomenclature seemed to be ahead of its time since the development of crystal structure determination, or crystallography, as a fiel d was still young. In 1959 the Hoogsteen base pair cocrystal structure emerged between 9-methyladenine and 1-methylthymine, which was paramount to understanding DNA which utilized Watson-Crick base pairing.37 It wasnt until 1971 that the elegance behind s upramolecular chemistry and the possibility of engineering crystalline forms came to fruition in the lit erature with Schmidts report on photodimerization in the solid state w ith cinnamic acids further studied by MacGillivray in 2000 utilizing a cocrysta l template as depicted in Figure 1.1.60-61 Figure 1.1: Single-crystal to single-crystal photodimeri zation within a cocrystal. 12

PAGE 30

Guatum Desiraju defined crystal engineering in 1989 as, the understanding of intermolecular interactions in the context of crystal packing and the utilization of such understanding in the design of new solids with desired physical and chemical properties.58 The importance of supramolecular chemistry was well defined at this point and more attention was paid to understandi ng intermolecular interactions and how to subsequently manipulate those interactions in crystalline materials. Margaret Etter helped to develop the field in the late 1980s and 1990s by st udying cocrystallization and analyzing H-bond patterns, which she termed motifs, that helped develop graph set analysis to define specific H-bonded motifs.62-68 Etters contributions helped establish the foundation for the current rapid growth of supramolecular chemistry and in 1991 she proposes three general rules with regards to H-bonding; 1) all good proton donors and acceptors are used, 2) six membered ring intramolecular H-bonds form before intermolecular H-bonds, and 3) the best donors and acceptors left after intramolecular Hbonding will be used for intermolecular H-bonding.63 She includes the following statement with regards to her work fore shadowing the importance of physicochemical property enhancement through supramolecular chemistry, We have also shown that hydrogen-bonded molecular aggregates can have unexpected properties as a result of the collective behavior of thes e weakly bound molecules .63 13

PAGE 31

1.4.2 Supramolecular Synthons As mentioned in section 1.2.3 superm olecules interact with one another via supramolecular synthons which are defined by Desiraju as, structural units within supermolecules which can be formed and/or assembled by known or conceivable synthetic operations involving intermolecular interactions .69 Typically complimentary H-bonds act as the supramolecular synthon, although other intermol ecular interactions are also included. There are two classificat ions of supramolecular synthons based on complimentarity and homogeneity, whic h are referred to as supramolecular homosynthons and supramolecular heterosynthons.70 Figure 1.2 below depicts examples of each with respect to carbamazepine (CBZ), an anti-convulsant API. a b Figure 1.2: a) AmideAmide supr amolecular homosynthon between CBZ molecules.71 b) AcidAmide supramolecular he terosynthon between aspirin and CBZ.72 14

PAGE 32

When there is self complimentarity between functional groups the result is a supramolecular homosynthon and when differe nt functional groups compliment each other the result is a supramolecular heterosy nthon. It should be noted that the term supramolecular mixed homosynthon can be used in the instance where two different carboxylic acids inte ract via an acid acid supramolecular homosynthon.73 Examples of supramolecular homosynthons are amide amide74 and acid acid75-76 dimers. Carboxylic acid dimers have been mentioned in the literature as early as 1897 when the association of acetic acid in benzene was recognized.77 Later work by F. T. Wall in the early 1940s reported the associ ation of benzoic, m-toluic and o-toluic acid with themselves in benzene as individual solu tions confirmed by vapor pressure osmometry.7879 Supramolecular heterosynthons can be exemplified by acid amide,68, 80-82 acid aromatic nitrogen (Narom)64, 70, 83-86, alcohol amine,87-90 and alcohol Narom.91-93 With crystal engineering in mind it was important to determine which supramolecular synthons would persist over others. Synthon competition was asse ssed by including the presence of other functionalitie s in order to establish whic h interactions take precedence in order to determine the hierarchy. Cocrysta ls have been particularly important in delineating these hierarchies with resp ect to alcohols, carboxylic acids, Narom, cyanos, and amides. Once Etter established some ground rules in terms of H-bonding patterns others started to experiment with multiple functi onal groups present at the same time during crystallization. Christer Aaker y studied whether or not the st rongest acid would interact with the strongest base and then if the next strongest acid went with the next strongest base and so on within thei r chosen set of molecules.94-97 He also studied ternary 15

PAGE 33

cocrystals by employing molecules with carboxylic acid, Narom, and amide functionality on the same molecule resulting in acid Narom and acid amide supramolecular heterosynthons.96, 98-99 More reports on this subject s oon followed where Bis reveals that alcohols prefer to interact with Narom while in the presence of cyano groups.100 Shattock then determined that carboxylic acids in the presence of Narom will prefer the supramolecular heterosynthon ove r the homosynthon and that the same theme holds true for alcohols in the presence of Narom.101 This study also attemped to distinguish between which supramolecular heterosynthon is preferred, however, the findings indicate that both heterosynthons occurred for the limited data se t utilized leaving this particular piece of the hierarchy open for debate for now. Charge assisted H-bonds are also categorized as supramolecular synthons28, 102 and are also important in te rms of crystal engineering. It has been observed that the carboxylate weakly acidic hydroxyl supramolecular heterosynthon is persistent in a set of fifteen zwitteri onic cocrystals with nutraceuticals.103-106 Figure 1.3 below depicts an earlier example of the carboxylate weakly acidic hydroxyl supramolecular heterosynthon betw een L-sarcosine and L-ascorbic acid which has been deposited into the Cambridge Structural Database (CSD).107 Figure 1.3: CarboxylateWeakly Acidic Hydroxyl supramolecular heterosynthon, CSD Refcode SERASC10 (SerineAscorbic Acid). 16

PAGE 34

Other supramolecular synthons in clude charge assisted H-bonds with other inorganic ions such as phosphate,108 interactions with chlorine of a hydrochloride salt (cocrystal of a salt),109 and also halogen bonding, Figure 1.4.110-116 Figure 1.4: Halogen bonded chain between tetrafluoro-1,4-diiodobenzene and piperazine, CSD Refcode DIVCUH. 1.4.3 The Cambridge Structural Database The CSD represents a large computerized archive of X-ray and neutron diffraction data of small (less than 50 0 non-H atoms) organic and metal organic molecules that began in 1965.117-118 This database package of software was developed by the Cambridge Crystallographic Data Cent re (CCDC) which includes software to visualize and analyze molecula r interactions including but not limited to bond distances, bond angles, symmetry elements, histogram generation, and 3D networks. The CSD contains 523,834 entries as of August 2010 and th is vast array of st ructural data has become an integral tool for crystal engineer ing. One simply needs to draw out a proposed interaction of interest and a plethora of information is likely to be discovered with statistical significance depending on the freque ncy of occurrence. Each entry is labeled with a six letter code, refcode, which may have numbers attached for repeat entries. The 17

PAGE 35

robustness of supramolecular synthons can be evaluated if enough data on that particular interaction is available. Database mining a ve ry useful tool with regards to solid-state chemistry since one can evaluate a very large pool of data not otherwise available from one source.9, 119-122 Indeed, much of the work devoted to understanding supramolecular synthons and their hierarchies mentioned above was supplemented and even guided by CSD analysis and it has now become routine that the initial approach to a crystal engineering experiment involves anal ysis of existing structural data.82, 100-102, 123 Crystal engineering has led to the discovery of materials with distinct applications such as porous materials124, non-linear optics, pharmaceuticals35, 125, and photographic materials.126 An example of the power of the CSD is represented in Figure 1.5, in which the average donor-acceptor distance of the carboxylic acid Narom supramolecular heterosynthon is revealed to be 2.639 through histogram generation. Figure 1.5: D-HA (DA di stance) in ngstroms for the carboxylic acidNarom supramolecular heterosynthon. CSD para meters; Aug. 2010 Update, Only Organics, R 0.075, 3D Coordinates Determined, and as drawn in white box. 18

PAGE 36

1.4.4 Crystal Structure Prediction Crystal form screening, polymorph c ontrol, and molecular design would be immensely facilitated if a solid-state chemist could accurately predict the crystalline structure of organic molecular crystals. In general, computational approaches search for the most thermodynamically stable structure. As remarked by Maddox, One of the continuing scandals in the physical sciences is that it remains in general impossible to predict the structure of even the simplest crystalline solids from a knowledge of their chemical composition ,127 it remains true that comple te and reliable prediction of a crystal structure has not been achieved.45, 128-135 The CCDC held the first crystal structure prediction workshop in 1999 to evaluate the current state of computational methods and the first blind test was initiated in which different computational groups were given a molecular diagram and asked to predict the crystal structure.128 There have been four blind tests to date and the results are increas ingly promising as the fourth test concluded with dramatic improvements in the success rate of prediction.128, 134, 136-137 The fourth test included fourteen groups with four targets in their sites. Thirteen successful predictions prevailed overall and for each target at le ast two groups were successful, while only one group got all four as their first choices. There is a long way to go but progress in this area is encouraging. 19

PAGE 37

1.5 Pharmaceutical Cocrystals 1.5.1 Synthesis Pharmaceutical cocrystals are defined as multiple component crystals in which at least one component is molecular and a solid at room temp erature (the coformer), and forms a supramolecular synthon with a mo lecular or ionic active pharmaceutical ingredient (API).48, 138 Cocrystals and pharmaceutical cocrys tals are identical in terms of synthetic routes. Traditional techniques for crystallization apply where supersaturation and subsequent nucleation from solution remains to be the most widely accepted method and in the case of multiple component crystals the supramolecular synthon between individual molecules helps determine the initial aggregate that leads to cocrystal growth. Grinding two solids together, or mechanochemi stry, to create a new crystalline phase is perhaps the simplest way to generate a multiple component stoichiometric adduct or cocrystal. It is ideal in terms of cost environmental impact, and difficulty, and furthermore, the addition of small amounts of solvent during the grinding process was shown to dramatically increase the kinetics of cocrystal formation in certain cases.36, 139140 Grinding can be done by hand wi th a mortar and pestle or mechanically with a ball mill or mixer mill. The melt/cool method involves melting one component followed by addition of the second component thereby dissol ving it followed by controlling the rate of cooling and may be performed in absen ce of unstable or high melting compounds.141-143 Slurry methods, which can be facilitated by sonication144, are useful as well and represent a method somewhere in between solvent dr op grinding and evaporative solution based techniques.48, 102, 109 Reaction crystallization is a solu tion based technique that has been specifically applied to cocrystal formation on the premise of mani pulating ternary phase 20

PAGE 38

diagrams. Reaction crystallization relies upon the cocrystal having a lower solubility in the solvent system involved resultin g in precipitation of the cocrystal.145 Supercritical carbon dioxide can also be used as a unique solvent system to produce cocrystals provided the components are soluble.146-147 Less common techniques involve vapor diffusion and layering with different solven ts/solutions which are more common for proteins and MOFs respectively. Each of the methods described above can produce pharmaceutical cocrystals and in some cases they will reach the same crystalline form,36 while in others they may produce polymorphs or even novel forms.148 Crystal engineering with pharmaceuticals via the supr amolecular synthon approach is successful due to the occurrence of functional groups w ith H-bonding capability, which is related to natures use of supramolecular chemistry to re gulate biological functi ons and indeed APIs contain lots of H-bonding f unctionalities, see Table 1.2.149 Table 1.2: Occurrence of H-Bonding function al groups in the top 100 prescription APIs. Functional Group % of Top 100 APIs Alcohol 39 Tertiary Amine 37 Carbonyl 35 Ether 33 Secondary Amine 31 Carboxylic Acid 30 Ester 22 Narom 12 Secondary Amide 11 Sulfonamide 3 21

PAGE 39

1.5.2 History The existence of pharmaceutical cocrystals goes very far back in time, however, their importance has only recently been appreciated as highlighted by Zaworotko and Shan as, long known but li ttle studied. 138 It should be noted that the following is not a comprehensive timeline but rather a description of important ev ents that led to the current state of pharmaceutical cocrystals. One of the first appearances of pharmaceutical cocrystals in the literature resides in a French patent describing molecular complexes between barbiturates and amino pyridines.150 Barbiturates are central nervous system depressants and years later, 1968 1974, more complexes with barbiturates surfaced.151154 Almost concurrently theophylline cocr ystals with chlorosalicylic acid and sulfathiazole are reported by Shefter.155-156 Theophylline acts as a bronchodilator and sulfathiazole is an antimicrobi al so it seems highly likely that synergism was factor of interest for bacterial lung infections. Thes e APIs seemed to have drawn attention to pharmaceutical cocrystals, although they were seen as complexes at the time, which remains a correct chemical description. Since then other theophylline, sulfa drug, and barbiturate pharmaceutical cocrystals con tinued to be reported throughout the 1980s early 1990s.157-163 Caira reported complexes with su lfonamides including benzoic acids and salicylic acids as coformers with synergism in mind.164-165 In 1992 Zerkowski and Whitesides report a complex between melami ne and cyanuric acid which would later become very important.166 In 2007 there was a pet food recall due to Chinese protein export contamination in which melamine:cyanuric acid cocrystals crystallized out in the kidneys of animals causing death in many instances.167 It is known that cyanuric acid can be produced by hydrolysis of melamine, whic h was added to the pet food to falsely 22

PAGE 40

mimic proteinogenic amino acids resulting in an inflated protein content analysis. In 2008 this same phenomenon caused the Chinese baby milk scandal that caused thousands of Chinese babies to get sick while some even lost their lives.168-169 Reverting back to 1993 the highly prescribed antidepressant Depakot e, which was actually a cocrystal of salt between two sodium hydrogen divalproate oli gomers, was patented due to its superior stability compared to the free acid or the monomeric sodium salt, Figure 1.6.170 Figure 1.6: Sodium hydr ogen divalproate oligomer. Throughout the 1990s and the early 2000s more examples of pharmaceutical cocrystals continued to emerge as the advent of crystal engineering began to take hold.68, 171-178 In 2003 and 2004 the importance of these pharmaceutical compositions began to take hold in the patent landscape with the re alization that the phys icochemical properties of a molecule can be significantly changed without covalently changing the molecule.179182 A seminal article by Zaworotko in 2003 sparked lots of interest in what is now the most studied API in terms of pharmaceuti cal cocrystals, the anticonvulsant CBZ. The interest in CBZ grew rapidly and has resulted in the production and analysis of dozens of multiple component forms.145, 183-188 Later in the year Zaworotko also reported multiple component pharmaceutical phases based on the carboxylic acidNarom supramolecular 23

PAGE 41

synthon with 4,4-bipyridine and the non-steroidal anti-inflammatories (NSAIDs) aspirin, ibuprofen and flurbiprofen.70 It wasnt until 2004 that the term pharmaceutical cocrystal is coined by Almarsson and Zaworotko and used from there on out.35 1.5.3 Crystal Form Impact and Phys icochemical Property Manipulation Over 90% of all pharmaceuti cal products, such as tablets, aerosols, capsules, suspensions, and suppositories contain drug in particulate, generally crystalline form .189 SmithKline Beechum R&D News Unique crystalline forms tend to exhib it distinctive physicochemical properties effecting the dissolution, manufacturing, physical stability, permeability, and oral bioavailability of an API.190-192 Aqueous solubility of APIs is indispensible for their optimal bioavailability and efficacy. The rate limiting step for absorption of an API in an oral dosage form can be the di ssolution of that API in the gastrointestinal (GI) tract. Some of the earliest dissolu tion experiments were conducte d and published by Arthur A. Noyes and Willis R. Whitney in 1897193 where they proposed that the rate of dissolution is proportional to the difference of concentration at time t and the saturation solubility governed by Eq. (1.2) below. dx = C(CS -x) Eq.(1.2) dt 24 In Eq.(1.2) C denotes a cons tant, x denotes the concentration at time t, and CS denotes the solubility. The constant, C, in this equati on assumes that all systems behave similarly however each individual compound will have different rates of diffusion based on its respective physical properti es including but not limite d to electrostatics. This

PAGE 42

mathematical model for their observation as a general law was m odified by Bruner and Tolloczko in 1900194 to include surface area (S) and then further by Nernst and Brunner in 1904195-196 to include a diffusion coefficient (D), the thickness of the diffusion layer (h), and the volume of the dissolution medium (V) as shown in Eq. (1.3). dx = DS (CS x) Eq.(1.3) dt Vh The advent of this equation was guided by Ficks second law which predicts how diffusion changes the concentration field i.e. changes in the concentration gradient.197 These relationships are related to how kinetic s can affect solubility and will dictate how fast an API will dissolve and subsequently be absorbed. The Biopharmaceutics Classification System (BCS) rates APIs on a scale from class I IV based on so lubility and permeability198 and when a compound is deemed class II, for example, it is said to have high permeability and low solubility. Class II compounds represent a situati on where absorption is primar ily limited by the rate of dissolution and since in Eq. (3) x is negligible compared to CS, sink conditions are created. It could also be inferred that the dissolution rate is more important than the thermodynamic solubility since absorption prevents saturation in GI fluids not to mention dissolution must occur before the thermodyna mic solubility can be reached. The FDA guidance on BCS class designation is highlighted on the next page,149 although it has recently been suggested by Zaki and Bergstr m that modification of this classification system may be necessary to reflect in vivo behavior more accurately.199 25

PAGE 43

Class I High Permeability, High Solubility Class II High Permeability, Low Solubility Class III Low Permeability, High Solubility Class IV Low Permeability, Low Solubility A drug substance is considered HI GHLY SOLUBLE when the highest dose strength is soluble in < 250 ml water over a pH range of 1 to 7.5. A drug substance is considered HIGHL Y PERMEABLE when the extent of absorption in humans is determined to be > 90% of an administered dose, based on mass-balance or in comparison to an intravenous reference dose. A drug product is considered to be RAPIDLY DISSOLVING when > 85% of the labeled amount of drug substance di ssolves within 30 minutes using USP (US Pharmacopeia) apparatus I or II in a volume of < 900 ml buffer solutions. When salt formation, amorphous disper sions, particle size reductions, and formulation changes fail to improve solubility and subsequent bioa vailability the more recent technology of pharmaceutical cocrys tallization can be employed. Amorphous compositions are typically more soluble than their crystalline counterparts but are much less stable in the solid state as they tend to revert to more thermodynamically stable crystalline compositions,192 therefore their use to improve so lubility is not desired if other crystalline options exist. Dissolution rate can be either increased or decreased depending on the particular crystalline form.143 In 2003 Transform Pharmaceuticals, Inc. reported cocrystals between cis-itraconazole, an anti -fungal, and carboxylic acids with improved dissolution profiles compared to pure API and a comparable profile of corystals with Lmalic acid or L-tartaric acid against the amorphous market ed form (Sporanox) in 0.1 N HCl at 25o C, Figure 1.7.200 26

PAGE 44

Figure 1.7: Dissolution profiles into 0.1 N HCl at 25 C for itraconazole pharmaceutical corystals. Squares = Marketed Form Diamonds = Pure API, Upside down triangles = L-Malic Acid Co crystal, Right side up triangles = Succinic acid Cocrystal, and Circles = L-Tartaric Acid Cocrystal. In 2004 Childs in conjunction with SSCI Inc. divulged the first pharmaceutical cocrystal with a hydrochloride salt, fluoxetine HCl with carboxylic acids, with improved intrinsic dissolution profiles, Figure 1.8a.109 Another important part of that study was the appearance of the spring and parachute behavi or of certain cocrystals where an initial spike in the dissolution profile is followed by a decrease toward the concentration of the pure API indicating dissociation of the cocr ystal, Figure 1.8b. Dissolution enhancement through pharmaceutical cocrystallization continued to be studied and reports have continuously surfaced since 2003.108, 143, 188, 201-216 Other important physical properties like stability to humidity211, 215, 217-218 and improved compressibility141, 219 have also been demonstrated with pharmaceutical cocrystals. 27

PAGE 45

a b Figure 1.8: a) Intrinsic dissolution prof iles for fluoxetine HCl pharmaceutical cocrystals in water at 10o C. b) Spring and parachute of fluoxetine HCl:succinic acid during powder dissolution in water at 20o C. 28

PAGE 46

Pharmaceutical cocrystals represent an opportunity to diversify the number of crystal forms of a given API and in turn fine tune or even customize its physicochemical properties without the need for chemical (covalent) modification. There is no longer any doubt that cocrystals can change the physicoc hemical properties of a given API, however, the way these changes affect the pharmacokinetic (PK) profile is not predictable or fully understood due in part to the limited number of animal studies repor ted in the literature. In fact, there are only thirteen case studies in the literature to date, including patent literature, which report pharmaceutical cocrystal PK studies in animals. In 2005 Transform Pharmaceuticals, Inc. applied for two patents, both of which included animal PK data on pharmaceutic al cocrystals. One for modafinil (antinarcoleptic) revealed a cocrystal with malonic acid to show an increase in Cmax (maximum plasma concentration) and AUC (area under the curve) of about 43% and 23% respectively when compared to the marketed form, Figure 1.9.220 Figure 1.9: Plasma concentration over time for the modafinil:malonic acid pharmaceutical cocrystal over 24 hrs compared to pure API. 29

PAGE 47

The other provided data for itraconazole cocrystals with tartaric acid revealing an even greater enhancement reflected by an approximate increase in both Cmax and AUC of 80% with a reduction of Tmax (time to reach Cmax) by 80% when compared to the marketed version which is an amorphous dispersion coated on beads.221 This work was followed by McNamara in 2006 when a sodium channel blocker was cocrystallized with glutaric acid. The resulting cocrystal s howed a ca. 14-fold increase in plasma Cmax when compared to the parent API at 50 mg/kg dose in dogs, Figure 1.10.201 Figure 1.10: Plasma concentration of the glutaric acid pharmaceutical cocrystal over time in dogs at 50mg/kg oral dosing compared to parent API, Open Circles = Cocrystal, Filled Circles = Pure API Also in 2006 Variankaval in associa tion with Merck & Co., Inc. published a phosphodiesterase 4 inhibitor cocrystal with Ltartaric acid. Due to the lack of single crystal X-ray diffraction data, the determination of cocrystal rather than salt was based upon pKa and solid state NMR data. They reported plasma concentrations compared to the parent API in rhesus monkeys revealing an approximate 15-fold increase in Cmax and 30

PAGE 48

23-fold increase in AUC.202 In 2007 Chen in association w ith Merck & Co., Inc. reported the first inorganic acid cocr ystal in which they claim solubility of > 250 mg/mL and excellent in vivo performance, however they show no data to back up such claims.108 2007 also produced an example in dogs where the CBZ:saccharin cocr ystal was slightly better than the marketed form of CBZ, Figure 1.11.203 Figure 1.11: Plasma concentration in dogs for the CBZ:saccharin pharmaceutical cocrystal in a capsule compared to the mark eted form immediate release tablet at a 200 mg dose equivalent for CBZ. In 2008 Bak produced an example in which an API: excipient interaction that led to the discovery of a cocrystal between sorbic acid and a vanilloid receptor 1 antagonist, AMG 517, that proved to have about an 8-fold increase in Cmax and AUC compared to the parent API in rats.204 A patent application in 2008 reve aled a fumaric acid cocrystal of tenofovir disproxil, a reverse transcriptase i nhibitor, which showed bioequivalence to the marketed form in rats.222 2009 was equally uneventful in producing PK data on cocrystals since only one patent application, whic h originated in Europe as EP 2009010 A1, described a C-glycoside derivativ e L-proline cocrystal for the treatment of diabetes that 31

PAGE 49

had better efficacy in mice.223 2010 has continued to produce animal studies and Cheney contributed to this area with a case study of lamotrigine salts and cocrystals where the cocrystals decreased the serum concentration in rats.143 Also in 2010 Stanton provides us more animal studies with AMG 517 cocrysta llized with carboxy lic acids and their corresponding amides in rats via oral gavage at 100 mg/kg resulting in 2.4 7.1 fold increase for all cocrystals in plasma AUC over six hours compared to the free base.206 An interesting part of this study was the concom itant measurement of former concentration as part of their powder dissolution measurem ents where increases in benzoic acid and benzamide concentrations nicely complimented the decrease in the free base, API, concentration implying that the cocrysta l dissociates over time, Figure 1.12.206 Figure 1.12: Powder dissolution profiles over 4 hrs for AMG 517 pharmaceutical cocrystals. AMG 517= triangles, benzoic ac id = open squares, benzamide = filled squares, cinnamic acid = open circl es, cinnamide = filled circles. 32

PAGE 50

Finally in 2010 the indomethacin:sa ccharin pharmaceutical cocrystal was analyzed in dogs at pH 1.2 and 7.4, which e nded up being comparable to the marketed form.216 These examples provide a solid foundation for pharmaceutical cocrystals as a tool to fine tune the physic ochemical properties of APIs without making or breaking a covalent bonds with the possibility of enhancing the PK performance. 1.5.4 Intellectual Property As mentioned in section 1.5.2 patent ac tivity on pharmaceutical cocrystals began to increase after the reali zation in 2003 that physicoche mical properties could be modified via cocrystallizati on and that these new crystall ine forms are new compositions of matter with unpredictable ne w physicochemical properties en abling them as patentable forms. This new intellectual property caught the attention of the pharmaceutical industry very rapidly and caused an in crease in patent activity.179-182, 224-225 Pharmaceutical cocrystals offer the advantage of life cycle management on old APIs as well as improved clinical performance. Similar to salt sc reening, cocrystal screening has now become another tool for preformulation chemists to search for novel crystalline forms with novel physicochemical properties expanding the sc ope of pipeline production. There are two examples of how important crystal forms are to intellectual property and while they are not cocrystals they still represent the importa nce of crystalline forms in general to the pharmaceutical industry. The first is the case of ranitidine. Ranitidine HCl, marketed as Zantac, was developed in the 1970s by Allen and Hanburys Ltd. as part of the Glaxo group, now GlaxoSmithKline (GSK), for antagonism of the histamine H2 receptor for the treatment of ulcers. The hydrochloride salt was 33

PAGE 51

discovered and patented by Glaxo in 1978.226 In 1980 a batch was found to have different PXRD and IR data and this new form was called form II227 for which two patents were granted in 1985 and 1987. Before expiration of the form I patent generic companies attempted to produce form I to be ready to market it by replicating example 32 of the form I patent but they claimed this example le d to form II and therefore the form II patent is invalid but Glaxo wins in litigation pr oving form I can be made by example 32 and claims form II seeds were present in the generic companies attempts.46 The generic companies then continue to pursue form I and Glaxo sues, still claiming form II is present, but this time they lose, however the litigation took till 1998 allowing Glaxo to continue to make money on form I longer than they should have. At that time this was the highest grossing API pulling in billions of dollars. The second is the case of ritonavir, an HIV protease inhibitor marketed by Abbott Laboratories. In 1996 ritonavir was marketed as Norvir and was available as an oral liquid or a semi-solid capsule. Both of thes e formulations containe d an aqueous ethanolic solution because the solid-state form wa s not bioavailable. During research and development only one crystal form was disc overed and 240 lots of the capsules were made without any problems.54 By the middle of 1998 however, some lots had failed dissolution testing due to a new polymorph with very redu ced solubility, which was characterized and found to be a conformational polymorph. 34

PAGE 52

This new, more stable but less soluble, fo rm II of ritonavir th en popped up throughout the manufacturing process putting a halt to the old formulation. This cost Abbott about $250 million dollars and lots of bad publicity.228 These examples of crystal form control highlight the importance of extensive polymor ph screens and, in the case of ritonavir, even a liquid formulation needs solid-state fo rm screening in order to avoid potential precipitation. 1.6 References Cited (1) Muller-Dethlefs, K.; Hobza, P., Chemical Reviews 2000, 100, 143. (2) van der Waals, D., Die Kontinuitat des gasformigenund flussigen Zustandes Amsterdam, 1881. (3) Margenau, H., Reviews of Modern Physics 1939, 11, 0001. (4) Debye, P., Physikalische Zeitschrift 1920, 21 178. (5) Keesom, W. H., Physikalische Zeitschrift 1921, 22, 129. (6) Bondi, A., Journal of Physical Chemistry 1964, 68, 441. (7) Elliott, J. R.; Suresh, S. J.; Donohue, M. D., Industrial & Engineering Chemistry Research 1990, 29, 1476. (8) Steed, J. W.; Atwood, J. L., Supramolecular Chemistry John Wiley and Sons: West Sussex, United Kingdom, 2009; Vol. 2. (9) Moulton, B.; Zaworotko, M. J., Chemical Reviews 2001, 101, 1629. (10) James, S. L., Chemical Society Reviews 2003, 32, 276. (11) Pauling, L., The Nature of the Chemical Bond Cornell University Press: Ithaca, New York, 1939. 35

PAGE 53

(12) Desiraju, G. R., Accounts of Chemical Research 2002, 35, 565. (13) Lehn, J. M., Science 1985, 227, 849. (14) Wolf, K. L.; Wolff, R., Angewandte Chemie 1949, 61, 191. (15) Lehn, J. M., Angewandte Chemie-International Edition in English 1988, 27, 89. (16) Fischer, E., Ber. Deutsch. Chem. Ges. 1894, 27. (17) Pedersen, C. J., Journal of the American Chemical Society 1967, 89 7017. (18) Pedersen, C. J.; Frensdor.Hk, Angewandte Chemie-International Edition 1972, 11 16. (19) Cram, D. J., Angewandte Chemie-International Edition in English 1988, 27, 1009. (20) Cram, D. J.; Cram, J. M., Science 1974, 183, 803. (21) Dietrich, B.; Lehn, J. M.; Sauvage, J. P., Tetrahedron Letters 1969, 2885. (22) Graf, E.; Kintzinger, J. P.; Lehn, J. M.; Lemoigne, J., Journal of the American Chemical Society 1982, 104, 1672. (23) Cram, D. J., Science 1983, 219, 1177. (24) Watson, J. D.; Crick, F. H. C., Nature 1953, 171, 737. (25) Byrn, S. R.; Pfeiffer, R. R.; Stowell, J. G., Solid-State Chemistry of Drugs 2nd ed.; SSCI, Inc.: West Lafayette, Indiana, 1999. (26) Davey, R.; Garside, J., From Molecules to Crystallizers: An Introduction to Crystallization Oxford University Press: New York, 2000. (27) Steed, J. W., Crystengcomm 2003, 5, 169. (28) Bis, J. A.; Zaworotko, M. J., Crystal Growth & Design 2005, 5, 1169. (29) Aakeroy, C. B.; Fasulo, M. E.; Desper, J., Molecular Pharmaceutics 2007, 4, 317. (30) Childs, S. L.; Stahly, G. P.; Park, A., Molecular Pharmaceutics 2007, 4, 323. 36

PAGE 54

(31) Desiraju, G. R., Crystengcomm 2003, 5, 466. (32) Dunitz, J. D., Crystengcomm 2003, 5, 506. (33) Trask, A. V., Molecular Pharmaceutics 2007, 4, 301. (34) Zukerman-Schpector, J.; Tiekink, E. R. T., Zeitschrift Fur Kristallographie 2008, 223, 233. (35) Almarsson, O.; Zaworotko, M. J., Chemical Communications 2004, 1889. (36) Weyna, D. R.; Shattock, T.; Vishweshwar, P.; Zaworotko, M. J., Crystal Growth & Design 2009, 9, 1106. (37) Hoogsteen, Acta Cryst 1959, 12 822. (38) Schmidt, J.; Snipes, J., International Journal of R adiation Biology and Related Studies in Physics Chemistry and Medicine 1967, 13, 101. (39) Whler, F., Justus Liebigs Ann. Chem. 1844, 51, 153. (40) Vanniekerk, J. N.; Saunder, D. H., Acta Crystallographica 1948, 1, 44. (41) Buck, J. S.; Ide, W. S., Journal of the American Chemical Society 1931, 53, 2784. (42) Anderson, J. S., Nature 1937, 140, 583. (43) Pekker, S.; Kovats, E.; Oszlanyi, G. ; Benyei, G.; Klupp, G.; Bortel, G.; Jalsovszky, I.; Jakab, E.; Borondics, F.; Kamaras, K.; Bokor, M.; Kriza, G.; Tompa, K.; Faigel, G., Nature Materials 2005, 4 764. (44) Hall, B.; Devlin, J. P., Journal of Physical Chemistry 1967, 71, 465. (45) Ling, A. R.; Baker, J. L., J. Chem. Soc. 1893, 63, 1314. (46) Bernstein, J., Polymorphism in Molecular Crystals Clarendon Press: Oxford, 2002. (47) McCrone, W. C.; Haleblian, J., Journal of Pharmaceutical Sciences 1969, 58, 911. 37

PAGE 55

(48) Cheney, M. L.; Weyna, D. R.; Shan, N.; Hanna, M.; Wojtas, L.; Zaworotko, M. J., Crystal Growth & Design 2010, 10, 4401. (49) Chen, S.; Guzei, I. A.; Yu, L., Journal of the Americ an Chemical Society 2005, 127, 9881. (50) Vippagunta, S. R.; Brittain, H. G.; Grant, D. J. W., Advanced Drug Delivery Reviews 2001, 48, 3. (51) Corradin.P, Chimica & L Industria 1973, 55 122. (52) Bernstein, J.; Hagler, A. T., Journal of the American Chemical Society 1978, 100 673. (53) Brittain, H. G., Polymorphism in Pharmaceutical Solids Marcel Dekker: New York, 1999; Vol. 95. (54) Bauer, J.; Spanton, S.; Henry, R.; Quick, J.; Dziki, W.; Porter, W.; Morris, J., Pharmaceutical Research 2001, 18, 859. (55) McMurray, J.; Fay, R. C., Chemistry. 4 ed.; Prentice Hall: Upper Saddle River, New Jersey, 2004. (56) Lapidus, S. H.; Stephens, P. W.; Aror a, K. K.; Shattock, T. R.; Zaworotko, M. J., Crystal Growth & Design 2010, 10, 4630. (57) Braga, D., Chemical Communications 2003 2751. (58) Desiraju, G. R., Crystal Engineering: The Design of Organic Solids Elsevier: 1989. (59) Pepinsky, R., Physical Review 1955, 100, 971. (60) Schmidt, G. M. J., Pure and Applied Chemistry 1971, 27 647. 38

PAGE 56

(61) MacGillivray, L. R.; Reid, J. L.; Ripmeester, J. A., Journal of the American Chemical Society 2000, 122, 7817. (62) Etter, M. C., Accounts of Chemical Research 1990, 23, 120. (63) Etter, M. C., Journal of Physical Chemistry 1991, 95, 4601. (64) Etter, M. C.; Adsmond, D. A., Journal of the Chemical Society-Chemical Communications 1990, 589. (65) Etter, M. C.; Adsmond, D. A.; Britton, D., Acta Crystallographica Section CCrystal Structure Communications 1990, 46, 933. (66) Etter, M. C.; Macd onald, J. C.; Bernstein, J., Acta Crystallographica Section BStructural Science 1990, 46, 256. (67) Etter, M. C.; Urbanczyklipkows ka, Z.; Ziaebrahimi, M.; Panunto, T. W., Journal of the American Chemical Society 1990, 112, 8415. (68) VidenovaAdrabinska, V.; Etter, M. C., Journal of Chemical Crystallography 1995, 25, 823. (69) Desiraju, G. R., Angewandte Chemie-Internati onal Edition in English 1995, 34, 2311. (70) Walsh, R. D. B.; Bradner, M. W.; Fleischman, S.; Morales, L. A.; Moulton, B.; Rodriguez-Hornedo, N.; Zaworotko, M. J., Chemical Communications 2003, 186. (71) Lisgarten, J. N.; Palmer, R. A.; Saldanha, J. W., Journal of Crystallographic and Spectroscopic Research 1989, 19, 641. (72) Vishweshwar, P.; McMahon, J. A.; Oliveira, M.; Peterson, M. L.; Zaworotko, M. J., Journal of the American Chemical Society 2005, 127, 16802. 39

PAGE 57

(73) Vishweshwar, P.; Beauchamp, D. A.; Zaworotko, M. J., Crystal Growth & Design 2006, 6, 2429. (74) Leiserowitz.L; Schmidt, G. M. J., Journal of the Chemical Society a -Inorganic Physical Theoretical 1969, 2372. (75) Etter, M. C., Journal of the American Chemical Society 1982, 104 1095. (76) Leiserowitz, L., Acta Crystallographica Section B-Structural Science 1976, 32, 775. (77) Beckman, Z. Physikal. Chem. 1897, 22, 610. (78) Wall, F. T.; Banes, F. W., Journal of the American Chemical Society 1945, 67, 898. (79) Wall, F. T.; Rouse, P. E., Journal of the American Chemical Society 1941, 63, 3002. (80) Leiserowitz, L.; Nader, F., Acta Crystallographica Section B-Structural Science 1977, 33, 2719. (81) Fleischman, S. G.; Kuduva, S. S.; McMahon, J. A.; Moulton, B.; Walsh, R. D. B.; Rodriguez-Hornedo, N.; Zaworotko, M. J., Crystal Growth & Design 2003, 3, 909. (82) McMahon, J. A.; Bis, J. A.; Vishwesh war, P.; Shattock, T. R.; McLaughlin, O. L.; Zaworotko, M. J., Zeitschrift Fur Kristallographie 2005, 220, 340. (83) Aakeroy, C. B.; Beatty, A. M.; Helfrich, B. A., Journal of the American Chemical Society 2002, 124, 14425. (84) Bhogala, B. R.; Vishweshwar, P.; Nangia, A., Crystal Growth & Design 2002, 2, 325. (85) Bond, A. D., Chemical Communications 2003, 250. 40

PAGE 58

(86) Vishweshwar, P.; Nangia, A.; Lynch, V. M., Journal of Organic Chemistry 2002, 67, 556. (87) Dey, A.; Kirchner, M. T.; Vangala, V. R.; Desiraju, G. R.; Mondal, R.; Howard, J. A. K., Journal of the American Chemical Society 2005, 127, 10545. (88) Hanessian, S.; Saladino, R.; Margarita, R.; Simard, M., Chemistry-a European Journal 1999, 5, 2169. (89) Mondal, R.; Howard, J. A. K.; Banerjee, R.; Desiraju, G. R., Crystal Growth & Design 2006, 6, 2507. (90) Vangala, V. R.; Mondal, R.; Broder, C. K.; Howard, J. A. K.; Desiraju, G. R., Crystal Growth & Design 2005, 5, 99. (91) MacGillivray, L. R.; Atwood, J. L., Journal of Solid State Chemistry 2000, 152, 199. (92) Papaefstathiou, G. S.; MacGillivray, L. R., Organic Letters 2001, 3, 3835. (93) Vishweshwar, P.; Nangia, A.; Lynch, V. M., Crystengcomm 2003 164. (94) Aakeroy, C. B., Acta Crystallographica Section B-Structural Science 1997, 53, 569. (95) Aakeroy, C. B.; Beatty, A. M., Australian Journal of Chemistry 2001, 54, 409. (96) Aakeroy, C. B.; Beatty, A. M.; Helfrich, B. A., Angewandte Chemie-International Edition 2001, 40, 3240. (97) Aakeroy, C. B.; Seddon, K. R., Chemical Society Reviews 1993, 22, 397. (98) Aakeroy, C. B.; Desper, J.; Smith, M. M., Chemical Communications 2007, 3936. (99) Aakeroy, C. B.; Salmon, D. J., Crystengcomm 2005, 7, 439. 41

PAGE 59

(100) Bis, J. A.; Vishweshwar, P.; Weyna, D.; Zaworotko, M. J., Molecular Pharmaceutics 2007, 4, 401. (101) Shattock, T. R.; Arora, K. K.; Vishweshwar, P.; Zaworotko, M. J., Crystal Growth & Design 2008, 8, 4533. (102) Kavuru, P.; Aboarayes, D.; Arora, K. K.; Clarke, H. D.; Kennedy, A.; Marshall, L.; Ong, T. T.; Perman, J.; Pujari, T.; Wojtas, L.; Zaworotko, M. J., Crystal Growth & Design 2010, 10, 3568. (103) Aakeroy, C. B.; Hughes, D. P.; Nieuwenhuyzen, M., Journal of the American Chemical Society 1996, 118, 10134. (104) Gilli, P.; Bertolasi, V.; Ferretti, V.; Gilli, G., Journal of the American Chemical Society 1994, 116, 909. (105) Russell, V. A.; Etter, M. C.; Ward, M. D., Chemistry of Materials 1994, 6, 1206. (106) Ward, M. D., Chemical Communications 2005, 5838. (107) Sudhakar, V.; Bhat, T. N.; Vijayan, M., Acta Crystallographica Section BStructural Science 1980, 36, 125. (108) Chen, A. M.; Ellison, M. E.; Peres ypkin, A.; Wenslow, R. M.; Variankaval, N.; Savarin, C. G.; Natishan, T. K.; Mathre, D. J.; Dormer, P. G.; Euler, D. H.; Ball, R. G.; Ye, Z. X.; Wang, Y. L.; Santos, I., Chemical Communications 2007 419. (109) Childs, S. L.; Chyall, L. J.; Dunla p, J. T.; Smolenskaya, V. N.; Stahly, B. C.; Stahly, G. P., Journal of the Americ an Chemical Society 2004, 126, 13335. (110) Cincic, D.; Friscic, T.; Jones, W., Journal of the American Chemical Society 2008, 130, 7524. 42

PAGE 60

(111) Aakeroy, C. B.; Beatty, A. M.; Lorimer, K. R., Structural Chemistry 1999, 10, 229. (112) Aakeroy, C. B.; Schulthei ss, N.; Desper, J.; Moore, C., Crystengcomm 2007, 9, 420. (113) Banerjee, R.; Desiraju, G. R.; Mondal, R.; Howard, J. A. K., Chemistry-a European Journal 2004, 10, 3373. (114) Metrangolo, P.; Meyer, F.; P ilati, T.; Resnati, G.; Terraneo, G., Angewandte Chemie-International Edition 2008, 47 6114. (115) Metrangolo, P.; Neukirch, H.; Pilati, T.; Resnati, G., Accounts of Chemical Research 2005, 38, 386. (116) Thakur, T. S.; Kirchner, M. T.; Blaser, D.; Boese, R.; Desiraju, G. R., Crystengcomm 2010, 12 2079. (117) Allen, F. H., Acta Crystallographica Section B-Structural Science 2002, 58, 380. (118) Allen, F. H.; Hoy, V. J., International Tables for Crystallography 2006; Vol. F. (119) Allen, F. H.; Motherwell, W. D. S., Acta Crystallographica Section B-Structural Science 2002, 58, 407. (120) Blatov, V. A.; Carlucci, L.; Ciani, G.; Proserpio, D. M., Crystengcomm 2004, 6, 377. (121) Infantes, L.; Motherwell, W. D. S., Zeitschrift Fur Kristallographie 2005, 220, 333. (122) Oswald, I. D. H.; Motherwell, W. D. S.; Parsons, S.; Pidcock, E.; Pulham, C. R., Crystallography Reviews 2004, 10, 57. 43

PAGE 61

(123) Clarke, H. D.; Arora, K. K.; Bass, H.; Kavuru, P.; Ong, T. T.; Pujari, T.; Wojtas, L.; Zaworotko, M. J., Crystal Growth & Design 2010, 10, 2152. (124) Russell, V. A.; Evans, C. C.; Li, W. J.; Ward, M. D., Science 1997, 276, 575. (125) Zaworotko, M., Journal of Pharmacy and Pharmacology 2006, 58 A91. (126) Taylor, L. D.; Warner, J. C. US 5,338,644A Process and composition for use in photographic materials c ontaining hydroquihones. 1994. (127) Maddox, J., Nature 1988, 335, 201. (128) Lommerse, J. P. M.; Motherwell, W. D. S.; Ammon, H. L.; Dunitz, J. D.; Gavezzotti, A.; Hofmann, D. W. M.; Leusen, F. J. J.; Mooij, W. T. M.; Price, S. L.; Schweizer, B.; Schmidt, M. U.; van Eijck, B. P.; Verwer, P.; Williams, D. E., Acta Crystallographica Section B-Structural Science 2000, 56 697. (129) Dunitz, J. D., Chemical Communications 2003, 545. (130) Ouvrard, C.; Price, S. L., Crystal Growth & Design 2004, 4, 1119. (131) Price, S. L., Advanced Drug Delivery Reviews 2004, 56, 301. (132) Price, S. L., Crystengcomm 2004, 6, 344. (133) Thakur, T. S.; Desiraju, G. R., Crystal Growth & Design 2008, 8, 4031. (134) Day, G. M.; Cooper, T. G.; Cruz-C abeza, A. J.; Hejczyk, K. E.; Ammon, H. L.; Boerrigter, S. X. M.; Tan, J. S.; Della Valle R. G.; Venuti, E.; Jose, J.; Gadre, S. R.; Desiraju, G. R.; Thakur, T. S.; van Eijc k, B. P.; Facelli, J. C.; Bazterra, V. E.; Ferraro, M. B.; Hofmann, D. W. M.; Neum ann, M. A.; Leusen, F. J. J.; Kendrick, J.; Price, S. L.; Misquitta, A. J.; Karame rtzanis, P. G.; Welch, G. W. A.; Scheraga, H. A.; Arnautova, Y. A.; Schmidt, M. U.; van de Streek, J.; Wolf, A. K.; 44

PAGE 62

Schweizer, B., Acta Crystallographica Section B-Structural Science 2009, 65, 107. (135) Price, S. L., Accounts of Chemical Research 2009, 42, 117. (136) Motherwell, W. D. S.; Ammon, H. L.; Dunitz, J. D.; Dzyabchenko, A.; Erk, P.; Gavezzotti, A.; Hofmann, D. W. M.; Leusen, F. J. J.; Lommerse, J. P. M.; Mooij, W. T. M.; Price, S. L.; Scheraga, H.; Sc hweizer, B.; Schmidt, M. U.; van Eijck, B. P.; Verwer, P.; Williams, D. E., Acta Crystallographica Section B-Structural Science 2002, 58, 647. (137) Day, G. M.; Motherwell, W. D. S.; Ammon, H. L.; Boerrigter, S. X. M.; Della Valle, R. G.; Venuti, E.; Dzyabchenko, A.; Dunitz, J. D.; Schweizer, B.; van Eijck, B. P.; Erk, P.; Facelli, J. C.; Bazt erra, V. E.; Ferraro, M. B.; Hofmann, D. W. M.; Leusen, F. J. J.; Liang, C.; Pantelid es, C. C.; Karamertzanis, P. G.; Price, S. L.; Lewis, T. C.; Nowell, H.; Torrisi, A.; Scheraga, H. A.; Arnautova, Y. A.; Schmidt, M. U.; Verwer, P., Acta Crystallographica Section B-Structural Science 2005, 61, 511. (138) Shan, N.; Zaworotko, M. J., Drug Discovery Today 2008, 13, 440. (139) Trask, A. V.; Jones, W., Crystal en gineering of organic co crystals by the solidstate grinding approach. In Organic Solid State Reactions Springer-Verlag Berlin: Berlin, 2005; Vol. 254, pp 41. (140) Shan, N.; Toda, F.; Jones, W., Chemical Communications 2002, 2372. (141) Karki, S.; Friscic, T.; Fabian, L.; Laity, P. R.; Day, G. M.; Jones, W., Advanced Materials 2009, 21, 3905. 45

PAGE 63

(142) Dhumal, R. S.; Kelly, A. L.; York, P.; Coates, P. D.; Paradkar, A., Pharmaceutical Research 2010, 27, 2725. (143) Cheney, M. L.; Shan, N.; Healey, E. R.; Hanna, M.; Wojtas, L.; Zaworotko, M. J.; Sava, V.; Song, S. J.; Sanchez-Ramos, J. R., Crystal Growth & Design 2010, 10, 394. (144) Friscic, T.; Childs, S. L.; Rizvi, S. A. A.; Jones, W., Crystengcomm 2009, 11, 418. (145) Rodriguez-Hornedo, N.; Nehru, S. J.; Seefeldt, K. F.; Pagan-Torres, Y.; Falkiewicz, C. J., Molecular Pharmaceutics 2006, 3, 362. (146) Padrela, L.; Rodrigues, M. A.; Velaga S. P.; Fernandes, A. C.; Matos, H. A.; de Azevedo, E. G., Journal of Supercritical Fluids 2010, 53, 156. (147) Padrela, L.; Rodrigues, M. A.; Vela ga, S. R.; Matos, H. A.; de Azevedo, E. G., European Journal of Pharmaceutical Sciences 2009, 38, 9. (148) Trask, A. V.; van de Streek, J.; Motherwell, W. D. S.; Jones, W., Crystal Growth & Design 2005, 5, 2233. (149) Rather, B.; Zaworotko, M. J., Chemical Communications 2003, 830. (150) Proce'de' de pre'paration de de'rive' s de combinasions oxyet aminopyridques. FR 769586, 1934. (151) Kim, S. H.; Rich, A., Proceedings of the National Academy of Sciences of the United States of America 1968, 60, 402. (152) Kiryu, S., Journal of Pharmaceutical Sciences 1971, 60, 699. (153) Voet, D., Journal of the American Chemical Society 1972, 94, 8213. (154) Hsu, I. N.; Craven, B. M., Acta Crystallographica Section B-Structural Science 1974, B 30, 843. 46

PAGE 64

(155) Shefter, E., Journal of Pharmaceutical Sciences 1969, 58, 710. (156) Shefter, E.; Sackman, P., Journal of Pharmaceutical Sciences 1971, 60, 282. (157) Giordano, F.; Bettinetti, G. P.; Lamanna, A.; Ferloni, P., Farmaco-Edizione Scientifica 1977, 32, 889. (158) Nakao, S.; Fujii, S.; Sakaki, T.; Tomita, K. I., Acta Crystallographica Section BStructural Science 1977, 33, 1373. (159) Aoki, K.; Ichikawa, T.; Koinuma, Y.; Iitaka, Y., Acta Crystallographica Section B-Structural Science 1978, 34, 2333. (160) Giuseppetti, G.; Tadini, C.; Bettin etti, G. P.; Giordano, F.; Lamanna, A., Farmaco-Edizione Scientifica 1980, 35 138. (161) Shimizu, N.; Nishigaki, S.; Nakai, Y.; Osaki, K., Acta Crystallographica Section B-Structural Science 1982, 38, 2309. (162) Wiedenfeld, H.; Knoch, F., Archiv Der Pharmazie 1986, 319, 654. (163) Zerkowski, J. A.; Seto, C. T.; Wierda, D. A.; Whitesides, G. M., Journal of the American Chemical Society 1990, 112 9025. (164) Caira, M. R., Journal of Crystallographic and Spectroscopic Research 1991, 21, 641. (165) Caira, M. R., Journal of Crystallographic and Spectroscopic Research 1992, 22, 193. (166) Zerkowski, J. A.; Seto, C. T.; Whitesides, G. M., Journal of the American Chemical Society 1992, 114, 5473. (167) He, L.; Liu, Y.; Lin, M.; Awika J.; Ledoux, D.; Li, H.; Mustapha, A., Sensing and Instrumentation for Food Quality and Safety 2008, 2, 66. 47

PAGE 65

(168) Scott McDonald, "Nearly 53,000 Chinese children sick from milk". Associated Press 2008. (169) Jane Macartney. "China baby milk scandal spreads as sick toll rises to 13,000". The Times (London). 2008 (170) Meade, E. M. Sodium H ydrogen Divalproate Oligomer. US 5,212,326, 1993. (171) Zerkowski, J. A.; Mathias, J. P.; Whitesides, G. M., Journal of the American Chemical Society 1994, 116, 4305. (172) Zerkowski, J. A.; Whitesides, G. M., Journal of the American Chemical Society 1994, 116, 4298. (173) Koshima, H.; Ding, K. L.; Chisaka, Y.; Matsuura, T., Tetrahedron-Asymmetry 1995, 6, 101. (174) AMOS, J. G.; INDELICATO, J. M.; PASINI, C. E.; REUTZEL, S. M. Bicyclic beta-lactam/paraben complexes. EP0637587 B1 1997. (175) Zerkowski, J. A.; MacD onald, J. C.; Whitesides, G. M., Chemistry of Materials 1997, 9, 1933. (176) BAJAJ, V.; MADAN, A. K. A process for preparation of urea complexes of vitamin e and its esters IN182620 A1, 1999. (177) Lynch, D. E.; Sandhu, P.; Parsons, S., Australian Journal of Chemistry 2000, 53, 383. (178) Vishweshwar, P.; Thaimattam, R.; Jaskolski, M.; Desiraju, G. R., Chemical Communications 2002, 1830. 48

PAGE 66

(179) Remenar, J.; Macphee, M.; Peters on, M. L.; Morissette, S. L.; Almarsson, Novel Conazole Crystalline Forms and Related Processes, Pharmaceutical Compositions and Methods WO 2003101392, 2003. (180) Almarsson, .; Hickey, M. B.; Pe terson, M.; Zaworotko, M. J.; Moulton, B.; Rodriguez-Hornedo, N. Pharmaceutical Cocrystal Compositions of Drugs such as Carbamazepine, Celecoxib, Olanzapine, Itraconazole, Topirimate, Modafinil, 5Fluorouracil, Hydrochlorothiazide, Acetaminophen, Aspirin, Flurbiprofen, Phenytoin, an Ibuprofen WO2004078161, 2004. (181) Childs, S. L. Novel Cocrystallization. WO2004064762, 2004. (182) Hickey, M. B.; Remenar, J. Novel Olanzapine Forms and Related Methods of Treatment WO2004089313, 2004. (183) Jayasankar, A.; Somwangthanaroj, A.; Shao, Z. J.; Rodriguez-Hornedo, N., Pharmaceutical Research 2006, 23, 2381. (184) Nehm, S. J.; Rodriguez-Spong, B.; Rodriguez-Hornedo, N., Crystal Growth & Design 2006, 6, 592. (185) Jayasankar, A.; Good, D. J.; Rodriguez-Hornedo, N., Molecular Pharmaceutics 2007, 4, 360. (186) Seefeldt, K.; Miller, J.; Alva rez-Nunez, F.; Rodriguez-Hornedo, N., Journal of Pharmaceutical Sciences 2007, 96, 1147. (187) Childs, S. L.; Wood, P. A.; Rodriguez-Hornedo, N.; Reddy, L. S.; Hardcastle, K. I., Crystal Growth & Design 2009, 9 1869. (188) Good, D. J.; Rodriguez-Hornedo, N., Crystal Growth & Design 2009, 9, 2252. 49

PAGE 67

(189) Valder, C.; Merrifield, D., Pharmaceutical Technology, SmithKline Beechum RD News 1996. (190) Byrn, S. R.; Pfeiffer, R. R.; Ste phenson, G.; Grant, D. J. W.; Gleason, W. B., Chemistry of Materials 1994, 6, 1148. (191) Berge, S. M.; Bighley, L. D.; Monkhouse, D. C., Journal of Pharmaceutical Sciences 1977, 66, 1. (192) Haleblian, J. K., Journal of Pharmaceutical Sciences 1975, 64, 1269. (193) Noyes, A.; Whitney, W., Journal of the American Chemical Society 1897, 19, 930. (194) Bruner, L.; Tolloczko, S., Zeitschrift Fur Physikalische Chemie--Stochiometrie Und Verwandtschaftslehre 1900, 35, 283. (195) Brunner, E., Zeitschrift Fur Physikalische Chemie--Stochiometrie Und Verwandtschaftslehre 1904, 47, 56. (196) Nernst, N., Zeitschrift Fur Physikalische Chemie--Stochiometrie Und Verwandtschaftslehre 1904, 47, 52. (197) Dokoumetzidis, A.; Macheras, P., International Journal of Pharmaceutics 2006, 321, 1. (198) Amidon, G. L.; Lennernas, H.; Shah, V. P.; Crison, J. R., Pharmaceutical Research 1995, 12, 413. (199) Zaki, N. M.; Artursson, P.; Bergstrom, C. A. S., Molecular Pharmaceutics 2010, 7, 1478. 50

PAGE 68

(200) Remenar, J. F.; Morissette, S. L.; Peterson, M. L.; Moulton, B.; MacPhee, J. M.; Guzman, H. R.; Almarsson, O., Journal of the American Chemical Society 2003, 125, 8456. (201) McNamara, D. P.; Childs, S. L.; Gi ordano, J.; Iarriccio, A. ; Cassidy, J.; Shet, M. S.; Mannion, R.; O'Donnell, E.; Park, A., Pharmaceutical Research 2006, 23, 1888. (202) Variankaval, N.; Wenslow, R.; Murry, J.; Hartman, R.; Helmy, R.; Kwong, E.; Clas, S. D.; Dalton, C.; Santos, I., Crystal Growth & Design 2006, 6, 690. (203) Hickey, M. B.; Peterson, M. L.; Scoppe ttuolo, L. A.; Morrisette, S. L.; Vetter, A.; Guzman, H.; Remenar, J. F.; Zhang, Z.; Tawa, M. D.; Haley, S.; Zaworotko, M. J.; Almarsson, O., European Journal of Pharmaceutics and Biopharmaceutics 2007, 67, 112. (204) Annette Bak; Gore, A.; Yanez, E.; Stanton, M.; Tufekcic, S.; Syed, R.; Akrami, A.; Rose, M.; Surapaneni, S.; Bostick, T.; King, A.; Neervannan, S.; Ostovic, D.; Koparkar, A., Journal of Pharmaceutical Sciences 2008, 97, 3942. (205) Cheney, M. L.; Weyna, D. R.; Shan, N.; Hanna, M.; Wojtas, L.; Zaworotko, M. J., Journal of Pharmaceutical Sciences 2010, Published Online DOI: 10.1002/jps.22434. (206) Stanton, M. K.; Kelly, R. C.; Collet ti, A.; Kiang, Y. H.; Langley, M.; Munson, E. J.; Peterson, M. L.; Roberts, J.; Wells, M., Journal of Pharmaceutical Sciences 2010, 99, 3769. (207) Basavoju, S.; Bostrom, D.; Velaga, S. P., Crystal Growth & Design 2006, 6, 2699. 51

PAGE 69

(208) Remenar, J. F.; Peterson, M. L. ; Stephens, P. W.; Zhang, Z.; Zimenkov, Y.; Hickey, M. B., Molecular Pharmaceutics 2007, 4, 386. (209) Zegarac, M.; Mestrovic, E.; Dumbovic, A.; Devcic, M.; Tudja, P. Pharmaceutically Acceptable Co-Cry stalline Forms of Sildenafil. WO 2007/080362 A1, 2007. (210) Shiraki, K.; Takata, N.; Takano, R.; Hayashi, Y.; Terada, K., Pharmaceutical Research 2008, 25, 2581. (211) Stanton, M. K.; Bak, A., Crystal Growth & Design 2008, 8, 3856. (212) Stanton, M. K.; Tufekcic, S.; Morgan, C.; Bak, A., Crystal Growth & Design 2009, 9, 1344. (213) Lee, H. G.; Zhang, G. G. Z.; Flanagan, D. R., Journal of Pharmaceutical Sciences 2010, n/a. (214) Takata, N.; Takano, R.; Uekusa, H.; Hayashi, Y.; Terada, K., Crystal Growth & Design 2010, 10, 2116. (215) Basavoju, S.; Bostrom, D.; Velaga, S. P., Pharmaceutical Research 2008, 25, 530. (216) Jung, M. S.; Kim, J. S.; Kim, M. S.; Alhalaweh, A.; Cho, W.; Hwang, S. J.; Velaga, S. P., Journal of Pharmacy and Pharmacology 2010, 62, 1560. (217) Trask, A. V.; Motherwell, W. D. S.; Jones, W., International Journal of Pharmaceutics 2006, 320 114. (218) Friscic, T.; Fabian, L.; Burley, J. C.; Reid, D. G.; Duer, M. J.; Jones, W., Chemical Communications 2008, 1644. (219) Rahman, Z.; Samy, R.; Sayeed, V. A.; Khan, M. A., Pharmaceutical Development and Technology 0 1. 52

PAGE 70

53 (220) Peterson, M.; Bourghol Hickey, M.; Oliveira, M.; Almarsson, O.; Remenar, J. Modafinil compositions. US 2005267209 A1, 2005. (221) Remenar, J.; MacPhee, M.; Peterson, M.; Morissette, S. L.; Almarsson, O. Novel Crystalline Forms of Conazoles and Me thods of Making and Using the Same. WO/2005/092884, 2005. (222) E. Dova; J. M. Mazurek; Anker, J. Tenofovir disoproxil hemi-fumaric acid cocrystal. WO/2008/143500 2008. (223) Imamura, M. Cocrystal of CGlycoside Derivative and L-Proline. WO 2007/114475 A1, 2007. (224) Wang, S.; Chen, J. Gossypol Co-crystals and the use thereof. WO2005094804. (225) Childs, S. L. Metronidazole cocr ystals and imipramine cocrystals. WO 2007067727 A2 2007. (226) Price, B. J.; Clitherow, J. W.; Brad shaw, J. W. Aminoalkyl furan derivatives. US 4128568, 1978. (227) Crookes, D. Amino Alkyl Furan Derivative. US 4521431, 1985. (228) Chemical and Engineering News Vol 85 No.25, 2007.

PAGE 71

Chapter 2: Mechanochemistry: Solvent Drop Grinding vs. Solution Evaporation for the Synthesis of Pharmaceutical Co crystals Involving Carbamazepine 2.1 Background Carbamazepine (CBZ) is an API with the brand name Tegretol that is used mainly as an anticonvulsant but al so as an analgesic and is known as 5 H dibenzo[ b,f ]azepine-5-carboxamide. Oral dosage forms consist of tablets or a suspension with eighty percent bioavailability despit e its low aqueous solubility, however, the absorption is slow in humans w ith a variable half life of twenty five to sixty five hours. CBZ represents a microcosm of the challenges and opportunities rela ted to crystal forms of APIs since it exhibits polymorphism1 to the extent of eight different polymorphs2-10 and readily forms a less soluble dihydrate when exposed to moisture.11 The amide functionality predisposes this molecule to various H-bondi ng possibilities. Therefore, it should not be surprising that CBZ was an ear ly candidate for cocrystallization studies12-13 and it is one of the few APIs for which there is published data con cerning bioavailability of a cocrystal, the 1:1 cocr ystal of CBZ and saccharin. This cocrystal does not form a hydrate and it exhibits improved bioavailabil ity in dogs when compared to Tegretol tablets, Figure 1.11.14 In fact, it has been proposed th at the CBZ:Nicotinamide cocrystal exhibits a 152-fold increase in aqueous solubility compared to the dihydrate.15 54

PAGE 72

Zaworotko12, 16 and others17 have reported that CB Z forms supramolecular adducts with a wide range of complementary cocrystal formers and a recent report addressed the preparation of CBZ cocrystals through f our different methodologies, including solvent drop grinding (SDG).17 The research herein addresses SDG as a methodology for the reproduction of cocrystals with CBZ that have been previously reported via solution crystallization using a nd a wide range of organic coformers and solvents with varying functionalities. Mechanochemistry in its earliest incarnat ion, grinding of solids, is a technique that dates back thousands of years in that tr ibal medicine utilized something similar to a mortar and pestle to process medicinally beneficial plants18 and even today a mortar and pestle symbolizes pharmacies and schools of pharmacy. To the best of my knowledge, the first report of mechanochemistry in the scientific literature of appeared over 160 years ago in 1844 when Whler prepared the 1:1 quinonehydroquinone cocrystal by kugelchen.19 In 1893 cocrystals of charge transf er complexes were studied and they were also prepared via grinding of solids.20 Today solid state grinding represents an attractive alternative to solution processes beca use it is an inherently Green approach to synthetic chemistry in that it offers a facile and low or no waste methodology for discovery or processing of new or existi ng compounds. Recent interest in grinding and cocrystals can be traced to the 1980s, when Etter et al demonstrated that dry grinding, also referred to as neat grinding, represents a viable methodology to prepare cocrystals of, for example. methyladenine and methylthymine .21-29 An important refinement to grinding came when a small but controlled amount of solvent was added during the grinding process. SDG was reported by Shan et al in 2002 and it became evident that the 55

PAGE 73

kinetics of cocrystal formation can be significantly enhanced through SDG.30 SDG can offer other advantages over solution crystall ization since dissolution of both cocrystal formers is not required, solvent interactions that might interfere with solute-solute interactions are more limited and we are take n to a region of the ternary phase diagram that might favor cocrystals over starting ma terials. Indeed, in certain cases SDG can produce cocrystals that are not readily obtainable through solution crystallization and SDG can be employed for control over polymorphism in caffeine:glutaric acid cocrystals.31-35 Two concomitant polymorphs of the caffeine:glutaric acid cocrystal result from solution but SDG with a non-polar solv ent afforded only form I whereas a polar solvent afforded only form II.36 SDG has been used in other situations to control polymorphism.37-38 Whereas there are numerous examples of cocrystals made by SDG that have also been grown from soluti on and characterized by single crystal X-ray diffraction, a direct comparison of the efficacy of SDG vs. solution crystallization over a wide range of cocrystal formers ha s not been systematically studied.39-42 This is addressed herein through studying cocrystals with the range of supramolecular synthons43 involving CBZ, Figures 2.1 and 2.2. Figure 2.1: Chemical diagram of CBZ. 56

PAGE 74

Figure 2.2: Chemical diagrams of the cocrystal formers employed. 2.2 Experimental Details 2.2.1 Materials Reagents were obtained from commercial sources and used as received. Solvents were obtained from commercial sour ces and distilled before use. 2.2.2 Methods Cocrystals were characterized by in frared spectroscopy (IR), X-ray powder diffraction (PXRD), and single crystal X-ray analysis where applicable. IR data was collected using a Nicolet Avat ar 320 FTIR instrument. PXRD data was collected using a Bruker AXS D8 discover X-ray di ffractometer equipped with GADDS (General Area Diffraction Detection System), a Bruker AXS HI-STAR area detector at a distance of 57

PAGE 75

15.05 cm as per system calibration, a copper sour ce, an automated x-y-z stage, and a 0.5 mm collimator. Data were collected over a 3.0-40.0 2 range at a step size of 0.05 2 SDG: Cocrystal formers were subjected to grinding with an agate mortar and pestle for 4 minutes and thereafter characterized by IR and PXRD. In the event that partial conversion was achieved a ball mill, SPE X 8000M Mixer/Mill, was used for 2 increments of 10 minutes with addition of solvent prior to each increment. The same stoichiometric ratios that the cocrystals exhibited after solution crystallization were used for grinding unless otherwise specified. The vol umes of solvents used during SDG were are follows: methanol (MeOH) 20 L, ethylacetate (EtOAc) 20 L, dimethylsulfoxide (DMSO) 4 L, water 5 L, toluene (Tol) 20 L, cyclohexane (Cychex) 20 L, chloroform 20 L, dimethylformamide (DMF) 4 L. Each solvent was used for each cocrystal experiment. Experimental details for each of the singl e crystal crystallizations have been previously published12, 14, 16-17 and the SDG experiments were conducted as described earlier. The following cocrystals were studied in terms of SDG: 2:1 cocrystal of CBZ and 4,4-bipyridine, 1; hydrate of the 1:1 cocrystal of CBZ and 4-aminobenzoic Acid, 2; 1:1 cocrystal of CBZ and 2,6-pyridinedicarboxylic Acid, 3; 2:1 cocrystal of CBZ and benzoquinone, 4; 2:1 cocrystal of CBZ and terephthalaldehyde, 5; 1:1 cocrystal of CBZ and saccharin, 6; 1:1 cocrystal of CBZ and nicotinamide, 7; 1:1 cocrystal of CBZ and aspirin, 8. 58

PAGE 76

2.3 Results and Discussion 2.3.1 Reproducibility The reliability of SDG to reproduce cocrysta ls of CBZ that were first prepared by slow evaporation was systematically analyze d. Cocrystal formers contained a wide range of functional groups which seemed to have no effect on reproducibility. Cocrystals 1-8 were all reproduced with at le ast one solvent, Table 2.1. The 100% rate of success for producing cocrystals obtained by solution furt her confirms the reliability of SDG for reproducing a specific form in a controlled fash ion with little waste although it remains to be seen if SDG and slurrying complement one another. Indeed, a recent study of slurrying using ethanol effected polymorphic transformation of the CBZ:isonicotinamide cocrystal.44 Experimental and calculated PX RD patterns and IR spectra for 1-8 are presented in Appendix 1. 59

PAGE 77

Nicotinamide Toluene EtOAc CHCl3 MeOH Cyclohex Aspirin Saccharin Terephthalaldehyde Benzoquinone 2,6 pyrdicarb Hyd Hyd4-Aminobenzoic acid 4,4BipyDMSO DMF Water CocrystalFormer *1 2 3 4 5 6 7 8 Table 2.1: Summary of SDG results for Cb z: Red = starting materials, Green = cocrystal formed, and Hyd = hydrate. Indicates Unconverted Starting Ma terials by Powder X-Ray Diffraction. 2.3.2 Solvent Choice Solvents with a range of polarity fr om non-polar to polar extremes were employed. Organic solvents like Cychex and Tol represent the non-polar side while water and DMSO represent the polar side. The middle of the range was covered with solvents like MeOH and EtOAc. The solvents used for sing le crystal growth were not always used in this screen however that could be useful in further studies. In the cases where that solvent was used the results were a positive match. DMF and DMSO appear particularly well suited for SDG. Given that DMF is slightly less prone to solvate formation when compared with DMSO45, it would probably be the solv ent of choice if only one SDG experiment were to be conducted. 60

PAGE 78

However, in the case of 8 only DMSO reproduced the cocrystal form obtained from solution. Nevertheless, it is cl ear that there is an underlyi ng trend: the more soluble the API and the cocrystal formers are in the solvent used for SDG, the more likely it is that the cocrystal will be produced. 2.3.3 Utility as a Screening Technique This set of SDG experiments involved more than one supramolecular synthon and is perhaps more representative of an industr ial situation for which crystal form screening would probably be conducted with multiple cocrystal formers and solvents. Our results complement those of Childs et al .,17 who utilized SDG, evaporation, Sonic SlurryTM, and reaction crystallization to screen for novel forms of CBZ with pharmaceutically accepted carboxylic acids while simultaneously screening for polymorphic behavior. Using sonication for co crystal formation is a relatively recent technique which was introduced into the literature by Bucar et al .46 While SDG can be used for polymorph screening, ot her techniques have also be en employed: supercritical fluid,47 reaction crystallization,17 and Sonic SlurryTM.17 2.4 Conclusions SDG is indeed a viable technique to genera te cocrystals since all 8 cocrystals that were prepared from solution were reproduced, often times with multiple solvents. Alternative methodologies which are also solvent based includ e crystallizations at high pressure48 and slurries49 but, as for slow evaporation, the volume of solvent necessary to perform these techniques is usually significant. Grinding or milling is typically used for particle size reduction as a means of improving the dissolution rate of APIs and there also 61

PAGE 79

exists a correlation between particle si ze reduction and bulk properties such as flowability, bulk density, mixing ability etc.50 However, cocrystals are new compositions of matter and the thermodynamic solubility of the API is therefore not only affected by particle size dynamics. Therefore, given that the newest generation of APIs tends to exhibit an increasing tendency towards low wa ter solubility, i.e. they are BCS class II drugs,51 the generation of pharmaceutical cocrysta ls could have important implications for drug development. That the cocrystals of this study and, by implication, cocrystals in general may be synthesized via SDG is th erefore relevant to both discovery and processing of cocrystals. In addition, SDG provides an eco-fri endly alternative to the use of relatively large amounts of solvent. In summary, SDG is a broadly useful, inexpensive, and Green approach to cocrystal formation. 2.5 References Cited (1) Brittain, H. G., Polymorphism in Pharmaceutical Solids Marcel Dekker: New York, 1999; Vol. 95. (2) Ceolin, R.; Toscani, S.; Gardette, M. F.; Agafonov, V. N.; Dzyabchenko, A. V.; Bachet, B., Journal of Pharmaceutical Sciences 1997, 86, 1062. (3) Chang, C. H.; Yang, D. S. C.; Yoo, C. S.; Wang, B. L.; Pletcher, J., Acta Crystallographica 1981, A37, C71. (4) Himes, V. L.; Mighell, A. D.; Decamp, W. H., Acta Crystallographica Section BStructural Science 1981, 37, 2242. (5) Lang, M. D.; Kampf, J. W.; Matzger, A. J., Journal of Pharmaceutical Sciences 2002, 91, 1186. 62

PAGE 80

(6) Lisgarten, J. N.; Palmer, R. A.; Saldanha, J. W., Journal of Crystallographic and Spectroscopic Research 1989, 19, 641. (7) Lowes, M. M. J.; Caira, M. R.; Lotter, A. P.; Vanderwatt, J. G., Journal of Pharmaceutical Sciences 1987, 76, 744. (8) Reboul, J. P.; Cristau, B.; Soyfer, J. C.; Astier, J. P., Acta Crystallographica Section B-Structural Science 1981, 37, 1844. (9) Reck, G.; Dietz, G., Crystal Research and Technology 1986, 21, 1463. (10) Rustichelli, C.; Gamberini, G.; Ferioli, V.; Gamberini, M. C.; Ficarra, R.; Tommasini, S., Journal of Pharmaceutical and Biomedical Analysis 2000, 23, 41. (11) Li, Y.; Chow, P. S.; Tan, R. B. H.; Black, S. N., Organic Process Research & Development 2008, 12, 264. (12) Fleischman, S. G.; Kuduva, S. S.; McMahon, J. A.; Moulton, B.; Walsh, R. D. B.; Rodriguez-Hornedo, N.; Zaworotko, M. J., Crystal Growth & Design 2003, 3, 909. (13) Vishweshwar, P.; McMahon, J. A.; Bis, J. A.; Zaworotko, M. J., Journal of Pharmaceutical Sciences 2006, 95, 499. (14) Hickey, M. B.; Peterson, M. L.; Scoppettuolo, L. A.; Morrisette, S. L.; Vetter, A.; Guzman, H.; Remenar, J. F.; Zhang, Z.; Tawa, M. D.; Haley, S.; Zaworotko, M. J.; Almarsson, O., European Journal of Pharmaceutics and Biopharmaceutics 2007, 67, 112. (15) Good, D. J.; Rodriguez-Hornedo, N., Crystal Growth & Design 2009, 9, 2252. (16) McMahon, J. A.; Bis, J. A.; Vishwesh war, P.; Shattock, T. R.; McLaughlin, O. L.; Zaworotko, M. J., Zeitschrift Fur Kristallographie 2005, 220, 340. 63

PAGE 81

(17) Childs, S. L.; Rodriguez-Hornedo, N.; Reddy, L. S.; Jayasankar, A.; Maheshwari, C.; McCausland, L.; Shipplett, R.; Stahly, B. C., Crystengcomm 2008, 10 856. (18) Brody, M.J. Marvin Samson Center for the Histor y of Pharmacy at the University of the Sciences in Philadelphia 2004 Mortar and Pestle from Renaissance to Present. (19) Whler, F., Justus Liebigs Ann. Chem. 1844, 51, 153. (20) Ling, A. R.; Baker, J. L., J. Chem Soc. 1893, 63, 1314. (21) Etter, M. C., Journal of Physical Chemistry 1991, 95, 4601. (22) Etter, M. C.; Adsmond, D. A., Journal of the Chemical Society-Chemical Communications 1990, 589. (23) Etter, M. C.; Frankenbach, G. M., Chem. Mater. 1989, 1, 10. (24) Etter, M. C.; Frankenbach, G. M.; Adsmond, D. A., Mol. Cryst. Liq. Cryst. 1990, 187, 25. (25) Etter, M. C.; Frankenbach, G. M.; Bernstein, J., Tetrahedron Letters 1989, 30, 3617. (26) Etter, M. C.; Reutzel, S. M., Journal of the Americ an Chemical Society 1991, 113, 2586. (27) Etter, M. C.; Reutzel, S. M.; Choo, C. G., Journal of the American Chemical Society 1993, 115, 4411. (28) Etter, M. C.; Urbanczyklipkows ka, Z.; Ziaebrahimi, M.; Panunto, T. W., Journal of the American Chemical Society 1990, 112, 8415. (29) Huang, K. S.; Britton, D.; Etter, M. C.; Byrn, S. R., Journal of Materials Chemistry 1997, 7, 713. 64

PAGE 82

(30) Shan, N.; Toda, F.; Jones, W., Chemical Communications 2002, 2372. (31) Jones, W.; Motherwell, S.; Trask, A. V., Mrs Bulletin 2006, 31, 875. (32) Trask, A. V.; Haynes, D. A.; Motherwell, W. D. S.; Jones, W., Chemical Communications 2006, 51. (33) Trask, A. V.; Jones, W., Top. Curr. Chem. 2005, 254, 41. (34) Cheng, W.-T.; Lin, S.-Y., International J. of Pharmaceutics 2007, 357, 164. (35) Bis, J. A.; Vishweshwar, P.; Weyna, D.; Zaworotko, M. J., Molecular Pharmaceutics 2007, 4, 401. (36) Trask, A. V.; Motherwell, W. D. S.; Jones, W., Chemical Communications 2004 890. (37) Trask, A. V.; Shan, N.; Motherwell, W. D. S.; Jones, W.; Feng, S. H.; Tan, R. B. H.; Carpenter, K. J., Chemical Communications 2005, 880. (38) Bis, J. A.; Vishweshwar, P.; Middleton, R. A.; Zaworotko, M. J., Crystal Growth & Design 2006, 6, 1048. (39) Friscic, T.; Fabian, L.; Burley, J. C.; Jones, W.; Motherwell, W. D. S., Chemical Communications 2006, 5009. (40) Friscic, T.; Jones, W., Faraday Discussions 2007, 136, 167. (41) Friscic, T.; Trask, A. V.; Motherwell, W. D. S.; Jones, W., Crystal Growth & Design 2008, 8, 1605. (42) Cincic, D.; Fr iscic, T.; Jones, W., Journal of the American Chemical Society 2008, 130, 7524. (43) Desiraju, G. R., Angewandte Chemie-Internati onal Edition in English 1995, 34, 2311. 65

PAGE 83

66 (44) ter Horst, J. H.; Cains, P. W., Crystal Growth & Design 2008, 8, 2537. (45) Nangia, A.; Desiraju, G. R., Chemical Communications 1999, 605. (46) Bucar, D. K.; MacGillivray, L. R., Journal of the American Chemical Society 2007, 129, 32. (47) Muthukumaran, P.; Chordia, L. Method and Apparatus of Screening Polymorphs of a Substance. WO2005105293, 2005. (48) Oswald, I. D. H.; Pulham, C. R., Crystengcomm 2008, 10 1114. (49) Takata, N.; Shiraki, K.; Takano, R.; Hayashi, Y.; Terada, K., Crystal Growth & Design 2008, 8, 3032. (50) Merisko-Liversidge, E.; Liversidge, G. G.; Cooper, E. R., European Journal of Pharmaceutical Sciences 2003, 18, 113. (51) Amidon, G. L.; Lennernas, H.; Shah, V. P.; Crison, J. R., Pharmaceutical Research 1995, 12, 413.

PAGE 84

Chapter 3: Cocrystallization and Pharmacokinetic Enhancement with Meloxicam 3.1 Background Traditional routines are resistant to change in the pharmaceutical industry but during the 1990s a major shift revolutionized the in vitro screening process from aqueous based manual screening to high throug hput screening in the name of efficiency and hopes of increased pipeline opportunities The high throughput process began with the utilization of dimethylsulf oxide (DMSO) as a medium wh ich led to numerous hits exhibiting very low aqueous solubilities.1 As a result the majority of APIs currently in development exhibit aqueous so lubilities less than 0.1 mg/mL.2 Meloxicam, 4-hydroxy-2methyl-N-(5-methyl-2-thiazolyl)-2H-1,2-benzot hiazine-3-carboxamide -1,1-dioxide, is a non-steroidal anti-inflammatory drug (NSA ID) originally developed by Boehringer Ingleheim in 2000. It is used for the following indications; rheumatoid and osteoarthritis,3 postoperative pain4-5 and fever.6 Oral doses comes as 7.5 or 15 mg tablets or a 7.5mg/mL suspension (no greater than 15 mg per day). It ex ists as a yellow solid that is practically insoluble in water7 and is considered a Class II dr ug (i.e. low solubility and high permeability) by the Biopharmaceutics Classification System (BCS).8 Meloxicam has variable aqueous solubility related to its pH dependent ionization states. Under acidic conditions meloxicam is present in solution in its cationic form, while in basic solutions it is present in its anionic form. When the molecule is neutral in charge it will either be in its zwitterionic or enolic form de pending on the polarity of the solvent.9-10 The different 67

PAGE 85

ionization states of meloxicam are shown in Figure 3.1. Due in part to its low solubility under acidic conditions the Tmax (time to reach maximum con centration) of meloxicam in the human body is typically four to six hours, while it can take more than two hours for the drug to reach its therapeu tic concentration in humans.11 Figure 3.1 Meloxicam and its ionization states. The slow onset of meloxicam prevents it from potential application for the relief of mild to medium level acute pain. To accel erate its onset of action, various complexes of meloxicam have been prep ared and evaluated with resp ect to aqueous solubility, including cyclodextrin inclusion complexes,12 various solvates,7 ethanolamine,13 ammonium and sulfate salts,9 or metal complexes with potassium and calcium.14 Preparation of polymorphic crystal forms of meloxicam10 has also been attempted, 68

PAGE 86

although unsuccessfully, to improve its dissolution profile.15-16 In spite of all the efforts that have been taken, a faster onset oral dosage form for meloxicam (i.e. 30 minutes or less) does not exist at this time. Unique crystalline forms tend to exhib it distinctive physicochemical properties effecting the dissolution, manufacturing, physical stability, permeability, and oral bioavailability of an API.17-19 A typical crystal form sel ection process comprises two stages of development after a target API has been selected: discovery of as many pharmaceutical crystal forms as possible and examination of the physicochemical properties of the newly discovered crystal forms. When salt formation, amorphous dispersions, particle size reduc tions, and formulation changes fail to improve solubility and subsequent bioavailability the more recent technology of pharmaceutical cocrystallization can be employed. A given API may form cocrystals with numerous pharmaceutically acceptable and/or approved materials, and these cocrystals could exhibit enhanced solubility20-23 and/or stability to hydr ation or compressibility.24 At the stage of crystal form discovery, two primary approaches are used. The more straightforward approach is largely based on trial-and-error (e.g. high throughput crystal form screening) and has been implemented to discover crystal forms including, but not limited to, salts,18 hydrates,25 solvates,26 and, more recently, cocrystals.27,28-29 The alternative approach for crys tal form discovery is the s upramolecular synthon approach,30 which recognizes supramolecular synthons31 as a design tool and can be more selective, time-efficient and cost-effective. The supramolecular synthon approach uses crystal engineering32-40 to carefully analyze the relevant supramolecular arrangements that an API might exhibit by utilizing the Ca mbridge Structural Database (CSD),41 and 69

PAGE 87

effectively prioritize all possible guest mol ecules for crystal form screening. Such an approach can be generally effective but has found particular success in generating pharmaceutical cocrystals.30, 42 Pharmaceutical cocrystals represent an opportunity to diversify the number of crystal forms of a given API and in turn fine tune or even customize its physicochemical properties without the need for chemical (covalent) modification. There is no longer any doubt that cocrystals can change the physicoc hemical properties of a given API, however, the way these changes affect the pharmacokinetic (PK) profile is not predictable and is not yet fully understood due in part to the li mited number of animal studies reported in the literature. 23, 43-51 These examples, discussed in Chapter 1, do however provide a solid foundation for pharmaceutical cocrystals as a tool to fine tune physicochemical and subsequent PK properties of APIs wit hout making or breaking a covalent bond. While the work presented herein was be ing performed a report on cocrystals of meloxicam with carboxylic acids by grinding surfaced.52 Compared to other drugs in the same class, such as piroxicam, meloxicam is preferred due to its ability to sel ectively inhibit cyclooxygenase-2 (COX-2).53-54 The only difference between these two APIs is a pyridine ring on the amide linkage for piroxicam versus a thiazole ring on meloxicam. Between a previous report on piroxicam cocrystals with carboxylic acids55 and CSD analysis, pharmaceutical cocrystallization of meloxicam with carboxylic acids56 represents a promising approach to diversify the crystal form portfolio to be used to improve the relevant aqueous solubility and accelerate the onset of action for acute mild to medium level pain relief. Indeed a recent report on PK analysis in rats for an aspirin:meloxicam cocrystal resulted in a 44 fold increase in kinetic solubility 70

PAGE 88

and a 4.4 fold increase in bioavailability.57 That study utilized previously reported toxicological information resulting from co-a dministration of meloxi cam and aspirin. The rate limiting step in absorption and subsequent bioavailability of a BCS class II API is solubility since as it dissolves and readily absorbs due to its high permeability sink conditions are created. Synthesis and rat PK analysis for a set of meloxicam pharmaceutical cocrystals is presented with this in mind. 3.2 Experimental Details 3.2.1. Materials Meloxicam was purchased from Jai Radhe Sales, India with a purity of 99.64% and was used without further purification. All other chemicals were supplied by SigmaAldrich and used withou t further purification. 3.2.2 Methods Meloxicam was reacted with 11 selected coformers: fumaric acid, succinic acid, maleic acid, malonic acid, gentisic acid, 4-hydroxybenzoic acid, adipic acid, (+)camphoric acid, glycolic acid, DL-malic acid, and -ketoglutaric acid. All coformers produced at least one meloxicam cocrystal except -ketoglutaric acid. The cocrystallization attempts re sulted in 10 cocrystals ( 1 10), many of which were prepared via multiple synthetic techniques including solvent-drop grinding58-59 and slurrying. Single crystals suitable for X-ra y diffraction were succe ssfully prepared for two of the new cocrystals ( 1 & 2). 71

PAGE 89

Synthesis of meloxicam:fumari c acid (2:1) cocrystal (1) (a) solvent-drop grinding 0.088 g (0.250 mmol) meloxicam was ball-mille d with 0.015 g (0.129 mmol) of fumaric acid and 50 L of THF for 30 minutes, generating 1 in ca. 100% yield; (b) slurry 0.880 g (2.50 mmol) meloxicam and 0.150 g (1.29 mmol) of fumaric acid were slurried in 3 mL of THF overnight sealed under ambient conditio ns at ca. 250 rpm. The resulting solid was filtered and washed with THF. 1 was isolated in ca. 81% yiel d; (c) solution crystallization 0.100 g (0.284 mmol) meloxicam and 0.330 g (0.284 mmol) of fumaric acid was dissolved in 9 mL of a THF and left to slowly evaporate resulting in single crystals of 1 (ca. 31% yield). Synthesis of meloxicam:succini c acid (2:1) cocrystal (2) (a) solvent-drop grinding 0.088 g (0.250 mmol) meloxicam was ball-mille d with 0.015 g (0.127 mmol) of succinic acid and 50 L of THF for 30 minutes, generating 2 in ca. 100% yield; (b) slurry 0.880 g (2.50 mmol) meloxicam and 0.150 g (1.27 mmol) of succinic acid were slurried in 3 mL of THF overnight sealed under ambient conditions at ca. 250 rpm. The resulting solid was filtered and washed with THF. 2 was isolated in ca. 78% yield; (c) solution crystallization 0.100 g (0.284 mmol) mel oxicam and 0.017 g (0.142 mmol) of succinic acid was dissolved in 10 mL of 1:1 THF and le ft to slowly evaporate. Single crystals of 6 (ca. 55% yield) grew concomitantly with meloxicam form I and succinic acid. Synthesis of meloxicam:maleic acid (1:1) cocrystal (3) (a) solvent-drop grinding 0.175 g (0.498 mmol) meloxicam was ball-mille d with 0.058 g (0.498 mmol) of maleic acid and 40 L of THF for 30 minutes, generating 3 in ca. 100% yield; (b) slurry 0.750 g (2.13 mmol) meloxicam and 0.248 g (2.13 mmol) of maleic acid were slurried in 2 mL of THF overnight sealed under ambient conditio ns at ca. 250 rpm. The resulting solid was 72

PAGE 90

filtered and washed with the same solvent employed for the slurry. 3 was isolated in ca. 92% yield. 3 can also be synthesized via ethyl acetate slurry. Synthesis of meloxicam:maloni c acid (1:1) cocrystal (4) (a) solvent-drop grinding 0.175 g (0.498 mmol) meloxicam was ball-mille d with 0.052 g (0.498 mmol) of malonic acid and 40 L of THF for 30 minutes, generating 4 in ca. 100% yield; (b) slurry 0.900 g (2.56 mmol) meloxicam and 0.266 g (2.56 mmol) of malonic acid were slurried in 2 mL of THF overnight sealed under ambient conditions at ca. 250 rpm. The resulting solid was filtered and washed with THF. 4 was isolated in ca. 88% yield. Synthesis of meloxicam:gentisic acid (1:1) cocrystal (5) (a) solvent-drop grinding 0.175 g (0.498 mmol) meloxicam was ball-mille d with 0.077 g (0.498 mmol) of gentisic acid and 40 L of chloroform or THF for 30 minutes, generating 5 in ca. 100% yield; (b) slurry 0.850 g (2.41 mmol) meloxicam and 0.373 g (2.41 mmol) of gentisic acid were slurried in 2 mL of chloroform overnight se aled under ambient condi tions at ca. 250 rpm. The resulting solid was filtered and washed with the same solvent employed for the slurry. 5 was isolated in ca. 85% yield. 5 can also be synthesized via slurry in ethyl acetate. Synthesis of meloxicam:4-hydroxybenzoic acid (1:1) cocrystal (6) solvent-drop grinding 0.175 g (0.498 mmol) meloxicam was ball-milled with 0.069 g (0.498 mmol) of 4-hydroxybenzoic acid and 40 L of THF for 30 minutes, generating 6 in ca. 100% yield. Synthesis of meloxicam:adipic acid (2:1) cocrystal (7) (a) solvent-drop grinding 0.088 g (0.250 mmol) meloxicam was ball-mille d with 0.018 g (0.123 mmol) of adipic acid and 50 L of THF for 30 minutes, generating 7 in ca. 100% yield; (b) slurry 0.880 73

PAGE 91

g (2.50 mmol) meloxicam and 0.180 g (1.23 mmol) of adipic acid were slurried in 3 mL of THF overnight sealed under ambient conditio ns at ca. 250 rpm. The resulting solid was filtered and washed with THF. 7 was isolated in ca. 80% yield. Synthesis of meloxicam:DL-malic acid (2:1) cocrystal (8) (a) solvent-drop grinding 0.088 g (0.250 mmol) meloxicam was ball-m illed with 0.017 g (0.127 mmol) of DLmalic acid and 50 L of THF for 30 minutes, generating 8 in ca. 100% yield; (b) slurry 0.880 g (2.50 mmol) meloxicam and 0.170 g (1.27 mmol) of DL-ma lic acid were slurried in 3 mL of THF overnight sealed under ambi ent conditions at ca. 250 rpm. The resulting solid was filtered and washed with THF. 8 was isolated in ca. 79% yield. Synthesis of meloxicam:(+)-camphoric acid (3:2) cocrystal (9) (a) solvent-drop grinding 0.233 g (0.663 mmol) meloxicam was ball-milled with 0.886 g (0.042 mmol) of DL-malic acid and 50 L of chloroform for 30 minutes, generating 9 in ca. 100% yield; (b) slurry 0.700 g (1.99 mmol) meloxicam and 0.266 g (1.33 mmol) of (+)camphoric acid were slurried in 4 mL of ch loroform overnight sealed under ambient conditions at ca. 250 rpm. The resulting solid was filtered and washed with chloroform. 9 was isolated in ca. 91% yield. Synthesis of meloxicam:glycolic acid (1:1) cocrystal (10) (a) solvent-drop grinding 0.175 g (0.498 mmol) meloxicam was ball-mille d with 0.038 g (0.498 mmol) of glycolic acid and 40 L of chloroform for 30 minutes, generating 10 in ca. 100% yield; (b) slurry 0.950 g (2.70 mmol) meloxicam an d 0.206 g (2.70 mmol) of glycolic acid were slurried in 2 mL of ethyl acetate overnight sealed under ambient conditions at ca. 250 rpm. The resulting solid was filtered and washed with ethyl acetate. 10 was isolated in ca. 92% yield. 74

PAGE 92

Crystal Form Characterization Quality single crystals for X-ray diffr action were obtained for two compounds, 1 and 2. Attempts to crystallize 3-10 did not afford crystals suita ble for single crystal X-ray crystallographic analysis. Si ngle crystal analysis for 1-2 was performed on a Bruker-AXS SMART APEX CCD diffractometer with monochromatized Cu K radiation ( = 1.54178 ). Data for 1 2 were collected at 100 K. Lattice parameters were determined from least-squares analysis, and reflec tion data were integrated using SAINT.60 Structures were solved by direct methods and refined by full matrix least squares based on F2 using the SHELXTL package.61 All non-hydrogen atoms were refined with anisotropic displacement pa rameters. All hydrogen atom s bonded to carbon, nitrogen, and oxygen atoms were placed geometrically and refined with an isotropic displacement parameter fixed at 1.2 times Uq of the atoms to which they were attached. Hydrogen atoms bonded to methyl groups were placed geom etrically and refined with an isotropic displacement parameter fixed at 1.5 times Uq of the carbon atoms. Powder X-Ray Diffraction (PXRD) : 1-10 were characterized using a D-8 Bruker X-ray Powder Diffractometer using Cu K radiation ( = 1.54178 ), 40 kV, 40 mA. Data were collected over an angula r range of 3 to 40 2 value in continuous scan mode using a step size of 0.05 2 value and a scan rate of 5 /min. Calculated PXRD : Calculated PXRD diffractograms were generated from the single crystal structures of 12 using Mercury 2.2 (Cambridge Crystallographic Data Centre, UK) for the following complexes and compared to the pattern obtained for the corresponding bulk sample. 75

PAGE 93

Differential Scanning Calorimetry (DSC) : DSC was performed on a Perkin Elmer Diamond Differential Scanning Calorimeter with a typical scan range of 25 C 280 C, scan rate of 10 C/min, and nitrogen purge of ca. 30 psi. Fourier Transform Infrared Spectroscopy (FT-IR) : FT-IR analysis was performed on a Perkin Elmer Spectrum 100 FT-IR spectrometer equipped with a solid-state ATR accessory. Pharmacokinetic Study Eight week old male Sprague Dawley rats with a jugul ar vein catheter were purchased from Charles River Laboratories, In c. and housed in a temperature-controlled room for at least 48 hours before the PK st udy. All animal experiments in the present study were approved by the Institutional Animal Care and Use Committee (IACUC). The rats, n = 5, were fasted over night and wei ghed immediately before dosing. Ten mg/kg of meloxicam or its equivalent (for cocrystals) were prepared, si eved to a particle size range between 53 and 75 m, and suspended in 1 ml of 5% polyethylene glycol 400 (PEG 400) with 95% methylcellulose solution (weight perc entage) and administered in a single dose suspension via oral gavage in a single dose. Serial blood samples (0.2 mL) were obtained from the catheter at 0, 0.25, 0.5, 0.75, 1, 2, and 4 hours after oral administration. Blood samples were centrifuged with an Eppendorf Ce ntrifuge at 3000 rpm, 4 C, for 10 min in order to obtain serum samples. All serum samp les were stored at -80 C for subsequent HPLC analysis. 76

PAGE 94

HPLC Analysis. A 12.5 g/mL piroxicam methanol solution wa s used as the internal standard (IS). 50 L of animal plasma sample was tr ansferred into an individual Eppendorf microcentrifuge tube and 200 L IS working solution was added. Each Eppendorf microcentrifuge tube was hand shaken well and the sample was allowed to sit for 20 min. Each Eppendorf microcentrifuge tube was shaken again and the sample was transfered into a 0.2 m Nylon-66 Microfilterfuge tube (Rainin, Oaklong, CA), and spun at 10,000 rpm for 4 min. Clear methanol solutions (200 L or less) with meloxicam were separated from serum proteins and 160 L of clear methanol solution wa s transferred into individual HPLC vials. HPLC analysis was carried out on a Perkin Elmer Instruments LLC comprising the following units: Series 200 Gradient Pump; 785A UV/VIS Detector; Series 200 Autosampler; NCI 900 Network Chromatography Interface and 600 Series Link. The machine was operated by Total Chrome Workstation (Perkin Elmer Instruments LLC). Sample holder temperature was at 4o C and a 250 x 4.6 mm x 1/4 Microsorb-MV 300-5 C-18 column was used. The analytes were eluted with a mixture of phosphate buffer (pH 3.0) and methanol (1/1, v/v). The temperature of the column was set at 40 C with a flow rate of 1 mL/m in, an injection volume of 20 L, and absorbance was measured at 360 nm. PK Parameters and Statistical Analysis Microsoft Excel 2007 was used to process th e PK data and generate statistics, one way analysis of variance (ANOVA) was used. 77

PAGE 95

3.3 Results and Discussion 3.3.1 Cambridge Structural Database Analysis As previously mentioned, a key step in generating pharmaceutical cocrystals is to analyze the target API from a crystal engineering perspective, i.e. to evaluate how the target molecule would form supramolecular synthons. This methodology partitions the target molecule by its functional groups and statistically62 examines the percentage of occurrence of supramolecular homoand hete rosynthons for these functional groups. The targeted supramolecular synthons are typically sustained via hydrogen bonds as they are strong and directional in nature This method is particularly beneficial as most APIs tend to be rich in functional groups that ar e capable of forming strong hydrogen bonds. Polymorphic form I63 of meloxicam indicates that meloxicam molecules form supramolecular chains that are sustained by sulfonyl-amide and thiazole-alcohol supramolecular heterosynthons, as shown in Figure 3.2. The ch ains are held together by various weak interactions, stacking along the a -axis in a slipped fashion. Thus for meloxicam cocrystallization, one or all of these supramolecular synthon motifs must be interrupted.55 Figure 3.2: Meloxicam supramolecular chains sust ained by sulfonyl:amide dimers and thiazole-alcohol supramolecula r synthons, CSD Refcode SEDZOQ. 78

PAGE 96

A CSD analysis64 was conducted to examine the occurrence of supramolecular synthon formation. First a search for an amid e-thiazole (5-membered ring, containing a nitrogen and a sulfur atom, linked to a primary amide) was conducted resulting in zero entries therefore the structure was narrowed dow n to an amino-thiazole functionality (5membered ring, containing a nitrogen and a su lfur atom, linked to a primary amine) resulting in five entries. When this struct ure was searched in th e presence of a carboxylic acid, primary amide, or alcohol moiety the resu lt was zero entries. The search was then narrowed further employing a simple thiazole (5-membered ring cont aining one nitrogen atom and one sulfur atom). The reliabil ity of supramolecular heterosynthon versus homosynthon formation between a thiazole and a carboxylic acid, primary amide, and alcohol were then examined in the CSD. Due to the inability of the thiazole to form a supramolecular synthon with itself; only the homosynthon formation of the car boxylic acid, primary amide, and alcohol moieties in the presence of a thiazole was examined. Any conclusions from the limited data set might not be statistically signifi cant due to the low number of hits for each search, however, heterosynthon formation was favored for carboxylic acids and alcohols, Table 3.1. 79

PAGE 97

Table 3.1: CSD statistics for thia zole supramolecular synthons.64 Search Type Total # of Entries Homosynthon (%) of total Heterosynthon (%) of total Thiazole (N as acceptor) Acid 22 31.8 40.9 Thiazole (S as acceptor) Acid 22 31.8 4.5 Thiazole (N as acceptor) Amide 15 60 13.3 Thiazole (S as acceptor) Amide 15 60 0 Thiazole (N as acceptor) Alcohol 80 7.5 10 Thiazole (S as acceptor) Alcohol 80 7.5 13.8 Based on the preference of supramolecu lar heterosynthon formation between thiazoles and carboxylic acids, a meloxicam cocr ystal screen with acidic coformers that are generally recognized as safe (GRAS) or pharmaceutically acceptable and/or approved was conducted. The focus of the study was ra ther narrow in scope and did not include coformers that only possessed alcohol moieties despite the potential for interaction based upon the CSD statistics. Meloxicam was ther eby reacted with fumaric acid, succinic acid, maleic acid, malonic acid, gentisic aci d, 4-hydroxybenzoic acid, adipic acid, (+)camphoric acid, glycolic acid, DL-malic acid, and -ketoglutaric acid. All coformers except -ketoglutaric acid produced at least one cocrystal. 80

PAGE 98

The cocrystallization attempts resulted in 10 crystal forms; namely, meloxicam:fumaric acid cocrys tal, meloxicam:succinic acid cocrystal, meloxicam:maleic acid cocrystal, meloxicam:malonic acid cocrys tal, meloxicam:gentisic acid cocrystal, meloxicam:4-hydroxybenzoic acid cocrystal, meloxicam:adipic acid cocrystal, meloxicam:(+)-camphoric acid cocrystal, me loxicam:glycolic acid cocrystal, and meloxicam:DL-malic acid cocrystal. Tabl e 3.2 contains chemical diagrams of the cocrystal formers, melting points, and p Ka information. 81

PAGE 99

Table 3.2: Molecular diagrams, p Ka information, and melting points for meloxicam and cocrystals 1 10. Name Molecular structure p Ka p Ka Ratio Meloxicam : Coformer MP of coformer ( o C) MP of cocrystal ( o C) Meloxicam See Figure 3.1 4.18 N/A N/A 254 N/A Fumaric Acid ( 1 ) HO O O OH 3.1 1.08 2:1 287 240 Succinic Acid ( 2 ) HO O O OH 4.19 -0.01 2:1 185 226 Maleic Acid ( 3 ) OH O O OH 1.93 2.25 1:1 137 192 Malonic Acid ( 4 ) HO O O OH 2.83 1.35 1:1 134 164 Gentisic Acid ( 5 ) HO OH O OH 2.98 1.20 1:1 203 237 4-Hydroxybenzoic Acid ( 6 ) O HO OH 4.58 -0.40 1:1 214 209 Adipic Acid (7 ) O HO O OH 4.42 -0.24 2:1 152 209 DL-Malic Acid ( 8 ) HO O OH OH O 3.40 0.78 2:1 131 215 (+)-Camphoric Acid ( 9 ) HO O O OH 4.70 -0.52 3:2 183 212 Glycolic Acid (10 ) OH O HO 3.83 0.35 1:1 75 163 82

PAGE 100

3.3.2 Crystal Structure Descriptions Meloxicam:Fumaric Acid (2:1) Cocrystal, 1 The asymmetric unit of the meloxicam:fumaric acid cocrystal ( 1) contains two meloxicam molecules and one fumaric acid mo lecule which crystallizes in the space group P In 1, the meloxicam dimer persists and links to adjacent dimers via fumaric acid molecules, creating an infinite supram olecular chain, Figure 3.3. The supramolecular heterosynthon comprising carboxylic acid and thiazole/NH moieties is a two point recognition observed between fumaric acid and meloxicam. Both O-H N and NHO hydrogen bonds are observed [O22-H22O N3: O N 2.68(3) H N 1.821 OH N 174.39 ; N2-H2 O21: N O 2.857(4) H O 1.976 N-H O 160.49 ]. The supramolecular chains of meloxicam and fumaric acid exhibited in 1 present as stacked layers, Figure 3.4. Figure 3.3: Supramolecular synt hons observed in Meloxicam:Fumaric Acid (2:1), Cocrystal 1. 83

PAGE 101

Figure 3.4: Supramolecular layers stacking in Meloxicam:Fumaric Acid (2:1), Cocrystal 1. Meloxicam: Succinic Acid (2:1) Cocrystal, 2 Preparation of the meloxi cam:succinic acid cocrystal ( 2) by solvent-drop grinding has been recently reported but without determination of its crystal structure.52 The calculated PXRD of 2 based upon this single crystal stru cture data matches that of the previously reported PXRD. 2 crystallizes in the space group P with the asymmetric unit containing one meloxicam molecule and half a succinic acid molecule. The crystal structure of 2 reveals that the meloxicam dimers are associated with adjacent dimers by succinic acid molecules forming infinite su pramolecular chains, Figure 3.5. Similar to previous meloxicam cocrystal structures, th e primary intermolecular interactions of 2 are hydrogen bonds between meloxi cam and succinic acid via the carboxylic acid to thiazole/NH supramolecu lar heterosynthon. The OH N and NH O=C hydrogen bonds are involved [O7-H7 N3: O N 2.863(4) H N 1.847 O-H N 173.63 ; N2H2 O6: N O 2.849(4) H O 1.993 N-H O 164.32]. The supramolecular chains of meloxicam and succinic acid in 2 are reminiscent of the supramolecular chains 84

PAGE 102

found in 1, Figure 3.5. As shown in Figure 4.6, supram olecular chains of succinic acid and meloxicam stack with an interplanar spacing of 3.386 As a result of overall packing comparison between meloxicam cocrys tal structures, it is observed that the crystal structures of 1 and 2 are isostructural. Table 3.3 li sts the crystallographic details for 1 and 2. Figure 3.5: Supramolecular synthons observed in Meloxicam:Succinic Acid (2:1), Cocrystal 2. Figure 3.6: Supramolecular layers stacking in Meloxicam:Succinic Acid (2:1), Cocrystal 2. 85

PAGE 103

Table 3.3: Crystal structure parameters for 1 and 2. 1 2 Chemical formula (C14H13N3O4S2)2 C4H4O4 (C14H13N3O4S2)2 C4H6O4 Formula weight 818.86 820.88 Crystal System Triclinic Triclinic Space group P P a () 7.145(5) 7.2315(4) b () 8.475(5) 8.4994(5) c () 15.088(9) 14.9383(8) (o) 82.250(9) 82.741(4) (o) 81.368(9) 80.061(3) (o) 70.519(9) 70.313(4) Vol (3) 848.1(7) 849.21(8) Dcal (g cm-3) 1.603 1.605 Z 1 1 Reflections collected 4399 5052 Independent reflections 3031 2719 Observed reflections 2298 1871 T (K) 100 100 R1 0.0451 0.0463 wR2 0.1067 0.0845 GOF 0.995 0.995 3.3.3 Meloxicam Crystal Forms: Cocrystals or Salts? Pharmaceutical cocrystals and salts are well defined and are typically classified as distinct subsets of crystal forms. Whether the meloxicam crystal forms generated by the 10 coformers in this study are cocrystals or salts has also been addressed. For crystal forms with single crystal XRD data (i.e. 12), the conclusion that 12 are cocrystals was drawn in a relatively simple and reliable ma nner. However, it was less straightforward to 86

PAGE 104

identify whether 310 are cocrystals or salts. The p Ka values for meloxicam are 1.09 and 4.18.65 The value of 1.09 is associated with the enolic OH group while the value of 4.18 is linked to the nitrogen atom on the sulf athiazole ring. The enolic OH is much less accessible from a crystal engineering perspec tive as it is involved in intramolecular hydrogen bonding to the neighbori ng ketone or NH moieties. In contrast, the nitrogen atom on the thiazole ring is the primary target for cocrystal or salt formation, as it could potentially sustain a supramolecular s ynthon with various hydrogen bond donors. pKa is widely accepted as a guidline to pred icting whether a salt or cocrystal will form.66-69 It is generally considered that, if pKa < 0, the resulting compound will be a cocrystal, whereas the result is typically a salt if pKa >3. For the region of pKa between 0 < pKa < 3, our ability to predict whether the resulting complex will be neutral or charged is limited.68 Indeed, ca. half of the pKa values reported herein fall into the range of 0 < pKa < 3 so how can one identify whether 310 are cocrystals or salts, especially where the FT-IR spectra may not provide adequate information? The pKa value of cocrystal 1, which has structural data was used as a reference to determine whether 310 are cocrystals or salts. As shown in Table 3.2, 1 possesses a pKa of 1.08 and proton transfer was not obser ved between meloxicam and fumaric acid. Although the molecular arrangemen t of various coformers may have an influence to the electron distribution and protona tion of meloxicam, with the single crystal XRD data of 2, it is reasonable to assert that all but one of the coformers involved in this study with pKa values close to or less than 1.08 would pot entially produce a cocrystal rather than a salt with meloxicam.70 Based on this, with the exception of 3, all crystal forms prepared in this study can be identified as meloxicam cocrystals. Since 3 remained questionable 87

PAGE 105

due to a p Ka of 2.25 further investigation into the FT-IR spectrum was performed to help identify the position of the proto n. The carbonyl group of maleic acid in 3 exhibits a distinct peak at 1716 cm-1, indicating that the carboxylic acid group is neutral rather than negatively charged.71 Therefore, proton transfer does not occur between meloxicam and maleic acid and it is deemed a co crystal in agreement with Myz et al .52 3.3.4 Cocrystal Stoichiometries Cocrystals 3 5 and 7 10 were prepared from both solution and solid-state grinding methods, while 6 was only produced by solvent dr op grinding. The absence of structural data for 3 10 is the result of unsuccessful solution-based growth attempts for single crystals. Nevertheless, polycrystalline powders of 3-10 were characterized by PXRD, FT-IR, and DSC, see Appendix 2. As mentioned in section 3.1 Myz et al reported cocrystals of meloxica m by grinding with succinic acid (1:1) and maleic acid (1:2) but the stoichiometric ratio presented here for the succinic acid cocrystal, 2 here, was 2:1 and for maleic cocrystal, 3 here, 1:1. Furthermore, no single crystal data was produced by Myz, but only FT-IR and PXRD data. Even w ithout the single crystal XRD data, the stoichiometries of 3-10 were determined by DSC or NMR in compliment with FT-IR and PXRD data as exemplified by 3. Based on the potential supramolecular interactions of meloxicam and the lack of ratio in the previous report, the most likely stoichiometry of meloxicam and maleic acid in 3 was proposed to be either 1:1, 2:1, or 1:2 (API:former). In order to determine the stoichiometry of 3 ethyl acetate slurries of physical mixtures of meloxicam and maleic acid in molar ratios of 2:1, 1:1, and 1:2 were performed at room temperature overnight. From each slurry experiment, the solid 88

PAGE 106

crystalline powder was separated, wash ed with ethyl acetate and dried for characterization. The absence of pure maleic ac id, water, or ethyl acetate was confirmed based on the DSC analysis, Figure 3.8, upon all three solid powders from the slurries. PXRD characterization i ndicated that the solids generated from the 1:1 and 1:2 slurries were identical to the cocrystal form from the initial solvent drop grind experiment conducted at a 1:1 ratio. In contrast, the 2:1 slurry produced a physical mixture of meloxicam form I and the cocrysta l as confirmed by PXRD, Figure 3.7. 1H NMR (nuclear magnetic resonan ce) analysis (400 MHz, d6-DMSO) on solids from the 1:1, 2:1, and 1:2 slurries confirmed that the stoich iometry of meloxicam and maleic acid in 3 is 1:1. PXRD, NMR, FT-IR, and DSC da ta can be found in Appendix 2. 46810121416182022242628303234363840 0 1000 2000 3000 4000 5000 Relative Intensity2 theta (deg)2:1 EtOAc slurry 1:2 EtOAc slurry 1:1 EtOAc slurry Meloxicam:Maleic Acid cocrystal Maleic Acid Meloxicam Form 1 Meloxicam Form 3 Figure 3.7: PXRD patterns for Mel oxicam:Maleic Acid slurries. 89

PAGE 107

Figure 3.8: DSC of Meloxicam: Maleic Acid slurries. 90

PAGE 108

3.3.5 In vivo Performance via Rat Pharmacokinetic Studies The PK profiles of 10 meloxicam:carboxyl ic acid cocrystals and pure meloxicam were determined and evaluated with respect to serum concentration in male SpragueDawley rats over four hours vi a a single oral dose of 10 mg /kg equivalent of meloxicam, Figure 3.9. The time versus concentration profile for many of the crystal forms continued to increase throughout most of the study. Cocrystals 7 and 8 had serum concentrations that were still increasing at four hours. An intrave nous (IV) leg was not conducted therefore bioavailability was not obtainable, however, Cmax, Tmax, and AUC over four hours are shown in Table 3.4. Cocrystals 2, 4, and 9 decreased Tmax significantly over the time of the study, although, since pure API and tw o of the cocrystals ar e still increasing at four hours this information could be misl eading with respect to long term action. In terms of onset of action the fifteen minute data point is intriguing as it reveals that seven of the cocrystals re ach higher serum concentrations; 6 is 2.73 fold higher, 3 is 2.42 fold higher, while 5 and 9 are ca. 2 fold higher. Th is could result in faster therapeutic levels in clinical studies and warrants further investigation. The Cmax values over four hours increased for five of the cocrystals compared to meloxicam with 7 and 8 reaching ca. 25% higher values. AUC values in creased for five of the ten cocrystals ( 2, 3, 6, 7, and 8) with the largest differences attributed to 2 and 6. 91

PAGE 109

Figure 3.9: Serum concentration fo r meloxicam and meloxicam cocr ystals in rats over 4 hours. 92

PAGE 110

Table 3.4: Pharmacokinetic data for meloxicam and meloxicam cocrystals. 93

PAGE 111

The therapeutic concentrat ion of meloxicam is achieved after ca. 2 hours in humans for the currently marketed oral dosage form11 and for comparison purposes the concentration of pure meloxicam as determined herein at 2 hours, 30.19 g/mL, will be used as a reference for further discussion al though it should be noted that efficacy was not assessed here. Cocrystals 1, 4, 5, 9, and 10 had serum concentrations equal to or less than pure meloxicam at 2 hours, while 2, 3, and 6 8 were greater. Figure 3.9 indicates five cocrystals have outperformed pure API at the two hour point. 2 achieved the highest serum concentration at 2 hours (44.02 g/mL), while 10 had the lowest concentration (20.77 g/mL). The meloxicam:fumaric acid cocrystal, 1, has an interesting PK profile in that it is relatively linear over 4 hours making it suitable for possible application as a controlled release form. Melting point is sometimes related to solubility due to the strength of the interactions involved in the cr ystalline lattice. The lower th e melting point of a substance the easier it may be to dissolve that substance in GI fluids therefor e correlations between cocrystal melting points and serum concentr ations, AUCs, and coformer melting points were investigated. It was found that only th e correlation between co crystal and coformer melting points exhibited a weak linear trend with an R2 value of 0.636. Predicted aqueous solubilities72 were also used to investigate correlations with th e PK data and no correlations were found. The lack of correlati ons with the PK data suggests that other physiological processes besides dissoluti on may be exerting their effects. 94

PAGE 112

3.4 Conclusions Nine meloxicam:carboxylic acid cocrys tals have been discovered and the previously reported maleic acid cocrystal ra tio was determined. Most of them can be synthesized by solvent drop grinding and slurry methods. Two of them, 1 and 2, produced single crystals suitable for X-ray diffraction resulting in reliable structural data. These two cocrystals are isostructural which is not surprising considering the only difference between these two coformers is a double bond. Interestingly two of the diacids here produced 1:1 stoichiometries while five others produced 2:1 stoi chiometries. In the case of 3 this could be explained by intram olecular hydrogen bonding due to the cis conformation of maleic acid resulting in the H-bonding functionalities being less available to partake in supramolecular synthons. Five of the cocrystals performed worse than pure meloxicam in PK analysis in rats and five others performed better with respect to Cmax and AUC. Cocrystal 8 achieved a 26.2% increase in Cmax compared to pure meloxicam and 2 achieved a 36.4% increase in AUC compared to pure meloxicam. At fift een minutes seven of the cocrystals reached a higher serum concentration with 6 exhibiting a 2.73 fold increas e, which could lead to faster onset in terms of therapeutic levels in clinical studie s and warrants further investigation of this and the other crystal forms that out perf ormed meloxicam at this time point. As an oral dosage form reaches the gast rointestinal (GI) trac t, it must first be dissolved in GI fluids befo re it can be absorbed. During this process many different physiological molecules such as enzy mes, hormones, second messengers, and 95

PAGE 113

immunological compounds may be present a nd actively inte rfering. Undoubtedly these physiological factors will also affect the ab sorption of the API. Since meloxicam is highly permeable (BCS class II) solubility is the limiting factor for absorption. However, permeability may also be affected by the use of cocrystals. In this case that effect may be insignificant or overlooked due to its inheren tly high permeability. It is well accepted that cocrystals represent a way to affect the solubility of an API, but how this affects the in vivo performance of that API remains to be fully understood or predictable. 3.5 References (1) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J., Advanced Drug Delivery Reviews 2001, 46, 3. (2) Dublin, C. H., Drug Delivery Technologies 2006, 6, 34. (3) Ahmed, M.; Khanna, D.; Furst, D. E., Expert Opinion 2005, 1, 739. (4) Thompson, J. P.; Sharpe, P.; Kiani, S.; Owen-Smith, O., British Journal of Anaesthesia 2000, 84, 151. (5) Roughan, J. V.; Flecknell, P. A., European Journal of Pain 2003, 7, 397. (6) Engelhardt, G.; Homma, D.; Schl egel, K.; Utzmann, R.; Schnitzler, C., Inflammation Research 1995, 44, 423. (7) Seedher, N.; Bhatia, S., AAPS PharmSciTech 2003, 4, Article 33. (8) Takagi, T.; Ramachandran, C.; Bermej o, M.; Yamashita, S.; Yu, L. X.; Amidon, G. L., Molecular Pharmaceutics 2006, 3, 631. (9) Luger, P.; Daneck, K.; Engel, W.; Trummlitz, G.; Wagner, K., European Journal of Pharmaceutical Sciences 1996, 4, 175. 96

PAGE 114

(10) Coppi, L.; Sanmarti, M. B.; Clavo, M. C. Crystalline Forms of Meloxicam and Processes for Thier Preparation and Interconversion. US 6,967,248 B2, 2005. (11) Davies, N. M.; Skjodt, N. M., Clinical Pharmacokinetics 1999, 36, 115. (12) Ghorab, M. M.; Abdel-Salam, H. M.; El-Sayad, M. A.; Mekhel, M. M., Aaps Pharmscitech 2004, 5. (13) Han, H. K.; Choi, H. K., European Journal of Pharmaceutics and Biopharmaceutics 2007, 65, 99. (14) Defazio, S.; Cini, R., Journal of the Chemical Society-Dalton Transactions 2002 1888. (15) Bock, T.; Saegmueller, P.; Sieg er, P.; Tuerck, D. Meloxicam for Oral Administration. US 6,869,948 B1, 2005. (16) Struengmann, A.; Freudensprung, B.; Klokkers, K. New Pharmaceutical Compositions of Meloxicam with Improve d Solubility and Bioavailability WO 99/09988 A1, 1999. (17) Byrn, S. R.; Pfeiffer, R. R.; Ste phenson, G.; Grant, D. J. W.; Gleason, W. B., Chemistry of Materials 1994, 6, 1148. (18) Berge, S. M.; Bighley, L. D.; Monkhouse, D. C., Journal of Pharmaceutical Sciences 1977, 66, 1. (19) Haleblian, J. K., Journal of Pharmaceutical Sciences 1975, 64, 1269. (20) Childs, S. L.; Chyall, L. J.; Dunlap, J. T.; Smolenskaya, V. N.; Stahly, B. C.; Stahly, G. P., Journal of the Americ an Chemical Society 2004, 126, 13335. (21) Banerjee, R.; Bhatt, P. M. ; Ravindra, N. V.; Desiraju, G. R., Crystal Growth & Design 2005, 5, 2299. 97

PAGE 115

(22) Good, D. J.; Rodriguez-Hornedo, N., Crystal Growth & Design 2009, 9, 2252. (23) Cheney, M. L.; Shan, N.; Healey, E. R.; Hanna, M.; Wojtas, L.; Zaworotko, M. J.; Sava, V.; Song, S. J.; Sanchez-Ramos, J. R., Crystal Growth & Design 2010, 10, 394. (24) Karki, S.; Friscic, T.; Fabian, L.; Laity, P. R.; Day, G. M.; Jones, W., Advanced Materials 2009, 21, 3905. (25) Khankari, R. K.; Grant, D. J. W., Thermochimica Acta 1995, 248, 61. (26) Vippagunta, S. R.; Brittain, H. G.; Grant, D. J. W., Advanced Drug Delivery Reviews 2001, 48, 3. (27) Zaworotko, M.; Arora, K., Pharm aceutical co-crystals: A new opportunity in pharmaceutical science for a long-known but little studied class of compounds. In Polymorphism in Pharmaceutical Solids, 2nd ed.; Brittain, H. G., Ed. Informa Healthcare: 2009. (28) Aaltonen, J.; Alleso, M.; Mirza, S.; Koradia, V.; Gordon, K. C.; Rantanen, J., European Journal of Pharmaceutics and Biopharmaceutics 2009, 71, 23. (29) Yin, S. X.; Grosso, J. A., Current Opinion in Drug Discovery & Development 2008, 11, 771. (30) Cheney, M. L.; Shan, N.; Healey, E. R.; Hanna, M.; Wojtas, L.; Zaworotko, M. J.; Sava, V.; Song, S.; Sanchez-Ramos, J. R., Crystal Growth & Design 2010, 10, 394. (31) Desiraju, G. R., Angewandte Chemie-Internati onal Edition in English 1995, 34, 2311. (32) Pepinsky, R., Physical Review 1955, 100, 971. 98

PAGE 116

(33) Desiraju, G. R., Crystal Engineering: The Design of Organic Solids Elsevier: 1989. (34) Schmidt, G. M. J., Pure and Applied Chemistry 1971, 27 647. (35) Shan, N.; Batchelor, E.; Jones, W., Tetrahedron Letters 2002, 43, 8721. (36) Shan, N.; Bond, A. D.; Jones, W., Crystal Engineering 2002, 5, 9. (37) Shan, N.; Bond, A. D.; Jones, W., Tetrahedron Letters 2002, 43, 3101. (38) Shan, N.; Jones, W., Tetrahedron Letters 2003, 44, 3687. (39) Bis, J. A.; Vishweshwar, P.; Weyna, D.; Zaworotko, M. J., Molecular Pharmaceutics 2007, 4, 401. (40) Shattock, T. R.; Arora, K. K.; Vishweshwar, P.; Zaworotko, M. J., Crystal Growth & Design 2008, 8, 4533. (41) Allen, F. H., Acta Crystallographica Section B-Structural Science 2002, 58, 380. (42) Almarsson, .; Zaworotko, M. J., Chemical Communications 2004, 1889. (43) Peterson, M.; Bourghol Hickey, M.; Oliveira, M.; Almarsson, O.; Remenar, J. Modafinil compositions. US 2005267209 A1, 2005. (44) Remenar, J.; MacPhee, M.; Peterson, M.; Morissette, S. L.; Almarsson, O. Novel Crystalline Forms of Conazoles and Me thods of Making and Using the Same. WO/2005/092884, 2005. (45) McNamara, D. P.; Childs, S. L.; Gior dano, J.; Iarriccio, A.; Cassi dy, J.; Shet, M. S.; Mannion, R.; O'Donnell, E.; Park, A., Pharmaceutical Research 2006, 23, 1888. (46) Variankaval, N.; Wenslow, R.; Murry, J.; Hartman, R.; Helmy, R.; Kwong, E.; Clas, S. D.; Dalton, C.; Santos, I., Crystal Growth & Design 2006, 6, 690. 99

PAGE 117

(47) Chen, A. M.; Ellison, M. E.; Peres ypkin, A.; Wenslow, R. M.; Variankaval, N.; Savarin, C. G.; Natishan, T. K.; Mathre, D. J.; Dormer, P. G.; Euler, D. H.; Ball, R. G.; Ye, Z. X.; Wang, Y. L.; Santos, I., Chemical Communications 2007, 419. (48) Hickey, M. B.; Peterson, M. L.; Scoppettuolo, L. A.; Morrisette, S. L.; Vetter, A.; Guzman, H.; Remenar, J. F.; Zhang, Z.; Tawa, M. D.; Haley, S.; Zaworotko, M. J.; Almarsson, O., European Journal of Pharmaceutics and Biopharmaceutics 2007, 67, 112. (49) Annette Bak; Gore, A.; Yanez, E.; St anton, M.; Tufekcic, S.; Syed, R.; Akrami, A.; Rose, M.; Surapaneni, S.; Bostick, T.; King, A.; Neervannan, S.; Ostovic, D.; Koparkar, A., Journal of Pharmaceutical Sciences 2008, 97, 3942. (50) E. Dova; J. M. Mazurek; Anker, J. Tenofovir disoproxil hemi-fumaric acid cocrystal. WO/2008/143500 2008. (51) Imamura, M. Cocrystal of C-Glycoside Derivative and L-Proline. WO 2007/114475 A1, 2007. (52) Myz, S. A.; Shakhtshneider, T. P.; Fucke, K.; Fedotov, A. P.; Boldyreva, E. V.; Boldyrev, V. V.; Kuleshova, N. I., Mendeleev Communications 2009, 19, 272. (53) Burdan, F., Toxicology 2005, 211 12. (54) Harirforoosh, S.; Aghazadeh-Habashi, A.; Jamali, F., Clinical and Experimental Pharmacology and Physiology 2006, 33, 917. (55) Childs, S. L.; Hardcastle, K. I., Crystal Growth & Design 2007, 7, 1291. (56) Cheney, M. L.; Weyna, D. R.; Shan, N.; Hanna, M.; Wojtas, L.; Zaworotko, M. J., Crystal Growth & Design 2010, 10, 4401. 100

PAGE 118

(57) Cheney, M. L.; Weyna, D. R.; Shan, N.; Hanna, M.; Wojtas, L.; Zaworotko, M. J., Journal of Pharmaceutical Sciences 2010, Online DOI: 10.1002/jps.22434. (58) Shan, N.; Toda, F.; Jones, W., Chemical Communications 2002, 2372. (59) Weyna, D. R.; Shattock, T.; Vishweshwar, P.; Zaworotko, M. J., Crystal Growth & Design 2009, 9, 1106. (60) Bruker SMART, SAINT-Plus, SADABS, XP and SHELXTL Madison, Wisconsin, USA, 1997. (61) Sheldrick, G. M. SHELXTL, University of Gottingen: Germany, 1997. (62) It must be noted that the CSD stru ctures have not been systematically populated with respect to balanced representation of the vari ous functional groups. Although the supramolecular synthon approach has been successfully used in the various pharmaceutical cocrystal studies, the CSD sta tistical result here could be biased. (63) Fabiola, G. F.; Pattabhi, V.; Manjunatha, S. G.; Rao, G. V.; Nagarajan, K., Acta Crystallographica Section C 1998, 54, 2001. (64) CSD version 5.31 (Aug 2010 Update) Search Parameters: 3D coordinates, R 0.075, and only organics. If contacts between donor and accepto r are used they were defined to be between 2 and 3.5 (65) Ki, H. M.; Choi, H. K., Archives of Pharmacal Research 2007, 30, 215. (66) Stahl, P. H.; Wermuth, C. G., Handbook of Pharmaceutical Salts: Properties, Selection, and Use WILEY-VCH: Zurich, 2002. (67) Bhogala, B. R.; Basavoju, S.; Nangia, A., Crystengcomm 2005, 7, 551. (68) Childs, S. L.; Stahly, G. P.; Park, A., Molecular Pharmaceutics 2007, 4, 323. 101

PAGE 119

102 (69) Delori, A.; Suresh, E.; Pedireddi, V. R., Chemistry-a European Journal 2008, 14 6967. (70) Identification of cocrystals based on pKa values is straight forward but also could be less rigorous. However, confirmation of relevant proton locations in 3-10 remains inconclusive using availa ble characterization techniques. (71) Silverstein, R. M.; We bster, F. X.; Kiemle, D. J., Spectroscopic Identification of Organic Compounds John Wiley & Sons: New York, 2005. (72) Scifinder Scholar 2007.

PAGE 120

Chapter 4: Crystalline Forms of (R,S ) Baclofen: A Zwitterionic Active Pharmaceutical Ingredient 4.1 Background Oral dosage forms approved by regulatory agencies, such as the Food and Drug Administration (FDA), are typically stable crystalline forms whether they be in tablet, capsule, or suspension form. Different crysta lline forms include, but are not limited to, salts1, hydrates2, solvates3, and, more recently, cocrystals4-5. Novel crystalline forms can tune solubility by changing the thermodynamic properties of the solid which is of seminal importance when considering oral dosage forms.6 Polymorphs, different packing arrangements of the same molecule, can also exhibit different sol ubilities although those differences are usually small due to the natu re of small differences in crystal packing.7 However, in some rare cases of conformationa l polymorphism the difference in solubility can be very substantial as in the case of ritonavir since the cr ystal packing is significantly different.8 Improving aqueous solubility through salt formation has been the gold standard approach for the pharmaceutical indus try when low solubility, ionizable, APIs are involved.9 Crystal packing of a salt is usually very different th an that of the conjugate acid or base which leads to different physicochemical pr operties such as rate of dissolution.10 Ionized molecules are generally far mo re water soluble than their neutral counterparts because they have a large in crease in dipole moment which provides the basis for this approach.11 Not including APIs which are peptide hormones, antibodies, 103

PAGE 121

proteins, polymeric, inorganic, or greater th an 1000 Da in molecular weight, there were 1356 compounds listed in the FDAs Orange B ook of approved drug products at the end of 2006.12 Of that total, 51.4% are salts with the majority of these salts formed from basic APIs where the most prevalent counterion is chloride (53.4% of these compounds). The most prevalent cation for acidic APIs was sodium (75.3% of these compounds). Logic dictates these as the most obvious choices si nce they are already prevalent in the human body. Different counterions will dictate differe nt crystalline forms which will present a range of physicochemical properties9-10 as in the case of diclofenac where the sodium salt is used for delayed release tablets and th e potassium salt for immediate release tablets since the two have significantly different dissolution rates.13 The focus of this study is a low solu bility zwitterionic API known as (R,S) baclofen or (R,S) -amino-(p-chlorophenyl)-butyric acid. Baclofen is an anti-spastic agent developed in the early 20th century for epilepsy that mimics -aminobutyric acid (GABA) and acts as an agonist for GABA b receptors which results in the reduction of excitatory neurotransmitter effects.14-15 It is a lipophilic deri vative of GABA that can permeate the blood brain barrier which is de livered as the racemate. However, the R enantiomer is more active since it can attain the same conformation as GABA.16 It was quickly determined that it had minimal e ffects on epilepsy but was recognized for its ability to reduce spasticity in certain patients. Baclofen is most widely used by patients with spasticity related problems like cerebral palsy17-19, dystonia20, and trigeminal neuralgia21 but is also currently being investigat ed for numerous other indications like spinal cord injuries22-23, binge eating24, alcohol dependence25, and opiate addiction.26 104

PAGE 122

There are oral and injectable dosage forms available at a maximum dose of 80 mg per day. Novartis marketed baclofen as Liores al and Medtronic has devised an intrathecal infusion pump for patients with severe spasticity.27 Zwitterions are neutral amphoteric molecules that carry both a formal positive and negative charge. In an aqueous solution the ionization state is determined by the pH value. Amphoteric molecules in water have an isoelectric point (pI) which is the pH at which the zwitterionic conformation is stablized. Under normal circumstances if the pH is lowered below the pI then the molecule is protonated with a net positive charge and the opposite is true if the pH rises above th e pI. With these pH changes the molecule becomes polarized and more soluble in wate r. Baclofen has a unique conformationally stabilized zwitterionic struct ure through intramolecular charge assisted H-bonding, Figure 4.1, that survives at a pH range of 5 8.5 creating a low solubility throughout this pH range.27 Figure 4.1: Baclofens zwitterionic structure. 105

PAGE 123

Currently baclofen precipita tes out at concentrations above 2 mg/mL in water or at pH conditions near the pI, 6.7, which can cause the intrathecal pump mentioned earlier to clog or create inaccurate dosing.27 By manipulating the crysta lline form of baclofen it is possible to increase the concentration in the infusion pump to allow for less physician visits involving surgical re fills and reduce the risk of precipitation and subsequent overdose if the pump clogs up and bursts. The degree of proton transfer is the dis tinguishing factor betw een cocrystals and salts and is affected by the ionization constants28, however, in the cas e of gabapentin, a zwitterionic API similar to bacl ofen, a multiple component crystalline form exists with 4hydroxybenzoic acid where there appear s to be partial proton transfer due to a disordered proton on a special position crystall ographically with 50% occupancy.29 This is a rare case which could potentially complicate in tellectual property (IP) arguments and patentability, however such rare cases cannot be ignored due to mounting evidence of the stability of such structures including but not limited to the conclusion of Mohamed et al 2009 where computational modeling revealed that a significant energy penalty is reduced when the presence of a disorder ed acidic proton is indicated.30 The study herein compares different multiple component crystal forms of (R,S) baclofen as sulfonate salts, on the basis of aqueous solubility under different pH conditions. The sulfonic acids used are depicted in Figure 4.2, which were chosen in order to break th e interaction represented in Figure 4.1 with the thought th at a strong enough acid would protonate the carboxylate. Structural comparisons, dissolution profile s in pure water, pH 1, and pH 7, and solubilities are discussed. 106

PAGE 124

Figure 4.2: Sulfonic acids paired with baclofen. 4.2 Experimental Details 4.2.1 Materials (R,S) Baclofen anhydrous was purchased from Spectrum and was used without further purification. All ot her chemicals were supplied by Sigma-Aldrich and used without further purification. 4.2.2 Methods Synthesis of (R,S) Baclofen monohydrate (1): (R,S) Baclofen was dissolved in 10% methanol, 0.68% acetic acid water with heat and left to evaporate. After one week colorless plates and needles were obtained with ca. 95% yield. Synthesis of (R,S) Baclofen:p-Phenolsulfonate (2): (R,S) Baclofen was dissolved in a 65% by weight aqueous p-phenolsulfonic acid so lution with heat to nearly saturated and then allowed to cool to room temperature. Colorless needles were obtained after cooling down in ca. 91% yield. 107

PAGE 125

Synthesis of (R,S) Baclofen:4 -Chlorobenzenesulfonate (3): (R,S) Baclofen and 4chlorobenzenesulfonic acid were dissolved in wa ter with heat at a 1:1 stoichiometric ratio and the resultant solution was left to eva porate affording colorless plates and needles after 5 days in ca. 95% yield. Synthesis of 4 and 5 are discussed in section 4.5.2. Crystal Form Characterization Single-Crystal X-ray Diffraction : Single crystals were obtained for 3 compounds. Single crystal analysis for 1-5 was performed on a Bruker-AXS SMART APEX CCD diffractometer with Mo K radiation ( = 0.71073 ) connected to a KRYO-FLEX low-temperature devi ce and collected at 100 K for 14 and 223 K for 5. Lattice parameters were determined from leas t-squares analysis and reflection data were integrated using SAINT.31 Structures were solved by di rect methods and refined by full matrix least squares based on F2 using the SHELXTL package.32 All non-hydrogen atoms were refined with anisotropic di splacement parameters. All hydrogen atoms bonded to carbon, nitrogen, and oxygen atoms were placed geometrically and refined with an isotropic displacement parameter fixed at 1.2 times Uq of the atoms to which they were attached. Hydrogen atoms bonded to methyl groups were placed geometrically and refined with an isotropic displacem ent parameter fixed at 1.5 times Uq of the carbon atoms. Powder X-Ray Diffraction (PXRD) : Powders were characterized by a D-8 Bruker X-ray Powder Diffractometer using a Cu K radiation ( = 1.54178 ), 40kV, 40mA. Data was collected over an angular range of 3 to 40 2 value in continuous scan mode using a step size of 0.05 2 value and a scan speed of 5.0 /min. 108

PAGE 126

Calculated PXRD : Calculated PXRD diffractograms were generated from the single crystal structures us ing Mercury 1.5 (Cambridge Cr ystallographic Data Centre, UK) for 1-5 for comparison to the bulk sample. Differential Scanning Calorimetry (DSC) : Thermal analysis was performed on a TA Instruments DSC 2920 Differential Scan ning Calorimeter. Aluminum pans were used for all samples and the instrument wa s calibrated using an indium standard. For reference, an empty pan sealed in the same way as the sample was used. Using inert nitrogen conditions, the samples were heated in the DSC cell from 30C to the required temperature (me lting point of the cocrystal) at a rate of 5C/min unless otherwise specified. Fourier Transform Infrared Spectroscopy (FT-IR) : FT-IR analysis was performed on a Perkin Elmer Spectrum 100 FT-IR spectrometer equipped with a solidstate ATR accessory. Thermogravimetric analysis (TGA): A Perkin Elmer STA 6000 Simultaneous Thermal Analyzer was used to conduct th ermogravimetric analysis. Open alumina crucibles were used to heat the samples fr om 30C to the required temperature at 10 C/min scanning rate under nitrogen stream. Dissolution and Solubility Particle Size: Crystals were ground up and sieved to maintain a particle size range between 53 and 75 m. 109

PAGE 127

High Performance Liquid Chromatography (HPLC) : Analysis was performed on a Shimadzu Prominence HPLC system comprising the following units: an SIL 20AHT autosampler; a SPD 20A UV/vis Detector; a CBM 20A Communications Bus Module; LC20 AT Liquid Chromatograph; DGU 20A5 Degasser. The system was at room temperature and a flow rate of 1 mL/min was used. The column was a Thermo Scientific Hypersil ODS C-18 (100 4.6 mm 5 m). The mobile phase consisted of a mixture of 0.01 M phosphate buffer (pH 3.5) with acetonitrile (4/1, v/v). Dissolution Study: Dissolution studies were performed on pure (R,S) baclofen, 2 and 3 allowing for the salt forms to be compar ed against the original API. Deionized water, pH 1 aqueous solution (0.1 N HCl, 37o C), and a 0.8 M sodium phosphate buffer pH 7 were used all at 37o C. The dissolution study was c onducted using an excess of freeflowing solid in solution in 25 mL of solvent; for (R,S) baclofen 175 mg/beaker was used in water, 875 mg/beaker was used in 0.1 M HCl, and 150 mg/beaker was used in pH 7 buffer; for 2, 3 g of solid was used in water, 3.5 g of solid was used in 0.1 M HCl, and 250 mg/beaker was used in pH 7 buffer; for 3, 188 mg of solid was used in water and 425 mg of solid was used 0.1 M HCl; The slurries were stirred with a magnetic stir bar at a rate of ca. 200-300 rpm. Aliquots were filtered with 0.45 m filters after 1, 5, 10, 15, 30, 60, 120, 240, 480, 720, and 1440 min (2880 min also for pH 7). The resulting solution was processed and the concentration of baclofen was measured using HPLC. The pH values of the resulting solutions and crystal fo rms of the solid in those solutions were also determined. The experiment was done in trip licate to allow for statistical analysis.33 pH Determination: pH was determined using a VWR SympHony pH meter model SP70P with a digital readout. 110

PAGE 128

Photographs (R,S) Baclofen monohydrate (1): 1 was viewed with a fully automated, upright Zeiss AxioImager Z.1 microscope with a 20x /0.70NA dry objective. (R,S) Baclofen:p-Phenolsulfonate (2): 2 was viewed with a fully automated, upright Zeiss AxioImager Z.1 microsc ope with a 20x /0.70NA dry objective, and Nomarski DIC contrasting prisms. Z-stacks of images were created at 0.5 micron step sizes using the AxioCam MRm CCD camera and Axiovision version 4.6.02 software suite. Images were then 3-dimensionally r econstructed using the iso-surface technique in Bitplanes (Zurich, Switzerland) Imaris software version 5.0.3. (R,S) Baclofen:4-Chlorob enzenesulfonate (3): 3 was viewed with an Olympus MIC-D digital microscope. 4.3 Results and Discussion 4.3.1 Pure Baclofen The hydrochloride salt of (R) bacl ofen which crystallizes in the P212121 space group was deposited into the Cambridge Structural Database (CSD) in 1982.16 Each baclofen molecule is protonated by HCl to have a neutral carboxylic acid group and an ammonium group which counterbalances the chloride anion. The ammonium group is involved in charge assisted hydrogen bonding with three adjacent chloride anions and the carbonyl of another baclofen molecule whose carboxylic OH group also interacts with 111

PAGE 129

one of the same chloride anions that intera cts with the ammonium. A bilayer sheet exists parallel to the ab plane and these sheets in terdigitate in a staggered fashion with each other through C-H interactions stacking along the c axis, Figure 4.3. Figure 4.3: Supramolecular arrangement of (R) Baclofen HCl from the CSD Refcode CRBMZC10. (R,S) Baclofen monohydrate crystals ( 1 ) were grown but provided poor diffraction data and a suitable /publishable structure was not obtained. Colorless plates (Figure 4.4) and needles were simply too sma ll, too thin, or twinned. PXRD data matched the previously publishe d data on the monohydrate.34 Figure 4.4: Single Crystals of (R,S) Baclof en monohydrate, 1, 20x magnification. 112

PAGE 130

4.3.2 Sulfonate Salts of (R,S) Baclofen a nd their Crystal Structure Descriptions The pKa values for (R,S) baclofen are 3.87 and 9.62 for the carboxylic acid and amino functional groups respectively.27 It is generally accepted as a guideline that the difference in the pKa of the base minus the pKa of the acid must be less than zero if the desired outcome is a neutra l complex (i.e. cocrystal).35-36 To generate a salt one would select two molecules with a difference in pKa of three or more units.11, 37 For the region in between ( pKa 0-3) the ability to predetermine whether the resulting complex will be neutral or charged is difficult.29-30, 38 It should also be noted that pKa is a solution-based measurement and does not always transl ate to solid state chemistry. The pKa values and the pKa values for the individual components of the supramolecular assemblies studied herein are provided in Table 4.1 and imply salt formation in each case. Table 4.1: pKa and pKa values. Chemical Name pKa pKa Baclofen 9.6227 NA p-Phenolsulfonic Acid -2.1939 11.81 4-Chlorobenzenesulfonic Acid -0.8340 10.45 (R,S) Baclofen:p-Phenolsulfonate monohydrate, 2: 2 crystallizes in the P21/n space group and contains one molecule of baclof en, p-phenolsulfonate, and water in the asymmetric unit, Figure 4.5b. An inversion center lies between R and S baclofen molecules and each baclofen contains ch arge assisted hydrogen bonds from the ammonium group to three different p-phe nolsulfonte molecules via oxygen bonded to 113

PAGE 131

sulfur [N11-H11A O22: N O 2.829(8) H O 2.159 N-H O 156.7 ; N11H11B O22: N O 2.887(8) H O 2.201 N-H O 129.2 ; N11-H11C O23: N O 2.832(9) HO 2.062 N-H O 148.1 ]. The carboxylic OH group on baclofen acts as a donor to a water molecule which then inturn hydrogen bonds to three different p-phenolsulfonate molecule s creating a 3D network, Figure 4.6, [O11H11A O1S: O O 2.705(8) H O 1.899 O-H O 160.0 ; O24-H24 O1S: O O 2.964(9) H O 2.160 O-H O 160.1 ; O1S-H1SAO23: O O 2.712(7) H O 1.655 O-H O 179.0 ; O1S-H1SB O24: O O 3.066(9) H O 2.057 O-H O 143.6 ]. The packing arrangement is similar to (R) baclofen HCl in the sense that staggered interdigitating phenyl rings are evident. a b Figure 4.5: a) Single Crystals of Bacl ofen:p-Phenolsulfonate monohydrate, 2.41 b) Asymmetric unit for 2. 114

PAGE 132

Figure 4.6: Overall supramolecular packin g motif for Baclofen:p-Phenolsulfonate monohydrate, 2. (R,S) Baclofen:4-Chlorobenzenesulfonate monohydrate, 3: This sulfonate salt crystallizes in the C2/c space group as needles, Figure 4.7, and contains one molecule of baclofen, 4-chlorobenzenesulfonate, and wate r in the asymmetric unit. Each different proton on the ammonium group interacts with three different 4-chlorobenzenesulfonate molecules via oxygen [N1-H1A O23: N O 2.911(2) H O 2.099 N-H O 111.2 ; N1-H1B O22: N O 2.810(2) H O 1.984 N-H O 158.4 ; N1-H1C O21: N O 2.828(2) HO 2.034 N-H O 155.9 ]. The carboxylic acid group of an (R) baclofen molecule interacts with two di fferent water molecules via the carbonyl, [O2 H41B-O41: O O 2.769 (2) H O 2.003 O H-O 160.41 o] and the OH group [O1-H10 O41: O O 2.535(2) HO 1.660 O-H O 169.4 ]. Those water molecules then interact in the same manne r with an (S) baclofen molecule, Figure 4.8. There are chains created by (R) baclofen molecules interacting by charge assisted hydrogen bonding with 4-chlorobenzenesulfona te molecules along the b axis that are 115

PAGE 133

connected to similar chains made by (S) baclof en molecules via water in the direction of the c axis. These stacked chains create sheets parallel to the bc plan e that stack along the a axis, Figure 4.9, with 21 screw axis along the b axis in between the sheets. Table 4.2 contains crystallogra phic information pertaining to 2-3. Indeed the strategy to use a strong acid to disrupt the intramolecular interac tion for baclofen was successful. TGAs for 1-3 can be seen in Appendix 3 and are relatively consistent with one water for each. All of them indicate that the water molecules are in channels due to the water coming off below 100o C,42 which is confirmed by the structures of 2 and 3. The water is held less tightly for 3 since in the TGA weight loss begins the moment heating begins from ambient temperature. Figure 4.7: Single Crystals of baclofen:4 -chlorobenzenesulfonate monohydrate, 3. 116

PAGE 134

Figure 4.8: Supramolecular interactions within the unit cell for baclofen:4chlorobenzenesulfonate monohydrate, 3. Figure 4.9: Overall suprmolecula r packing motif for baclofen:4chlorobenzenesulfonate monohydrate, 3, looking down the b axis. 117

PAGE 135

Table 4.2: Crystal structure parameters for salts 2-3. Compound 2 3 Chemical formula C10H13Cl NO2C6H5O4S H2O C10H13Cl NO2 C6H4O3SCl H2O Formula weight 405.89 424.33 Crystal System Monoclinic Monoclinic Space group P21/ n C2/c a () 13.837(9) 29.694(4) b () 5.584(3) 5.610(8) c () 23.750(13) 23.438(3) (o) 90 90 (o) 102.381(14) 101.453(2) (o) 90 90 Vol (3) 1792.5(19) 3826.8(9) Dcal (g cm-1) 1.504 1.473 Z 4 8 Z 1 1 Reflections collected 3273 8778 Independent reflections 2231 4303 Temperature (K) 100 100 R1 0.0829 0.0424 w R2 0.2044 0.1071 GOF 0.98 1.046 118

PAGE 136

4.3.3 Pure Water Dissolution In order to dissolve a solid into any solv ent system the lattice energy of the solid must be over come in order for solvent molecu les to interact and solvate each individual molecule within the solid. Melting point is sometimes indicative of a general solubility trend when dealing with non-i onizable compounds but in the case of salts this can be misleading since a salt will typically have a high melting point and a high aqueous solubility as in the case of sodium chloride whose melting point is ca. 800o C with an aqueous solubility of 360 mg/mL.43 The melting points for the multiple component crystalline forms studied here are shown in Table 4.3. Table 4.3: Melting Points as dete rmined by DSC (See Appendix 3). Compound Melting Point of Supramolecular Complex (oC) Melting Point of Former (oC) Baclofen NA 206-208 1 NA 198.37 2 181.47 6.4 3 190.61 102 Dissolution profiles were measured at 37o C in order to mimic physiological conditions although it should be pointed out that these we re not USP (United States Pharmacopeia) validated protocols. The particle size was controlled by sieves with a range of 53 75 m for all dissolution profiles in this and the ne xt two sections. The goa l of these dissolution experiments herein were aimed at understanding the effect of different pH environments with regard to oral and in trathecal dosing. Figures 4.10 and 4.11 show the twenty four hour and one hour dissolution profiles for baclofen, 2, and 3 in pure water. Ranking the forms from lowest to highest solubility in water results in 3 at 3.88 mg/mL, baclofen at 119

PAGE 137

5.01 mg/mL, and 2 at 60.22 mg/mL. The standard deviation for these profiles is low and 2 is the clear winner if 1st prize is for the greatest solubility. 2 reached a value ca. 20 times higher than pure baclofen partially due to the decrease in pH to 2.40 initially and 1.86 at twenty four hours associat ed with the dissolution of the form since baclofen is more soluble under these pH conditions. Tabl e 4.4 includes the pH values recorded. Interestingly 3 retained a slightly lower solubility compar ed to baclofen even over a full day despite the fact that the pH decreased to < 3. Both salts did not have greater solubility in water which is somewhat surprising considering the sustained lower pH of both salt forms in water. It should be pointed out that the only difference between 2 and 3 is an alcohol on the aromatic ring of p-phenolsulfonate in 2, which is a chloride in 3. The hydrogen bonding capability in 2 fueled by the strong dipole mo ment for an alcohol is likely the cause for such a dramatic difference in solubility with a minor difference in molecular structure due to solvation by water. Table 4.4: pH values for water dissolution. Chemical Name pH Initial pH 24hr Baclofen 5.70 6.58 Bac:p-Phenolsulfonate 2.40 1.86 Bac:4-Chlorobenzenesulfonate 2.96 2.71 120

PAGE 138

Figure 4.10: 24 hour dissolution profiles in water at 37o C. Figure 4.11: 1 hour dissolution profiles in water at 37o C. 121

PAGE 139

PXRD data was collected on the solid re maining after dissolution, Appendix 3, for baclofen, 2, and 3. For baclofen the powder was partia lly amorphous and exhibited peaks primarily from the original anhydrous material and also 1, which is expected since the formation of the hydrate can take days at 25o C. For 2 the post dissolution PXRD pattern matched the salt as indicated by the di ssolution profile remaining stable. For 3 the post dissolution PXRD pattern also matched th e salt as indicated by no change in the dissolution profile where it could have been possible to precipitate out 1. 4.3.4 pH 1 (0.1 N HCl) Dissolution The dissolution profiles in 0.1 N HCl are shown in Figures 4.12 and 4.13 for twenty four hours and 1 hour respectively. The or der of lowest to high est solubility is the same as in water however the differences between the forms are not the same. The solubilities from the low end are; 3 at 9.36 mg/mL, baclofen at 30.79 mg/mL, and 2 at 59.16 mg/mL. Each solubility dramatically increased in 0.1 N HCl, when compared to water, except for 2 which was not significantly different. As it is in the single crystal structures, in a solution at a pH of 1 baclof en will be protonated leading to a neutral carboxylic acid and an ammonium cation, which in turn causes the molecule to be polar and therefore much more water soluble since the zwitterionic structure is defeated. This phenomenon explains the increases in solubi lity under low pH conditions. The solubility of 2 remained virtually the same compared to the water dissolution since the pH was decreased to < 2 in water afte r twenty four hours. The pH values for powder dissolution in 0.1 N HCl are listed in Table 4.5. 122

PAGE 140

Table 4.5: pH values for 0.1 N HCl dissolution. Chemical Name pH Initial pH 24hr Baclofen 3.17 3.23 Bac:p-Phenolsulfonate 1.20 0.97 Bac:4-Chlorobenzenesulfonate 1.22 0.95 Figure 4.12: 24 hour dissolution profiles in 0.1 N HCl at 37o C. 123

PAGE 141

Figure 4.13: 1 hour dissolution profiles in 0.1 N HCl at 37o C. Figures 4.12 and 4.13 are more representative of an environment that an oral dosage form would experience since the stomach contains HCl and the average human body temperature is 37o C. The equilibrium solubility has not been reached in 0.1 N HCl for baclofen at one day since the curve continues to climb slightly. Interestingly the post dissolution solids contained the same phases as in water, as shown in Appendix 3, for each compound except for baclofen which was considered more crystalline due to the absence of the amorphous hump between 5 and 20o 2 theta. Also, it should be pointed out that the HCl salt of baclofen was not indicated by any of the PXRD data. 4.3.5 pH 7 Sodium Phosphate Buffer Dissolution Intrathecal administration is currently used for baclofen in cases of severe spasticity when the patient is unresponsive to oral therapy or central nervous system (CNS) side effects, like sedation, become too great, like sedation for example. Use of an 124

PAGE 142

implanted pump allows for a large reducti on of dose while maintaining efficacy with reduced side effects, however, surgically implanted pumps have their own inherent risks which is why they are reserved for patients that do not derive sufficient benefit from oral therapy. Plasma levels associated with intr athecal administration ar e 100 times lower than those obtained by oral doses. Current dosi ng for maintenance therapy associated the implanted pumps ranges from 12 2003 g/day. Current refill inte rvals last ca. sixty to ninety days and Albright et al acknowledges the importance of increasing the solubility to reduce pump refill intervals which is highlighted in the intrathecal package insert, certain populations of patients require hi gher daily doses of bac lofen, and obtaining a desirable length of time between refills for these patients becomes difficult. 27, 44 Compounding the problem, strict aseptic conditions must be us ed during refills to avoid bacterial infection. Therefore improvement of the solubility could reduce the frequency of refills and overall health care costs by reducing physician visits. With this in mind solubility of the baclofen salts discovered wa s tested in a pH 7 buffer since the pH of the cerebral spinal fluid (CSF) is 7.30 to 7.3645 and the current formulation for intrathecal administration comes in the pH range of 5 7. The solubilities from lowest to highest were as follows; 4.10 mg/mL for baclofen, 5.51 mg/mL for baclofen:p-phenolsulf onate, and 5.81 mg/mL for baclofen:4chlorobenzenesulfonate. The dissolution prof iles, Figures 4.14 and 4.15 for twenty four hours and 1 hour respectively, reveal the salts having relatively low increases compared the other solvent systems tested. 125

PAGE 143

2 exhibits a profile with an in itial boost in solubility that is lost as the concentration is moves toward that of pure baclofen after two days. Two day time courses were employed for this solvent system since equilibrium wa s not reached for baclofen after one day in 0.1 N HCl. Figure 4.14: 24 hour dissolution profiles in pH 7 buffer at 37o C. Figure 4.15: 1 hour dissolution profiles in pH 7 buffer at 37o C. 126

PAGE 144

While baclofen seems to have reached equilibrium, 3 appears to be still increasing after two days which is surpri sing considering the small difference in molecular structure between 2 and 3. The interaction between ions is clearly stronger with respect to 3 at pH 7, which could be related to so lvation related to the polar na ture of the hydroxyl group on 2 instead of the chlorine on 3. Both salts resulted in ca. 66% increase in solubility compared to baclofen after one hour alt hough it is not known why between five and ten minutes 3 remained constant. It is possible that this was experimental error although each of the three samples agreed as indi cated by the very sm all error bars. Compound 3 has the best performance of the two salts since it maintains a higher solubility and the concentration of API does not come do wn although it should be noted that longer studies would be necessary to test the ability of this form to sustain long periods of time in an intrathecal implanted pump. A pH of 7 remained constant throughout the entire dissolution experiment and interactions with buffer are also likely affecting the results. Both the formation of phosphate salts and solute -solute interactions between all the components present can affect solubility. Further studies would be necessary to optimize buffer concentration, which could be lowered since the current approved formulation for intrathecal administ ration exhibits a pH range of 5 7. The PXRD of the solid remaining after the disso lution experiment can be referred to in Appendix 3. For baclofen the PXRD patte rn has an amorphous hump and peaks corresponding to 1 and baclofen. PXRD data for 2 primarily matches 1 that is slightly mixed with baclofen. For 3 the PXRD pattern is not what one would expect after looking at the dissolution profile. Since the profile does not decrease towards baclofens value it would be expected that the pattern would matc h the original salt as it does in the other 127

PAGE 145

two solvent systems tested, however, the pattern is consistent with 1 containing some amorphous content. This could mean that the sustained solubility for baclofen could be related to excess 4-chlorobenzenesul fonic acid that is present while 1 is precipitating out or phosphate salt formation could be occu rring. The excess powder present that is amorphous could be the source of baclofen fuel ing this trend. It would be useful to further experiment with 3 to determine if at a certain concentration known to be soluble would result in 1 precipitating out over time. 4.4 Conclusions for Baclofen Salts pH has a tremendous effect on the solubility of baclofen but predicting such an affect for baclofen sulfonate salts be comes very difficult if not impossible. Physiologically speaking some level of aqueous solubility is necessary for an API to be effective as a therapeutic agent. pH has a clear effect on ionizable APIs which becomes more complicated when zwitterions are involv ed. The more ionization states a molecule possesses the more difficult predicting solubil ity becomes, but it is possible to control which ionization state the API exhibits in the solid state throu gh crystal engineering which could be translated to solution. The existing literature shows salts, in general, will have a larger increase in aque ous solubility and that cocrystals are more amenable to crystal engineering, however, in this case using the strategy of u tilizing a strong acid to disrupt the uniquely stabilized zwitterionic structure of baclofen was successful. The solubility difference pertaining to the sulfonate salts was dramatic in water and at pH 1 which maintained a trend of lowe st to highest solubility of 3, baclofen, then 2. Although a twenty fold increase in solubility was achieved with 2 in those solvent systems and p128

PAGE 146

phenolsulfonic acid (median lethal dose (LD50) = 1900 mg/kg in rats) has been used in dental applications implying some level of safety,46-47 the solubility was not as dramatically increased or stable at pH 7. For intrathecal administration 2 is not a useful candidate since the pH range of 5 -7 must be maintained without precipitation of the API as 1. 3 was the most interesting since it retained a lower solubility in water and at pH 1 compared to baclofen but maintained the highest solubility at pH 7. This ca. 66% increase in solubility compared to pure baclofen makes 3 a promising candidate to study further and could lead to fewer refills for intratheca l pumps reducing surgery associated risks and potential precipitation at concentrations above 2 mg/mL, which can clog the pump leading to an underdose or overdose due to bur sting. The safety of sulfonates in CSF is beyond the scope of this work but potentia l therapeutic benefits may exist. 4Chlorobenzenesulfonic acid has an LD50 > 500 mg/kg in rats but toxicological studies would be necessary upon further developm ent once the physicochemical properties are completely understood. There is a con cern of pH reduction in CSF with 3, however, this can be controlled either by the natural CSF buffering capabilities45 which utilize endogenous bicarbonate or by addition of a bicarbonate buffer to the intrathecal formulation. 4.5 Baclofen Lactam Polymorphis m: High and Low Z Structures 4.5.1 Background Z can be strictly defined as the number of formula units in the unit cell divided by the number of independent general positions.48 In 2006 structures with a Z > 1 accounted for 8.8% of the CSD 49. Despite their low frequency of occurrence there has 129

PAGE 147

been recent interest concerning structures with Z > 1 with some debate over their occurrence in the CSD50 and why they occur.48, 51-54 Whether high Z prime structures are simply kinetic structures which have not re ached thermodynamic equilibrium or are just the best option energetically is an importa nt question, however, a CSD analysis by Steed along with other reports claim that some high Z structures are in f act the most stable form 51-52, 55, which has been supported by computational modeling.56-57 It is recognized that high Z structures occur more frequently for homochiral molecu les such as steroids and nucleosides 53 and also for chiral space groups.48 On rare occasions chiral, or Sohnke, space groups can be assigned to racemates, i.e. kryptoracemates, which are likely to contain pseudosymmetric elements between enantiomers.54 The appearance of pseudosymmetry in struct ures with Z > 1 has been shown to occur 27% of the time, where as for Z = 2 structures it occurs 83% of the time.48, 57-58 Pseudosymmetry, which can be described as molecules or intermolecular aggregates coming very close to a symmetry element wit hout actually fulfilling it is likely influenced by what has been called frustration between favorable p acking and highly directional (strong) supramolecular synthons.55 It would not be su rprising that molecules of a certain shape and size simply cannot satisfy either demand completely, which could result in a compromise between the two, leading to more than one symmetry equivalent in the asymmetric unit and indeed Steed has highlight ed the factors that may lead to a high Z structure48; 1) irregular, non-self complementary shape 2) small number of strong intermolecular interacting functionalities 3) frustration between overall packing and strong intermolecula r interactions and 4) strong self complimentar y functionality with a 130

PAGE 148

resolved chiral center. Also, when there are multiple energetically favorable conformations for a molecule a higher te ndency toward polymorphism and Z > 1 structures would be expected. There are a multiple examples containing polymorphic systems where at least one of the polymorphs have a Z > 1 51-52, 56, 59-64, but there are also cases where each structure in a polymorphic se t contains a Z > 1, as in the case of cholesterol 48, 65. A case of polymorphism including a high Z structure for (R,S) baclofen lactam, or (R,S) 4-(4-chlorophenyl)-2 -pyrrolidone, Figure 4.16, is presented herein. Baclofen lactam is a dehydration product of baclofen and can be isolated as a synthetic intermediate on the way to baclofen66 or used for the synthesis of a proposed GABAergic prodrug.67 It has also been shown that a lyophiliz ed formulation with baclofen containing polyvinylpyrrolidone K30 (PVP K30), a hydrophilic excipient used as a wet binder or disintegrant, had a significant effect on th e degradation of baclofen to the lactam 68. Form I of (R,S) Baclofen lactam has been pr eviously deposited in the CSD (refcode ZUWKOR) and exhibits Z = 1.69 The R form of baclofen la ctam prepared by resolution using (2R, 3R)-(+)-tartaric acid was also report ed in the same article as Form I. It is reported herein how attempts to prepare single crystals of (R,S) baclof en or cocrystals of (R,S) baclofen afforded two new polymorphs of the lactam, Forms II and III. 131

PAGE 149

Figure 4.16: Baclofen lact am chemical diagram. 4.5.2 Synthesis Form II, was prepared by dissolvi ng 300mg of (R,S) baclofen in hot 3bromopyridine and the solution was left at room temperature to evaporate. Large colorless blocks were present after 128 days. Form III, was prepared by dissolving a 1:2 mola r ratio of glutaric acid (92.75mg, 0.702mmol) to (R,S) baclofen (300mg, 1.40mmol ) in hot dimethylformamide (DMF) and the solution was left to evaporate at room te mperature. Colorless plates and blocks were present after 38 days. 4.5.3 Cambridge Structural Database Statistics The CSD was surveyed for the occurrence of high Z structures and the statistics are presented in Table 4.6. Z > 1 structur es account for 8.94% of all structures and 11.90% of organics. These values are consistent with the of 8.8% and 11.5%, respectively, provided by Steed in 2006.49 Steed also reported that 14.6% of crystals that 132

PAGE 150

adopt chiral space groups exhibit high Z values, as might be expected.70 As Z increases the occurrence of structures drops off consider ably and only 0.08% of structures exhibits Z > 4. Table 4.6: CSD statistics for Z > 1.71 Z value No restrictions % of total Only organics % of only organics > 1 46827 8.94 26706 11.90 > 2 4868 0.93 3133 1.40 > 3 2542 0.49 1731 0.77 > 4 441 0.08 322 0.14 4 2531 0.48 1726 0.77 = 4 2090 0.39 1404 0.63 A search was undertaken with greater rest rictions (3D coordinates determined, R 0.05, only organics, and Z 4) to get a more deta iled perspective of Z 4 structures, Table 4.7. This afforded only 175 hits amounting to 0.17% of the total amount of entries with these same restrictions. This value is much lower than the 0.77% in Table 4.6 for Z 4 (only organics) indicating that as higher standards are applied the occurrence of high Z structures is even less lik ely than indicated by the raw se arch. Each of the 175 entries in Table 4.7 was analyzed for chiral centers for racemates, and for kryptoracemates. It was found that 23.30% of Z 4 structures are racemates whereas only 3.43% were kryptoracemates, which has been reported to be very rare.54 133

PAGE 151

Table 4.7: CSD Refined statistics for Z 4.72 Total % out of 104,996 # Racemates (% of total) #Kr y ptoracemates (% of total) 175 0.17 41 (23.30) 6 (3.43) 4.5.4 Crystal Structure Descriptions Form II (4) Form II is a racemate that crystallizes in P -1 with four molecules in the asymmetric unit. The asymmetric unit is depict ed in Figure 4.17 where the bc plane is the plane of the page. The four different molecule s in the asymmetric unit are labeled A D. Each molecule forms an amide dimer with another baclofen lactam molecule but not necessarily with the opposite en antiomer. A (proton at the chir al center is coming toward you), B, and D are in the R configurati on, while C is in the S configuration. Figure 4.17: 50% probability ORTEP diagram fo r the asymmetric unit of Form II. 134

PAGE 152

The hydrogen bond distances and angles are presented in Table 4.8. A is paired with another R molecule (R-R dimer) [N41-H41 O61: N O 2.849(4) H O 1.92 NH O 170.0 ], B is paired with C, an S molecule (R-S dimer) [N1-H1 O21: N O 2.919(4) H O 2.09 N-HO 163.1; N21-H21 O1: N O 2.827(4) HO 1.97 N-H O 177.6 ], and D is paired with a nother R molecule (R-R dimer) [N61H61 O41: N O 2.925(4) HO 1.96 N-H O 158.2 ]. These hydrogen bond distances are close to that of Form I (2.925 ), the primary amide dimer (2.95 see Appendix 3), and the secondary amide dime r which includes lactams (2.87 see Appendix 3). The B-C (R-S) dimer appears to have a pseudosymmetric center of inversion. There are centers of inversion on the corners of the unit cell, the middle of each axis, the center of the unit cell, and at the center of each face. These inversion centers are between dimeric pair s of S-S and R-R or R-S and RS. It is rather unusual that the R-S pairing is maintained but not around an inversion center as is typical of racemates. Figure 4.18 highlights the odd pairi ng of enantiomers. Selected bond angles and distances are provided in Table 4.9. The bond distances and angles around the stereogenic carbon atom for A D range as follows; A: 1.512(5) 1.536(5) and 101.4(3)o 118.4(3)o B: 1.515(4) 1.560(5) and 102.0(3)o 118.0(3)o, C: 1.463(6) 1.537(5) and 104.5(3)o 117.3(3)o, and D: 1.516(5) 1.556(5) and 103.0(3)o 118.2(3)o. Each symmetry independent molecule va ries slightly from the next but C, which is the only S molecule in the asy mmetric unit, has a relatively short C-C bond between the chiral carbon and the carbon conne cted to nitrogen in the five member ring. This could be a result of packing forces. 135

PAGE 153

Figure 4.18: Pairing between R and S bacl ofen lactam molecules in Form II. Form III (5) Form III crystallizes in P 21/c and like form I only has one molecule in the asymmetric unit. This low Z polymorph is a racemate and contains the amide dimer observed in Forms I and II with hydrogen bonding parameters as follows; [N1-H2NO1: N O 2.915(3) HO 1.91 N-H O 168.6 ], Table 4.8. The amide dimer is depicted in Figure 4.19. 136

PAGE 154

Figure 4.19: Lactam dimer in Form III. These hydrogen bond distances are close to th at of Form I (2.925 ), the primary amide dimer from a CSD survey (2.95 see Appendi x 3), and the secondary amide dimer from a CSD survey which includes lactams (2.87 see Appendix 3).The bond angles and distances, Table 4.8, around the chir al carbon atom range from 103.0(3) 115.5(3)o and 1.512(4) 1.540(4) respectively. The bond distances and angles are comparable to Forms I and II. The torsion a ngle (C3-C4-C7-C10) of 123.2(3)o, Table 4.9, is smaller than that in Form II but larger than that in Form I. The puckering of the five member ring is reminiscent of Form II but the planes of the rings are closer to pe rpendicular. There is an inversion center between R and S molecules and a 21 screw axis along b. Form III was obtained in the presence of glut aric acid during an unsuccessf ul cocrystallization attempt with (R,S) baclofen. The heat involved most likely caused th e dehydration of baclofen to the lactam and the presence of glutaric acid served as an accidental additive affecting nucleation of Form III. Additives have been shown to affect the outcome of crystallizations, for example there have been failed cocrystal attempts where the cocrystal former has impacted polymorphism and even induced high Z structures. 60, 63 Crystallographic data for Forms II and III can be found in Table 4.10. 137

PAGE 155

Table 4.8: Selected bond distances and angles for Forms II (4) and III (5) [ and ]. Bond lengths () Bond angles () _____________________________________________________ Form II C(4)-C(7) 1.515(4) C(4)-C(7)-C(8) 118.0(3) C(7)-C(8) 1.539(5) C(4)-C(7)-C(10) 113.0(3) C(7)-C(10) 1.560(5) C(8)-C(7)-C(10) 102.0(3) C(24)-C(27) 1.537(5) C(30)-C(27)-C(28) 104.5(3) C(27)-C(30) 1.463(6) C(30)-C(27)-C(24) 115.9(4) C(27)-C(28) 1.528(5) C(28)-C(27)-C(24) 117.3(3) C(44)-C(47) 1.517(5) C(48)-C(47)-C(44) 118.2(3) C(47)-C(48) 1.516(5) C(48)-C(47)-C(50) 103.0(3) C(47)-C(50) 1.556(5) C(44)-C(47)-C(50) 113.5(3) C(64)-C(67) 1.512(5) C(64)-C(67)-C(68) 118.4(3) C(67)-C(68) 1.526(5) C(64)-C(67)-C(70) 114.8(3) C(67)-C(70) 1.536(5) C(68)-C(67)-C(70) 101.4(3) Form III C(4)-C(7) 1.512(4) C(4)-C(7)-C(10) 113.3(3) C(7)-C(10) 1.538(4) C(4)-C(7)-C(8) 115.5(3) C(7)-C(8) 1.540(4) C(10)-C(7)-C(8) 103.0(3) Table 4.9: Selected torsion angles (o) for Forms I, II (4), and III (5). ________________________________________ Form II A C(45)-C(44)-C(47)-C(48) 171.7(3) B C(5)-C(4)-C(7)-C(8) 175.7(4) C C(25)-C(24)-C(27)-C(28) 144.9(4) D C(65)-C(64)-C(67)-C(68) 167.3(4) Form III C(3)-C(4)-C(7)-C(8) -118.3(3) C(3)-C(4)-C(7)-C(10) 123.2(3) 138

PAGE 156

Table 4.10: Crystallographic data for forms II (4) and III (5). Compound Form II Form III Formula C10H10ClNO C10H10ClNO Formula Weight 195.64 195.64 Crystal System Triclinic Monoclinic Space Group P-1 P21/c a () 9.7937(10) 10.936(4) b () 13.2904(14) 8.932(4) c () 16.2963(17) 9.878(4) 69.985(2) 90 79.546(2) 90.202(8) 68.724(2) 90 V (3 ) 1853.2 (3) 964.9(7) Z 2 4 Z 4 1 D (g cm-1 ) 1.402 1.347 Temperature (K) 100 (2) 223(2) range 1.33 28.28 1.86 to 28.28 Limiting indicies -10 h 13, -17 k 16, -20 l 13 -13 h 12, -8 k 11, -12 l 12 Reflns measd 9112 4571 Reflns unique / Rint 7342, 0.0189 2114, 0.0394 Reflns observed 5157 935 Tmin, Tmax 0.361, 1.000 0.781, 1.000 Goodness of fit on F2 1.021 0.970 Completeness to 28.28o, 79.8% 28.28o, 88.7% R1, w R2 [I>2sigma(I)] R1 = 0.0699, wR2 = 0.1632 R1 = 0.0574, wR2 = 0.1343 139

PAGE 157

4.5.5 Conclusions for Baclofen Lactam Polymorphs Forms II and III of (R,S) baclofen lactam both came serendipitously. Form II came from an attempt to grow single crystals of pure (R,S ) baclofen where the applied heat in the presence of 3-bromopyridine cause d dehydration of baclofen to the lactam and subsequent crystallization. Fo rm II crystals did not form for 128 days therefore whether or not this is a kinetic product remains to be seen and requires further study. It is possible however that a high Z structure can be the most stable form. 55-57 One factor previously reported to lead to a high Z structure, ha ving a small number of strong intermolecular functionalities, is confirmed for Form II. This high Z polymorph contains a peculiar arrangement of enantiomers and has larger to rsion angles than the other polymorphs. Z 4 structures only account for a mere 0.48% of the CSD and any further discoveries of high Z structures will add to the understanding of w hy and how these solid-state arrangements come about. 4.6 References Cited (1) Berge, S. M.; Bighley, L. D.; Monkhouse, D. C., Journal of Pharmaceutical Sciences 1977, 66, 1. (2) Khankari, R. K.; Grant, D. J. W., Thermochimica Acta 1995, 248, 61. (3) Vippagunta, S. R.; Brittain, H. G.; Grant, D. J. W., Advanced Drug Delivery Reviews 2001, 48, 3. (4) Aaltonen, J.; Alleso, M.; Mirza, S.; Koradia, V.; Gordon, K. C.; Rantanen, J., European Journal of Pharmaceutics and Biopharmaceutics 2009, 71, 23. 140

PAGE 158

(5) Yin, S. X.; Grosso, J. A., Current Opinion in Drug Discovery & Development 2008, 11, 771. (6) Cheney, M. L.; Shan, N.; Healey, E. R.; Hanna, M.; Wojtas, L.; Zaworotko, M. J.; Sava, V.; Song, S. J.; Sanchez-Ramos, J. R., Crystal Growth & Design 2010, 10, 394. (7) Pudipeddi, M.; Serajuddin, A. T. M., Journal of Pharmaceutical Sciences 2005, 94, 929. (8) Bauer, J.; Spanton, S.; Henry, R.; Quick, J.; Dziki, W.; Porter, W.; Morris, J., Pharmaceutical Research 2001, 18, 859. (9) Serajuddin, A. T. M., Advanced Drug Delivery Reviews 2007, 59, 603. (10) Berge, S. M.; Bighley, L. D.; Monkhouse, D. C., Journal of Pharmaceutical Sciences 1977, 66, 1. (11) Stahl, P. H.; Wermuth, C. G., Handbook of Pharmaceutical Salts: Properties, Selection, and Use WILEY-VCH: Zurich, 2002. (12) Paulekuhn, G. S.; Dressman, J. B.; Saal, C., Journal of Medicinal Chemistry 2007, 50, 6665. (13) Fini, A.; Fazio, G.; Hervas, M. J. F.; Holgado, M. A.; Rabasco, A. M., European Journal of Pharmaceutical Sciences 1996, 4, 231. (14) Bowery, N. G., Trends in Pharmacological Sciences 1982, 3, 400. (15) Misgeld, U.; B ijak, M.; Jarolimek, W., Progress in Neurobiology 1995, 46, 423. (16) Chang, C. H.; Yang, D. S. C.; Y oo, C. S.; Wang, B. C.; Pletcher, J.; Sax, M.; Terrence, C. F., Acta Crystallographica Section B-Structural Science 1982, 38, 2065. 141

PAGE 159

(17) Albright, A. L., Journal of Child Neurology 1996, 11, 77. (18) Kolaski, K.; Logan, L. R., Neurorehabilitation 2007, 22, 383. (19) Krach, L. E., Developmental Medicine and Child Neurology 2009, 51 106. (20) Greene, P., Clinical Neuropharmacology 1992, 15, 276. (21) Chole, R.; Patil, R.; Degwekar, S. S.; Bhowate, R. R., Journal of Oral and Maxillofacial Surgery 2007, 65, 40. (22) Kumru, H.; Kofler, M.; Valls-Sole, J.; Portell, E.; Vidal, J., Neurorehabilitation and Neural Repair 2009, 23, 921. (23) Schreiber, A. L.; Fried, G. W.; Formal, C. S., Neuromodulation 2009, 12, 310. (24) Berner, L. A.; Bocarsly, M. E.; Hoebel, B. G.; Avena, N. M., Behavioural Pharmacology 2009, 20, 631. (25) Addolorato, G.; Leggio, L.; Agab io, R.; Colombo, G.; Gasbarrini, G., International Journal of Clinical Practice 2006, 60 1003. (26) Heinrichs, S. C.; Leite-Morris, K. A.; Carey, R. J.; Kaplan, G. B., Behavioural Brain Research 2010, 207, 353. (27) Sigg, J., Sonntag, J., Li, J., Solubili ty and Stability of Intrathecal Baclofen Solutions at High Concentrations. Im plications for Chronic Use in the Synchromed Infusion System. Medtronic, UC200701657 EN NP7533, 2006. (28) Brittain, H. G., Pharmaceutical Technology 2007, 31, 78. (29) Reddy, L. S.; Bethune, S. J.; Kampf, J. W.; Rodriguez-Hornedo, N., Crystal Growth & Design 2009, 9, 378. (30) Mohamed, S.; Tocher, D. A.; Vickers, M.; Karamertzanis, P. G.; Price, S. L., Crystal Growth & Design 2009, 9, 2881. 142

PAGE 160

(31) Bruker SMART, SAINT-Plus, SADABS, XP and SHELXTL Madison, Wisconsin, USA, 1997. (32) Sheldrick, G. M. SHELXTL, University of Gottingen: Germany, 1997. (33) It is noted that this dissolution study did not use the standard USP apparatus but it is believed that our dissolution stud y produced reasonably comparable results. (34) Mirza, S.; Miroshnyk, I.; Rantane n, J.; Aaltonen, J.; Harjula, P.; Kiljunen, E.; Heinamaki, J.; Yliruusi, J., Journal of Pharmaceutical Sciences 2007, 96 2399. (35) Mohamed, S.; Tocher, D. A.; Vickers, M.; Karamertzanis, P. G.; Price, S. L., Crystal Growth and Design doi:10.1021/cg9001994. (36) Bhogala, B. R.; Basavoju, S.; Nangia, A., Crystengcomm 2005, 7, 551. (37) Delori, A.; Suresh, E.; Pedireddi, V. R., Chemistry-a European Journal 2008, 14 6967. (38) Childs, S. L.; Stahly, G. P.; Park, A., Molecular Pharmaceutics 2007, 4, 323. (39) EPA, Initial Risk-Based Prio ritization of High Production Volume (HPV) Chemicals Hydroxybenzenesulfoni c Acid (CASRN 1333-39-7) 2009. (40) Scifinder Scholar 2007. (41) Samples were viewed with a fully automated, upright Zeiss AxioImagerZ.1 microscope with a 20x /0.70NA dry object ive, and Nomarski DIC contrasting prisms. Z-stacks of images were crea ted at 0.5 micron step sizes using the AxioCam MRm CCD camera and Axiovi sion version 4.6.02 software suite. Images were then 3-dimensionally recons tructed using the iso-surface technique in Bitplanes (Zurich, Switzerland) Imaris software version 5.0.3. 143

PAGE 161

(42) Clarke, H. D.; Arora, K. K.; Bass, H.; Kavuru, P.; Ong, T. T.; Pujari, T.; Wojtas, L.; Zaworotko, M. J., Crystal Growth & Design 2010, 10, 2152. (43) Sodium Chloride Material Safety Data Sheet, J.T. Baker. (44) Albright, A. L.; Awaad, Y.; Muhone n, M.; Boydston, W. R.; Gilmartin, R.; Krach, L. E.; Turner, M.; Zidek, K. A.; Wright, E.; Swift, D.; Bloom, K., Journal of Neurosurgery 2004, 101, 64. (45) Mitchell, R. A.; Carman, C. T.; Seve ringhaus, J. W.; Richardson, B. W.; Singer, M. M.; Shnider, S., Applied Physiology 1965, 20, 443. (46) Hamaguchi, F.; Tsutsui, T., Japanese journal of pharmacology 2000, 83, 273. (47) Miyachi, T.; Tsutsui, T., Odontology 2005, 93, 24. (48) Steed, J. W., Crystengcomm 2003, 5, 169. (49) Anderson, K. M.; Afarinkia, K.; Yu, H. W.; Goeta, A. E.; Steed, J. W., Crystal Growth & Design 2006, 6, 2109. (50) Allen, F. H., Acta Crystallographica Section B-Structural Science 2002, B58, 380. (51) Anderson, K. M.; Steed, J. W., Crystengcomm 2007, 9, 328. (52) Desiraju, G. R., Crystengcomm 2007, 9, 91. (53) Bishop, R.; Scudder, M. L., Crystal Growth & Design 2009, 9, 2890. (54) Fabian, L.; Brock, C. P., Acta Crystallographica Section B-Structural Science 2010, 66, 94. (55) Anderson, K. M.; Goeta, A. E.; Steed, J. W., Crystal Growth & Design 2008, 8, 2517. (56) Das, D.; Banerjee, R.; Mondal, R.; Ho ward, J. A. K.; Boese, R.; Desiraju, G. R., Chemical Communications 2006, 555. 144

PAGE 162

145 (57) Gavezzotti, A., Crystengcomm 2008, 10, 389. (58) Collins, J. G., Yale Journal of Bi ology and Medicine 1993, 66, 457. (59) Dunitz, J. D.; Bernstein, J., Accounts of Chemical Research 1995, 28, 193. (60) Rafilovich, M.; Bernstein, J., Journal of the Americ an Chemical Society 2006, 128, 12185. (61) Sarma, B.; Roy, S.; Nangia, A., Chemical Communications 2006, 4918. (62) Nichol, G. S.; Clegg, W., Crystengcomm 2007, 9, 959. (63) Nath, N. K.; Nangia, A., Crystengcomm 2011. (64) Yu, L.; Reutzel-Eden s, S. M.; Mitchell, C. A., Organic Process Research & Development 2000, 4, 396. (65) Hsu, L. Y.; Nordman, C. E., Science 1983, 220, 604. (66) Yoshifuji, S.; Kaname, M., Chemical & Pharmaceutical Bulletin 1995, 43, 1302. (67) Wall, G. M.; Baker, J. K., Journal of Medicinal Chemistry 1989, 32, 1340. (68) Cutrignelli, A.; Denora, N.; Lopedota, A.; Trapani, A.; Laquintana, V.; Latrofa, A.; Trapani, G.; Liso, G., International Journal of Pharmaceutics 2007, 332, 98. (69) Caira, M. R.; Nassimbeni, L. R.; Scott, J. L.; Wildervanck, A. F., Journal of Chemical Crystallography 1996, 26, 117. (70) Brock, C. P.; Dunitz, J. D., Chemistry of Materials 1994, 6, 1118. (71) CSD Conquest 1.12 (August 2010 update) 523,834 total entries. Search parameter: organics only 224,378 total entries. (72) CSD Conquest 1.12 August 2010 update. Search parameters: 3D coordinates, R 0.05, only organics, and Z 4 total entries 104,996. One duplicate removed, KOVBIG.

PAGE 163

Chapter 5: Summary and Future Directions 5.1 Summary Crystal engineering has become another to ol for solid-state chemists in general and biological sciences in part icular to control self assembly of molecules w ith biological activity and is based upon understanding str ong intermolecular interactions. Utilizing highly directional and str ong hydrogen bonding can be used to affect physicochemical properties of materials, especially crystal line pharmaceutical formulations, with emphasis upon solubility and subsequent de livery of therapeutic compounds. This contribution has exemplified methodologies to generate new multiple component crystalline forms of pharmaceuticals with new properties such as solubility and pharmacokinetics. Mechanochemistry with the addition of small amounts of solvent, or solvent drop grinding (SDG), has been shown to be consistent with traditional solution based crystal growth for the generati on of cocrystals with carbamazepine. Dimethylformamide and dimethylsulfoxide have been shown to be the most effective solvents to grind with and subsequently produce the same cocrystal obtained by solution crystallizations. Solvents in which both molecules employed, API and cocrystal former, have some solubility have also been shown to successfully generate cocrystalline forms via SDG. 146

PAGE 164

This method is not only Green by nature, usin g microliter amounts of solvent, but also inexpensive and time efficient. These propert ies make it attractive for crystal form screening and the patentability of new cr ystalline forms drives the pharmaceutical industrys recent interest in this technique. SDG was also successfully used to ge nerate cocrystals with meloxicam, a commonly prescribed NSAID for mild to mode rate pain. The use of this medicine for acute pain could exist if a fa ster onset formulation was developed. With this in mind ten carboxylic acid cocrystals were synthesized, many by slurry methods as well, and their respective pharmacokinetics in rats was asse ssed with a single dose administration via oral gavage as a suspension. St ructural data in the form of single crystal X-ray diffraction data was produced for two of the co crystals, meloxicam:fumaric acid and meloxica:succinic acid. These two structures we re isostructural which is not surprising considering these two coformers only differ by a double bond. Five of these cocrystals outperformed meloxicam in terms of AUC and Cmax. In terms of a faster onset the earliest time point recorded was fifteen minutes and the meloxicam:4-hydroxybenzoic acid cocrystal exhibited a 2.73 fold increase in se rum concentration. The solubility of a largely prescribed anti-spastic agent, baclofen, ha s been affected by the generation of two sulfonate salts, one with pphenolsulfonic acid and one w ith 4-chlorobenzenesulfonic acid. The current intrathecal formulation has lim ited solubility, 2 mg/mL, at the pH range necessary for perispinal administration. This limits the length of tim e between surgically implanted pump refills and any concentration greater than that will lead to precipitation of baclofen monohydrate over time. Dissolution profiles were generated in three different solvent systems at 37o C, two of which were pH controlled. Pure water powder 147

PAGE 165

dissolution resulted in acidic pH and a 20 fold increase in solubility compared to pure baclofen for the p-phenolsulfonate salt afte r twenty four hours with a dramatic increase after only one minute. The 4-chlorobenzenesul fonate salt maintained a slightly lower solubility over one day compared to baclofen in water although the pH remained less than three throughout the study simila r to the p-phenolsulfonate sa lt. For pure baclofen the pH range sustained was 5.70 6.58 over twenty f our hours in water and it is known that a zwitterionic state is maintained at this pH range. In 0.1 N HCl the pH remained close to one for both salts and the order of least solubl e to most soluble remained the same as in water, which is unsurprising since an acidic pH was maintained. There was also a 20 fold increase in solubility with the p-phenolsul fonate salt, however, pure baclofen was much more soluble at this pH due to protonation resulting in disruption of the zwitterionic state bringing baclofen to its cationic form. The 4-chlorobenznensulfona te salt was one third lower in solubility when compared to baclofen. In pH 7 sodium phosphate buffer the order of highest to lowest solubility cha nged to the 4-chlorbenzenesulfonate salt, pphenolsulfonate salt, then baclofen. The magn itude of change was much less dramatic and only the 4-chlorobenzenesulfonate salt ma intained higher solubility than baclofen, which was still increasing after forty eight hours. The p-phenolsulf onate salt decreased towards pure baclofen making it unsuitable fo r further development. More experiments are necessary to determine if the 4-chlorobe nzenesulfonate salt is safe for intrathecal administration and stable over long periods of time without precipitation of baclofen monohydrate. 148

PAGE 166

Baclofen lactam polymorphs, Forms II and III, were discovered during crystal growth experiments with pure baclofen and cocr ystallization attempts with glutaric acid respectively. Both forms are conformationa l polymorphs while Form II has a high Zof four. It was found that only 0.39% of en tries in the CSD have a Z = 4. The conformational flexibility of b aclofen lactam enables Form II to have this high Z and an odd pairing of enantiomers including the pres ence of pseudosymmetry is exhibited. Form II was the result of dehy dration of baclofen upon crystalliz ation experiments targeted at growing single crystals of pure baclof en. Form III was the result of a failed cocrystallization experiment w ith glutaric acid and, similarly to Form II, the dehydration of baclofen to the lactam resulted from heating the solution. Supramolecular synthons for Forms II and III are consistent with Form I exhibiting the amide dimer. Multiple component pharmaceutical crys tal forms have been shown to be produced by multiple methods and these new crystalline entities can have tunable physicochemical properties that can lead to tunable pharmacokinetics. These conclusions have added tools to basic pharmaceutical sciences and shown that solid-state forms can be engineered. New multiple component cr ystalline forms can help remedy problems such as solubility, stability, and bioavailability and even extend the life cycle of currently marketed APIs with new patent protection. 149

PAGE 167

5.2 Future Directions Supramolecular chemistry has grown as a field impacting both academia and industry. It is now well accepte d that self assembly of or ganic and inorganic compounds can be rationally designed and cont rolled based upon the understanding of supramolecular interactions. Ma nipulation of intermolecular forces to obtain desirable characteristics has expanded its presence to include multiple facets of industrial application. Pharmaceutical sciences in particul ar has taken a deep interest in optimizing and developing novel crystalline forms of active pharmaceutical ingredients (APIs) without the use of covalent changes in order to retain desired pharmacological action while improving clinical performance. The future of this area will be more focus on understanding how the physicochemical proper ties of novel crystalline materials are impacted based of the parameters chosen for design, i.e. which salt or cocrystal former will lead to what change. In terms of pharmaceuticals, in vitro in vivo correlations need much more time and effort to understand how the bioavailability and pharmacological action will be affected. Utilizing synergistic multiple component pharmaceutical crystal forms could lead to lower doses and a s ubsequent reduction in adverse events. To understand these effects more animal studies across multiple species will be required to harness the full potential of multiple component pharmaceutical crystal forms including but not limited to salts and cocrystals. This manuscript has shown that SDG can be a useful technique for cocrystal screening with two BCS class II APIs. In the future this technique can be utilized much more frequently and even routinely for crystal form discovery. It is likely that the crystal forms discovered will also be amenable to later stage development and processing. This 150

PAGE 168

is important with respect to traditional large scale batches for the production of kilogram quantities. The fact that multiple solvents can produce the same cocrystal provides the chemical engineers more options in the even t that a particular solvent is considered dangerous or simply fails to be reproducible upon scale up. This method is attractive and could become routine due to its cost effectiv eness, eco-friendliness, and time efficiency. Two BCS class II compounds have been shown to be amenable to cocrystal formation through SDG while corresponding so lution based methods produced the same cocrystals. Beyond that meloxicam cocrystals have been shown to have tunable PK properties compared to the pure API. A carbamazepine cocrystal has also been shown by others to have improved PK performance th erefore BCS class II compounds in general could be future targets for co crystallization experiments wh ere the goal is tunable PK performance and potentially tunabl e clinical performance. Futu re work could be aimed at using more animal models with various AP Is in this class an d a broader choice of coformer to systematically asses this t echnology for BCS class II compounds as a group. For the zwitterionic API baclofen the use of strong acids has been used to disrupt a conformationaly stabilized zwitterionic state and change its pH dependent solubility. In particular it has been revealed that the pr oblem of limited solubili ty at physiological pH can be increased with this strategy. Future work should be done to see if this rationale can be used for other zwitterionic APIs with si milar limitations. Thereafter or concurrently the in vitro data presented here should be tested for correlations with in vivo testing. 151

PAGE 169

152 By learning from and complimenting natures template of non-covalent intermolecular interactions the field of cocr ystallization will become more and more important as the number of non-ionizable lipophilic drug candidates persists. Few cocrystal formulations are currently marketed as such and the development of scale up and processing will become the next challenge for this area. It is also possible that polypeptide based APIs may be stabilized to de gradation before they reach their targets by manipulating strong hydrogen bonding interactions. Other unstable formulations may also benefit from non-covalent modifications The application of novel drug delivery and tunable physicochemical properties could eventu ally reach a level of predictability in order to customize PK profiles for a particular need whether it be controlled release or targeted delivery.

PAGE 170

Appendices Appendix 1: Experimental Data for Carbamazepine Tue Jan 24 16:44:13 2006 drw Cbz 80 100%T 4,4'-dipyridyl 60 80 %T drw 1A 1/20/06 90 %T Fri Jan 20 15:44:57 2006 drw 1b 90 %T Fri Jan 20 15:58:16 2006 drw 1d 90 %T Fri Jan 20 1605:43 2006 drw 1e 95 100%T Fri Jan 20 1605:43 2006 drw 1e 95 100%T Fri Jan 20 16:12:32 2006 drw 1f 80 90 %T Fri Jan 20 16:20:04 2006 drw 1g 100%T Fri Jan 20 16:20:04 2006 drw 1g 100%T 1000 2000 3000 4000 Wavenumbers (cm-1)Cbz 4,4-Bipy MeOH EtOAc DMSO Water Tol Cychex CHCl3DMF Figure 1.1A: FT-IR data for SDG with Carbamazepine and 4,4-Bipyridine. 153

PAGE 171

05101520 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2530354045 2 tRelative IntensityCbz: 2: hetaCocrystal Calculated DMF CHCl3Cyclohexane Toluene Water DMSO EtOAc MeOHBipy 1 Figure 1.2A: PXRD data for SDG with Carbamazepine and 4,4-Bipyridine. 154

PAGE 172

Tue Jan 24 16:44:13 2006 drw Cbz 80 90 100%T 4-aminobenzoic acid 88 90 92 94 96 98 %T Tue Feb 14 10:43:56 2006drw 2b-C 75 80 85 90 95 %T 1000 2000 3000 4000 Wavenumbers (cm-1) Cbz 4-Ambz DMSO Figure 1.3A: FT-IR data for SDG with Ca rbamazepine and 4-Aminobenzoic Acid. 051015202530354045 -100 0 100 200 300 400 500 600 700 800 900 1000 Relative Intensity2 theta04-Ambz/Cbz/H2O DMSO Cbz 4-Ambz Figure 1.4A: PXRD data for SDG with Carbamazepine and 4-Aminobenzoic Acid. 155

PAGE 173

624.41 647.70 80124 1382.88 1676.69Tue Jan 24 16:44:13 2006 drw Cbz 80 90 100%T 749.91 1253.68 1683.72Thu Feb 02 10:43:18 2006 drw 3 (2,6pyrdicarb) 80 85 90 95 %T 61853 646.02 68095 751.62 768.97 1226.26 1653.62 1735.94Thu Feb 02 10:39:51 2006 drw 3h 60 70 80 90 %T 1000 2000 3000 4000 Wavenumbers (cm-1) 1374.59Cbz Pyrdicarb DMF Figure 1.5A: FT-IR data for SDG with Carbamazepine and 2,6Pyridinedicarboxylic Acid. 051015202530354045 0 200 400 600 800 1000 Relative intensity2 theta0Cbz-Pyrdicarb DMF Cbz Pyrdicarb Figure 1.6A: PXRD data for SDG with Carbamazepine and 2,6Pyridinedicarboxylic Acid. 156

PAGE 174

888.75 94191 1072.46 1083.05 1305.82 159031 1642.47Wed Feb 15 16:24:35 2006drw bnzqn 70 80 90 100%T 624.41 647.70 801.24 1382.88 1676.69Tue Jan 24 16:44:13 2006 drw Cbz 70 80 90 100%T 625.94 648.57 76927 803.67 885.09 1070.14 1389.51 1490.79 1590.74 1639.83 1672.80Wed Feb 15 16:52:54 2006drw 4h 20 40 60 80 100%T 1000 2000 3000 4000 Wavenumbers (cm-1)Bzqn Cbz DMF Figure 1.7A: FT-IR data for SDG with Carbamazepine and Benzoquinone. 051015202530354045 0 200 400 600 800 1000 1200 Relative intensity2 thetaoCbz-Bzqn DMF Cbz Bzqn Figure 1.8A: PXRD data for SDG with Carbamazepine and Benzoquinone. 157

PAGE 175

Cbz Tere MeOH EtOAc DMSO H2O CHCl3DMF Tue Jan 24 16:44:13 2006 drw Cbz 80 100%T Fri Feb 17 09:18:56 2006drw 5 terptalal 90 100%T Wed May 03 10:47:11 2006drw5aw 80 100%T Wed May 03 10:52:37 2006drw5bw 90 100%T Wed May 03 10:59:49 2006drw5cw 80 100%T Wed May 03 11:03:34 2006drw5dw 80 100%T Wed May 03 11:12:26 2006drw5gw 80 100%T Wed May 03 11:15:25 2006drw5hw 80 100%T 1000 2000 3000 Wavenumbers (cm-1) Figure 1.9A:FT-IR data for SDG with Ca rbamazepine and Terephthalaldehyde. 051015202530354045 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 Relative intensity2 thetaoCbz-Tere DMF CHCl3H2O DMSO EtOAc MeOH Cbz Tere Figure 1.10A: PXRD data for SDG with Carbamazepine and Terephthalaldehyde. 158

PAGE 176

Cbz Sacc MeOH EtOAc DMSO DMF 624.41 647.70 725.62 801.24 1382.88 1488.91 1676.69Tue Jan 24 16:44:13 2006 drw Cbz 60 80 100%T 621.09 630.19 64534 702.05 757.42 774.01 79456 898.52 1054.74 1118.95 113830 1161.06 1174.59 1256.70 1296.12 1332.82 1458.51 1592.51 171589Thu Mar 02 14:32:19 80 100%T 624.70 648.04 671.71 703.85 719.79 747.38 758.09 76928 787.67 807.44 902.04 1113.42 1135.20 1159.68 1177.45 1249.35 1308.56 1421.24 1461.89 1489.99 1574.07 1645.10 1728.68Thu Mar 02 14:57:37 80 100%T Thu Mar 02 14:57:37 80 100%T Thu Mar 02 14:57:37 80 100%T Thu Mar 02 14:57:37 80 100%T 1000 1500 2000 Wavenumbers (cm-1) Figure 1.11A: IR data for SDG wi th Carbamazepine and Saccharin. 051015202530354045 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000 3200 Relative Intensity2 thetaoCbz-Sacc DMF DMSO EtOAc MeOH Cbz Sacc Figure 1.12A: PXRD data for SDG with Carbamazepine and Saccharin. 159

PAGE 177

624.41 647.70 725.62 764.01 80124 1382.88 1488.91 1676.69Tue Jan 24 16:44:13 2006 drw Cbz 60 70 80 90 100%T 644.12 776.65 828.04 1028.22 1201.58 133952 1392.18 1421.67 1574.60 1592.38 161524 1673.17Fri Apr 07 09:21:19 2006drw7 nicotinamide 90 92 94 96 98 100 %T 613 26 624.51 640.42 705.54 719.46 765.12 804.51 1383.40 1412.99 1491.50 158533 1619.15 1658.11 1682.91Fri Apr 07 09:51:38 2006drw7h 70 80 90 100%T 1000 1500 2000 Wavenumbers (cm-1)Cbz Nic DMF Figure 1.13A: Ft-IR data for SDG with Carbamazepine and Nicotinamide. 051015202530354045 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 Relative Intensity2 thetaoCbz-Nic DMF Cbz Nic Figure 1.14A: PXRD data for SDG with Carbamazepine and Nicotinamide. 160

PAGE 178

051015202530354045 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 Relative Intensity2 theta Cocrystal Calculated DMSO Cbz:Asp 1:1 Figure 1.15A: IR data for SDG wi th Carbamazepine and Aspirin. Figure 1.16A: PXRD data for SDG with Carbamazepine and Aspirin. 616.93 627.97 651.07 667.66 699.81 720.69 751.21 769.06 776.14 804.41 924.12 960.67 101000 1023.84 1084.47 120454 1218.30 1254.53 1278.35 1301.72 1366.59 1420.20 147092 1490.10 1551.98 1576.09 160362 164329 1693.60 1748.13 2360.73DMSO 98 94 96 100%T 624.41 647.70 725.62 764.01 801.24 870.15 1382.88 1488.91 1594.48 1604.85 1676.69 3465.03Tue Jan 24 16:44:13 2006 drw Cbz 70 80 90 100%T 623.53 64306 666.08 703.75 753.55 789.58 803.05 83899 915.02 969.83 1012.00 1037.86 1093.30 113428 1181.43 1217.98 1303.09 1366.13 1418.17 1455.47 1604.30 1680.88 1749.84Mon May 08 10:17:52 2006drw asp 70 80 90 100%T 3000 2000 4000 Wavenumbers (cm-1)DMSO Cbz Asp 1000 161

PAGE 179

Appendix 2: Experimental Data for Meloxicam Figure 2.1A: Melting point correlatio n between meloxicam cocrystals and coformers. 510152025303540 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 3600 3800 Cocrystal Cocrystal Calculated Beta Fumaric Acid Alpha Fumaric Acid Meloxicam Form I Meloxicam Form III2 theta (deg)Relative Intensity (counts) Figure 2.2A: PXRD data for 1. 162

PAGE 180

Figure 2.3A: FT-IR data for 1. Figure 2.4A: DSC data for 1. 163

PAGE 181

510152025303540 0 500 1000 1500 2000 2500 Relative Intensity (counts)2 theta (deg)Cocrystal Cocrystal Calculated Succinic Acid Meloxicam Form I Meloxicam Form III Figure 2.5A: PXRD data for 2. Figure 2.6A: FT-IR data for 2. 164

PAGE 182

Figure 2.7A: DSC data for 2. Figure 2.8A: FT-IR data for 3. 165

PAGE 183

Figure 2.9A: [1H] NMR Spectrum For 1:1 Meloxicam: Maleic Acid Slurry in EtOAc. 166

PAGE 184

Figure 2.10A: [1H] NMR Spectrum For 1:2 Meloxicam: Maleic Acid Slurry in EtOAc. 167

PAGE 185

Figure 2.11A: [1H] NMR Spectrum For 2:1 Meloxicam: Maleic Acid Slurry in EtOAc. 168

PAGE 186

510152025303540 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2 theta (deg) Meloxicam:Malonic Acid MALNAC03 Malonic Acid Malonic Acid Meloxicam Form I Meloxicam Form IIIFigure 2.12A: PXRD data for 4. Figure 2.13A: FT-IR data for 4. 169

PAGE 187

Figure 2.14A: DSC data for 4. 510152025303540 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2 theta (deg) Meloxicam:Gentisic Acid BESKAL01Gentisic Acid Gentisic Acid Meloxicam Form I Meloxicam Form III Figure 2.15A: PXRD data for 5. 170

PAGE 188

Figure 2.16A: FT-IR data for 5. Figure 2.17A: DSC data for 5. 171

PAGE 189

510152025303540 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 2 theta (deg)Meloxicam:4-hydroxybenzoic acid 4-Hydroxybenzoic Acid Hydrate 4-Hydroxybenzoic Acid Meloxicam Form I Meloxicam Form III Figure 2.18A: PXRD data for 6. Figure 2.19A: FT-IR data for 6. 172

PAGE 190

Figure 2.20A: DSC data for 6. 46810121416182022242628303234363840 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 Meloxicam:Adipic Acid Adipic Acid Form 2 Calculated Adipic Acid Form 1 Calculated Meloxicam Form 1 Meloxicam Form 32 theta (deg)Relative IntensityFigure 2.21A: PXRD data for 7. 173

PAGE 191

4000036003200280024002000180016001400120010008006500 300 35 40 45 50 55 60 65 70 75 80 85 90 951 cm-1 %T 3288.98 2967.05 1930.39 1703.77 1620.20 1599.2 1549.45 1530.40 1448.39 373.51 1348.28 1325.41 1268.26 1215.99 1182.80 1153.97 13607 1119.70 1064.57 1044.49 992.49 963.90 937.76 89883 845.39 820.32 785.00 760.47 730.79 704.79 678.13 Figure 2.22A: FT-IR data for 7. Figure 2.23A: DSC data for 7. 174

PAGE 192

46810121416182022242628303234363840 0 500 1000 1500 2000 Relative Intensity2 theta (deg)Meloxicam:DL-Malic Acid DL-Malic Acid Calculated Meloxicam Form 1 Meloxicam Form 3 Figure 2.24A: PXRD data for 8. Figure 2.25A: FT-IR data for 8. 4000036003200280024002000180016001400120010008006500 200 25 30 35 40 45 50 55 60 65 70 75 80 85 90 929 cm-1 %T 3533.21 3288.06 3134.91 3028.66 2473.40 1872.66 1696.29 1631.79 1604.62 1550.82 1528.59 1450.50 137761 1343.18 1310.98 1264.49 1248.88 1211.94 1180.9 1152.90 1135.44 1120.36 106516 1044.98 1009. 937.66 881.09 843.76 808.93 777.33 757.19 730.18 701.93 678.16 659.95 175

PAGE 193

Figure 2.26A: DSC data for 8. 46810121416182022242628303234363840 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2 theta (deg)Relative Intensity (counts)Meloxicam:(+)Camphoric Acid (+)Camphoric Acid Meloxicam Form I Meloxicam Form III Figure 2.27A: PXRD data for 9. 176

PAGE 194

Figure 2.28A: FT-IR data for 9. Figure 2.29A: DSC data for 9. 177

PAGE 195

510152025303540 0 500 1000 1500 2000 2500 Meloxicam:Glycolic A c Glycolic Acid Meloxicam Form I Meloxicam Form III 2 theta (deg) Figure 2.30A: PXRD data for 10. Figure 2.31A: FT-IR data for 10. 178

PAGE 196

Figure 2.32A: DSC data for 10. 179

PAGE 197

Appendix 3: Experimental Data fo r Baclofen and Baclofen Lactam Figure 3.1A: TGA for (R,S) Baclofen Monohydrate. Figure 3.2A: DSC for (R,S) Baclofen Monohydrate. 180

PAGE 198

Figure 3.3A: TGA for (R,S) Baclofen:p-Phenolsulfonate. Figure 3.4A DSC for (R,S) Baclofen:p-Phenolsulfonate. 181

PAGE 199

Figure 3.5A: TGA for (R,S) Baclof en:4-Chlorobenzenesulfonate. Figure 3.6A: DSC for (R,S) Baclof en:4-Chlorobenzenesulfonate. 182

PAGE 200

4681012141618202 0 50 100 150 200 250 300 350 400 450 500 550 600 650 2242628303234363840 Relative Intensity (counts)2 thet a (deg)Post Dissolution pH 7 Post Dissolution Water Post Dissolution pH1 Baclofen HCl Calculated Baclofen Monohydrate Calculated Baclofen Figure 3.7A: PXRD data for baclofen dissolution. 46810121416182022242628303234363840 0 100 200 300 400 500 600 700 800 Post Dissolution pH7 Post Dissolution Water Post Dissolution pH1 Phenolsulfonate:Baclofen Calculated Baclofen HCl Calculated Baclofen Monohydrate Calculated Baclofen2 theta (deg)Relative Intensity (counts)Figure 3.8A: PXRD data for Baclofen:p-Phenolsulfonate dissolution. 183

PAGE 201

46810121416182022242628303234363840 0 100 200 300 400 500 600 700 800 900 1000 Post Dissolution pH 7 Post Dissolution Water Post Dissolution pH 1 4-Chlorobenzenesulfonate-Baclofen Calculated 4-chlorobenzenesulfonic acid Baclofen HCl Calculated Baclofen Monohydrate Calculated Baclofen2 theta (deg)Relative Intensity (counts)Figure 3.9A: PXRD data for Baclofen:4-Chl orobenzenesulfonate in pH 7 sodium phosphate buffer. Figure 3.10A: Histogram for amide-amide dimer, 1st contact. CSD version 5.31 Aug. 2010 update with search parameters; 3D coordinates, organics, and R 0.05. Contact between donor and acceptor defined as 2.7 3.3 184

PAGE 202

Figure 3.11A: Histogram for amide-amide dimer, 2nd contact. CSD version 5.31 Aug. 2010 update with search parameters ; 3D coordinates, organics, and R 0.05. Contact between donor and acceptor defined as 2.7 3.3 Figure 3.12A: Histogram for secondary am idesecondary amide dimer (includes lactams), 1st contact. CSD version 5.31 Aug. 2010 update with search parameters; 3D coordinates, organics, and R 0.05. Contact between donor and acceptor defined as 2.5 3.1 185

PAGE 203

Figure 3.13A: Histogram for secondary am idesecondary amide dimer (includes lactams), 2nd contact. CSD version 5.31 Aug. 2010 update with search parameters; 3D coordinates, organics, and R 0.05. Contact between donor and acceptor defined as 2.5 3.1 186

PAGE 204

187 About the Author David R. Weyna received his Bachelor of Science degree in chemistry with a concentration in biochemistry in 2001 from the University of North Carolina at Wilmington in Wilmington, North Carolina While at UNCW undergraduate research was supported by the Charles Cahill Award. This undergraduat e research, regarding the synthesis of a Succinyl Phosphate analog, wa s published in 2007 in a peer reviewed journal. Two years were then spent at the Medical University of South Carolina in the biomedical sciences doctoral program focusing on pharmaceutical sciences. He then transferred to the University of South Flor ida (USF) under the advisory of Dr. Michael Zaworotko to continue doctoral studies. Wh ile at USF pharmaceutical cocrystallization and other multiple component pharmaceutical cr ystal form discovery was pursued with emphasis on physicochemical property and pharmacokinetic enhancement. During 2006 and 2007 his research was presented at meeti ngs of the American Chemical Society. Also in 2007 he began work at Thar Pharmaceuticals Inc. concurrently while finishing the second half of the doctoral program at USF with particular focus on pharmaceutical cocrystal discovery and clinical improvement of active pharmaceutical ingredients. He is a co-inventor on thirte en patent applications and his work has been published in the following peer reviewed journals; Phosphorus Sulfur, Silicon, and Related Elements, Molecular Pharmaceutics, Crystal Grow th and Design, and the Journal of Pharmaceutical Sciences.


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 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2011 flu ob 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004855
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Weyna, David Rudy.
0 245
Crystal engineering of multiple component crystal forms of active pharmaceutical ingredients
h [electronic resource] /
by David Rudy Weyna.
260
[Tampa, Fla] :
b University of South Florida,
2011.
500
Title from PDF of title page.
Document formatted into pages; contains 204 pages.
Includes vita.
502
Disseration
(Ph.D.)--University of South Florida, 2011.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
520
ABSTRACT: Enhancing the physicochemical properties of solid-state materials through crystal engineering enables optimization of these materials without covalent modification. Cocrystals have become a reliable means to generate novel crystalline forms with multiple components and they exhibit different physicochemical properties compared to the individual components. This dissertation exemplifies methodologies to generate cocrystals of active pharmaceutical ingredients (API's) based upon knowledge of supramolecular interactions (supramolecular synthons), while focusing on enhanced delivery through in vitro and in vivo processes with both salts and cocrystals respectively. The utility of mechanochemistry involving small amounts of an appropriate solvent, or solvent drop grinding (SDG), has been shown to reliably reproduce cocrystals with the anti-convulsant carbamazepine that were originally obtained by solution crystallization. This technique has been confirmed as a reliable screening method using solvents in which both components exhibit some solubility. The benefits of this technique lie in the time and cost efficiency associated with it as well as its inherently small environmental impact making it a "Green" method. SDG was also used as an efficient way to discover cocrystals of the anti-inflammatory meloxicam with carboxylic acids after analysis of existing reports and the analysis of structural data from the Cambridge Structural Database (CSD) to guide the choice of coformer. It has been shown that SDG can be used to screen for cocrystalline forms that are also obtainable by solution crystallization which is important in later stage development and manufacturing including but not limited to large scale up processes. Single crystals suitable for single crystal X-ray diffraction were obtained with meloxicam and two of the coformers, fumaric and succinic acid. Some of the meloxicam cocrystals exhibited enhanced pharmacokinetic (PK) profiles in rats exemplifying significantly higher serum concentrations after only fifteen minutes and consistently higher exposure over the time studied while others maintained lower exposure. This reveals that cocrystals can fine tune the PK profile of meloxicam in order to reduce or enhance exposure. Two different sulfonate salts, 4-hydroxybenzenesulfonate (p-phenolsulfonate) and 4-chlorobenzenesulfonate, of the anti-spastic agent (R,S) baclofen were developed by strategically interrupting the intramolecularly stabilized zwitterionic structure of baclofen. This zwitterionic structure results in low solubility associated with physiological pH required for intrathecal administration. Structural data for both salts in the form of single crystal X-ray diffraction data was successfully obtained. Solubility based on baclofen was assessed and shown to increase in pure water and at pH's 1 and 7. Only the 4-chlorobenzenesulonate salt maintained an increased solubility over two days at pH 7 making it a viable candidate for further study in terms of intrathecal administration. During crystallization experiments with (R,S) baclofen two polymorphic forms of the baclofen lactam were generated, Forms II and III. Both forms are conformational polymorphs confirmed by single crystal X-ray diffraction and Form II has a Z' of 4 with an unusual arrangement of enantiomers.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor:
Zaworotko, Mike J.
653
Crystal Engineering
Pharmaceutical Cocrystal
Pharmacokinetics
Physicochemical Properties
Zwitterion
690
Dissertations, Academic
z USF
x Chemistry, Pharmaceutical
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.4855