Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Study of templation and molecular encapsulation using highly stable and guest-selective self-assembling… Chapman, Robert Glen 1997

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1997-195619.pdf [ 18.06MB ]
Metadata
JSON: 831-1.0059500.json
JSON-LD: 831-1.0059500-ld.json
RDF/XML (Pretty): 831-1.0059500-rdf.xml
RDF/JSON: 831-1.0059500-rdf.json
Turtle: 831-1.0059500-turtle.txt
N-Triples: 831-1.0059500-rdf-ntriples.txt
Original Record: 831-1.0059500-source.json
Full Text
831-1.0059500-fulltext.txt
Citation
831-1.0059500.ris

Full Text

Study of Templation and Molecular Encapsulation Using Highly Stable and Guest-Selective Self-Assembling Structures by ROBERT GLEN CHAPMAN B.Sc, The University Prince Edward Island, 1991 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March, 1997 ©Robert G. Chapman, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by,the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. ' Department of The University of British Columbia Vancouver, Canada Date DE-6 (2/88) Abstract The use of a rigid bowl-shaped molecule (tetrol 1) as the building block for the construction of covalently linked and noncovalently linked molecular capsules is presented in this thesis. First, as part of an exploration into the driving forces for formation carceplex 2«guest, the number of guest or template molecules that can be incorporated into carceplex 2»guest was greatly expanded. This reaction was found to be extremely selective towards the template molecule used; indeed, the relative templating abilities varied by six orders of magnitude. Second, the discovery of a reversible self-assembling structure composed of two molecules tetrol 1, which encapsulates a guest molecule in the resulting cavity, is presented (complex 3«guest). This complex was found to have great relevance to the formation of carceplex 2«guest as follows: Complex 3»guest provides organization of two molecules of tetrol 1 about a guest molecule such that the guest acts as a template for nonreversible covalent bond formation with bromochloromethane to give carceplex 2»guest. Complex 3«guest itself represents a new prototype for a family of switchable (by adjusting pH) self-assembling structures that involve molecular encapsulation and is highly guest selective. Third, complex 3»guest led to the discovery of related complexes that differ in the number and type of linkage that interconnects the two bowls of the complex. These complexes, together with complexes 3»guest and carceplexes 2«guest provide good model systems for the study of noncovalent interactions between molecules, as small changes in the template molecule leads to large changes in the stability of the complexes. Finally, carceplex 2»guest, complex 3»guest and some related compounds provided unique environments to study mobility and orientation of their bound guest molecule. Some potential applications of self-assembling structures based on tetrol 1 are presented. ii Complex 3'guest R = CH 2CH 2Ph Solvent or Solute = Guest Carceplex 2»guest iii Table of Contents Abstract ii Table of Contents iv List of Schemes xi List of Figures xii List of Tables xv List of Abbreviations xvii Acknowledgments xix 1. THE IMPORTANCE OF TEMPLATION IN SUPRAMOLECULAR CHEMISTRY 1 A. Introduction 1 i. Supramolecular Chemistry 2 ii. Molecular Recognition, Self-Assembly and Templation 3 iii. Forefathers of Supramolecular Chemistry 7 B. Templation 8 i. Introduction 8 ii. Classification of Templates 9 iii. Catenanes and Rotaxanes 10 iv. Template Formation of Catenanes 11 v. Template Effects in the Formation of 2-Catenanes 14 vi. Template Effects in the Formation of a Rotaxane 18 vii. Self-Replicating Systems 20 viii. Template Effects in a Self-Replicating System 21 ix. "Positive" and "Negative" Templates 24 x. Template Effects in the Formation of Cyclic Porphyrins 25 xi. Anions as Templates 27 iv C. Templates Used to Make Materials 28 i. Molecularly Imprinted Polymers 28 ii. Templates in Crystal Engineering 29 iii. Template Molecules Used to Create Zeolites 30 iv. Template Formation of Tubules 31 D. Molecular Encapsulation : 31 i. Self-Assembly of Cavities Capable of Molecular Encapsulation 31 ii. Rebek's Tennis B all 33 iii. Urea Based Calixarene Dimer 34 iv. Cyclocholates 36 v. Shinkai's Heterodimer 37 vi. Encapsulation in Cucurbituril 38 vii. Molecular Encapsulation in Cyclodextrins 39 viii. CD Dirnerization 41 ix. CD's and C6o 4 2 x. Encapsulation of 6-Methyl Glucopyranoside 43 E. Carceplexes 44 i. Introduction 44 ii. Synthesis of the First Molecular Container Compound 46 iii. The Synthesis of the First Fully Characterized Carceplex 47 iv. Properties of Carceplex 2a»Guest 50 v. Other Carceplexes 52 vi. Endohedral Complexes of Fullerenes 54 vii. Related Compounds 55 viii. One Pot Synthesis of Hemicarceplexes and a Carceplex 55 F. Conclusions 57 G. References 58 2. TEMPLATE EFFECTS IN FORMATION OF A CARCEPLEX 66 A. Introduction 66 v B. Results and Discussion 68 i. Selectivity in the Formation of Carceplex 2a»guest 68 a. Screening for Suitable Template Molecules 68 b. Competition Experiments Between Successful Template Molecules 70 c. Template Trends for Formation of Carceplex 2a»Guest 74 d. Template Molecules That Led to Mixtures 78 ii. The Guest Determining Step 80 a. Reaction Intermediates in the Formation of Carceplex 2a»Guest... 80 b. Disappearance of Tetrol 81 c. The Importance of Base in the Formation of Carceplexes 82 d. Effect of Solvent on the Formation of Carceplexes 83 e. A,C-bis-bridged 104»DMSO and Tris-bridged 105»DMSO 86 f. A,B-bis-bridged 103 and Monobridged 102 Intermediates 87 g. The Guest Determining Step for Formation of Carceplex 2a»Guest 88 iii. Templation Requirements in Formation of a Hemicarceplex 89 iv. Crystal Structure of Carceplex 2»Guest 90 a. Crystal Structure of Carceplex 2b»Pyrazine 90 b. Crystal Structure of Carceplex 2a»DMA 92 C. Conclusion 93 D. Experimental 95 i. General Experimental 95 ii. Carceplexes 2a»Guest 96 iii. Carceplex 2a»Guest Mixtures 116 iv. Carceplexes 2b»Guest 118 v. Carceplex Intermediates 121 vi. Competition Experiments 125 E. References 132 3. SELF-ASSEMBLING CAVITANDS 133 A. Self-Assembly of Carceplex 2*Guests 133 B. Results and Discussion 134 vi i. Investigations into the Formation of Complexes of Tetrol 1 134 ii. Determination of Complex la*Guest Using Deuterium NMR 136 iii. *H NMR Characterization of Complex 3b»Pyrazine 138 iv. Requirements of Base in the Formation of Complex 3b»Pyrazine 142 v. Formation of Complex 3b#CHCl3 143 vi. Charged Hydrogen Bonds in Complex 3b»Pyrazine 149 vii. Is Complex 3b»guest pH-Switchable? 151 viii. Stability of Complex 3b»Pyrazine in DMSO 152 ix. Electrospray Mass Spectrometry of Complex 3b»Guest 153 x. Orientation of Pyrazine in Complex 3b»Pyrazine 158 xi. Crystal Structure of Complex 3b»Pyrazine 160 xii. Selectivity of Complex 3b»Guest Towards Successful Templates in the Formation of Carceplex 2a*Guest 167 a. Introduction 167 b. Thermodynamic Parameters for Complex 3b«Guest in CDCI3 170 xiii. Selective Guest Binding of Other Related Complexes in CDCI3 174 a. Introduction 174 b. Complexation Properties of Tetraprotio 108 175 c. Complexation Properties of Monol 109 176 d. Complexation Properties of A,B-diol 110a and A,C-diol 111 178 e. Complexation Properties of Triol 112 and Monobridged 102 181 xiv. Preorganized Hosts that Exhibit Binding Under Neutral Conditions 186 a. Complexation Properties of A,B-bis-bridged 103 186 b. Synthesis and Binding Properties of A,B-bis-bridged Tetraprotio 123a 188 c. Orel's for AB-bis-bridged 103»Guest and Tetraprotio 123a#Guest 192 d. Kap for AB-bis-bridged 103»Guest 193 e. Other Hinged Compounds 194 xv. Nitrobenzene-^ as a Solvent for Complexation 196 xvi. Relative Stabilities of Complexes»Methyl Acetate 201 a. Introduction 201 b. Determination of the Relative Stability Constants of Complexes»Methyl Acetate 202 c. Thermodynamic Parameters for Complexes»Methyl Acetate 206 xvii. Absolute Stabilities of Complex tetra-OMe 125»Guest 210 a. The Nature of the Free Species 210 vii b. Determination of an Absolute Stability Constant Using Tetra-OMe 125 212 c. Interesting Features of the Guest Binding in tetra-OMe 125«guest 216 d. Calculation of the Absolute Stability Constants for Other Complexes*Methyl Acetate 218 xviii. Energetics for Formation of Carceplex 2a»guest 219 xix. Computations 221 C. Summary 223 D. Experimental 226 i. General 226 ii. Synthesized Compounds 227 iii. Crystal Structure of Complex 3b*pyrazine 236 iv. General procedure for determining the 11/2 for decomplexation of pyrazine from complexes 237 v. Complexation Experiments 238 a. General procedure for determination of Orel's for complex 3b»guest in CDC13: 238 b. Determination of Orel's for A,C-diol-complex 116»guest and A,B-diol-complex 115»guest 241 c. Determination of Orel's for triol-complex 117a»guest and monobridged complex 118»guest 245 d. Determination of A'rei's for A,B-bis-bridged complex 103»guest and tetraprotio 123a»guest 251 vi. Complexation Experiments in Nitrobenzene-^ 5 256 a. General procedure for determination of Orel's for complex 3b*guest in Nitrobenzene-^: 256 b. Determination of Orel's for tetra-OMe 125»guest in Nitrobenzene-J5 at 333 K 259 vii. Determination of the relative stability constants for complexes'methyl acetate 264 E. References 267 viii 4. MOBILITY OF HOST AND GUEST IN CARCEPLEX 2-GUEST AND RELATED COMPOUNDS 270 A. Orientation of Guest Molecules in Carceplex 2»Guests and Related Compounds 270 i. Orientation of Guest Molecules in Carceplex 2»Guests 270 ii. Sensitivity of the X H NMR 5 of the Host Signals of Carceplex 2a»Guests to the Nature of the Incarcerated Guest Molecule 273 iii. Sensitivity of the  lU NMR 5 of Host and Guest Signals of Carceplex 2»Guest to a Change of Pendent Group 275 iv. Orientation of Guest Molecules in Other Two-Bowl Complexes»Guest.. 277 B. Variable Temperature *H NMR Spectroscopy of Two-Bowl Complexes»Guest 280 i. Introduction 280 ii. Variable Temperature  lH NMR Experiments with Tetraprotio 123a»Pyrazine 282 iii. Chiral Carceplex 286 iv. Twistomers with DMSO as Guest 290 v. Dynamic  lH NMR of Carceplex 2a« 1,4-Thioxane 296 vi. Dynamic *H NMR of Carceplex 2b»l,4-Dioxane and Complex 3b»l,4-Dioxane 299 vii. Summary 302 C. Experimental 303 i. Synthesis of Carceplex 2»Guest 303 ii. Supplementary Tables 306 D. References 313 5. CONCLUSIONS AND FUTURE APPLICATIONS 314 A. Carceplex 2a»Guest 314 B. Larger Self-Assembling Structures 316 i. Introduction 316 ii. Polymeric Side-to-Side Assemblies 317 iii. Polymeric Tail-to-Tail Assemblies 318 ix iv. Other Assemblies 319 v. Monolayers 320 vi. Incarceration of Reactive Intermediates in Carceplex 2b»Guest 321 C. References 323 x List of Schemes Scheme 1.1. Metal Binding Hosts 8 Scheme 1.2. Metal Templated Catenane Synthesis 11 Scheme 1.3. 2-Catenane Synthesis Using Organic Templates 12 Scheme 1.4. Catenane Synthesis 15 Scheme 1.5. Thermodynamic Analysis of a Catenane System 17 Scheme 1.6. Amide Rotaxane Synthesis 19 Scheme 1.7. Rebek's Adenine Reaction 22 Scheme 1.8. "Positive" and "Negative" Templates 25 Scheme 1.9. Formation of Cyclic Dimers and Trimers of Porphyrins 26 Scheme 1.10. Formation of a Calix[4]arene Heterodimer 37 Scheme 1.11. Cucurbituril Complex 39 Scheme 1.12. Synthesis of a Cavitand; Yields are for R = CH2CH2Ph 46 Scheme 1.13. First Synthesis of a Carceplex 47 Scheme 1.14. Synthesis of Carceplex 2»Guest 49 Scheme 1.15. Synthesis of a Thioacetal Bridged Carceplex 52 Scheme 1.16. Synthesis of Calix[4]arene-Based Carceplex 53 Scheme 2.1. First Soluble Carceplexes 66 Scheme 2.2. The Competition Reaction 72 Scheme 2.3. The Formation of a Hemicarceplex 90 Scheme 3.1. Hydrogen Bonding of Phenols in Dipolar Nonhydroxylic Solvents 135 Scheme 3.2. Encapsulation of Dioxane-rfg 138 Scheme 3.3. Complex 3b«guest 139 Scheme 3.4. Synthesis and Complexation of Tetraprotio 123a 189 xi List of Figures Figure 1.1. The Reaction to Form Carceplex 2a«Guest 2 Figure 1.2. Tobacco Mosaic Virus 6 Figure 1.3. Metal Templation 9 Figure 1.4. Catenane and Rotaxane 10 Figure 1.5. Molecular Mosaic Pattern of a Self-Complementary Cyclophane 13 Figure 1.6. Schematic Representation of Self-Complementary Template-Based Autocatalysis 20 Figure 1.7. Tecton Assembly 30 Figure 1.8. Rebek's Tennis Ball 34 Figure 1.9. Rebek's Urea Based Capsule 35 Figure 1.10. Formation of Cyclocholates 36 Figure 1.11. Cyclodextrin Illustration 40 Figure 1.12. CD Metallic Dimer 41 Figure 1.13. y-CD Dimer Encapsulating C60 42 Figure 1.14. Encapsulation of 6-Methyl Glucopyranoside 44 Figure 1.15. Molecular Prison 45 Figure 1.16. Gate functionalized Hemicarceplex 56 Figure 2.1. Template Requirements for Carceplex 2a«Guest Formation 67 Figure 2.2. Graph of Solvent Parameters versus ln(Template Ratios) 78 Figure 2.3. Schematic of Carceplex Intermediates 86 Figure 2.4. Stereo View of the Crystal Structure of Carceplex 2b«Pyrazine 91 Figure 2.5. Cutaway CPK Model of the Crystal Structure of Carceplex 2b«Pyrazine 91 Figure 2.6. Inter-bowl Planes of Carceplex 2a»DMA and Carceplex 2a«Pyrazine 93 Figure 3.1. ! H NMR of Complex 3b»Pyrazine 141 Figure 3.2. Titration of Tetrol l b and Pyrazine with DBU 142 Figure 3.3. Titration of Tetrol with DBU 144 Figure 3.4. Formation of Complex 3b«CHCl3 in Nitrobenzene-^  146 Figure 3.5. Solvent Suppression of Tetrol l b complex in 50/50 v/v CHCI3/CDCI3 148 Figure 3.6. lH NMR of [complex 3b»pyrazine][TBA]4 in Acetone-^ at 200 K 151 Figure 3.7. ESMS of [Complex 3b»Pyrazine] [TB A]4 156 Figure 3.8. Determination of Charge using ESMS 157 Figure 3.9. Orientation of Pyrazine 159 Figure 3.10. Asymmetric Complex 3»Pyrazine 160 Figure 3.11. Crystal Structure of Complex 3b«pyrazine 162 Figure 3.12. Crystal Structures of Carceplex 2b«Pyrazine and Complex 3b»Pyrazine 164 Figure 3.13. View of the Planes Connecting the Aryl Ethers of Carceplex 2b»Pyrazine and the Phenols/Phenoxides of Complex 3b»Pyrazine 165 Figure 3.14. Kre\ for complex 3b»pyrazine versus complex 3b»dioxane at 298 K 169 Figure 3.15. Graph of ln(KK\) versus ln(TR's) 170 Figure 3.16. Graph of Rln(/C"rel Complex 3b«guest) versus 1/T (K) 172 Figure 3.17. Cavitands 175 Figure 3.18 , Tetraprotio-complex 113»guest 176 xii Figure 3.19. Complexes of Cavitands. 178 Figure 3.20. Schematic Representation of the Dipole Interactions of A,B-diol-complex 115»pyridine, Complex 3b»pyridine, and A,C-diol-complex 116*pyridine 180 Figure 3.21. Formation of Mono-bridged complex 118«guest 181 Figure 3.22. Asymmetric triol-complex 117b«pyrazine 183 Figure 3.23. ln(Afrei of triol-complex 117a«guest and monobridged complex 118»guest) versus ln(Xre] of complex 3b»guest) 185 Figure 3.24. Complexation in the Presence and Absence of Base 187 Figure 3.25. NMR Characterization of tetraprotio 123a 191 Figure 3.26. Graphs of ln^el tetraprotio 123a»guest) and \n(Kre\ A,B-bis-bridged complex 103»guest) versus ln(Afrei Complex 3b»guest) 193 Figure 3.27. Monobridged 102»Guest Neutral Binding 195 Figure 3.28. Graph of ln(/Trei of Complex 3b*guest in CDCI3) versus ln(Kre\ of Complex 3b»guest in Nitrobenzene-Js) at 298 K 198 Figure 3.29. Graph of \n(Kre\) Versus 1/T (K) in Nitrobenzene-^  199 Figure 3.30. ] H NMR Spectra of the Relative Stability of Complexes«methyl acetate in Nitrobenzene-^  203 Figure 3.31. Schematic Illustration of the two Orientations of Methyl Acetate in Asymmetric Complex (Tetrol lb)(Triol 112)»Methyl Acetate 204 Figure 3.32. Synthesis of tetra-OMe 125 205 Figure 3.33. Graph of Rln(Complex 1'methyl acetate / Complex 2«methyl acetate) versus 1/T (K) in Nitrobenzene-^  207 Figure 3.34. The Tetrol and DBU in Nitrobenzene-^ 5 at Ambient Temperature 212 Figure 3.35. The effect of O2 and N 2 gases on the ! H NMR Spectra of tetra-OMe 125 in Nitrobenzene-^  215 Figure 3.36. Correlation Between log(Template Ratios For Carceplex 2a»Guest) and log(Theoretical Calculated Energies of Complex 3b«Guest) 222 Figure 3.37. Plot of ln[complex 3b«pyrazine] versus time at 273 K, r^  = 0.99 237 Figure 3.38. A,C-diol-complex 116»guest and A,B-diol-complex 115«guest 242 Figure 4.1. Schematic Representation of the Orientation of DMA in Carceplex 2a«DMA and Pyrazine in Carceplex 2b«Pyrazine as Determined by X-ray Crystallography 271 Figure 4.2. Carceplex 2«Guest 275 Figure 4.3. Schematic Representation of the Lower-Bowl of tetraprotio 123a»guest and carceplex 2»guest 278 Figure 4.4. Schematic Representation of the Lower-Bowl of tetraprotio 123a»guest Illustrating the Favored Orientation of Pyrazine 280 Figure 4.5. Variable Temperature NMR Spectra of tetraprotio 123a»pyrazine in CDCI3 284 Figure 4.6. Tetraprotio 123a»pyrazine 285 Figure 4.7. Variable Temperature lH NMR Spectra of Carceplex 2b«(fl)-(-)-2-butanol in CDCI3. . . . 288 Figure 4.8. Schematic Representation of the Diastereomeric Twistomers of Carceplex 2b«(/?)-(-)-2-butanol ' 289 Figure 4.9. Schematic Representation of Host Systems Investigated by Cram et al 291 Figure 4.10. Schematic Representation of the Two Diastereomers of Carceplex 2d»DMSO 293 Figure 4.11. Variable Temperature lH NMR of Carceplex 2d«DMSO in CDCI3 295 Figure 4.12. Variable Temperature J H NMR Spectra of Carceplex 2a«l,4-thioxane in CDCI3 298 xiii Figure 4.13. Interconversion of the Chair Conformations of 1,4-Thioxane 299 Figure 4.14. Variable Temperature 'H NMR Spectra of Carceplex 2b*l,4-dioxaxe and Complex 3b«l,4-dioxaxe in CDCI3 301 Figure 5.1. Side-to-Side Assemblies 318 Figure 5.2. Polymeric Rods Based on Tail-to-Tail 132 319 Figure 5.3. Five-bowl Assembly 133 320 Figure 5.4. Formation of Self-assembled Monolayers of Tetrol 1 320 Figure 5.5. An Alternative Route to Carceplex 2a*Guest 322 xiv List of Tables Table 1.1. Effects of Spacer on Rebek's Template 23 Table 1.2. Template Effects on Cyclic Porphyrin Oligomers 27 Table 2.1. Successful Template Molecules 69 Table 2.2. Unsuccessful Guests 70 Table 2.3. Carceplex 2a«Guest Yields and Competition Experiments 73 Table 2.4. Disappearance of Tetrol at 25 'C 81 Table 2.5. Effect of Base on Formation of Carceplex 2a»guest 83 Table 2.6. Solvents for the Carceplex Reaction Using DBU as the Base 84 Table 2.7. Template Ratios For Carceplex Intermediates 88 Table 2.8. List of Competition Results 128 Table 2.9. Additional guests added to Competition Experiments 129 Table 2.10. Cross Check Experiments 131 Table 3.1. Observed ionic species from the ESMS of complex [3b»pyrazine][TBA]4 155 Table 3.2. Comparison of KXQ\ versus Template Ratios 168 Table 3.3. Thermodynamic data for Complex 3b»Guest in CDCI3 172 Table 3.4. Orel's of A,B-diol-complex 115«guest and A.C-diol-complex 116"guest at 253 K. 180 Table 3.5. Orel's of Triol-complex 117a«guest and Monobridged 118«guest at 273 K 185 Table 3.6. #rel's o f tetraprotio 123a«guest and A,B-bis-bridged complex 103*guest at 298 K 192 Table 3.7. Kap of AB-bis-bridged 103'guest at 298 K 194 Table 3.8. Orel's of Complex 3b»guest in Nitrobenzene-^  and CDCI3 at 298 K 198 Table 3.9. Thermodynamic data for Complex 3b»Guest in Nitrobenzene-d/5 and CDCI3 199 Table 3.10. Orel's of Various Complexes»Methyl Acetate in Nitrobenzene-^ 5 205 Table 3.11. Thermodynamic Data for KTQ\'S of Various ComplexesrMethyl Acetate in Nitrobenzene-^  207 Table 3.12. KT&\ and Ks for tetra-OMe 125'guest at 333 K in Nitrobenzene-^ 5 214 Table 3.13. ^ap's of Various Complexes»Methyl Acetate in Nitrobenzene-^  219 Table 3.14. Template Ratios and KKi's for A,B-bis-bridged 103 and Tetrol lb Determined at 298 K 220 Table 3.15. Quantities of Host and Guest Added to lH NMR Samples at 298 K in CDCI3 239 Table 3.16. Integration results for Arrei's for Complexes 3b»guest at 298 K in CDCI3 240 Table 3.17. *H NMR assignments for complexes 3b»guest in CDCI3 241 Table 3.18. Kre\'s A,C-diol-complexes 116«guest (3.44 mM) at 253 K in CDCI3 242 Table 3.19. ] H NMR Assignments for A.C-diol-complexes 116'guest at 253K in CDCI3 243 Table 3.20. KTe]'s for A,B-diol-complexes 115»guest (3.44 mM) at 253 K in CDCI3 244 Table 3.21. lH NMR Assignments for A,B-diol-complexes 115«guest at 253 K in CDCI3 244 Table 3.22. Integration results for ATrei's of Triol-complexes 117a»guest at 273 K in CDCI3 246 Table 3.23. ! H NMR Assignments for Triol-complexes 117a»guest at 273 K in CDCI3 247 Table 3.24. Integration results for Orel's of Monobridged complexes 118'guest at 273 K in CDCI3 : 249 XV Table 3.25. 'H NMR Assignments of Monobridged complexes 118«guest at 273 K in CDCI3 250 Table 3.26. Integration results for KTe\'s A,B-bis-bridged complex 103»guest at 298 K in CDCI3 251 Table 3.27. lU NMR Assignments for A,B-bis-bridged complexes 103»guest at 298 K in CDCI3 252 Table 3.28. Integration results for KK\'s for tetraprotio 123a«guest at 298 K in CDCI3 254 Table 3.29. ! H NMR Assignments of tetraprotio 123a«guest at 298 K in CDCI3 255 Table 3.30. Integration results for Abel's f° r Complexes 3b»guest at 298 K in nitrobenzene-^  257 Table 3.31. *H NMR assignments for complex 3b»guest in nitrobenzene-^  at 298 K 258 Table 3.32. Integration results for ZTrei's for tetra-OMe 125»guest at 333 K in Nitrobenzene-^  260 Table 3.33. ! H NMR Assignments for tetra-OMe 125»guest at 298 K in Nitrobenzene-^  262 Table 3.34. Integration results for ATrei's for complex 1'methyl acetate/ complex 2»methyl acetate in Nitrobenzene-^  at 296 K 265 Table 3.35 'H NMR Chemical Shifts for Methyl Acetate Encapsulated in Various Complexes in Nitrobenzene-^  at Ambient Temperature 266 Table 4.1. 8 (ppm) of Guest Molecules in Carceplexes 2a»Guest at Ambient Temperature in CDCI3 272 Table 4.2. lH NMR A8 (ppm) of Host Signals for Carceplexes 2a«guest in CDCI3 at Ambient Temperature 274 Table 4.3. Comparison of the 8 (ppm) in the ^H NMR Spectra of Carceplexes 2a»guest versus Carceplexes 2b»guest in CDCI3 at ambient temperate 276 Table 4.4. *H NMR Assignments for tetraprotio 123a»guest at 298 K in CDCI3 278 Table 4.5. Activation Barrier for the Interconversion of Twistomers of tetraprotio 123a*pyrazine in CDCI3 286 Table 4.6. Activation Barrier for the Interconversion of Twistomers of Carceplex 2b«(/?)-(-)-2-butanol in CDCI3 289 Table 4.7. Free energies of Activation for Rotation of DMSO about the C2 axes of the Host 290 Table 4.8. 8 (ppm) of Host Signals in the lH NMR Spectra of Carceplex 2a»Guest in CDCI3 at Ambient Temperature : 306 Table 4.9. 8 (ppm) of Guest Signals in the *H NMR Spectra of Carceplex 2a»guest in CDCI3 at Ambient Temperature 308 xvi List of Abbreviations Anal. - microanalysis aq. - aqueous Bn - benzyl [C6H5CH2-] c - concentration calcd - calculated CHB - charged hydrogen bond COSY - correlation spectroscopy AS - change in chemical shift AG° - Gibbs free energy AH0 - enthalpy of reaction AS° - entropy of reaction 8 - chemical shift DBU - l,8-diazabicyclo[5.4.0]undec-7-ene DCI-MS - desorption chemical ionization mass spectrometry or spectrum dioxane - 1,4-dioxane DME - 1,2-dimethoxyethane DMF - A^N-dimethylformamide DMSO - dimethyl sulfoxide equiv. - equivalents GDS - guest determining step IR - infrared (spectroscopy) J - coupling constant Kap . apparent stability constant ^rel . relative stability constant Ks . absolute stability constant xvii LSIMS liquid secondary ion mass spectrometry lit. literature M parent mass (mass spectra) or molar, moles per liter (concentration) m/z mass-to-charge ratio Me methyl [CH3-] mp melting point NBS N-bromosuccinimide NMP 1 -methylpyrrolidin-2-one NMR nuclear magnetic resonance (spectroscopy) NOE nuclear Overhauser effect P para pet. ether petroleum ether Ph phenyl [-C6H5] ppm parts per million r correlation factor r.t. room temperature % retention factor or retardation factor SAS self-assembling structure t\l2 half-life Tl longitudinal relaxation time T c coalescence temperature THF tetrahydrofuran xviii Acknowledgments First and foremost, I wish to thank my Ph.D. research supervisor, Professor John Sherman, for his enthusiastic support and encouragement throughout my stay at UBC. I would like to thank all the past and present members of the Sherman group for providing a stimulating and enjoyable work environment. Special thanks to Bruce, Janet, Adam, Ashley, Ayub and Naveen for their helpful discussions regarding my research projects. I am indebted to Dr. Martin Tanner, Ayub Jasat, Adam Mezo and Ashley Causton for proof-reading this thesis. I would also like to thank the Natural Sciences and Engineering Research Council of Canada, and the University of British Columbia (Graduate Fellowship) for financial support. The assistance of the staff of the NMR Laboratory, Mass Spectrometry Laboratory, Microanalysis Laboratory (Mr. Peter Borda), Glass Shop (Mr. Steve Rak) is greatly appreciated. Finally, I would like to thank my parents for their support and my wife, Heather, for her endless encouragement and help throughout the years. xix 1 The Importance of Templation in Supramolecular Chemistry A . Introduction Molecular recognition, self-assembly and templation are integral to many chemical and biological processes. These phenomena play pivotal roles in the formation of compounds called carceplexes, which are the subject of this thesis. Carceplexes are closed surface container compounds that incarcerate a guest molecule within their interior. The guest can not escape the confines of the carceplex unless one or more covalent bonds are broken. Chapter 1 provides further definitions and reviews some of the pertinent literature. Emphasis is placed on the template formation of supramolecular structures as well as the self-assembly of molecules to form cavities capable of molecular recognition. Chapter 2 introduces the importance of noncovalent interactions in the template formation of a carceplex. Chapter 3 expands upon these template effects by illustrating the importance of self-assembly in carceplex formation; this will include a discussion of a new series of reversible complexes that involve molecular encapsulation. Chapter 4 explores the dynamic behavior of carceplexes and some related complexes. Finally, Chapter 5 explores the implications of this thesis. In 1989 Cram and Sherman published a paper entitled "Carcerand Interiors Provide a New Phase of Matter".1 (Carcerands are the shell or exterior of carceplexes, i.e., no guest.) They found that the shell closure of two molecules of tetrol l a with four molecules of bromochloromethane produced carceplex 2a«guests, whereby a molecule of solvent became permanently entrapped within the interior of the host forming a molecular container compound called a carceplex (Figure 1.1).1-3 We became interested in this reaction because it brought together seven molecules via the formation of eight new carbon-oxygen bonds in yields greater than 60%. Also, it was apparent that the reaction to form carceplex 1 2a»guest required a template molecule because all carceplexes that were isolated contained a molecule of entrapped solvent, and no carceplexes were formed when the solvent was to large to serve as a guest. This thesis investigates the driving forces for the formation of carceplex 2a»guest in order to determine the importance of the template molecule in the formation of carceplex 2a»guest and to determine why a reaction that creates eight new bonds proceeds in greater than 60% yield. Figure 1.1. The Reaction to Form Carceplex 2a»Guest. 2 Tetrol la Complex 3a«guest Carceplex 2a*guest R = CH 2CH 2Ph Solvent or Solute = Guest i. Supramolecular Chemistry Advancements in synthetic chemistry over the past century have demonstrated an ever increasing mastery over the formation of covalent bonds to build molecules.4,5 The techniques developed by synthetic chemists to construct covalent bonds have culminated in the total synthesis of vitamin B126 and palytoxin (molecular weight = 2680).7 In the 1970's, the traditional boundaries of synthetic chemistry were crossed as chemists began to extend their efforts from connectivity using the covalent bond, to organization of molecules using noncovalent interactions to form complexes or "supermolecules". The use of noncovalent forces to join molecules together to form supermolecules created a new division of science called "supramolecular chemistry". 2 Supramolecular chemistry is a term coined by Jean-Marie Lehn to describe the study of supermolecules.8 Lehn formally defines supramolecular chemistry as "chemistry beyond the molecule, referring to the organized entities of higher complexity which result form the association of two or more chemical species held together by intermolecular forces".8-10 Supermolecules or Umbermolekiile11 is a term used to describe these entities of higher organization resulting from the association of multiple chemical species via noncovalent interactions.10 The structure and characteristic properties of the supermolecule are distinct from the properties of the chemical species or subunits of which it is composed.12 Therefore, the development of supramolecular chemistry holds promise for the discovery of new and exciting supermolecules with interesting properties. ii. Molecular Recognition, Self-Assembly and Templation The construction of a supermolecule involves two important processes: molecular recognition and self-assembly. Both of these processes are fundamental to life itself, because in nature they are integral to the function biological assemblies such as enzymes, antibodies, and membranes. Molecular recognition is the process by which some molecules select and bind other molecules in a structurally well defined pattern of intermolecular forces.9 For example, a substrate is selectively recognized by an enzyme and is bound in a specific orientation in the enzyme's active site. Likewise, cytosine recognizes guanine in duplex DNA. The functionality of one molecule complements the functionality of the other and the two molecules associate or bind with one another by sharing their noncovalent information. Often, one of the molecules has a convergent binding site (the host or receptor) and the other a divergent binding site (the guest or substrate) but by no means is molecular recognition limited to such host-guest or receptor-substrate systems. 3 The concept of molecular recognition was described as early as 1894 by Emil Fischer in his lock-and-key theory.13 According to Fischer's original idea, molecular recognition is similar to the complementarity of a lock and a key. The lock is the molecular receptor and the key is the substrate being recognized to form a receptor-substrate complex. Although this idea simplifies molecular recognition, it vividly emphasizes the complementarity necessary between the two chemical species involved in the recognition process. Over 100 years later, research in molecular recognition remains at the forefront of scientific thinking. As an example of its importance, advances in molecular recognition may one day enable scientists to predict the three-dimensional structure of a protein from its primary sequence (the protein folding problem). These predictions would aid in the study of natural enzymes and also allow scientists to develop new enzymes with unique catalytic abilities. The second process that aids in the construction of the supermolecule is self-assembly. The term self-assembly occurs frequently in the literature, and it has been the subject of numerous reviews.14-16 Although many definitions exist for self-assembly,17 the definition by Whitesides is the most appropriate for this thesis. Whitesides defines self-assembly as "the spontaneous assembly of molecules into structured, stable, noncovalently joined aggregates."14 The structural integrity of the self-assembled aggregate is maintained after its formation because it represents the thermodynamically most stable structure. There are four properties of self-assembling structures that are important: (1) The properties of the self-assembled aggregate are unique from the properties of the subunits from which it is composed. (2) The reversibility of the self-assembly causes improperly formed assemblies or mismatched subunits to be eliminated from the final structure (a type of error checking). (3) All the information necessary for forming a self-assembled structure is contained in the individual subunits. (4) The subunits bind cooperatively to form the most stable structure. Self-assembly is inextricably linked to the phenomenon of molecular recognition because 4 the recognition of the individual components of the aggregate by each other guide the construction of the supermolecule, or self-assembling structure. ^  Research in the field of self-assembly continues in three main areas: (1) the design of self-assembling structures that mimic natural structures, (2) the design of solids and solution structures that act as molecular devices, and (3) the study of the noncovalent forces that govern molecular recognition and therefore the self-assembly process.17 The work presented in this thesis is directed towards the latter of these directions; namely to uncover the noncovalent forces that are at play in a reaction to form carceplexes. The carceplexes studied in this thesis are formed as a result of directed self-assembly15 in which a template molecule aids in the formation of the product. Directed self-assembly is a type of self-assembly defined by Lindsey as "the case where a temporary scaffolding agent, jig, or template, participates as a structural element in the assembly process but does not itself appear in the final product"15. The external element may play a thermodynamic role by stabilizing the association of subunits or destabilizing an undesirable aggregate. It also can play a kinetic role by directing the association of subunits along a specific reaction pathway.15 The use of directed self-assembly to describe the formation of carceplexes is not completely accurate because the template molecule becomes part of the product due to mechanical entrapment. Nevertheless, the template plays a crucial role in the formation of the carceplexes investigated in this thesis. Molecular recognition, self-assembly, and templation are the cornerstone of supramolecular chemistry. The formation of the tobacco mosaic virus (TMV) is a good example of a biomolecule that encompasses these components of supramolecular chemistry and is illustrated in Figure 1.2.19 The TMV is composed of a 6400 base strand of RNA enclosed in a protein sheath which is made up of 2130 identical wedge-shaped protein subunits. Nature's use of multiple copies of the same building block reduces the amount of information necessary to create a self-assembling structures such as the TMV. If the components of the TMV are separated, they can be spontaneously reassembled in vitro to 5 regenerate the active virus. The process of reformation of the active virus involves the self-assembly of the 2130 protein subunits around the strand of R N A , which acts as a template. Molecular recognition between protein subunits causes their self-assembly with each other and with the strand of R N A , which is necessary for the formation of the virus. Imperfectly formed subunits are excluded from the final structure as a result of the reversible nature of this self-assembly process. Chemists have gained a wealth of information about the self-assembly and molecular recognition processes that are at work in nature by studying biological systems such as the T M V . The beauty of such processes inspires us to learn more. Presently, we can embark on an iterative process of trying to devise nonnatural assemblies with high complexity, delineate the noncovalent interactions that govern their formation, and redesign the system using the newly found knowledge to create even more advanced assemblies. During this process, we continue to re-check nature's examples, both to help in our design and to help understand nature as our science becomes more sophisticated. Figure 1.2. Tobacco Mosaic Virus. A 8 C D Model for the self-assembly of T M V : (A) initiation of self-assembly, R N A threads into the central hole of the protein disc and transforms it into (B) the helical lock-washer form; (C) self-assembly of additional discs; (D) one of the R N A tails is continually pulled through the central hole to aid in self-assemble of further discs . 1 9 6 iii. Forefathers of Supramolecular Chemistry Many people have contributed to the explosive growth in the area of supramolecular chemistry in the past three decades, but it was Donald Cram, Jean-Marie Lehn, and Charles Pedersen who were the significant pioneers in the field. For their efforts they were awarded the Nobel Prize in chemistry in 1987. Their work focused on designing and building host molecules with cavities of various sizes, shapes and degrees of rigidity that were capable of selectively binding metal ions as shown in Scheme 1.1. The crown ethers such as 18-crown-6 (4) developed by Pedersen's group are polyether rings that reorganize to form a spherical cavity upon complexation of an alkali metal ions.20-21 Cryptands such as 2,2,2-cryptand 5 are bicyclic compounds that reorganize to form an ellipsoidal cavity during metal ion complexation.22*24 Spherand 6 developed in Cram's group contains aryl ether oxygen donors that are organized in a rigid macrocyclic ring.25 Spherands are said to be fully preorganized for binding because they have a defined cavity in the uncomplexed state, as the rigidity of the molecule precludes conformational mobility (Scheme l . l) . 2 6 Cram defines the principle of preorganization as " the more highly hosts and guests are organized for binding and low solvation prior to their complexation, the more stable will be their complexes."27 Preorganization differentiates spherands from the crowns and the cryptands. The significance of preorganization to our work is developed further later in this chapter. Metal ion binders such as the crown ethers, cryptands and spherands provide the first examples of synthetic receptors designed to bind substrates for studying molecular recognition. From this building block a variety of novel molecular hosts were soon developed for binding substrates other than metal ions. Often, these new molecular hosts systems were found to require the use of a template molecule to aid in their synthesis. 7 Scheme 1.1. Metal Binding Hosts. B. Templation i. Introduction Templation is an integral part of supramolecular chemistry as it often aids in the construction of complex molecular structures. Indeed, the prototypical example of a template in nature is DNA, which functions as its own template. The word template has been defined by Busch:28 "a chemical template organizes an assembly of atoms, with respect to one or more geometric loci, in order to achieve a particular linking of the atoms. Templates are distinguished from reagents because they effect the macroscopic geometry of 8 the reaction and not the intrinsic chemistry".28 The formation of a crown ether serves as an excellent illustration of the action of the template molecule.20 As seen in Figure 1.3, the potassium ion acts as the template for the formation of 18-crown-6 by binding to the precursor acyclic polyether thereby organizing the reactive ends of the molecule in a conformation that allows for efficient covalent bond formation. Therefore, a chemical template manipulates a reaction pathway in order to achieve a particular linking of atoms, a process that is often referred to as a template effect. Figure 1.3. Metal Templation. ii. Classification of Templates Busch classified templates as either thermodynamic or kinetic.28 A thermodynamic template shifts the equilibrium of a reversible reaction by binding to one of the products formed in a reaction, thereby shifting the equilibrium in favor of this product. The reversibility of the reaction can often lead to high chemical yields. On the other hand, kinetic templates operate in irreversible reactions. The kinetic template organizes the reactive groups of the forming structure with respect to geometry and orientation, facilitating the formation of a single product. 17,28-31 The irreversible nature of the reaction 9 requires the kinetic template to stabilize the transition state of the rate determining step of the reaction. There are numerous examples of the use of templates in synthesis. 15-17,28-31 A large body of work has been done on the use of metal ions as templates especially for the synthesis of ligands and macrocycles such as the crown ethers (Figure 1.3).31 In the 1980's, the first examples of non-metal ions as templates for the synthesis of supramolecular structures appeared in the literature.28 This chapter will focus on the more recent examples of non-metal ions as templates for the formation of supramolecular structures emphasizing template effects in supramolecular synthesis. It first begins by introducing the historically interesting template formation of catenanes and rotaxanes. Mechanically joined molecules such as catenanes (7) and rotaxanes (8) provide challenging syntheses for supramolecular chemists. The name "catenane" was derived from the Latin word "catena" meaning chain32 due to their topological resemblance (schematically represented in Figure 1.4). For the n-catenanes, the n macrocycles are mechanically joined to each other but are not covalently bound. The prefix indicates the number of rings involved in the catenane i.e. 7 is a 2-catenane. Rotaxanes resemble catenanes because two or more molecules are mechanically linked but are not covalently bound. The name "rotaxane" was derived for the Latin word "rota" meaning wheel and "axis" meaning axle.32 In rotaxane 8, a dumbbell-shaped component is encircled by a macrocyclic component; the escape of the Figure 1.4. Catenane and Rotaxane. in. Catenanes and Rotaxanes bulky groups at the ends of the dumbbell-macrocyclic component is prevented by the shaped component (schematically represented in Figure 1.4). If the bulky groups are small 2-catenane 7 rotaxane 8 10 enough to allow the cyclic molecule to escape, the prefix pseudo is added to give pseudorotaxanes. Multiple rings threaded onto a single axle are referred to as a polyrotaxane. A variety of template studies have been performed recently with catenanes and rotaxanes and this work is presented below. Both the mechanical confinement of their components and the templation involved in their formation make catenanes and rotaxanes highly relevant to carceplexes. Catenanes can be prepared using either metal ion templates or organic templates. The first efficient synthesis of a 2-catenane (10) was done by Sauvage et al. who used the tetrahedral coordination properties of Cu(I) to form an ordered molecular assembly of ligands around the Cu(I) template to give assembly 9 (schematically represented in Scheme 1.2).3 3 Here, the Cu(I) holds two rigid aromatic ligands in an interwoven conformation (9) while macrocyclization of one or both of the ligands occurs. The Cu(I) template is then removed to give the free 2-catenane. The Cu(I) template does not affect the intrinsic chemistry of this reaction but merely increases the likelihood that topological selectivity will occur. Sauvage later used this strategy to produce a trefoil knot ( l l ) 3 3 (Scheme 1.2) as well as 3-catenanes.34 Scheme 1.2. Metal Templated Catenane Synthesis. iv. Template Formation of Catenanes (O) 10 2> 9 Trefoil Knot 11 11 Stoddart and coworkers have developed efficient procedures for the preparation of a large variety of catenanes and rotaxanes by using a non-metal template.16 Stoddart's synthesis of 2-catenane 16 involved the use of macrocyclic polyethers that contain n-electron-rich aromatics as the templates for the formation of tetracationic cyclophanes that contain 7C-electron-deficient aromatic units (Scheme 1.3). The favorable %-% interactions Scheme 1.3. 2-Catenane Synthesis Using Organic Templates. 16 [14*15] 0 -O -o-i-.-.-.-h —0 12 between the electron rich aromatic template molecule (15) and the electron poor aromatic rings of the cyclophane precursors (14 and/or 12) promote an interwoven complex ([14*15][PF4]3) that cyclizes to form 2-catenane 16 in 70% yield.35 In this synthesis, template molecule 15 becomes part of the formed catenane. Similar templated procedures using n-n interactions were used to construct rotaxanes36, switchable pseudorotaxanes37, pseudopolyrotaxanes38 and self-assembled macrocycles forming a molecular mosaic pattern (Figure 1.5).39 Figure 1.5. Molecular Mosaic Pattern of a Self-Complementary Cyclophane. Molecular Mosaic 13 v. Template Effects in the Formation of 2-Catenanes Stoddart has investigated the templation requirements for the formation of four tetracationic cyclophanes 19-22 with two different template molecules (Scheme 1.4)40 This investigation into the template requirements for 2-catenane synthesis is analogous to the templation study performed on carceplex 2a*guest and therefore a detailed discussion of this system follows. Stoddart performed competition experiments to indicate the relative preference for one template molecule over another. The competition experiments were carried out under identical conditions with equimolar amounts of the diphenyl template 17 and the dinapthyl template 18 as well as the macrocyclic components of either 19, 20, 21 or 22 (Scheme 1.4). The product ratios of the isolated 2-catenanes were determined by *H NMR. They found that dinapthyl template 18 was preferred over diphenyl template 17 in all cases and the magnitude of this selectivity depended upon which tetracationic cyclophane (19-22) was being cyclized. The greatest selectivity they observed was 99:1 for dinapthyl template 18 over diphenyl template 17 in the template formation of tetracationic cyclophane 21.40 14 Scheme 1.4. Catenane Synthesis. © 2[PFJ 1 1 equiv Br Br or Br (gag 2[PF6] »aa A equiv 1 I -mi Br p o o o o 17 0^  o diphenyl template 1 equiv + dinapthyl template 1 equiv 1. DMF, 20°C, 14 d 2. NH 4 PF6 /H 2 0 © mm i© IT'l m^1 4[PF6] Tetracationic Cyclophane Component of [2] Catenane Product ratio: 19 20 21 1.5:1 6:1 99:1 3:1 = Newly formed covalent bond 15 When they explored the reaction further, they found the selectivity for 2-catenane formation was under kinetic control and not thermodynamic control (Scheme 1.5). After the initial covalent bond is formed to produce the tricationic intermediates, self-assembly of this tricationic species with each of diphenyl template 17 and dinapthyl template 18 then occurs. The presence of a complex between the tricationic intermediate precursor to 19 and both diphenyl template 17 and dinapthyl template 18 were detected by FAB mass spectrometry as well as by lH NMR. The binding of the tricationic intermediate with either diphenyl template 17 or dinapthyl template 18 is controlled by their respective equilibrium constants K and K'. The magnitudes of the respective equilibrium constants K and K' are expected to be of similar magnitude because each tricationic intermediate (19, 20, and 21) contains at least one 4,4-bipyridinium dication unit that is strongly bound by both diphenyl template 17 and dinapthyl template 18. As long as complexation and decomplexation events are rapid with respect to ring closure then, the formation of the final bond to produce the tetracationic catenanes (occurs with a rate constant of k 2 for the diphenyl template and a rate constant of k3 for the dinapthyl template) determines the product ratio which was found to 1.5:1 for dinapthyl 2-catenane 24 to diphenyl 2-catenane 23. They reasoned that the selectivity was due to the dinapthyl polyether complex with the tricationic species lowering the Gibbs free energy of activation for the transition states for ring closure relative to that of the diphenyl complex resulting in the formation of more dinapthyl 2-catenane 24 (k3 is greater than k2). Stoddart et al. expected the same transition state stabilization would exist in the formation of catenanes containing tetracation cyclophane 20 and 21. Possibly, their conclusions would be more compelling if the experiment was performed on a system that had a greater selectivity than 1.5:1 and the actual equilibrium constants K and K' had been measured. 16 Scheme 1.5. Thermodynamic Analysis of a Catenane System. vi. Template Effects in the Formation of a Rotaxane The synthesis of rotaxanes that differ in their central axle component was the subject of a recent template effect study by Vogtle and coworkers.41 In this reaction (Scheme 1.6), a cycloamide-based cyclophane template (26) is believed to bind an axle component (25) which then reacts with a primary amine blocking group (27) to form complex 29. The formation of complex 29 is designed to be stabilized by N-H to carbonyl hydrogen bonding and TZ-TZ interactions between the macrocyclic template 26 and the axle molecules 25a-f. Reaction of complex 29 with another equivalent of 27 produces rotaxanes 30a-f while reaction of the unthreaded axle leads to compounds 31a-f. A range of axle components 25a-f that vary in their types of hydrogen bond donor abilities (carbonyls versus sulphones), shape selectivity (meta versus para-phenylene units), and size of the aromatic ring (six versus five-membered) were chosen in order to optimize the yield for rotaxane formation. Stronger noncovalent interactions as described above should lead to a greater formation of complex 29 and result in higher yields of rotaxane 30 versus production of the undesired barbell molecule 31. As seen in Scheme 1.6, the formation of rotaxane 30e proceeded with the highest yield over the five other axle molecules, and it also gave the lowest yield for the formation of barbell shaped molecule 31e. The authors concluded that the formation of rotaxane 30 is tolerant to a variety of axle molecules, which greatly expands the variations of Unking molecules that can be used to create rotaxanes.41 18 Scheme 1.6. Amide Rotaxane Synthesis. a b o d e f % Yield of 11 19 6 7 41 4 Rotaxane 30 % Yield of 83 44 85 39 10 17 "Axle" 31 19 vii. Self-Replicating Systems Self-replicating and autocatalytic systems are other important areas of supramolecular chemistry that have received a lot of attention due to their sweeping implications about the origin of life.42 Replication is often thought of as a biological event whereby one generation passes on its hereditary information from one generation to the next. The seminal work of Kiedrowski has furthered our knowledge of the processes that control replicating systems.43-45 Chemists have recently developed a number of simple chemical models that are capable of self-replication 4 6 - 5 0 In a schematic example of a self-replicating system (Figure 1.6),46 molecule A reacts with molecule B, due to their complementarity, to form the template molecule T. The self-complementarity of template molecule T to A and B leads to formation of termolecular complex T: A:B. Complex T: A:B changes the reaction between A and B from intermolecular to intramolecular and generally increases the rate at which A and B react to form complex 2T. Complex 2T then dissociates to form two molecules of template T and the cycle repeats itself.46 Figure 1.6. Schematic Representation of Self-Complementary Template-Based Autocatalysis. H I HI T B < + o A, B -o T+A + B T:A:B o--o 2T covalently bonded -IP g>-noncovalently bonded 20 viii. Template Effects in a Self-Replicating System Rebek and coworkers have created a number of self-replicating systems based on adenine recognition of an imide derivative of Kemp's triacid.46>51>52 In one example, they looked at the template accelerated formation of adduct 35 from an adenine derivative (32) containing a primary amine group and a second adenine derivative (33) containing an activated ester illustrated in Scheme 1.7.53 The seven different template molecules (36-42, Table 1.1) used all contained two recognition sites for adenine but these binding sites were separated by a different spacer located between the carbazole units of each template molecule. Spacers were chosen that varied the distance, geometry and rigidity of the template molecule. The use of these template molecules resulted in rate accelerations ranging from one to 160 fold for the coupling reaction between 32 and 33 to produce 35. Generally, the high effective molarities of the activated ester and the amine groups when held in close proximity in the termolecular complex with the template resulted in faster reactions. Rebek and coworkers concluded that the most effective template molecule was better able to stabilize the tetrahedral intermediate within complex 34 (Scheme 1.7). Templates 40 and 41 have the most complementary surface (distance and rigidity) that allows for stabilization of the transition state of 34 and therefore leads to the fastest rate enhancements. The other templates (36-39 and 42) did not cause large accelerations in the rate of reaction because they lacked either the proper distance or rigidity to stabilize the transition state 34- In another study, Rebek and coworkers showed that the rate of a reaction could be impeded by recognition of the reactants to form a complex. The complex formed in this example did not have the suitable geometry or distance to allow the reactive ends of the reactants to reach each other and undergo reaction.54 In the self-replicating systems studied by Rebek and others, effective turnover by the catalyst is often impeded by product inhibition. The design of template molecules that are only complementary to the 21 transition state of these reactions and not the product would prevent such product inhibition. Scheme 1.7. Rebek's Adenine Reaction. 22 23 ix. "Positive" and "Negative" Templates Sanders and coworkers have developed efficient syntheses for the formation of a series of cyclic porphyrin oligomers using a variety of pyridine-based template molecules.55"59 They used the template formation of cyclic porphyrins to further classify template molecules as either "positive" or "negative". A "positive" template directs a reaction to form a particular product while a "negative" template will disfavor the formation of a particular product.29 Rebek's templates 3 7 - 4 2 , discussed in the previous section, represent positive templates. The formation of cyclic porphyrin 4 5 is also an excellent example of the use of positive and negative templates. For example, meso-tetra(4-pyridyl)porphyrin 4 4 acts as a negative template by inhibiting the formation of both cyclic dimer 4 6 (only 9% yield) and other higher oligomers (13% yield) as well as a positive template by accelerating the formation of cyclic tetramer 4 5 (78% yield) as seen in Scheme 1.8.29'58 The yields of higher oligomers, cyclic dimer 4 6 and cyclic tetramer 4 5 are 65, 25, and 10 %, respectively, in the absence of a template.58 Thus, meso-tetra(4-pyridyl)porphyrin 4 4 prevents intramolecular coupling of 4 3 and favors the intermolecular coupling. In principle, any reaction that can be templated to form a specific product can also be templated to prevent that product. The design of negative template molecules that prevent unwanted side reactions may help form the product desired in higher yields and is therefore important to supramolecular synthesis. 24 Scheme 1.8. "Positive" and "Negative" Templates. x. Template Effects in the Formation of Cyclic Porphyrins The Sanders group studied the template effects for the formation of a cyclic trimer of porphyrins. In one of their experiments, they used a series of six template molecules (47-52) that range in size, shape and zinc binding sites (Table 1.2), and looked at the product distribution of cyclic dimer 46 versus cyclic trimer 54 (Scheme 1.9).58 They found that the use of a noncomplementary template such as pyridine (47) led to formation of both the cyclic dimer and cyclic trimer in approximately equal amounts. When they used bipy 48 as a template molecule there was an overall increase in yield with a large preference for dimeric host 46 to which bipy template 48 is complementary. The use of templates 50 and 51 increased the yield of cyclic trimer 54 relative to pyridine (47) but also substantially decreased the yield of cyclic dimer 46. The highly complementary template 52 increased the yield of cyclic trimer 54 and decreased the yield of cyclic dimer 46 to the greatest extent within the series of templates, but it was only slightly better than the bifunctional template 50 and 51. These results suggest that the selective formation of cyclic trimer 54 with the templates 50-52 was largely the result of these templates 25 preventing the formation of cyclic dimer 46 (Scheme 1.9). In this sense, the bipy 48 acts as a positive template for the production of cyclic dimer 46 and a negative template for the production of cyclic trimer 54. Templates 50-52 are acting in the opposite direction, as positive templates for the production of cyclic trimer 54 and negative templates for the production of cyclic dimer 46. Overall, Sanders and coworkers found that the choice of template molecule often dictated the major product that would be formed in the reaction.58 Scheme 1.9. Formation of Cyclic Dimers and Trimers of Porphyrins. Cyclic Trimer 26 Table 1.2. Template Effects on Cyclic Porphyrin Oligomers. 47 48 49 50 51 52 % Yield Cyclic 23 72 27 7 8 6 Dimer 46 % Yield Cyclic 34 4 34 43 44 52 Trimer 54 xi. Anions as Templates Anions have also been used as templates for the synthesis of a variety of ion cages.60-61 The organization of polyoxometalates of vanadium, molybdenum and tungsten around anions such as CI", CO3 , CIO4", N3" and SO4 to form host-guest structures that resemble the structures of carceplexes has been studied by Muller and coworkers62-63 Here the anionic guest is crucial to the formation of the caged structure. In the absence of the anionic template there are no such cages formed. The structure of the cage was also found to be dependent upon the anionic template used. Similar investigations into the template effects of halide ions for the formation of macrocyclic mercury complexes were explored by Hawthorne et al. 6 4 27 C. Templates Used to Make Materials i. Molecularly Imprinted Polymers The use of templates to create molecular imprints in polymers is another active area of supramolecular chemistry.65'66 The preparation of a molecularly imprinted polymer involves the formation of a cross-linked polymer around a template molecule. After the removal of the template molecule, a functionalized cavity remains that is complementary to the template molecule used in its synthesis. This cavity is capable of selectively recognizing the template molecule and even capable of resolving enantiomers of chiral templates. The process of creating the molecularly imprinted polymer resembles the production of antibodies in biological systems. As in antibodies, the shape of the cavity, the spatial arrangement of functional groups, and the flexibility of the binding site all determine the extent of molecular recognition. The transfer of molecular recognition information from the template to the polymer is a quick and efficient means of creating a selective molecular host without the laborious task of synthesizing a host in a step wise manner as is done in many supramolecular systems.67 The disadvantage of imprinted polymers is that they are difficult to characterize because they lack homogeneity within their binding cavities. This limits the molecular recognition information that one can obtain from binding studies, which is essential to the detailed characterization of these systems. Molecular imprinting has been done on surfaces such as silica gel6 8 and monolayers on gold. 6 9- 7 0 Imprinted polymers have found uses in resolution of enantiomers,71'72 asymmetric catalysis 7 3 mimicry of antibodies74 and selective transport across membranes.75 28 ii. Templates in Crystal Engineering Molecular recognition between molecules is far from restricted to solution chemistry. The goal of crystal engineering is to create solid state structures with useful functions. As with supramolecular chemistry in solution, the noncovalent interactions and molecular recognition between molecules during crystal packing are not well understood.12 Although the study of self-assembling structures in solution differs from their solid state equivalents, the intermolecular interactions that govern the formation of each are essentially the same. Often the growth of a crystal will lead to the inclusion of a guest molecule within the interstitial space of the packed crystal. Such inclusion of a guest molecule is often called enclathration, and the solid state structure is described as an inclusion compound. Wuest et al. have developed structure-directing molecules know as tectons such as 55 to aid in the self-assembly of three-dimensional networks that form large chambers.76 The tecton is designed with directional hydrogen bonding sites that direct their aggregation to form predictable structures. The use of tecton 55 led to the formation of a diamondoid network 56 in the presence of a suitable enclathrate molecule or template molecule (Figure 1.7). When the crystal was grown with tecton 55 from CH3CH2COOH/hexane/MeOH or CH3COOH/hexane/MeOH, a non-diamondoid network was formed due to the competitive hydrogen bonding of the acid molecule with tecton 55. The use of larger acid molecules in the solvent mixture as in (CH3(CH2)2COOH/hexane/MeOH) or (CH 3(CH 2) 3COOH /hexane/MeOH) led to the predicted diamond crystal structure 56. Although the crystallization was performed in a mixed solvent, only the acid molecules were incorporated into the cavity, demonstrating that the self-assembly of tectons are template-dependent.76 29 0 Figure 1.7. Tecton Assembly. tecton 55 diamondoid network 56 iii. Template Molecules Used to Create Zeolites Zeolites, especially silica-based zeolites, are another type of material whose structure is highly dependent upon the structure-directing agents, or template molecules, used in their synthesis.77 The use of a variety of rigid polycyclic template molecules77 and liquid crystals as template assemblies have expanded the types of zeolite structures that can be made. The applications of zeolites in applied chemistry and engineering disciplines are continually expanding from the traditional catalysis and adsorbent technology to more recent micro-reaction chambers.78-80 Templation will play a key role in the development of such new molecular sieve lattices. Presently it is difficult to predict what structure of zeolite would be produced from a given template molecule. The prediction of the final structure of the zeolite, however, may be aided by computational approaches. Lewis et al. have developed a program for the de novo design of template molecules that are "computationally grown" in the desired inorganic framework.81 They have successfully worked backwards from known zeolite structures to create the template molecules that were 30 shown experimentally to form these structures. In the future, such computations may vastly expand our knowledge about the templation of zeolites. iv. Template Formation of Tubules Electronically conductive polymer nanostructures such as fibrils and tubules have been successfully synthesized within the pores of nanoporous membranes.82 The nanoporous membrane acts as a mold or template for the formation of the nanostructure. After its formation, the nanostructure can be separated from the nanoporous template. The nature of the nanostructure is dependent upon chemical make up and size of the nanoporous membrane employed in the synthesis. These structures have applications in bioencapsulation and biosensors.82 D. Molecular Encapsulation i. Self-Assembly of Cavities Capable of Molecular Encapsulation The phenomenon of self-assembly is important to biological and materials sciences alike because of its widespread use in the formation of structures such as cell membranes and monolayers. The driving forces for the formation of a self-assembling structure are a multitude of noncovalent interactions, such as hydrogen bonds, van der Waals, electrostatic, and n-K interactions, that bring the molecules together in a defined aggregate. Numerous one dimensional and two dimensional self-assembled systems14'83 are known,84-85 but relatively few self-assembling structures are known to form in three dimensions and create cavities capable of encapsulating guest molecules. Here, we examine a number of self-assembling structures that demonstrate the formation of internal cavities capable of molecular recognition in solution. 31 The design of self-assembling structures that contain a cavity capable of encapsulating one or more guest molecules has received a lot of attention recently because of potential applications, including drug delivery devices and miniature reaction chambers.27 The construction of traditional molecular hosts used for molecular recognition has generated sophisticated compounds such as spherands,27 cryptophanes86 and modified cylodextrins87 which are capable of binding ions or molecules of various sizes. These systems often require multistep syntheses and also require the design of an opening that will allow for guest entrance and egress. A more appealing and economic method of creating a host-guest system would be to self-assemble the host around the guest molecule. Often, the encapsulated molecule acts as a template by aiding in this self-assembly of molecular components around the template molecule. The result is a well-defined three dimensional spherically-shaped aggregate. Here the word template is used in the formation of a noncovalent and reversible assembly of molecules instead of in the formation of a covalent bond. Such flexibility was left in Busch's definition28 of the word "template" which stated: "a template organizes an assembly of atoms, with respect to one or more geometric loci, in order to achieve a particular linking of the atoms". The word "linking" can be extended to both covalent and noncovalent bonds. Our discovery of complex 3b«guest (Figure 1.1) represents one of the few self-assembling structures that is capable of selective, reversible molecular encapsulation, and is the subject of Chapter 3. A number of researchers are actively exploring related types of self-assembling structures and they are briefly reviewed below. 32 ii. Rebek's Tennis Ball Rebek and coworkers have created a number of self-assembling systems that form cavities which are capable of binding neutral molecules that vary in size from that of methane to substituted adamantanes.88-90 A common design feature of Rebek's self-assembling structures is the use of two di-substituted glycoluril substituents connected by a rigid spacer such as durene, as shown in structure 71 (Figure 1.8).91 The concave nature of compound 71 allows for its self recognition to form a reversible dimeric capsule that encapsulates a guest molecule within its interior (complex 71»71»guest). Complex 71»71»guest has the same shape and symmetry as the cover of a tennis ball (Figure 1.8). Rebek's complexes such as 71»71»guests are amenable to NMR characterization where the decomplexation rates are generally in slow exchange on the *H NMR timescale, and thus signals for bound guest are distinguishable from free guest. Moreover, integration of the *H NMR gives the stoichiometry of the complex, the N-H signals are shifted downfield indicating hydrogen bonding, and the encapsulated guest molecules have large upfield shifts due to the shielding by the aromatic host. Rebek et al. have also used X-ray crystallography and mass spectrometry to further characterize their complexes. Binding constants ([71«71»guest]/([free host][free 71]) for complex 71»71»guests were determined: CHC13 (0.04 M"1), CH 2C1 2 (4 M"1), ethylene (280 M"1), and C H 4 (300 M' 1) in CDCI3; complex 71«71»xenon was also shown to form in the presence of the noble gas Xe in CDCI3 but no binding constant was reported.92 The *H NMR of complex 71 in CDCI3 shows two sets of signals in the presence of a suitable guest such as methane. One set of signals is due to complex 71»71»CH4 while the other set of signals is thought to correspond to the empty dimer. The possibility that the empty species contains water or dissolved gases in the cavity could not be excluded.91 Derivatives of compound 71 that are large enough to encapsulate substituted adamantanes, ferrocenes and two molecules of benzene derivatives have been synthesized by this group.89 The incorporation of two 33 molecules in Rebek's larger self-assembled host molecules now provides the opportunity for performing bimolecular reactions within complex interiors. i i i . Urea Based Calixarene Dimer Rebek and coworkers have recently described the dimerization of self-complementary calix[4]arenes 72 through intermolecular hydrogen bonding of the urea functionalities incorporated into the upper rim of the calix[4]arene 9 3 - 9 5 The urea moieties form hydrogen bonds (as indicated by their downfield shifts in the H N M R of the N H signals) in a directional cyclic array as shown in Figure 1.9. The resulting complex 72»72«guest was found to selectively bind the following molecules in order of increasing binding constants: ethylbenzene < p-xylene < o-xylene < toluene < chloroform = benzene (the binding constant of benzene was reported as 2.3 x 10 M in [Djolp-xylene). Rebek and coworkers were able to perform desorption mass spectrometry and obtain molecular ion peaks for their complexes with the same relative intensities as that found in solution. 9 3 A series of related calix[4]arene dimers were studied in a variety of solvents by Bohmer and coworkers but no evidence of molecular encapsulation of guest molecules within the cavities of the dimers was reported. 9 6 Instead, Bohmer and coworkers focused on the formation of heterodimers that resulted when calixarenes that differ in the urea groups at the 34 upper rim were mixed together in solution. These studies found that a more or less statistical equilibrium controlled the dimerization process. Similarly, Reinhoudt studied the dimerization of two bis-functionalized calixarenes.97 Figure 1.9. Rebek's Urea Based Capsule. 35 iv. Cyclocholates Bonar-Law and Sanders have reported a similar self-assembled structure to those discussed in the previous two sections. The dimerization of cyclocholate 73 via hydrogen bonding of amides groups in the C-ring of the steroid led to the formation of a capsule-like assembly 73*73 (Figure 1.10).98 Cyclocholates are rigid macrocycles formed by the condensation of multiple units of cholic acid. Dimerization of the cyclocholates was evident by the large downfield shifts of the N-H protons in the lU NMR and the N-H and carbonyl stretching frequency in the IR. Both vapor pressure osmometry (VPO) and freezing point depression experiments in benzene indicated molecular weights consistent with the proposed dimer 73*73. No guest binding studies were reported but the dimerization constant for 73*73 was determined to be 3 x 104 M" 1 in CDCI3. 9 8 Figure 1.10. Formation of Cyclocholates. 73*73 36 v. Shinkai's Heterodimer Shinkai et al. reported the self-assembly of two calixarenes functionalized at their upper rims to form heterodimer 74*75 (Scheme 1.10)." The proposed structure of heterodimer 74*75 is based on VPO measurements, fluorescence spectroscopy of the stilbazole unit of 75 and the increased solubility of calixarene 74 in CDCI3 when calixarene 75 is present. No guest binding studies were reported. Later, Reinhoudt et al. reported the formation of a similar heterodimer via the association of a calix[4]arene functionalized at the bottom rim with 4-pyridyls and another calix[4]arene functionalized at the upper rim with carboxylic acid groups.100 Scheme 1.10. Formation of a Calix[4]arene Heterodimer. 74*75 3 7 vi. Encapsulation in Cucurbituril Recently, Kim et al. reported the switchable assembly of a molecular container capable of reversibly binding guests molecules such as T H F . 1 0 1 They found the solubility of cucurbituril 76 increased dramatically in aqueous solutions of alkali metal salts such as sodium sulfate. They grew crystals from this solution and the X-ray crystal structure indicated that a rigid molecular container had formed whereby two molecules of Na + ions and five H2O molecules formed a cap-like structure at the top and at the bottom of cucurbituril (77, Scheme 1.11). Addition of THF to a D 2 0 solution of cucurbituril and Na2SC»4 resulted in the formation of a complex that was in slow exchange on the *H NMR timescale and had a formation constant (Kf) of 5.1 x 102 M" 1 . An X-ray crystal structure of complex 77»THF indicated that the host had the same shape and symmetry as it did in the absence of guest. Furthermore, complex 77«THF was found to be "switchable" by altering the pH of the solution. Thus, addition of trifluoroacetic acid to a solution of complex 77»THF results in decomplexation as indicated by the decrease in bound THF and the increase in uncomplexed THF. Addition of Na2CC>3 to this sample regenerated complex 77»THF. Similar binding experiments were performed with cyclopentanone (Kf = 2.2 x 103 M"1), benzene (Kf = 2.7 x 101 M"1) and furan (Kf = 7.1 x 103 M"1). This study of the binding properties of cucurbituril expands the range of molecules that can be incorporated into its cavity from traditional ammonium-based guests102 to neutral guests about the size of benzene. 38 Scheme 1.11. Cucurbituril Complex. 2Na + 5H O 2 76 77»T H F vu. Molecular Encapsulation in Cyclodextrins Cyclodextrins (CDs) are "lamp-shade" shaped cyclic oligomers of glucose.103 They are soluble in water and contain a chiral hydrophobic cavity with openings at both ends. The size of the cavity and its openings depend upon the number of glucose subunits in the cyclodextrin; (denoted by the prefix a (6), (3 (7), and y (8)) (Figure 1.11). CDs bind a number of hydrophobic guest molecules in aqueous solution where the strength of binding is often determined by the hydrophobic effect. Thirty years of investigation into the binding properties has revealed a large variety of molecules that bind to CDs 8 7 and the search for new substrates still continues.104 Porphyrins, especially metallophorphyrins, are important components of natural systems such as in cytochromes where they aid in electron transport. The versatility of porphyrins make them interesting building blocks for the formation of supramolecular structures. The encapsulation of porphyrins in CD complexes has been studied by Lawrence105-106, Nolte107 and others.87'108"111 The binding constants of porphyrins to CDs was demonstrated to increase when multiple CD's are covalently linked together with an appropriate spacer. The use of a covalently linked tetramer of CD's (78) as illustrated schematically in Figure 1.11 formed one of the 39 strongest complexes with porphyrins. For example, both neutral porphyrin 7 9 and zinc porphyrin 8 0 form complexes 7 8 * 7 9 and 7 8 * 8 0 with binding constants (Kb) of = 108 M" 1 in D 2 O . 1 0 6 Binding constants for dimers of CD's (two covalently linked CD's) with porphyrins are significantly smaller (e.g. Kb ~ 104 M" 1 ) . 1 0 6 The binding properties of porphyrins in CDs holds promise as new catalytic systems. Figure 1.11. Cyclodextrin Illustration. primary face 7 8 7 9 , X = C0 2 H, M = H,H 7 8 * 7 9 8 0 , X = C0 2 H, M = Zn 7 8 * 8 0 40 viii. CD Dimerization Klufers et al. discovered that mixing 6-CD 81, lithium hydroxide and a source of copper(JJ) ions in an aqueous solution resulted in the formation of blue crystals. X-ray crystallography showed that two 6-CD 81's are connected via metallic bridges to give a 6-CD dimer 81*81 that is schematically illustrated in Figure 1.12.112 B-CD dimer 81*81 contained four copper(II) ions that connect the two CD's (Figure 1.12). In addition, a number of intramolecular and intermolecular lithium ion salt bridges and charged O"- • H-O hydrogen bonds were found in the crystal structure. The charged hydrogen bond distances ranged from 2.46-2.60 A indicating strong hydrogen bonds.113"115 Seven molecules of water were encapsulated within the interior of the cavity of B-CD dimer 81*81, each of which completed the tetrahedral coordination sphere of a lithium ion. A similar y-CD dimer structure was found to form in the presence of lead(II) ions; this dimer had a higher metal to CD ratio and represents the first lead(II) carbohydrate complex known to date.116 Figure 1.12. CD Metallic Dimer. 6-CD 81 6-CD dimer 81*81 41 ix. C D ' s and C60 Yoshida et al. discovered that an aqueous solution of y-CD 82 could selectively extract C60 from a mixture of fullerenes in toluene. 1 1 7 Remarkably, only C6o (7 A, spherically-shaped) and not C70 (7 A by 8 A oval-shaped) was extracted into an aqueous solution of y-CD (there is a 9 A circular opening at the secondary face). Both a - C D and (3-C D failed to extract any fullerenes from the toluene solution. They found encapsulation of C60 by two molecules of y-CD 82 forming complex 82»82»C60 resulted in the solubilization of C60 into aqueous solution (Figure 1.13). Furthermore, evidence for complex 82«82»C60 formation included: C N M R chemical shifts of host and guest, and their relative integration, *H N M R chemical shifts of the y-CD and elemental analysis. Other researchers have developed means of purifying fullerenes by the aid of 1:1 binding in bowl-shaped compounds such as calix[8]arenes 1 1 8, calix[6]arenes 1 1 9 and cyclotriveratrylenes 1 2 0. Figure 1.13. y-CD Dimer Encapsulating C60-2 y -CD 82 C60 82»82«C 6 o 42 x. Encapsulation of 13-Methyl Glucopyranoside Aoyama et al. reported the encapsulation of a B-methyl glucopyranoside between two molecules of octol 83. 1 2 1 Here octol 83 extracts the normally insoluble B-methyl glucopyranoside into CDCI3 or CCI4 from aqueous solution to form complex 83»83*8-methyl glucopyranoside where there are two molecules of octol 83 to one B-methyl glucopyranoside (Figure 1.14). The 2:1 stoichiometry was confirmed by VPO measurements. The *H NMR spectrum of complex 83«83«B-methyl glucopyranoside exhibited an upfield shift of 3.58 ppm for the methoxy group of B-methyl glucopyranoside, indicating that it is strongly shielded by the aromatic host. Incidentally, a 3.58 ppm shift of the methoxy group of B-methyl glucopyranoside indicates that it is most likely held deep in the cavity of complex SS^^-methyl glucopyranoside and not at the equator of the molecule as the illustration in Aoyama's paper indicated (see Chapter 4). Also, the *H NMR spectrum of complex 83»83»8-methyl glucopyranoside exhibited a complex spectrum with a multitude of peaks, indicating that the B-methyl glucopyranoside guest is frozen out on the *H NMR time scale inside the host. This complex also showed remarkable selectivity for B-methyl glucopyranoside over a-methyl glucopyranoside. 43 Figure 1.14. Encapsulation of 8-Methyl Glucopyranoside. 83 83»83»guest R = ( C H 2 ) I Q C H 3 guest = B-methyl-glucopyranoside E . Carceplexes i. Introduction Mechanically joined supramolecular structures, which include both catenanes and rotaxanes, have attracted the attention of chemists for years because of their novel structures, challenging syntheses, and potentially useful properties. In 1983, Cram proposed another technique for mechanically joining together molecules whereby a rigid closed surface spherical host such as 84 acts as a molecular prison for the entrapment of a guest molecule within its interior (Figure 1.15).26 These types of highly preorganized compounds had never been made before and would likely have interesting chemical and physical properties. 44 Figure 1.15.Molecular Prison. Cram's success in creating the spherand demonstrated the importance of preorganization within supramolecular hosts (see page 8). Cram has continued to develop other molecular host systems that encompass his idea of preorganization.25-27 His synthesis of cavitands (rigid macrocyclic molecules with an enforced cavity) greatly expanded upon the versatility of preorganized molecular hosts, because unlike spherands, cavitands are capable of binding either ions or neutral organic molecules 2 6 Cavitand 89 (tetrabromo-bowl) can be synthesized on a multigram scale via a three step synthesis. The first step of the synthesis involves the condensation reaction between resorcinol (86) and hydrocinnamaldehyde (85a) to yield cyclic tetramer 87 (the cyclic tetramer can incorporate a variety of pendent groups or feet by using different aldehydes122"125). Octol 87 is then brominated with A -^bromosuccinimide (NBS) to give tetrabromo-octol 88. Finally, bridging of the adjacent phenols in the conformationally flexible tetrabromo-octol 88 with methylenes gives cavitand 89. The methylene bridging reaction has been successfully applied to a diverse range of bromooctols that differ in their pendent group; also, a variety of bridges can be introduced between the adjacent phenols.122"124 Furthermore, tetrabromo-bowl 89 can be further modified by exchanging the aryl bromide with a variety of functional groups, which makes cavitands attractive components for supramolecular synthesis. Cavitands provided the stepping stone into the synthesis of the first carceplexes 45 because joining two of these hemi-spherical molecules together would form a spherical molecule with a cavity large enough to entrap small molecules in its interior. Scheme 1.12. Synthesis of a Cavitand; Yields are for R = CH2CH2PI1. 4 R C H O + 85a, R = CH 2CH 2Ph 85b, R = C H 3 4 10T 86 HC1 6 9 % 87a, and 87b /Y-Bromosuccinimide 50% CH2ClBr H^ K 2 C 0 3 52% 89a and 89b 88a and 88b ii. Synthesis of the First Molecular Container Compound In 1985, Cram reported the synthesis of the first molecular container compounds known to permanently entrap small molecules or ions within their interiors.2 Cram called these compounds carcerands which he defined as "closed surface sphere-like organic host compounds with structures rigid enough to contain enforced interiors of sufficient volumes to incarcerate guest atoms, molecules or ions. Carceplexes are carcerands containing guests in their inner phases that are unable to depart without covalent breaking bonds of the host."126 The shell closure reaction between tetrabenzylchloro-cavitand 90a and tetrabenzylthiol-cavitand 91a gave the first carceplexes 92a«guests (Scheme 1.13) that were characterized as mixtures due to their insolubility. Both mass spectrometry and elemental analysis indicated that carceplex 92a«guests contained most chemical species 46 present in the reaction medium including both neutral molecules, gases and ions. In 1990, the more soluble cavitands 90b and 91b were subjected to the shell closure reaction which resulted in the formation of carceplex 92b»guest. Carceplex 92b»guest was found to encapsulate only neutral molecules (such as iV,A/-dimethyl acetamide (DMA) and 2-butanone) that were present in the reaction medium.1 2 7'1 2 8 Scheme 1.13. First Synthesis of a Carceplex. 90a,b R R R R R R R R 91a,b 92 92a, R = C H 3 92b, R = C H 2 C H 2 C H 2 C H 2 C H 3 iii. The Synthesis of the First Fully Characterized Carceplex In a parallel study, Cram's group prepared a series of acetal-linked carceplexes, 2a»guests.1'3 Tetrol la, the starting material for formation of carceplex 2a»guest, was synthesized on a multi-gram scale via the treatment of tetrabromo-bowl 89 with n-butyllithium followed by quenching with trimethylborate and further oxidation with hydrogen peroxide and hydrolysis with base to give tetrol la in 53% yield (Scheme 47 1.14).3 The successful bridging of adjacent phenols in tetrabromo-octols (such as 88) led to the idea to intermolecularly bridge the phenols of two molecules of tetrol la with bromochloromethane.129 Thus, the intermolecular bridging reaction between two molecules of tetrol la was attempted under high dilution conditions using bromochloromethane as the bridging material and cesium carbonate as the base (Scheme 1.14). The reaction, when performed in neat dimethyl formamide (DMF), dimethyl acetamide (DMA) and dimethyl sulfoxide (DMSO) gave carceplexes 2a»DMF, carceplexes 2a»DMA and carceplexes 2a»DMSO, (Scheme 1.14) in 49, 54 and 61% yields, respectively.3 These yields are remarkable high for a reaction that joins seven molecules together and makes eight new carbon-oxygen bonds. Moreover, Cram and Sherman also discovered three interesting things about this reaction: (1) No carceplex was formed when the reaction was run in A -^formylpiperidine (NFP), a molecule too big for the interior of carceplex 2a«guest; (2) A 10% yield of DMA carceplex was obtained when they ran the reaction in a mixture of NFP and DMA (99.5/0.5 molar ratio) as solvent; and (3) A 5:1 ratio of carceplex 2a»DMA and carceplex 2a»DMF resulted when they ran the reaction in a mixture of DMA and DMF (50/50 v/v) as solvent.3 Taken together, these three results suggested two things: (1) The reaction to form carceplex 2a»guest required a template molecule because no carceplex was isolated without a guest. (2) The carceplex reaction demonstrates selectivity when given the choice of two suitable templates as demonstrated with the 5:1 preference for formation of carceplex 2a*DMA over that of carceplex 2a»DMF. These results are the premise of this thesis. Further investigations into the extent of the template requirements for carceplex 2a«guest are the subject of Chapter 2. Incidentally, the insolubility of carceplex 92a»guest (pendent group = methyl) and the solubility of carceplex 92b»guest (pendent group = pentyl) and carceplex 2a«guest (pendent group = phenethyl) created the presumption that large soluble pendent groups were a necessity for solubility of these closed surface compounds. Cram and others have 48 since created a large number of compounds that contain large pendent groups such as phenethyls, pentyls, and unidecyls based on this solubility lore. 2 7- 1 3 0 Fraser et al. synthesized the first soluble carceplex that contained pendent methyl groups (carceplex 2b»pyrazine, Scheme 1.14).131 They found the solubility properties of cavitands such as tetrol 1 and carceplex 2»guest were largely independent of the pendent group used. Furthermore, the smaller methyl pendent group of carceplex 2b»pyrazine provided more stable crystals than those of carceplex 2a»DMA. The greater stability of the crystals and the smaller methyl groups of carceplex 2b»pyrazine made the determination of its X-ray crystal structure easier and also provided better resolution.131 Scheme 1.14. Synthesis of Carceplex 2»Guest. tetrabromo-bowl 89 R R R R Tetrol l a R = CH2CH2PI1 carceplex 2a«guest Tetrol l b R = CH3 carceplex 2b«guest 49 iv. Properties of Carceplex 2a»Guest. The inner phase of carceplex 2a»guest was described by Cram as a new phase of matter.1 Indeed, the physical properties of a molecule entrapped within another molecule were largely unknown until the discovery of carceplexes. To investigate the properties of the inner phase, Cram and Sherman measured the energy barrier for the rotation of the amide bond in DMA and DMF in their respective carceplexes. They found these energy barriers are intermediate between what is observed in the gas and the liquid phases of their amides, neat. Also, the infrared spectra of carceplex 2a«DMA and carceplex 2a»DMF showed carbonyl stretching frequencies that are intermediate between that of gas and liquid phases for these molecules. These experiments suggested that the inner phase of the carceplex is intermediate between that of a liquid and that of a gas.3-129 The entrapped guest in the carceplex demonstrated solvent dependent *H NMR chemical shifts indicating they can communicate with their external environment.3-129 Also, carceplex 2a»DMA and carceplex 2a»DMF are separable via silica gel chromatography further illustrating guest communication with external influences.3-129 In the crystal structure of carceplex 2a»DMA , the DMA molecules are largely ordered. The six non-hydrogen atoms of DMA are roughly coplanar with the carbonyl oxygen located near the center of the carceplex (between the two bowls) and a methyl group of DMA extending deep into both the upper and lower bowls. The DMA guest molecule of neighboring carceplexes have the same orientation with respect to the C4 axis of the carceplex indicating possible communication of DMA molecules in the solid phase.3-129 A more detailed description of the crystal structure of carceplex 2a»DMA is given in Chapter 2. Cram and Sherman also studied the rotation of the guest within the carceplex using variable temperature *H NMR spectroscopy. All guests were found to rotate rapidly (on the *H NMR timescale) about the C4 axis of the host down to -38 °C. However, rotation 50 with respect to the C2 axis was found to be guest dependent. For example, C2 rotation for DMF in carceplex 2a»DMF was fast down to -38 °C while C2 rotation for DMA in carceplex 2a»DMA was slow up to 175 °C. On the other hand, the C2 rotation of DMSO in carceplex 2a»DMSO could be frozen out at 2 °C giving an energy barrier of 13 kcal/mol. Chapter 4 of this thesis expands upon these findings by taking an in-depth look at guest and host dynamics of carceplexes and some related complexes. Cram's group also prepared carceplex 95»guest via a two step synthesis from tetrathiol 93. In their synthesis, tetrathiol 93 was reacted with bromochloromethane to give the corresponding tetrachloromethylsulfide 94 in 62% yield. Shell closure of 93 and 94 in neat DMA gave carceplex 95«DMA in 22% yield (Scheme 1.15).132. Later, Paek et al. synthesized carceplex 95»DMA in 5% yield by intermolecularly bridging two molecules of tetrathiol 93 with diiodomethane.133 51 Scheme 1.15. Synthesis of a Thioacetal Bridged Carceplex CH 2 ClBr R = C H 2 C H 2 P h or (CH^^CHg 93 Cs 2 C0 3 carceplex 95»guest v. Other Carceplexes Very few true carceplexes have been synthesized to date. Carceplex 2«guests and carceplex 92»guests represented the only carceplexes until 1993 when Muller synthesized polyoxyvandate compounds that incarcerated anions (see section xi on page 27). 6 3 In 1994, Reinhoudt et al. synthesized carceplex 97»guest via combination of a calix[4]arene and a cavitand (Scheme 1.16).130 They observed high yields for the shell closure of 96 in neat solvents to form carceplex 97«guest where the guests was a molecule of solvent. Guests entrapped in carceplex 97»guest included DMA, DMF and 7V-methyl-2-52 pyrrolidinone(NMP). These compounds did not exchange their guests molecules even under prolonged periods of heating, which confirms that they are indeed carceplexes. Later, Reinhoudt et al. used a doping procedure whereby a guest molecule is added to a solvent (5-10 mol % based on solvent) which is itself a poor template for carceplex 97»guest formation thereby allowing for screening various template molecules.134 The new incarcerated guest molecules included: l,5-dimethyl-2-pyrrolidinone, 2-butanone, ethylmethylsulphoxide, thiolane 1-oxide, and 3-sulfolene. Chemical modification of carceplex 97«guest occurred without the loss of its guest molecule and was found to modify the orientation of the incarcerated guest molecule. For example, reaction of carceplex 97»guest with thiourea led to formation of carceplex 98#guest.134 Incidentally, the carceplexes created by Reinhoudt et al. show a unique type of stereoisomerism (named carcerism) due to the restricted internal rotation of the incarcerated guest molecule.130 Reinhoudt's group has also created monolayers of their calix[4]arene-based carceplexes on a gold surface.135 Scheme 1.16. Synthesis of Calix[4]arene-Based Carceplex. 96 X = O, carceplex 97«guest X = S, carceplex 98#guest 53 vi. Endohedral Complexes of Fullerenes Fullerenes are spherical closed surface aromatic carbon compounds discovered in 1985 whose prototypical member is C6o 1 3 6 The founders of this entirely new area of chemistry, Kratos, Smalley and Curl, were awarded the 1996 Nobel prize in chemistry. The fullerenes represent one of the most intensely studied compounds of the last decade.1 3 7'1 3 8 Besides their promising applications as superconductors and semiconductors, fullerenes can encapsulate atoms such as noble gases139-140 and a large variety of metal ions. 1 4 1- 1 4 2 Difficulties in the extraction of endohedral complexes of fullerenes and their subsequent purification have created problems with their isolation. The endohedral complexes of larger fullerenes such as M@C82,' M@Cso, and M@C74 traditionally are easier to isolate than their M@C60 counter parts, but recently the use of aniline as an extraction solvent led to successful isolation of significant quantities of M@C60- 1 4 1 The formation of endohedral complexes of fullerenes with noble gases is traditionally done by heating samples of the fullerene in the presence of a nobel gas at high pressures (600 °C, 2500 atm., 5 hours).139 Also, mass spectroscopic collision experiments have been used to form the endohedral complexes of fullerenes and noble gases.139 The formation of endohedral complexes of fullerenes and various metals can be prepared by arc-heating MxOy/graphite rods in a low pressure helium atmosphere.141 X-ray diffraction has been used to confirm that the metal is indeed inside the fullerene.143 The interior of endohedral complexes of fullerenes are only large enough to entrap atoms or ions thus making them possibly the smallest carceplexes that can be formed. The promising properties of fullerenes will continue to keep them at the forefront of scientific research. 1 The symbolism M@Cg2 is used to represent a metal ion entrapped in the interior of fullerene (Cg2>. 54 vii. Related Compounds It is beyond the scope of this thesis to review all the molecular hosts that have the ability to bind molecules. Instead, the reader is referred to a number of recent reviews of the topic. 8 8' 1 2 9' 1 4 4 The hemicarceplexes are worthy of mention here because of their similarity to carceplexes and because this thesis presents work that builds on that done with hemicarceplex systems. Hemicarceplexes differ from carceplexes in that they can reorganize to create holes or portals which allows for egress of guests. By definition, hemicarceplexes must be kinetically stable at ambient temperatures to allow for their isolation and subsequent characterization. A large number of hemicarceplexes have been synthesized to date and this subject has recently been reviewed.27 Tetrol l a has been used to create a large portion of the known hemicarceplexes via the use of large inter-bowl bridges which include: o-xylene, naphthalene, hexamethylene, tetramethylene, and 2-butyne.27'129 Activation energies for decomplexation has been calculated for a number of such hemicarceplexes.27'129 The interior of hemicarceplexes allow for unique chemical reactions to occur where bulk solvent or external reagents can not directly participate.145"150 This environment has allowed for some unique chemistry, probably the most famous example being the room temperature stabilization of cyclobutadiene.150 The use of hemicarceplexes continues to provide exciting examples of unique chemistry. viii. One Pot Synthesis of Hemicarceplexes and a Carceplex. An example of a hemicarceplex with potential for new and interesting chemistry was recently synthesized by Paek et al. They described the one pot synthesis of a hemicarceplex with a functional group located at the portal of the host.151 The reaction between tetrol lc and triol 99 (1:1), bromochloromethane and potassium carbonate in 55 DMA as solvent gave a mixture of products which included carceplex 2c»DMA (14% yield), asymmetric hemicarceplex 100 # DMA (6% yield), and hemicarceplex 101»DMA (18% yield) as shown in Figure 1.16. Although related compounds of carceplex 2c«DMA3 and hemicarceplex 1 0 1 » D M A 1 5 2 had been synthesized previously in greater yields, this represented the first synthesis of a gate functionalized hemicarceplex ( 1 0 0 ) . The hydroxyl of asymmetric hemicarceplex 1 0 0 may act as a point for attachment of functional groups that may serve as a switch for complexation and decomplexation of encapsulated guest molecules; it may also be used to attach this compound to a polymer backbone to create new supramolecular materials. In Chapter 3, we further explore the driving forces for the formation asymmetric hemicarceplex 100»guest. Figure 1 .16 . Gate functionalized Hemicarceplex. carceplex 2c«guest R = ( C H 2 ) I Q C H 3 hemicarceplex 1 0 1 56 F. Conclusions This chapter has illustrated that the formation of sophisticated supramolecular assemblies is often facilitated by maximizing favorable noncovalent interactions between the reactants; template molecules are often used for this purpose. It is important for chemists to learn as much as possible about these noncovalent interactions because they dominate the properties of many biological assemblies, including enzymes, cell membranes and viruses. A wealth of information about noncovalent interactions is presently available but even more is required in order for chemists to develop supramolecular assemblies that approach the complexity of those found in natural systems. The study of self-assembling structures that can form cavities capable of molecular encapsulation is an expedient means of obtaining valuable information about the noncovalent interactions that governs self-assembly. The reversible nature of these systems allow examination of the thermodynamic properties of the system. Such information increases our knowledge of the importance of particular interactions such as hydrogen bonds, n-iz interactions and van der Waals interactions. This information can then be used to develop more complex assemblies of molecules with useful properties, and it can also be used to refine the parameters used for noncovalent interactions in computer modeling programs. 57 G. References 1. Sherman, J. C ; Cram, D. J. J. Am. Chern. Soc. 1989, 111, 4527-4528. 2. Cram, D. J.; Karbach, S.; Kim, Y. H.; Baczynskyj, L.; Kallemeyn, G. W. J. Am. Chern. Soc. 1985, 107, 2575-2576. 3. Sherman, J. C ; Knobler, C. B.; Cram, D. J. J. Am. Chern. Soc. 1991, 113, 2194-2204. 4. Nicolaou, K. C ; Sorensen, E. J. Classics in Total Synthesis; VCH: Weinheim, 1996. 5. Corey, E. J.; Cheng, X. M. The Logic of Chemical Synthesis; Wiley: New York, 1989. 6. Woodward, R. B. Pure Appl. Chern. 1973, 33, 145-177. 7. Armstrong, R. W.; Beau, J.-M.; Cheon, S. H.; Christ, W. J.; Fujioka, H.; Ham, W.-H.; Hawkins, L. D.; Jin, H.; Kang, S. H.; Kishi, Y.; Martinelli, M. J.; McWhorter, W. W., Jr.; Mizuno, M.; Nakata, M.; Stutz, A. E.; Talamas, F. X.; Taniguchi, M.; Tino, J. A.; Ueda, K.; Uenishi, J.-L; White, J. B.; Yonaga, M. J. Am. Chern. Soc. 1989, 111, 7525-7530. 8. Lehn, J.-M. Science 1985, 227, 849-856. 9. Lehn, J.-M. Angew. Chern. Int. Ed. Engl. 1988, 27, 89-112. 10. Lehn, J.-M. Angew. Chern. Int. Ed. Engl. 1990, 29, 1304-1319. 11. Wolf, K. L.; Frahm, H.; Harms, Z. Phys. Chern. Abt. B. 1937, 36, 17. 12. Desiraju, G. R. Angew. Chern. Int. Ed. Engl. 1995, 34, 2311-2327. 13. Fischer, E. Ber. Dtsch. Chern. Ges. 1894, 27, 2985-2993. 14. Whitesides, G. M.; Mathias, J. P.; Seto, C. T. Science 1991, 254, 1312-1319. 15. Lindsey, J. S. New J. Chern. 1991,15, 153-180. 16. Philp, D.; Stoddart, J. F. Angew. Chern. Int. Ed. Engl. 1996, 35, 1154-1196. 17. Fredricks, J. R.; Hamiliton, A. D. Chapter 16; Hydrogen Bonding Control of Molecular Self-assembly: Recent Advances in Design, Synthesis, and Analysis, Gokel, G. W., Vol. Ed.; in the series Comprehensive Supramolecular Chemistry; Lehn, J.-M.; Atwood, J. L.; Davies, J. E. D.; MacNicol, D. D.; Vogtle, F., Series Ed.; Pergamon: New York, 1996; Vol. 9, pp 565-594. 18. Bradley, D. Chern. Soc. Rev. 1995, 379-382. 19. Stryer, L. Biochemistry; 2nd ed.; W. H. Freeman: New York, 1988, pp 724-727. 58 20. Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 7017-7036. 21. Pedersen, C. J. Synthetic Multidentate Macrocyclic Compounds; Academic: New York, 1978. 22. Dietrich, B.; Lehn, J.-M.; Sauvage, J.-P. Tetrahedron Lett. 1969, 2885-2888. 23. Lehn, J.-M. Struct. Bonding (Berlin) 1973,16, 1-69. 24. Dietrich, B. Cryptate Complexes; Dietrich, B., Ed.; Academic: London, 1984; Vol. 2. 25. Cram, D. J. Angew. Chem. Int. Ed. Engl. 1986, 25, 1039-1057. 26. Cram, D. J. Science 1983, 219, 1177-1183. 27. Cram, D. J.; Cram, J. M. Container Compounds and Their Guests; The Royal Society of Chemistry: Cambridge, 1994; Vol. No. 4. 28. Busch, D. H. J. Inclusion Phenom. Mol. Recognit. Chem. 1992,12, 389-395. 29. Anderson, S.; Anderson, H. L.; Sanders, J. K. M. Acc. Chem. Res. 1993, 26, 469-475. 30. McMurray, T. J.; Raymond, K. N.; Smith, P. H. Science 1989, 244, 938-942. 31. Busch, D. H.; Vance, A. L.; Kolchinski, A. G. Chapter 1; Molecular Template Effect: Historical View, Principles, and Perspectives; Gokel, G. W., Vol. Ed.; in the series Comprehensive Supramolecular Chemistry; Lehn, J.-M.; Atwood, J. L.; Davies, J. E. D.; MacNicol, D. D.; Vogtle, F., Series Ed.; Pergamon: New York, 1996; Vol. 9, pp 1-42. 32. Schill, G. Catenanes, Rotaxanes and Knots; Acedemic: New York, 1971. 33. Chambron, J. C ; Dietrich-Buchecker, C. O.; Nierengarten, J. F.; Sauvage, J. P. Pure Appl. Chem. 1994,66, 1543-1550. 34. Kern, J.-M.; Sauvage, J.-P.; Weidmann, J.-L. Tetrahedron 1996, 52, 10921-10934. 35. Brown, C. L.; Philp, D.; Stoddart, J. F. Synlett 1991, 462-464. 36. Ashton, P. R.; Ballardini, R.; Balzani, V.; Belohradsky, M.; Gandolfi, M. T.; Philp, D.; Prodi, L.; Raymo, F. M.; Reddington, M. V.; Spencer, N.; Stoddart, J. F.; Venturi, M.; Williams, D. J. J. Am. Chem. Soc. 1996,118, 4931-4951. 37. Ballardini, R.; Balzani, V.; Credi, A.; Gandolfi, M. T.; Langford, S. J.; Menzer, S.; Prodi, L.; Stoddart, J. F.; Venturi, M.; Williams, D. J. Angew. Chem. Int. Ed. Engl. 1996, 35, 978-981. 38. Zhu, S. S.; Carroll, P. J.; Swager, T. M. J. Am. Chem. Soc. 1996,118, 8713-8714. 59 39. Ashton, P. R.; Claessens, C. G.; Hayes, W.; Menzer, S.; Stoddart, J. F.; White, A. J. P.; Williams, D. J. Angew. Chern. Int. Ed. Engl. 1995, 34, 1862-1865. 40. Amabilino, D. B.; Ashton, P. R.; Perez-Garcia, L.; Stoddart, J. F. Angew. Chern. Int. Ed. Engl. 1995, 34, 2378-2380. 41. Vogtle, F.; Jager, R.; Handel, M.; Ottens-Hildebrandt, S.; Schmidt, W. Synthesis 1996, 353-356. 42. Fox, S. W.; Dose, K. Molecular Evolution and the Origin of Life; Dekker: New York, 1977. 43. von Kiedrowski, G.; Wlotzka, B.; Helbing, J.; Matzen, M.; Jordan, S. Angew. Chern. Int. Ed. Engl. 1991, 30, 423-426. 44. von Kiedrowski, G.; Wlotzka, B.; Helbing, J. Angew. Chern. Int. Ed. Engl. 1989,28, 1235-1237. 45. von Kiedrowski, G. Angew. Chern. Int. Ed. Engl. 1986, 25, 932-935. 46. Wintner, E. A.; Conn, M. M.; Rebek, J., Jr. Acc. Chern. Res. 1994, 27, 198-203. 47. Wintner, E. A.; Conn, M. M.; Rebek, J., Jr. J. Am. Chern. Soc. 1994,116, 8877-8884. 48. Bolm, C ; Bienewald, F.; Seger, A. Angew. Chern. Int. Ed. Engl. 1996, 35, 1657-1659. 49. Orgel, L. E. Nature 1992, 358, 203-209. 50. Lee, D. H.; Granja, J. R.; Martinez, J. A.; Severin, K.; Ghadiri, M. R. Nature 1996, 382, 525-528. 51. Rebek, J., Jr. Chern. Rev. 1996, 96, 255-263. 52. Reinhoudt, D. N.; Rudkevich, D. M.; de Jong, F. J. Am. Chern. Soc. 1996,118, 6880-6889. 53. Hue, I.; Pieters, R. L; Rebek, J., Jr. / . Am. Chern. Soc. 1994, 116, 10296-10297. 54. Conn, M. M.; Wintner, E. A.; Rebek, L, Jr. Angew. Chern. Int. Ed. Engl. 1994, 33, 1577-1581. 55. Anderson, H. L.; Sanders, J. K. M. J. Chern. Soc, Perkin Trans. 1 1995,18, 2223-2229. 56. Anderson, H. L.; Anderson, S.; Sanders, J. K. M. J. Chern. Soc, Perkin Trans. I 1995,18, 2231-2245. 57. Anderson, S.; Anderson, H. L.; Sanders, J. K. M. J. Chern. Soc, Perkin Trans. 1 1995,18, 2247-2254. 60 58. Anderson, S.; Anderson, H. L.; Sanders, J. K. M. J. Chem. Soc., Perkin Trans. 1 1995,18, 2255-2267. 59. Mackay, L. G.; Anderson, H. L.; Sanders, J. K. M. J. Chem. Soc, Perkin Trans. 1 1995,18, 2269-2273. 60. Kaufmann, D. E.; Otten, A. Angew. Chem. Int. Ed. Engl. 1994, 33, 1832-1834. 61. Midler, A. J. Mol. Struct. 1994, 325, 13-35. 62. Miiller, A.; Krickemeyer, E.; Dillinger, S.; Bogge, H.; Stammler, A. J. Chem. Soc, Chem. Commun. 1994, 2539-2540. 63. Miiller, A.; Diemann, E.; Krickemeyer, E.; Che, S. Naturwissenschaften 1993, 80, 77-78. 64. Zheng, Z.; Knobler, C. B.; Hawthorne, M. F. J. Am. Chem. Soc. 1995,117, 5105-5113. 65. Wulff, G. Angew. Chem. Int. Ed. Engl. 1995, 34, 1812-1832. 66. Wulff, G. Template Induced Control of Stereochemistry for the Synthesis of Polymers; Wulff, G., Ed.; Kluger Academic Publishers: Netherlands, 1995; Vol. 473, pp 13-19. 67. Tanabe, K.; Takeuchi, T.; Matsui, J.; Ikebukuro, K.; Yano, K.; Karube, I. J. Chem. Soc, Chem. Commun. 1995, 2303-2304. 68. Wulff, G.; Heide, B.; Helfmeier, G. J. Am. Chem. Soc. 1986,108, 1089-1091. 69. Chailapakul, O.; Crooks, R. M. Langmuir 1993, 9, 884-888. 70. Chailapakul, O.; Crooks, R. M. Langmuir 1995,11, 1329-1340. 71. Wulff, G. Biorecognition in Molecular Imprinted Polymers: Concept, Chemistry and Applications; Wulff, G., Ed.; Plenum: New York, 1993, pp 363-381. 72. Andersson, L. I.; Ekberg, B.; Mosbach, K. Bioseparation and Catalysis in Molecular Imprinted Polymers; Andersson, L. I.; Ekberg, B.; Mosbach, K., Ed.; Plenum: New York, 1993, pp 383-394. 73. Gamez, P.; Dunjic, B.; Pinel, C ; Lemaire, M. Tetrahedron Lett. 1995, 36, 8779-8782. 74. Hedborg, E.; Winquist, F.; Lundstrom, I.; Andersson, L. I.; Mosbach, K. Sensors Actuators A 1993, 37-38, 796-799. 75. Mathew-Krotz, J.; Shea, K. J. J. Am. Chem. Soc. 1996,118, 8154-8155. 76. Simard, M.; Su, D.; Wuest, J. D. / . Am. Chem. Soc. 1991,113, 4696-4698. 77. Zones, S. I.; Nakagawa, Y.; Yuen, L. T.; Harris, T. V. J. Am. Chem. Soc. 1996, 118, 7558-7567. 61 78. Xu, T.; Munson, E. J.; Haw, J. F. J. Am. Chern. Soc. 1994,116, 1962-1972. 79. Pitchumani, K.; Warder, M.; Cui, C ; Weiss, R. G.; Ramamurthy, V. Tetrahedron Lett. 1996,37, 6251-6254. 80. Pitchumani, K.; Corbin, D. R.; Ramamurthy, V. J. Am. Chern. Soc. 1996,118, 8152-8153. 81. Lewis, D. W.; Willock, D. J.; Catlow, C. R. A.; Thomas, J. M.; Hutchings, G. J. Nature 1996, 382, 604-606. 82. Martin, C. R. Acc. Chern. Res. 1995, 28, 61-68. 83. Whitesides, G. M.; Simanek, E. E.; Mathias, J. P.; Seto, C. T.; Chin, D. N.; Mammen, M.; Gordon, D. M. Acc. Chern. Res. 1995, 28, 37-44. 84. Fan, E.; Yang, J.; Geib, S. L; Vicent, C ; Garcia-Tellado, F.; Tecilla, P.; Hamilton, A. D. Macromol. Symp. 1994, 77, 209-217. 85. Vreekamp, R. H.; van Duynhoven, J. P. M.; Hubert, M.; Verboom, W.; Reinhoudt, D. N. Angew. Chern. Int. Ed. Engl. 1996, 35, 1215-1218. 86. Garel, L.; Dutasta, J.-P.; Collet, A. Angew. Chern. Int. Ed. Engl. 1993, 32, 1169-1171. 87. Breslow, R. Acc. Chern. Res. 1991, 24, 159-164. 88. Rebek, L, Jr. Pure Appl. Chern. 1996, 68, 1261-1266. 89. Kang, J.; Rebek, J. J. Nature 1996, 382, 239-241. 90. Meissner, R. S.; Rebek, J. J.; de Mendoza, J. Science 1995, 270, 1485-1488. 91. Branda, N.; Wyler, R.; Rebek, J. J. Science 1994, 263, 1267-1268. 92. Branda, N.; Grotzfeld, R. M.; Valdes, C ; Rebek, J. J. J. Am. Chern. Soc. 1995, 117, 85-88. 93. Shimizu, K. D.; Rebek, J., Jr. Proc. Natl. Acad. Sci. 1995, 92, 12403-12407. 94. Hamann, B. C ; Shimizu, K. D.; Rebek, J., Jr. Angew. Chern. Int. Ed. Engl. 1996,55, 1326-1329. 95. Castellano, R. K.; Rudkevich, D. M.; Rebek, J., Jr. / . Am. Chern. Soc. 1996, 118, 10002-10003. 96. Mogck, O.; Bohmer, V.; Vogt, W. Tetrahedron 1996, 52, 8489-8496. 97. Scheerder, J.; Vreekamp, R. H.; Engbersen, J. F. J.; Verboom, W.; van Duynhoven, J. P. M.; Reinhoudt, D. N. J. Org. Chern. 1996, 61, 3476-3481. 98. Bonar-Law, R. P.; Sanders, J. K. M. Tetrahedron Lett. 1993, 34, 1677-1680. 99. Koh, K.; Araki, K.; Shinkai, S. Tetrahedron Lett. 1994, 35, 8255-8258. 62 100. Vreekamp, R. H.; Verboom, W.; Reinhoudt, D. N. J. Org. Chem. 1996, 61, 4282-4288. 101. Jeon, Y.-M.; Kim, J.; Whang, D.; Kim, K. / . Am. Chem. Soc. 1996,118, 9970-9971. 102. Cintas, P. J. Inclusion Phenom. Mol. Recognit. Chem. 1994,17, 205-220. 103. Szejtli, J. Cyclodextrins and Their Inclusion Complexes; Akademiai Kiado: Budapest, 1982. 104. Maletic, M.; Wennemers, H.; MacDonald, D. Q.; Breslow, R.; Still, W. C. Angew. Chem. Int. Ed. Engl. 1996, 35, 1490-1492. 105. Lawrence, D. S.; Jiang, T.; Levett, M. Chem. Rev. 1995, 95, 2229-2260. 106. Jiang, T.; Li, M.; Lawrence, D. S. J. Org. Chem. 1995, 60, 7293-7297. 107. Venema, F.; Rowan, A. E.; Nolte, R. J. M. J. Am. Chem. Soc. 1996, 118, 257-258. 108. Mosseri, S.; Mialocq, J. C ; Perly, B.; Hambright, P. J. Phys. Chem. 1991, 95, 4659-4663. 109. Breslow, R.; Halfon, S.; Zhang, B. Tetrahedron 1995, 51, 377-388. 110. Breslow, R.; Zhang, B. J. Am. Chem. Soc. 1996,118, 8495-8496. 111. Wenz, G. Angew. Chem. Int. Ed. Engl. 1994, 33, 803-822. 112. Fuchs, R.; Habermann, N.; Klufers, P. Angew. Chem. Int. Ed. Engl. 1993, 32, 852-854. 113. Hibbert, F.; Emsley, J. Advances in Physical Organic Chemistry 1990, 26, 255-379. 114. Emsley, J. Chem. Soc. Rev. 1980, 9, 91-124. 115. Novak, A. Struct. Bonding 1974,18, 177-216. 116. Klufers, P.; Schuhmacher, J. Angew. Chem. Int. Ed. Engl. 1994, 33, 1863-1865. 117. Yoshida, Z.-L; Takekuma, H.; Takekuma, S.-L; Matsubara, Y. Angew. Chem. Int. Ed. Engl. 1994,53, 1597-1599. 118. Atwood, J. L.; Koutsantonis, G. A.; Raston, C. L. Nature 1994, 368, 229-231. 119. Araki, K.; Akoa, K.; Ikeda, A.; Suzuki, T.; Shinkai, S. Tetrahedron Lett. 1996, 37, 73-76. 120. Steed, J. W.; Junk, P. C ; Atwood, J. L. J. Am. Chem. Soc. 1994,116, 10346-10347. 63 121. Kikuchi, Y.; Tanaka, Y.; Sutarto, S.; Kobayashi, K.; Toi, H.; Aoyama, Y. / . Am. Chem. Soc. 1992,114, 10302-10306. 122. Gibb, B. C ; Chapman, R. G.; Sherman, J. C. J. Org. Chem. 1996, 61, 1505-1509. 123. Tunstad, L. M.; Tucker, J. A.; Dalcanale, E.; Weiser, J.; Bryant, J. A.; Sherman, J. C ; Helgeson, R. C ; Knobler, C. B.; Cram, D. J. J. Org. Chem. 1989, 54, 1305-1312. 124. Timmerman, P.; Verboom, W.; Reinhoudt, D. N. Tetrahedron 1996, 52, 2663-2704. 125. Weinelt, F.; Schneider, H.-J. J. Org. Chem. 1991, 56, 5527-5535. 126. Maverick, E.; Cram, D. J. Chapter 12; Carcerands and Hemicarcerands: Hosts that Imprison Molecular Guests; Vogtle, F. Vol. Ed.; in the series Comprehensive Supramolecular Chemistry; Lehn, J.-M.; Atwood, J. L.; Davies, J. E. D.; MacNicol, D. D.; Vogtle, F., Ed.; Pergamon: New York, 1996; Vol. 2, pp 367-418. 127. Bryant, J. A.; Blanda, M. T.; Vincenti, M.; Cram, D. J. J. Chem. Soc, Chem. Commun. 1990, 1403-1405. 128. Bryant, J. A.; Blanda, M. T.; Vincenti, M.; Cram, D. J. J. Am. Chem. Soc. 1991,113, 2167-2172. 129. Sherman, J. C. Tetrahedron 1995, 51, 3395-3422. 130. Timmerman, P.; Verboom, W.; van Veggel, F. C. J. M.; van Duynhoven, J. P. M.; Reinhoudt, D. N. Angew. Chem. Int. Ed. Engl. 1994, 33, 2345-2348. 131. Fraser, J. R.; Borecka, B.; Trotter, J.; Sherman, J. C. J. Org. Chem. 1995, 60, 1207-1213. 132. Helgeson, R. C ; Knobler, C. B.; Cram, D. J. J. Chem. Soc, Chem. Commun. 1995, 307-308. 133. Jung, J.; Ihm, H.; Paek, K. Bull. Korean Chem. Soc. 1996,17, 553-556. 134. van Wageningen, A. M. A.; van Duynhoven, J. P. M.; Verboom, W.; Reinhoudt, D. N. / . Chem. Soc, Chem. Commun. 1995, 1941-1942. 135. Huisman, B.-H.; Rudkevich, D. M.; van Veggel, F. C. J. M.; Reinhoudt, D. N. J. Am. Chem. Soc. 1996,118, 3523-3524. 136. Kroto, W. H.; Heath, J. R.; O'Brien, S. C ; Curl, R. F.; Smalley, R. E. Nature 1985,318, 162-163. 137. Goroff, N. Acc. Chem. Res. 1996, 29, 77-83. 64 138. Avent, A. G.; Birkett, P. R.; Christides, C ; Crane, J. D.; Darwish, A. D.; Hitchcock, P. B.; Kroto, H. W.; Meidine, M. F.; Prassides, K.; Taylor, R.; Walton, D. R. M. / . Mol. Struct. 1994, 325, 1-11. 139. Patchkovskii, S.; Thiel, W. J. Am. Chern. Soc. 1996,118, 7164-7172. 140. Saunders, M.; Jimenez-Vazquez, H. A.; Cross, R. J. / . Am. Chern. Soc. 1994, 116, 2193-2194. 141. Kubozono, Y.; Maeda, H.; Takabayashi, Y.; Hiraoka, K.; Nakai, T.; Kashino, S.; Emura, S.; Ukita, S.; Sogabe, T. J. Am. Chern. Soc. 1996,118, 6998-6999. 142. Braun, T. ACE - Models Chern. 1995,132, 245-263. 143. Takata, M.; Umeda, B.; Nishibori, E.; Sakata, M.; Saito, Y.; Ohno, M.; Shinohara, H. Nature 1995, 377, 46-48. 144. Izatt, R. M.; Pawlak, K.; Bradshaw, J. S. Chern. Rev. 1995, 95, 2529-2586. 145. Farran, A.; Deshayes, K. D. / . Phys. Chern. 1996,100, 3305-3307. 146. Farran, A.; Deshayes, K.; Matthews, C ; Balanescu, I. J. Am. Chern. Soc. 1995, 117, 9614-9615. 147. Parola, A. J.; Pina, F.; Maestri, M.; Armaroli, N.; Balzani, V. New J. Chern. 1994,18, 659-661. 148. Pina, F.; Parola, A. J.; Ferreira, E.; Maestri, M.; Armaroli, N.; Ballardini, R.; Balzani, V. J. Phys. Chern. 1995, 99, 12701-12703. 149. Robbins, T. A.; Cram, D. J. J. Am. Chern. Soc. 1993,115, 12199. 150. Cram, D. J.; Tanner, M. E.; Thomas, R. Angew. Chern. Int. Ed. Engl. 1991, 30, 1024-1027. 151. Paek, K.; Joo, K.; Kim, M.; Kim, Y. Bull. Korean Chern. Soc. 1995,16, 411-478. 152. Cram, D. J.; Tanner, M. E.; Knobler, C. B. J. Am. Chern. Soc. 1991,113, 11X1-1121. 65 2. Template Effects in Formation of a Carceplex A. Introduction Carceplex 2a«guest was originally prepared by Cram et al. via a coupling reaction between two molecules of tetrol la with bromochloromethane in 49-61% yield (Scheme 2.I).1-2 The high yields obtained in these reactions is remarkable, given that a total of seven molecules are brought together forming eight new C-0 bonds (see Chapter 1, section Eiii). Furthermore, not only is the presence of a template molecule, such as DMSO, a prerequisite for formation of carceplexes, but also these reactions exhibit some degree of selectivity toward the template molecule itself.1"3 For example, a 5:1 ratio of carceplex 2a»DMA and carceplex 2a»DMF resulted when the reaction was run in a 1:1 molar ratio of DMA and DMF. 2 Scheme 2.1. First Soluble Carceplexes. ' K C4 axis •>- C2 axis Ph Ph Ph Ph Tetrol la Carceplex 2a*Guest It was these early accounts from the UCLA group that drew our attention to this area of supramolecular chemistry. Namely, we wanted to explore the template 66 requirements and template selectivity found in the reaction to form carceplex 2a»guest as depicted in Figure 2.1. For example, what types of molecules would be successful templates for formation of carceplex 2a»guest (Figure 2.1)? Are amides and sulfoxides the only suitable templates? How big, small, polar or nonpolar can template molecules be? The accessibility of carceplex 2a#guest and its facile characterization by ] H NMR made these assemblies particularly attractive models for studying noncovalent interactions between molecules. With this in mind, we began to address these questions by expanding the range of templates molecules that led to formation of carceplex 2a#guest. This was achieved by doping a solvent that is itself not a suitable template with a small amount of a suitable template for the formation of carceplex 2a»guest, as described in the present chapter. Figure 2.1. Template Requirements for Carceplex 2a»Guest Formation. R R R R R R polymer, decomposition R = C H 2 C H 2 P h 67 B. Results and Discussion i. Selectivity in the Formation of Carceplex 2a»guest a Screening for Suitable Template Molecules The relatively large dipolar l-methylpyrrolidin-2-one (NMP) was chosen as the bulk solvent for screening potential template molecules because (1) it is polar enough to allow for formation of carceplex 2a«guest and (2) it is itself a poor template molecule.4 Using conditions" slightly modified from those described in the literature,2 seventy potential template molecules were screened by incorporating them as co-solvent (~ 5 mole % based on NMP) into the reaction mixture and the reaction was run over two days at 60 °C. Of these 70 template molecules, 40 molecules were found to be suitable templates for formation of carceplex 2a«guest (Table 2.1) whilst 30 molecules were unsuitable templates (Table 2.2). Guests that did not lead to formation of carceplex 2a»guest were classified in terms of their size, basicity or polarity as summarized in Table 2.2. The failure of the basic secondary amines to form carceplex 2a«guest, however, may not be due to their unsuitability as template molecules but could be due to degradation of the bridging material bromochloromethane during the reaction. To eliminate the possibility of the latter, we performed a series of control experiments with 1 mol % pyrazine and 1 mol % of the four secondary amines as guests for formation of carceplex 2a»guest, which resulted in formation of carceplex 2a»pyrazine. Thus, the inability of secondary amines to form " Typical Conditions were 0.1 mmol of tetrol la, 10 mmol of potassium carbonate, 1 mmol of bromochloromethane, 50 mmol of guest (5 mole % based on NMP) in 100 mL of NMP. The reaction mixture was heated to 60 °C for two days. High dilution conditions were unnecessary and the less expensive potassium carbonate worked just as well as cesium carbonate. 68 carceplex 2a»secondary amines is a result of these guests being unsuitable templates and not due to degradation of bromochloromethane by the amines. T a b l e 2.1. Successful Template Molecules O W / Q O O O U 0 ^ 0 o o C H / C H , O. .0 C H 3 / C H 3 C H 3 ' C H 3 O O O o o C H 3 v X OH N ' T H , J \ I 3 C H 3 ^ CH 3 CH 3 C H , 1 . N H C H 3 C H 3 O N C H . H ^ C H ^ ^ O) C H 3 \ C H 3 Q s N - C H 3 iOS C H ^ O ^ C H s C H C H 3 C N CH 3CH 2OH CH 2C1 2 C H 2 C l B r C H 2 I 2 The boxed molecules formed carceplex 2a»guest but were characterized as mixtures: 69 Table 2.2. Unsuccessful Guests. Too Large c V^ c H 3 1 o 0° C H 3 O N C H 3 0-ci o C 2 H 5 O / \ H 3 co OCH3 R F C 2 H 5 s A 2 N H C 2 H 5 o / ~ ~ \ CI Cl O , . J~0 o CC1 4 F F Too Polar Too Apolar Too Basic O N H ' ^ N H HO O H H 2 0 N H Q N H \ / b. Competition Experiments Between Successful Template Molecules The diverse range of incorporated guest molecules (Table 2.1) prompted us to investigate further their templating abilities. To do this it would be necessary to directly compare two template molecules head-to-head in a series of competition experiments, and determine their product ratios. Competition experiments (Scheme 2.2) were performed on 34 of the successful template molecules by giving the carceplex a choice of entrapping two different guest molecules during its formation (the results for the four dihalomethane guests 70 and CHCI3 are presented separately in Chapter 3; competition experiments were not performed on tetrahydropyran). Generally, the two template molecules were added to the reaction mixture (1 mole % guests based on NMP for conditions A and 5 mole % guests for conditions B) and the carceplex reaction was run as before. After completion of the reaction, the mixture of two carceplex products were isolated from polymeric side products, and the product ratios for each pair were determined by integration of the unique host and guest peaks in the lH NMR spectra. The competition reactions provided a hierarchical listing of the template molecules studied, in terms of their abilities as templates. With this template map in hand, further competition experiments were performed between guests that had similar template abilities. Thus, a precise ordering of the templates, according to their "template ratios" was determined.111 The 34 successful template molecules, their yields, and their template ratios are listed in Table 2.3. Each of the template ratios were referenced to the poorest template (NMP) which was arbitrarily given a value of l . 4 m The individual template ratios that were determined along with a number of cross check experiments are listed in the experimental section at the end of this chapter (page 125). A sample calculation is also provided. 71 Scheme 2.2. The Competition Reaction. 2 carceplex 2a»guest#2 R = CH 2 CH 2 Ph 72 Table 2.3. Carceplex 2a»Guest Yields and Competition Experiments. Guest yield % a Template Ratiob conditions13 1 pyrazine 87 1,000,000 A 2 methyl acetate 75 470,000 A 3 1,4-dioxane 68 290,000 A 4 dimethyl sulfide 52 180,000 A 5 ethyl methyl sulfide 67 130,000 A 6 dimethyl carbonate 52 73,000 A 7 DMSO 63 70,000 A 8 1,3-dioxolane 64 38,000 A 9 2-butanone 75 37,000 A 10 pyridine 46 34,000 A 11 dimethyl sulphone 60 19,000 A 12 1,4-thioxane 55c 14,000 A 13 2,3-dihydrofuran 38 13,000 A 14 furan 54 12,000 A 15 tetrahydrofuran 50 12,000 A 16 pyridazine 30 8,600 A 17 acetone 51 6,700 A 18 thiophene 23 5,800 A 19 1,3-dithiolane 43 4,400 A 20 ± 2-butanol 47 2,800 A 21 benzene 43 2,400 A 22 2-propanol 74 1,500 A 23 pyrrole 73 1,000 B 24 tetrahydrothiophene 34 410 B 25 1,3-dioxane 45 200 B 26 acetamide 26 160 B 27 trioxane 24 100 B 28 acetonitrile 35 73 B 29 ethanol 38 61 B 30 ethyl acetate 17 45 B 31 diethyl ether 14d 21 B 32 dimethylacetamide 15 20 B 33 dimethylformamide 4 7 B 34 NMP 5e 1 B a Yield refers to the reaction run with only one guest (see experimental). b Conditions A: 1 mol % guests, 2 days 60 °C. Conditions B: 5 mol % guests, 1 day ambient temperature, 2 days 60 °C (see experimental). Errors in each template ratio are estimated at < ±20%; for more on errors see cross check experiments in the text and in the experimental. c Characterized as a mixture of carceplex 2a» 1,4-thioxane (86%) and carceplex 2a« 1,4-dioxane (14%). d Characterized as a mixture of carceplex 2a«diethyl ether (96%) and carceplex 2a»bromochloromethane (4%). e Reaction run in neat NMP. 73 These competition experiments revealed that the reaction to form carceplex 2a»guest is subject to a template effect such that the best guest (pyrazine) is one million times more efficient as a template than the poorest guest (NMP).4 Thus, template ratios reflect the relative rates of the guest determining step (GDS); the GDS is the step in the carceplex reaction where the guest becomes entrapped under the conditions used for carceplex formation. So, for example, the transition state for the GDS is 8.3 kcal/mol lower in energy in the presence of pyrazine than in the presence of NMP, which results in the 10^  fold difference in their template or product ratios.1V The superior templating ability of pyrazine over that of NMP is also illustrated in the 75% yield of carceplex 2a«pyrazine and < 1 % yield of carceplex 2a«NMP when the reaction was carried out with only a stoichiometric amount of pyrazine (0.01 mol % based on NMP). c. Template Trends for Formation of Carceplex 2a*Guest The template ratios determined in Table 2.3 and the unsuccessful guests listed in Table 2.2 demonstrate some definite trends for formation of carceplexes. Many of these trends are discussed further in Chapter 3. First, there are definite size restrictions for incarceration in carceplex 2a*guest. For instance, the largest successful template, NMP, is the only template to have seven non-hydrogen atoms, whilst all other successful template molecules contain six or less non-hydrogen atoms. For the unsuccessful templates, the majority are simply too large for the interior of the carceplex. It is fairly remarkable just how selective the carceplex is towards the size of its guest molecule. For example, benzene is a reasonably good template molecule, but fluorobenzene exhibits no template effect; carceplex formation is apparently sensitive to the substitution of a hydrogen (van der Waals 1V The 8.3 kcal/mol difference in activation energy is calculated using the following equation: AAG* = -RTln(kp V r a z i n e /kNMP)> where kjsiMP 1 S 'he rate constant for the formation of carceplex 2a*NMP and kpyrazine is the rate constant for the formation of carceplex 2a»pyrazine; k p y r a z i n e / k N M P = 106-74 radius =1.2 A) 5 for a fluorine (van der Waals radius = 1.47 A), although this differentiation could additionally be electronic in nature.5 Furthermore, 1,4-dioxane is 21 times better than the slightly larger 1,4-thioxane. The substitution of the second sulfur for an oxygen in the case of 1,4-thioxane to 1,4-dithiane results in a unsuccessful template. The formation of carceplexes is very sensitive to the addition of a single methylene. For example, methyl acetate is 10,000 times better than ethyl acetate, and 1,3-dioxolane is 190 times better than 1,3-dioxane. The addition of an oxygen can also create differences in template abilities. For example, dimethyl sulfide is 2.5 times better than DMSO which is itself 3.7 times better than dimethyl sulphone. Also note, the template abilities of THF, 2,3-dihyrdofuran, and furan are very similar indicating that the addition of hydrogens to furan does not effect the template ability of the guest. Possibly, the small size of furan does not fill the cavity of the forming carceplex allowing for additional substitution of hydrogens without affecting the template abilities. Naturally, electronics may play a role; furan may have favorable %-TI and CH-n interactions with the forming cavity, while tetrahydrofuran may have more favorable van der Waals interactions. In contrast to furan-related guest molecules, substantial differences in template abilities exist between pyrazine, pyridine and benzene which only differ by the substitution of a nitrogen(s) for a C-H(s). Pyrazine is 29 times better template than pyridine which is itself 14 times better than benzene; possibly the introduction of an extra atom (hydrogen) causes enough steric strain in the transition state of the GDS to account for these differences in template abilities/ Second, polarity of the template molecules is also an important factor. Though the polarity of the successful templates spans a diverse range from that of DMSO (er = 46)6 to that of benzene (er = 2)6, both highly polar molecules such as water and highly apolar molecules such as cyclopentane are unsuitable templates. v The selectivity difference between pyrazine, pyridine and benzene could also be electronic in nature. 75 Third, high symmetry tends to increase the templating power of guests as is evident with 1,4-dioxane which is a 1400 times better template for the carceplex formation than 1,3-dioxane. Similarly, pyrazine (1,4-diazabenzene) is 120 times better than pyridazine (1,2-diazabenzene). Finally, cyclic guests are much better than acyclic guests. For example, 1,4-dioxane is the third best template while dimethoxyethane is an unsuitable template. Furthermore, THF is 570 times better than diethyl ether. The superior templating ability of cyclic guest molecules relative to acyclic guest molecules may be due to the greater loss of freedom that acyclic guests have to endure (entropically disfavored) in the cavity of the forming carceplex. Overall, the template ability of guest molecules for carceplex formation undoubtedly depends upon maximizing favorable van der Waals, electrostatic, 7C-7C and CH-7C interactions while minimizing unfavorable steric strain between the template molecule and the interior of the forming carceplex. Another general trend that was observed from Table 2.3 is that the yields usually increase with increasing template abilities of the guest molecules. For example, the best guest (pyrazine) has the highest yield and the poorest guests (NMP and DMF) have the lowest yields. There are many exceptions to this trend. Therefore, the yield alone is not strictly indicative of the template ability of the guest molecules for formation of carceplex 2a«guest. For example, thiophene gives a lower yield than pyrrole even though thiophene is a better template. This may be due to a small fraction of the guest (e.g., thiophene) slowly reacting with bromochloromethane thus degrading the bridging reagent. This would not affect the template ratios determined because the yields of both carceplex 2a»thiophene and carceplex 2a»guest would be reduced by an equal amount. Furthermore, lower than expected yields may also be the result of a slower rate of reaction for formation of carceplexes subsequent to the GDS. For example, large guest molecules (e.g., benzene) may cause steric distortion in the host thereby misaligning the phenoxides resulting in slow formation of the final bridge of the forming carceplex; this would allow polymerization to 76 compete effectively with formation of the carceplex. If this does occur, it would directly affect the template ratios determined for guests such as benzene. For the worst case, benzene, the 30% lower yield compared to 2-propanol might manifest itself in a lower apparent ratio by as much as 30% since 30% of the benzene carceplex intermediates may polymerize. This yield is only slightly outside of our 20% error. There was no correlation found between our template ratios (Table 2.3) and solvent parameters such as acceptor number, dielectric constant,6 dipole moment,6 Ej£ or Hildebrand's 5.7 The graph of ln (template ratios) versus these five solvent parameters results in correlations of r < 0.14 (three of these plots are shown in Figure 2.2). Therefore, the polarity of the template is not a dominant factor in the template abilities of the guest molecules (solvophobic effects are addressed in Chapter 3). Additionally, there were no apparent similarities between template ratios and the theoretical MM2 calculated energies for carceplex 2a»guest.4 More sophisticated theoretical studies of carceplex 2»guest and related compounds were performed through a collaborative effort with Ken Houk's group at UCLA the results of which will be discussed in the next chapter. We also demonstrate in the next chapter that the template ratios for carceplex formation correlate with the selectivity observed for formation of a reversible complex of tetrol lb. 77 Figure 2.2. Graph of Solvent Parameters versus ln(Template Ratios). Graph of Acceptor Number and Dielectric Constant versus ln(Template Ratios) 0.0 2.0 4.0 6.0 8.0 10.0 ln(Template Ratios) • acceptor number • dielectric constant 12.0 14.0 ET versus ln(Template Ratios) 0.7 j 0.6 l-< <D O 0.5 --s & 0.4 -OH <> •4—* C 0.3 <D j> "3 0.2 -0.1 -0 --0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 ln(Template Ratios) d. Template Molecules That Led to Mixtures A number of the carceplexes in Table 2.3 were initially isolated as a mixture with carceplex 2a«NMP. These guests included: acetamide, trioxane, acetonitrile, ethanol, ethyl 78 acetate, dimethylacetarnide and dimethylformamide. These mixtures, however were readily separated via silica gel chromatography to give carceplex 2a«NMP and the respective carceplex 2a»guest. All carceplexes characterized in this thesis have the same Ryin 3:1 CHCl3:hexanes except for carceplexes 2a»NMP and 2a»DMA. It is believed the that large size of NMP and DMA create steric distortion of the carceplex shell which allows for their separation from other carceplexes 2a»guest via careful silica gel chromatography. This is consistent with the crystal structure of carceplex 2a»DMA which indicates that DMA causes steric distortion of the carceplex shell as shown later in this chapter (page 92). There were also a number of guest molecules that led to an inseparable mixture of carceplexes. CHCI3, diethyl ether, and tetrahydropyran furnished an inseparable mixture of carceplexes 2a»guest, 2a»CH2BrCl, and 2a»NMP. Although carceplex 2a»NMP could be separated from these mixtures, carceplex 2a»CH2BrCl could not be separated. Generally, a guest molecule is a fairly poor template when the small amount of CH 2BrCl present in the reaction (1 mmol Cf^BrCl compared to 50 mmol guest) effectively competes with it. The carceplex 2a»CH2BrCl impurity in principle can be avoided by use of a larger bridging material such as methylene ditosylate (MDT).8-9 Nevertheless, carceplexes 2a»CHCl3, 2a»tetrahydropyran, and 2a»diethyl ether were characterized as mixtures by ! H NMR and matrix assisted laser desorption ionization (MALDI) mass spectrometry. Carceplex 2a»diethyl ether was included in our competition experiments because it only yielded a small impurity (< 4%) of carceplex 2a»CH2BrCl while carceplexes 2a»tetrahydropyran and 2a»CHCl3 were not included because of the large amount (> 20%) of the carceplex 2a»CH2BrCl impurity. The reactions carried out with 1,4-thioxane as guest also resulted in an inseparable mixture of carceplexes 2a» 1,4-thioxane and 2a» 1,4-dioxane (7:1 respectively), because 1,4-thioxane contains a small amount of 1,4-dioxane as an impurity (-1%). The superior templating ability of 1,4-dioxane over that of 1,4-thioxane (see Table 2.3) for carceplex formation gives 14% yield of carceplex 2a» 1,4-dioxane. 1,4-Thioxane was, nonetheless, 79 included in our competition experiments and the dynamic lH NMR spectra of carceplex 2a»l,4-thioxane is addressed in detail in Chapter 4. There are also a number of carceplexes 2a*guest that were characterized only by !H NMR spectroscopy that deserve mention in this thesis. Imidazole, 1,2,4-triazole, 2,3-dihydropyran, and cyclopentadiene all gave carceplexes 2a»guest in < 5% yields. Also, methanol gave a mixture of carceplexes 2a»guest in < 5% yield; lH NMR of this material indicated the majority was carceplex 2a«NMP and carceplex 2a»bromochloromethane, but there were also signals at -0.58 and -0.62 ppm that could not be assigned and may be due traces of entrapped MeOH. Although much had been discovered about the formation of carceplex 2a»guest, many questions remain unanswered. For example, which step is the guest determining step? ii. The Guest Determining Step a Reaction Intermediates in the Formation of Carceplex 2a»Guest The reaction to form carceplex 2a«guest is an excellent model for studying noncovalent interactions between molecules because small changes in template molecules result in large changes in the propensity for incarceration. The selectivity observed in this reaction is the result of relative rates of carceplex 2a»guest formation at the guest determining step (GDS, see page 74). Perhaps the most obvious question that arises is which step in this reaction is the GDS? Is it formation of the fourth O -CH2 -O inter-bowl bridge? In order to address these questions, we needed to learn more about the intermediates formed in the carceplex reaction. 80 b. Disappearance of Tetrol We were interested to see if the selectivity observed for the formation of carceplexes was the result of the formation of the first inter-bowl (O-CH2-O) bridge. To answer this question, we looked at the rate of formation of the first bridge in the carceplex reaction to form the monobridged intermediate by following the disappearance of the starting material (tetrol la) using both our best guest (pyrazine) and our poorest guest (NMP). We found that the disappearance of tetrol was much faster in the presence of 1 mol % pyrazine in NMP than in neat NMP (Table 2.4). This indicates that pyrazine accelerates the formation of the first bridge in the carceplex reaction relative to that of NMP. (The formation of each bridge is presumably irreversible.) Apparently, favorable van der Waals and electrostatic interactions between the walls of the forming carceplex and the template result in a lowering of the activation energy for the formation of the first bridge in the presence of pyrazine relative to NMP. However, these findings do not provide information about the actual GDS. We therefore focused our attention on the isolation of reaction intermediates. Table 2.4. Disappearance of Tetrol la at 25 °C Time (hr.) % Tetrol Remaining % Tetrol Remaining Guest = Pyrazine Guest = NMP 6 13 83 12 0 80 48 0 10 Initially, we attempted to isolate these carceplex reaction intermediates by performing the reaction under our usual conditions except the reaction was stopped prematurely. Evaluation of the reaction mixture indicated that the product mixture consisted predominately of unreacted tetrol la and product carceplex 2a»guest. No other intermediates were isolated. Thus, in order to isolate carceplex reaction intermediates, we would have to design a synthetic pathway that would favor the formation of these 81 intermediates, for example by protection of tetrol followed by bridging and deprotection. Alternatively, we performed the carceplex reaction in various solvents using various bases in the hope of obtaining intermediates. c. The Importance of Base in the Formation of Carceplexes. The two obvious questions that needed to be addressed here were (i) Are carbonates the only suitable bases? (ii) What is the limiting pK a range of these suitable bases? Additionally, do metal cations play any significant role in the formation of carceplexes by forming salt bridges between two tetrol molecules prior to covalent bond formation. To answer these questions a series of standardized reactions were run in NMP using pyrazine as the template molecule and bromochloromethane as the bridging reagent, and in the presence of various organic and inorganic bases (Table 2.5). The results summarized in Table 2.5 indicate that weak bases such as pyrazine, NaOAc and pyridine are too weak to lead to carceplex formation whereas exceedingly strong bases such as NaH are also unsuitable for carceplex formation. The successful bases for carceplex 2a»pyrazine formation range in pK a from 6-19 (pKa values of conjugate acids in H2O). Although the amine bases triethylamine and morpholine fall in the correct pK a range, they are not suitable for formation of carceplex 2a»guest. This may be the result of triethylamine and morpholine being too weakly basic in NMP, the solvent used for the reaction. The successful formation of carceplex 2a#pyrazine with l,8-diazabicyclo[5.4.0]undec-7-ene (DBU) as base indicates that metal salt bridges are not necessary for carceplex formation. Moreover, this discovery is particularly noteworthy in that it greatly expands the number of potential solvent systems that may now be explored for carceplex formation due to the greater solubility of DBU in organic solvents. 82 Table 2.5. Effect of Base on Formation of Carceplex 2a«guest. Base # equivalents of Base pK a* % yield carceplex 2a*pyrazine pyrazine 60 0.65 0, recovered tetrol NaOAc 30 5 0, recovered tetrol pyridine 30 5.25 0, recovered tetrol NaHC0 3 30 6.35 78 morpholine 30 8.21 0, recovered tetrol triethylamine 30 10 0, recovered tetrol K2CO3 30 10 80 DBU 60 12-13 60 L i C 5 H 5 30 15 5 KOH 200 15.7 80 t-BuOK 30 19 60 NaH 30 40 0, recovered tetrol * p K a of conjugate acid in H2O. d. Effect of Solvent on the Formation of Carceplexes Although the above experiments indicate that carceplex formation is sensitive to the base employed, they did not provide any evidence about carceplex reaction intermediates. We therefore explored a variety of solvents systems for running the carceplex reaction using DBU as the base (Table 2.6). Generally, the reaction to form carceplexes proceeds slower in these solvents than in NMP (often too slow to be of any practical value). Nonetheless, when acetone, THF and benzene were used as bulk solvents trace amounts of carceplex reaction intermediates were detected. It is interesting to note that the highly apolar solvent cyclohexane led to carceplex formation. 83 Table 2.6. Solvents for the Carceplex Reaction Using DBU as the Base. Solvent Guest tetrol la CH 2BrCl Time % Yield mM Equivalents (days) carceplex 2a»guest acetone3 acetone 1.3 12 2 10 % & trace of intermediates acetone3 acetone 0.8 10/ day 8 65 % & trace of intermediates THF 3 THF 1.3 12 2 0 % & trace of intermediates THF 3 THF 0.8 10/ day 8 20 % & trace of intermediates benzene3 benzene 0.4 10/day 10 30 % & trace of intermediates CHCI3 3 pyrazine 0.4 10/day 10 0% & recovered tetrol toluene13 pyrazine 1.0 10/day 5 25% cyclohexaneb methyl- 0.4 10/day 6 60% acetate xylenes15 pyrazine 1.0 10/day 8 0% tetrachloro- pyrazine 1.0 10/day 2 0% ethyleneb nitrobenzene*3 pyrazine 1.0 10/day 2 70% 3 Reaction mixture was refluxed. b Reaction mixture was stirred at 60 °C. Of the solvents employed, nitrobenzene proved to be the most successful because it produced carceplexes in good yields and was itself not a competitive guest. When tetrol lb (pendent group = methyl), 2.1 equiv. DBU, CH2I2, and 5 mol % DMSO as guest were stirred at 60 °C in nitrobenzene for 2 hours a mixture of carceplex intermediates resulted (Figure 2.3). This reaction yielded 27% of monobridged 102, 16% of A,B-bis-bridged 103 and 18% of recovered tetrol lb. Other reaction products included A,C-bis-bridged 104»DMSO, tris-bridged 105«DMSO, and carceplex 2a»DMSO. These intermediates were formed as mixtures with a small percentage of CH2I2 as guest and therefore could not be fully characterized. It was, however, possible to isolate the A,C-bis-bridged 104»DMSO and tris-bridged 105»DMSO species from the reaction of tetrol lb with 84 CH2BrCl, and potassium carbonate in DMSOv l (at ambient temperature for 18 hours), in 4.6% and 22% yields, respectively. In addition, 7.6% of carceplex 2b»DMSO was isolated. Incidentally, the use of methyl footed tetrol lb over that of phenethyl footed tetrol l a greatly simplified the ! H NMR spectra of the carceplex intermediates. The separation of these carceplex reaction intermediates via silica gel chromatography was often difficult. For example, the separation of monobridged 102 from tetrol lb was done using regular phase silica gel chromatography (CHCl3:methanol (9:1, v/v) as the eluent). These two compounds appear as overlapping spots on a regular phase TLC plate. Therefore, progress of the regular phase silica gel chromatography was followed by reverse phase TLC. By reverse phase TLC, two clear and distinct spots are observed for tetrol lb (R/= 0.6) and monobridged 102 (R/= 0.3) using acetone/water (80/20, v/v) as the eluent. Nevertheless, the isolation of the carceplex reaction intermediates as initial mixtures followed by separation via silica gel chromatography was found to be much more efficient than alternative approaches using more direct methods such as protection and deprotection steps. v i Sherman reported in his Ph.D. thesis that a build up of tris-bridged carceplex 2a«DMSO formed when the reaction was run in DMSO. 1 0 85 Figure 2.3. Schematic of Carceplex Intermediates. 2b»guest 102 103 104»guest 105»gues t R = C H 3 e. A,C-bis-bridged 104»DMSO and Tris-bridged 105'DMSO. The carceplex intermediates A,C-bis-bridged 104»DMSO and tris-bridged 105»DMSO each contained a tightly encapsulated molecule of D M S O which remained encapsulated after work-up. The stability of A,C-bis-bridged 104-DMSO and tris-bridged 105»DMSO was tested by recording their lH N M R spectra in deuterated nitrobenzene as a function of time and temperature. No noticeable changes in their ' H N M R spectra were observed even at temperatures as high as 120 °C up to 48 hours! Therefore, the activation energy for egress of D M S O from the holes in 104«DMSO and 105«DMSO is extremely large and these compounds can be regarded as true carceplexes. We further examined 86 104«DMSO and 105»DMSO to test if the GDS occurred before or after their formation. This was necessary because of the possibility that both 104»DMSO and 105»DMSO could exchange their guest molecule under our reaction conditions. We thus performed competition experiments with 104»DMSO and 105»DMSO in the presence of pyrazine, a vastly superior template molecule, to furnish the corresponding carceplex 2a«guest. In both cases, these reactions resulted in the isolation of only carceplex 2b»DMSO, which shows that the GDS occurs during or prior to the formation of these two intermediates. /. A,B-bis-bridged 103 and Monobridged 102 Intermediates The A,B-bis-bridged 103 and monobridged 102 intermediates isolated from the reaction carried out in nitrobenzene as solvent did not contain an encapsulated molecule of DMSO. Examination of these two intermediates by CPK models7" indicates that they have significant openings that allow for easy guest escape. Does the guest determining step for carceplex formation occur prior to the formation of these intermediates? We performed competition experiments using these two intermediates and tetrol lb at ambient temperature using six of the templates from Table 2.3 (results from Table 2.3 were obtained at 60 °C using tetrol la). The results of these competition studies are presented in Table 2.7. The template ratios between tetrol lb and monobridged 102 correlated well (r2 = 1.0) while the template ratios for tetrol lb and A,B-bis-bridged 103 showed a poor correlation (r2 = 0.7). These results strongly suggest that the GDS for carceplex formation occurs after the formation of first bridge, because monobridged 102 showed the same selectivity for the incarceration of templates molecules as tetrol lb. On the other hand, the poor correlation between tetrol lb and A,B-bis-bridged 103 indicates that the step which determines the selectivity in the carceplex v i i Corey-Pauling-Koltun (CPK) are space filling models that provide qualitative information about steric interactions in host guest systems. 1 1 87 reaction occurs during or prior to formation of this intermediate. It is also possible that the A,B-bis-bridged 103 intermediate does not form at all in NMP, and the reaction goes exclusively through the A,C-bis-bridged 104 intermediate, but this is unlikely. Table 2.7. Template Ratios For Carceplex Intermediates. Guest Template Ratios Template Ratios Template Ratios tetrol lb monobridgedl02 A,B-bis-bridged 103 pyrazine 860.0 1130.0 40.0 1,4-dioxane 177.0 245.0 5.4 DMSO 18.8 21.0 17.0 pyridine 13.5 17.0 19.0 acetone 1.8 2.0 3.0 benzene 1.0 1.0 1.0 g. The Guest Determining Step for Formation of Carceplex 2a*Guest Combining the results for the monobridged 102 and A,B-bis-bridged 103 with those of the A,C-bis-bridged 104»DMSO and the tris-bridged 105«DMSO, we conclude that the formation of the second O - C H 2 - O bridge in the carceplex reaction (either A,B-bis or A,C-bis) is the guest determining step. After formation of this second O - C H 2 - O bridge the guest is irreversibly entrapped under our reaction conditions, and subsequent bridging leads to the corresponding product (carceplex 2a»guest). Therefore, pyrazine reduces the activation energy of the transition state for formation of the second O - C H 2 - O bridge by 8.3 kcal/mol relative to that of NMP, which results in the one million to one product ratio (template ratio) observed in this reaction. 88 iii. Templation Requirements in Formation of a Hemicarceplex Our investigation into the template requirements for the formation of carceplex 2a»guest led to a similar investigation into the template requirements for the formation of hemicarceplex 107»guest.12 Hemicarceplex 107»guest resembles carceplex 2a#guest except it has one of the four O-CH2-O inter-bowl bridges replaced by two hydrogens. The absence of this acetal creates a slot-shaped portal such that guests can enter or escape. The reaction to form hemicarceplex 107»guest is statistically more demanding than the reaction to form carceplex 2a»guest because there is the potential to form the first bridge between the triol 106 molecules incorrectly.13 Chopra and Sherman measured the template ratios for the production of hemicarceplex 107»guest with seven guest molecules from Table 2.3 and found a good correlation with the template ratios measure for the formation of carceplex 2a»guest.12 Therefore, the driving forces for the formation of carceplex 2a»guest and hemicarceplex 107»guest are similar. The favorable van der Waals, electrostatic and n-rt interactions between the template and the forming host compound overwhelm any disparity that might be caused by the lower symmetry of hemicarceplex 107»guest. Also, the much greater than statistical yields observed for formation of hemicarceplex 107»guest suggest that hydrogen bonding may play an integral role in aligning the bowls prior to covalent bond formation. The exploration of such a hydrogen bonded complex is the subject of the next chapter. 89 Scheme 2.3. The Formation of a Hemicarceplex. triol 106 hemicarceplex 107«guest iv. Crystal Structure of Carceplex 2»Guest. a Crystal Structure of Carceplex 2b»Pyrazine. The crystal structure determination of carceplex 2a»pyrazine14 (Figure 2.4 & Figure 2.5) in our lab by Fraser et al. has provided further insight as to why pyrazine is the best template for carceplex formation.14 Pyrazine nicely complements the interior of the carceplex. Pyrazine sits in carceplex 2b»pyrazine such that its nitrogens are at the equator and its C-H bonds extend into the bowls. This arrangement maximizes van der Waals interactions, electrostatic interactions and CH-7C hydrogen bonds between pyrazine and the interior of the carceplex. Also, the four O-CH2-O inter-bowl bridges of carceplex 2a«pyrazine are all conformationally equivalent yielding a symmetric shell with a twist of 21°. This 21° helical twist of the shell of the carceplex allows all eight of the aryl oxygen's involved in the inter-bowl bridges to be conjugated into the aromatic rings of the carceplex which could lead to as much as 16-24 kcal/mol of stability (8 x 2-3 kcal/mol per interaction); it also prevents steric strain between the intra-bowl O-CH2 -O bridges of the top and bottom bowls.14 90 b. Crystal Structure of Carceplex 2a*DMA. In contrast to the crystal structure of carceplex 2b»pyrazine, the crystal structure of carceplex 2a»DMA reported by Cram et al. the O-CH2-O inter-bowl bridges are conformationally non-equivalent, which results in a distorted shell.2 This distortion is shown in diagrams A and C in Figure 2.6 where the four O-CH2-O inter-bowl bridges of carceplex 2a#DMA are shown in two perpendicular views; the same views are shown for the symmetric carceplex 2b*pyrazine (diagrams B and D, Figure 2.6). The O-CH2-O inter-bowl bridges of carceplex 2b,pyrazine are all equivalent and twist in the same direction (diagram B) while in carceplex 2a»DMA two of these O-CH2-O inter-bowl bridges twist in one direction while the other two O-CH2-O inter-bowl bridges are distorted. This asymmetry in carceplex 2a»DMA causes the planes connecting the two sets of four inter-bowl oxygens to be 5.2° from being parallel, whereas in the crystal structure of carceplex 2b»pyrazine these planes are only 0.3° from being parallel (Figure 2.6). The differences in the these two crystal structures was attributed to the change of the guest molecule and not due to the differences in the pendent groups, although the latter cannot be ruled out. Nevertheless, the differences found in the crystal structures of these two compounds helps explain the 50,000:1 template ratio observed for pyrazine:DMA for carceplex 2a»guest formation. It is pyrazine's favorable van der Waals and other noncovalent interactions which result in a lowering of the activation energy at the GDS in the formation of carceplex 2a«pyrazine relative to that of carceplex 2a»DMA such that a template ratio of 50,000:1 results. The complementarity of pyrazine to the interior of the carceplex is illustrated in the crystal structure of carceplex 2b»pyrazine in Figure 2.4 and Figure 2.5. The space filling view of Figure 2.5 illustrates just how nicely pyrazine complements the interior of the carceplex. It should be noted that, although the carceplex is far from the transition state of the GDS for carceplex formation, the complementarity of 92 pyrazine to the cavity of the carceplex clearly has some relevance to the templation ability of pyrazine. Figure 2.6. Inter-bowl Planes of Carceplex 2a»DMA and Carceplex 2a»Pyrazine. A B Partial view of the crystal structures of carceplex 2a*DMA (A, top view and C, side view) and carceplex 2b»pyrazine (B, top view and D, side view). The planes which connect the aryl ethers of carceplex 2a*DMA and carceplex 2b»pyrazine are drawn in. C. Conclusion The carceplex reaction has proven to be an excellent model for the study of noncovalent interactions between molecules because small changes in guest size, shape, and/or electronic properties results in vast differences in incarceration propensities. For example, the importance of van der Waals interactions is exemplified in the crystal structure of carceplex 2b»pyrazine where the host and the guest are good complements to each other. 93 The substitution of a nitrogen of pyrazine with a methine group to give pyridine disrupts the complementarity of pyrazine with the forming cavity as illustrated by the 29 fold decrease in the template ratio. Furthermore, substitution of both nitrogen's of pyrazine for two methines to give benzene results in a 420 fold decrease in the template. The formation of the second O -CH2 -O inter-bowl bridge was found to be the step responsible for the selectivity (GDS) observed in the formation of carceplex 2a»guest. Pyrazine's favorable noncovalent interactions with the forming carceplex lowers the Gibbs free energy of activation for this step by 8.3 kcal/mol relative to NMP. The ability of the forming carceplex 2a»guest to recognize subtle changes in guest properties ultimately led to a million-fold difference in template ratios between the poorest template (NMP) and the best template (pyrazine). These two carceplexes and the 32 other carceplexes that complete our competition experiments have led to valuable information about noncovalent interactions that can be applied in the design of more complex systems. The high efficiency of the reaction to form carceplex 2a»guests (up to 87% yield) and the much greater than statistical yield obtained in the formation of hemicarceplex 107»guest strongly suggest a preorganization of starting materials prior to covalent bond formation. This phenomenon is further explored in the next chapter. 94 D. Experimental i. General Experimental Nitrobenzene and NMP were stirred over BaO for 24 hours, distilled under reduced pressure, and stored under N 2 over 4 A molecular sieves prior to use. All other commercially available reagents were used as purchased without further purification unless stated otherwise. Desorption chemical ionization (DCI) and liquid secondary ion mass spectrometry (LSIMS) mass spectra were recorded on a Kratos Concept II HQ and matrix assisted laser desorption ionization (MALDI) mass spectra were recorded on a VG Tofspec in reflectron mode. Silica gel (BDH, 230-400) was used for column chromatography. Silica gel thin-layer chromatography was performed on Aldrich glass-backed plates (silica gel 60, F254, 0.25 mm). Microanalyses were performed by Mr. P. Borda of the UBC Microanalytical Laboratory on a Carlo-Erba CHN elemental analyzer, model 1106 or Fisons CHN-O elemental analyzer, model 1108. Melting points were measured on a Mel-Temp II apparatus. ! H NMR spectra were recorded on a Bruker WH-400 spectrometer in CDCI3 at ambient temperature using the residual lH as a reference (7.24 ppm) unless noted otherwise. The following guests cause top and bottom asymmetry of their respective carceplex 2a«guest due to slow rotation on the *H NMR timescale of the guest about the C 2 axis of the carceplex: ethyl methyl sulfide, 2-butanone, methyl acetate, ethyl acetate, and 2-butanol. The assignments of the protons on the top and on the bottom are denoted H x and H x ' . A number of carceplexes 2a»guest had conformational restriction of the guest molecule within the interior of the host. For example, the interconversion of the chair conformation of 1,4-thioxane is in slow exchange on the ! H NMR timescale (both the axial and equatorial hydrogens are observed). Guests like 1,4-thioxane that exhibit this type of dynamic lH NMR are explored further in Chapter 4. Two additional carceplexes (carceplex 2d»DMSO and carceplex 2b»(ft)-(-)-2-butanol) are reported in Chapter 4. The 95 following carceplexes 2a»guest were assigned based on COSY spectroscopy: carceplex 2a»2-propanol, 2a»(+)-2-butanol, 2a»l,3-dioxane. The following carceplexes 2a«guest were assigned based on homonuclear decoupling experiments: carceplex 2a,acetamide and 2a»NMP. ii. Carceplexes 2a»Guest A C4 axis c b >- C 2 axis H c 2a»pyrazine General Procedure A : A mixture of tetrol la (102 mg, 0.100 mmol), pyrazine (420 mg, 5.3 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) was stirred at 60 °C for 24 h. An additional 1.0 mmol of CH 2BrCl were added 96 and the reaction was stirred for an additional 24 h at 60 °C. The reaction mixture was concentrated in vacuo, water (50 mL) was added, and the slurry was acidified with 2 M HC1. The slurry was extracted with CHCI3 (3 x 60 mL), and the combined organic extracts were washed with saturated aq. NaHCC»3 (30 mL) and brine (30 mL), and dried over anhydrous MgSC»4. Silica gel (0.5 g) was added to the CHCI3 solution and the solvent removed in vacuo. The silica gel-absorbed sample was dry loaded onto a silica gel gravity column (20 g) and eluted with CHCl3:hexanes (3:1), affording 2a»pyrazine as a white solid which was recrystallized from CHC^/EtOAc and dried at 110 °C (0.1 mm Hg) for 24 h (97 mg, 87%): mp > 250 °C; IH NMR (CDCI3, 400 MHz): 8 7.13-7.24 (m, 40H, H a , H b , and H c), 6.93 (s, 8H, H d), 6.47 (s, 8H, H e), 6.02 (d, J = 7.5 Hz, 8H, Hf), 4.90 (t, J = 7.9 Hz, 8H, H g), 4.26 (d, / = 7.5 Hz, 8H, Hh), 4.07 (s, 4H, C4H4N2), 2.66 (m, 16H, Hi), 2.51 (m, 16H, Hj). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 100), 2082 (M + ; 80), 2162 ((M»C 4 H 4 N 2 ) + ; 90), 2204 ((M»C 4 H 4 N 2 + CH(CH 3) 2) +; 25). Anal. Calcd for C136H1 i 6 0 2 4 N 2 : C, 75.54; H, 5.41; N, 1.30. Found: C, 75.41; H, 5.35; N, 1.23. 2a» l ,4 -d ioxane Application of procedure A with 1,4-dioxane (0.45 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCb/EtOAc, 74 mg (68%) of 2a« 1,4-dioxane as a white solid: mp > 250 °C; 97 A H NMR (CDCI3, 400 MHz): S 7.11-7.24 (m, 40H, H a , H b , and H c), 6.72 (s, 8H, H d), 6.57 (s, 8H, H e), 6.19 (d, J = 7.6 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.46 (d, J = 7.6 Hz, 8H, H h), 2.63 (m, 16H, Hj), 2.43 (m, 16H, Hj), -0.28 (br, 8H, C 4 H 8 0 2 ) . MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 90), 2082 (M +; 100), 2170 ( (M»C 4 H 8 0 2 ) + ; 30). Anal. Calcd for C i 3 6 H i 2 o 0 2 6 : C, 75.26; H, 5.57. Found: C, 74.93; H, 5.61. 2a*dimethyl sulfide Application of procedure A with dimethyl sulfide (0.39 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 56 mg (52%) of 2a#dimethyl sulfide as a white solid: mp > 250 °C; A H NMR (CDCI3, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , and H c), 6.73 (s, 8H, H d), 6.57 (s, 8H, H e), 6.21 (d, J = 7.5 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.47 (d, J = 7.5 Hz, 8H, Hh), 2.64 (m, 16H, Hi), 2.43 (m, 16H, Hj), -1.24 (s, 6H, (CH3)2SO). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 100), 2082 (M + ; 50), 2144 ((M^CH3)2S)+; 65). Anal. Calcd for C 1 3 4 H 1 i 8 0 2 4 S : C, 75.05; H, 5.55. Found: C, 74.65; H, 5.72. 2a»dimethyl carbonate Application of procedure A with dimethyl carbonate (0.44 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 98 mL) gave, after recrystallization from CHCl3/EtOAc, 57 mg (52%) of 2a»dimethyl carbonate as a white solid: mp > 250 °C; X H NMR (CDCI3, 400 MHz): d 7.12-7.24 (m, 40H, H a , H b , and H c), 6.67 (s, 8H, H d), 6.54 (s, 8H, H e), 6.16 (d, J = 7.2 Hz, 8H, Hf), 4.92 (t, J = 7.9 Hz, 8H, H g), 4.38 (d, J = 7.2 Hz, 8H, H h), 2.66 (m, 16H, Hj), 2.44 (m, 16H, Hj), -0.76 (s, 6H, CO(OCH3)2). MS (MALDI) m/z (rel intensity): 2196 ((M»CH 3 OC0 2 CH 3 + Na +) +; 100); Calcd for C i 3 5 H i i 8 0 2 7 « N a + = 2195. Anal. Calcd for C i 3 5 H i i 8 0 2 7 : C, 74.64; H, 5.47. Found: C, 74.98; H, 5.52. 2a»l,3-dioxolane Application of procedure A with 1,3-dioxolane (0.37 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH2J3rCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 69 mg (64%) of 2a« 1,3-dioxolane as a white solid: mp > 250 °C; *H NMR (CDCI3, 400 MHz): S 7.11-7.24 (m, 40H, H a , H b , and H c), 6.75 (s, 8H, H d), 6.56 (s, 8H, H e), 6.19 (d, J = 7.5 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.37 (d, J = 7.5 Hz, 8H, Hh), 2.64 (m, 16H, Hi), 2.45 (m, 16H, Hj), 0.65 (s, 2H, OCH 2 0, 0.30 (s, 4H, OCH 2 CH 2 0). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 100), 2082 (M +; 80), 2156 ((M-C 3H 60 2) +; 55). Anal. Calcd for C i 3 5 H i i 8 0 2 6 : C, 75.19; H, 5.52. Found: C, 74.96; H, 5.48. 99 2a»pyridine Application of procedure A with pyridine (0.43 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K2CO3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 50 mg (46%) of 2a»pyridine as a white solid: mp > 250 °C; *H NMR (CDCI3, 400 MHz): 8 7.14-7.24 (m, 40H, H a , H b , and H c), 6.91 (s, 8H, H d), 6.49 (s, 8H, H e), 6.34 (t, J = 7.3 Hz, IH, C 5 H 5 N ) , 6.03 (d, J = 7.4 Hz, 8H, Hf), 4.89 (t, / = 7.6 Hz, 8H, H g), 4.14 (d, J = 7.4 Hz, 8H, H h), 4.02 (d, J = 5.2 Hz, 2H, C5H5N), 2.73 (m, 2H, C 5 H 5 N), 2.66 (m, 16H, Hi), 2.50 (m, 16H, Hj). MS ( D C I , isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 90), 2082 (M + ; 100), 2161 ((M«C 5H 5N) +; 55). Anal. Calcd for C137H117O24N: C, 76.13; H, 5.46; N, 0.65. Found: C, 75.99; H, 5.38; N, 0.64. 2a«dimethyl sulphone Application of procedure A with dimethyl sulphone (490 mg, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 66 mg (60%) of 2a»dimethyl sulphone as a white solid: mp > 250 °C; IH NMR (CDCI3, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , and H c), 6.76 (s, 8H, H d), 6.57 (s, 8H, H e), 6.14 (d, J = 1.6 Hz, 8H, Hf), 4.88 (t, J = 1.9 Hz, 8H, H g), 4.49 (d, J = 1.6 Hz, 8H, Hh), 2.63 (m, 16H, Hi), 2.45 (m, 16H, Hj), -0.89 (s, 6H, (CH 3) 2S0 2). 100 MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 90), 2082 (M + ; 100), 2176 ((M«(CH 3 ) 2 S0 2 ) + ; 25). Anal. Calcd for C i 3 4 H i i 8 0 2 6 S : C, 73.95; H, 5.46; S, 1.47. Found: C, 73.90; H, 5.40; S, 1.57. 2a«l,4-thioxane Application of procedure A with 1,4-thioxane (0.49 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 60 mg (55%) as a mixture of 2a» 1,4-thioxane (86%) and 2a» 1,4-dioxane (14%) as a white solid: mp > 250 °C; 1H NMR (CDCI3, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , and H c), 6.73 (s, 8H, H d), 6.57 (s, 8H, H e), 6.16 (d, J = 7.6 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.80 (d, J = 7.6 Hz, 8H, Hh), 2.63 (m, 16H, Hi), 2.43 (m, 16H, Hj), -0.01 (br, 2H, C 4H 8OS), -0.17 (br, 2H, C 4H 8OS), -1.55 (br, 2H, C 4H 8OS), -1.60 (br, 2H, C 4 H 8 OS). MS (MALDI) m/z (rel intensity): 2211 ((M»C 4H 8OS + Na +) +; 100); Calcd for Ci36Hi20O 25S«Na+ = 2212. Anal. Calcd for C i 3 6 H i 2 0 O 2 5 S : C, 74.71; H, 5.53. Found: C, 74.62; H, 5.61. 2a»2,3-dihydrofuran Application of procedure A with 2,3-dihydrofuran (0.40 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 16 mg (38%) of 2a»2,3-dihydrofuran as a white solid: mp > 250 °C; 101 1H NMR (CDCI3, 400 MHz): S 7.11-7.24 (m, 40H, H a , H b , and H c), 6.78 (s, 8H, H d), 6.55 (s, 8H, H e), 6.16 (d, J = 7.5 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.34 (d, J = 7.5 Hz, 8H, Hh), 2.64 (m, 16H, Hj), 2.46 (m, 16H, Hj), 1.90 (br, 1H, C 4 H 6 0) , 1.19 (br, 1H, C 4 H 6 0) , 0.29 (t, J = 9.3 Hz, 2H, C 4 H 6 0) , -0.25 (m, 2H, C 4 H 6 0) . MS (MALDI) m/z (rel intensity): 2177 ((M»C 4 H 6 0 + Na +) +; 100); Calcd for Ci36Hii8025-Na+ = 2175. Anal. Calcd for C136H118O25: C, 75.89; H, 5.53. Found: C, 75.79; H, 5.46. 2a»furan Application of procedure A with furan (0.38 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K2CO3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 58 mg (54%) of 2a»furan as a white solid: mp > 250 °C; !H NMR (CDCI3, 400 MHz): 3 7.12-7.24 (m, 40H, H a , H b , and H c), 6.84 (s, 8H, H d), 6.52 (s, 8H, H e), 6.12 (d, J = 7.4 Hz, 8H, Hf), 4.88 (t, J = 7.8 Hz, 8H, H g), 4.16 (d, J = 7.4 Hz, 8H, Hh), 3.09 (br, 2H, C 4 H 4 0) , 3.06 (br, 2H, C 4 H 4 0) , 2.65 (m, 16H, Hi), 2.48 (m, 16H, Hj). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 90), 2082 (M +; 100), 2150 ((M«C 4 H 4 0) + ; 50). Anal. Calcd for C136H116O25: C, 75.96; H, 5.44. Found: C, 75.73; H, 5.36. 2a»tetrahydrofuran Application of procedure A with tetrahydrofuran (0.43 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 102 mL) gave, after recrystallization from CHCl3/EtOAc, 54 mg (50%) of 2aHetrahydrofuran as a white solid: mp > 250 °C; !H NMR (CDC13, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , and H c), 6.76 (s, 8H, H d), 6.56 (s, 8H, H e), 6.16 (d, J = 7.6 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.40 (d, J = 7.6 Hz, 8H, H h), 2.63 (m, 16H, HO, 2.45 (m, 16H, Hj), -0.30 (br, 4H, C 4 H 8 0) , -1.26 (br, 4H, C 4 H 8 0) . MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 95), 2082 (M + ; 100), 2154 ((M«C 4 H 8 0) + ; 75). Anal. Calcd for C136H120O25: C, 75.82; H, 5.61. Found: C, 75.51; H, 5.56. 2a»pyridazine Application of procedure A with pyridazine (0.38 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K7CO3 (1.4 g, 10 mmol), and CH 2BrCl (65 (il, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 13 mg (30%) of 2a»pyridazine as a white solid: mp > 250 °C; A H NMR (CDCI3, 400 MHz): 8 7.13-7.24 (m, 40H, H a , H b , and H c), 6.91 (s, 8H, H d), 6.49 (s, 8H, H e), 6.03 (d, J = 7.5 Hz, 8H, Hf), 4.89 (t, J = 7.8 Hz, 8H, H g), 4.64 (m, 2H, C 4 H 4 N 2 ) , 4.39 (m, 2H, C 4 H 4 N 2 ) , 4.10 (d, J = 7.5 Hz, 8H, H h), 2.66 (m, 16H, Hj), 2.51 (m, 16H, Hj). MS (MALDI) m/z (rel intensity): 2202 ((M«C 4 H 4 N 2 + K + ) + ; 100); Calcd for Ci36Hi i 6 0 2 4 N2«K + = 2201. Anal. Calcd for Ci3 6Hii 602 4N2: C, 75.54; H, 5.41; N, 1.30. Found: C, 75.50; H, 5.50; N, 1.30. 103 2a»acetone Application of procedure A with acetone (0.39 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 55 mg (51%) of 2a«acetone as a white solid: mp > 250 °C; !H NMR (CDCI3, 400 MHz): S 7.11-7.24 (m, 40H, H a , H b , and H c), 6.74 (s, 8H, Hd), 6.57 (s, 8H, H e), 6.19 (d, J = 7.5 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.36 (d, J = 7.5 Hz, 8H, Hh), 2.63 (m, 16H, Hi), 2.44 (m, 16H, Hj), -1.63 (s, 6H, (CH 3) 2CO). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 90), 2082 (M +; 100), 2140 ((M»(CH 3) 2CO) +; 40). Anal. Calcd for C135H118O25: C, 75.76; H, 5.56. Found: C, 75.95; H, 5.56. 2a»thiophene Application of procedure A with thiophene (0.42 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 25 mg (23%) of 2a»thiophene as a white solid: mp > 250 °C; !H NMR (CDCI3, 400 MHz): 5 7.13-7.24 (m, 40H, H a , H b , and H c), 6.88 (s, 8H, H d), 6.51 (s, 8H, H e), 6.05 (d, J = 7.4 Hz, 8H, Hf), 4.88 (t, J = 7.9 Hz, 8H, H g), 4.26 (d, J = 7.4 Hz, 8H, Hh), 3.73 (s, 4H, C4H4S), 2.65 (m, 16H, HO, 2.49 (m, 16H, Hj). 104 MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 100), 2082 (M +; 60), 2166 ((M»C 4H 4S) +; 20). Anal. Calcd for C136H1 i 6 0 2 4 S : C, 75.40; H, 5.40. Found: C, 75.50; H, 5.55. 2a«l,3-dithiolane Application of procedure A with 1,3-dithiolane (0.44 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 47 mg (43%) of 2a« 1,3-dithiolane as a white solid: mp > 250 °C; !H NMR (CDCI3, 400 MHz): 8 7.13-7.24 (m, 40H, H a , H b , and H c), 6.78 (s, 8H, H d), 6.56 (s, 8H, H e), 6.11 (d, J = 7.5 Hz, 8H, Hf), 4.86 (m, 16H, Hg and H h), 2.63 (m, 16H, Hi), 2.45 (m, 16H, Hj), -0.59 (s, 2H, SCH2S), -0.93 (br, 2H, SCHaxHeqCHaxHeqS), -1.05 (br, 2H, SCH^HeqCH^HeqS). MS (MALDI) m/z (rel intensity): 2212 ((M»C 3 H 6 S 2 + Na +) +; 100); Calcd for C i 3 5 H n 8 0 2 4 S 2 « N a + = 2212. Anal. Calcd for C i 3 5 H n 8 0 2 4 S 2 : C, 74.08; H, 5.44. Found: C, 74.30; H, 5.42. 2a»methyl acetate Application of procedure A with methyl acetate (0.42 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 82 mg (75%) of 2a»methyl acetate as a white solid: mp > 250 °C; *H NMR (CDCI3, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , H c , Ha>, H b ' , and H c 0, 6.71 (s, 4H, H d or Hd>), 6.69 (s, 4H, H d or H d0, 6.57 (s, 8H, H e), 6.19 (m, 105 8H, H f and H f ) , 4.89 (t, J = 7.8 Hz, 8H, H g and %), 4.35 (d, / = 7.4 Hz, 4H, H h or Hh'), 4.31 (d, / = 7.4 Hz, 4H, H h or H h0, 2.65 (m, 16H, Hj and Hr), 2.44 (m, 16H, Hj and Hj>), -0.88 (s, 3H, CH 3 C0 2 CH 3 ) , -2.40 (s, 3H, CH 3 C0 2 CH 3 ) . MS (MALDI) m/z (rel intensity): 2195 ((M»CH 3 C0 2 CH 3 + K + ) + ; 100); Calcd for C i 3 5 H i i 8 0 2 6 « K + = 2 1 9 5 . Anal. Calcd for C i 3 5 H i i 8 0 2 6 : C, 75.19; H, 5.52. Found: C, 75.06; H, 5.45. 2a»ethyl methyl sulfide Application of procedure A with ethyl methyl sulfide (0.48 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 74 mg (67%) of 2a«ethyl methyl sulfide as a white solid: mp > 250 °C; X H NMR (CDC13, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , H c , H a ' , Hb>, and Hc>), 6.69 (s, 8H, H d and Hd>), 6.59 (s, 8H, H e), 6.19 (d, J = 7.4 Hz, 4H, H f or H f ) , 6.17 (d, J = 7.4 Hz, 4H, H f or H f ) , 4.89 (m, 8H, H g andHg'), 4.59 (d, J = 7.4 Hz, 4H, H h or Hh'), 4.39 (d, J = 7.4 Hz, 4H, H h or Hh'), 2.65 (m, 16H, H ; and Hi'), 2.43 (m, 16H, Hj andHj-), -0.10 (q, J = 7.6 Hz, 2H, CH 3 SCH 2 CH 3 ) , -2.29 (s, 3H, CH 3 SCH 2 CH 3 ) , -3.23 (t, J = 7.6 Hz, 3H, CH 3 SCH 2 CH 3 ) . MS (MALDI) m/z (rel intensity): 2181 ((M»CH 3SCH 2CH 3 + Na +) +; 100); Calcd for C i 3 5 H i 2 o 0 2 4 S « N a + = 2182. Anal. Calcd for C i 3 5 H i 2 o 0 2 4 S : C, 75.12; H, 5.60. Found: C, 75.18; H, 5.68. 2a»2-butanone Application of procedure A with 2-butanone (0.47 mL, 5.3 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 106 mL) gave, after recrystallization from CHCl3/EtOAc, 81 mg (75%) of 2a»2-butanone as a white solid: mp > 250 °C; A H NMR (CDCI3, 400 MHz): 5 7.13-7.24 (m, 40H, H a , H b , H c , Ha>, H b S and Hc>), 6.71 (s, 4H, H d or Hd>), 6.70 (s, 4H, H d or Hd>), 6.58 (s, 8H, H e), 6.16 (d, J = 7.3 Hz, 8H, H f and Hf), 4.88 (m, 8H, H g and Hg>), 4.40 (m, 8H, H h and Hh'), 2.65 (m, 16H, H; and Hi-), 2.44 (m, 16H, Hj and Hj>), -0.05 (q, J= 7.2 Hz, 3H, C H 3 C H 2 C O C H 3 ) , -2.36 (s, 3H, C H 3 C H 2 C O C H 3 ) , -3.43 (t, J = 7.2 Hz, 3H, CH3CH2COCH3),. MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 100), 2082 (M + ; 95), 2154 ((M»CH 3CH 2COCH 3) +; 25). Anal. Calcd for C136H120O25: C, 75.82; H, 5.61. Found: C, 75.80; H, 5.61. 2a«(±)2-butanol General Procedure B: A mixture of tetrol la (102 mg, 0.100 mmol), (±)2-butanol (2.5 mL, 27 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) was stirred at rt for 24 h. An additional 1.0 mmol of CH2BrCl were added and the reaction was stirred at 60 °C for an additional 48 h. The reaction mixture was concentrated in vacuo, water (50 mL) was added, and the slurry was acidified with 2 M HC1. The slurry was extracted with CHCI3 (3 x 60 mL), and the combined organic extracts were washed with saturated aq. NaHCC>3 (30 mL) and brine (30 mL), and dried over anhydrous MgSC»4. Silica gel (0.5 g) was added to the CHCI3 solution and the solvent removed in vacuo. The silica gel-absorbed sample was dry loaded onto a silica gel gravity column (20 g) and eluted with CHCl3:hexanes (3:1), affording 2a»(±)2-butanol as a white solid which was recrystallized from CHCl3/EtOAc and dried at 110 °C (0.1 mm Hg) for 24 h (51 mg, 47%): mp > 250 °C; 107 IH NMR (CDCI3, 400 MHz): S 7.12-7.24 (m, 40H, H a , H b , H c , Ha>, H b ' , and Hc>), 6.73 (s, 4H, H d or Hd>), 6.71 (s, 4H, H d or Hd'), 6.58 (s, 8H, H e), 6.16 (d, J = 7.3 Hz, 8H, H f and Hf>), 4.89 (t, / = 7.9 Hz, 8H, H g and Hg>), 4.46 (d, J = 7.3 Hz, 4H, H h or Hh>), 4.37 (d, J = 7.3 Hz, 4H, H h or Hh'), 2.65 (m, 16H, Hj and Hi-), 2.44 (m, 16H, Hj and Hj-), 0.91 (br, IH, CH 3 CHOHCH 2 CH 3 ) , -0.93 (br, 2H, CH 3 CHOHCH x H y CH 3 ) , -1.32 (br, IH, CH 3 CHOHCH x H y CH 3 ) , -3.27 (d, J = 6.0 Hz, 3H, CH 3 CHOHCH 2 CH 3 ) , -3.47 (t, J = 7.3 Hz, 3H, CH 3 CHOHCH 2 CH 3 ) , MS (MALDI) m/z (rel intensity): 2180 ((M»CH 3CHOHCH 2CH 3 + Na +) +; 100); Calcd for C i 3 6 H i 2 2 0 2 5 » Na + = 2179. Anal. Calcd for C i 3 6 H i 2 2 0 2 5 : C, 75.75; H, 5.70. Found: C, 75.35; H, 5.70. 2a»ethyl acetate Application of procedure B with ethyl acetate (4.9 mL, 58 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 15 mg (14%) of 2a«ethyl acetate as a white solid: mp > 250 °C; ! H NMR (CDCI3, 400 MHz): 6 7.13-7.24 (m, 40H, H a , H b , H c , Ha>, Hb>, and H c 0, 6.67 (s, 4H, H d or Hd>), 6.58 (s, 4H, H d or Hd>), 6.50 (s, 8H, H e), 6.14 (d, J = 7.6 Hz, 4H, H f and Hf), 6.12 (d, / = 6.9 Hz, 4H, H f and Hf), 4.90 (t, / = 7.8 Hz, 8H, H g and Hg>), 4.51 (d, 4H, J = 7.6 Hz, H h or Hh'), 4.44 (d, 4H, J = 6.9 Hz, H h or Hh>), 2.65 (m, 16H, Hj and H r), 2.44 (m, 16H, Hj and Hj>), 0.58 (q, J= 7.1 Hz, 3H, CH 3 CH 2 C0 2 CH 3 ) , -2.50 (s, 3H, C H 3 C H 2 C 0 2 C H 3 ) , -2.63 (t, / = 7.1 Hz, 3H, CH 3 CH 2 C0 2 CH 3 ) . 108 MS (MALDI) m/z (rel intensity): 2193 ((M«CH 3 CH 2 C0 2 CH 3 + Na +) +; 100); Calcd for Ci3 6Hi2o026 ,Na+ = 2193. Anal. Calcd for Ci 3 6Hi2o0 26: C, 75.25; H, 5.58. Found: C, 75.04; H, 5.55. 2a»benzene Application of procedure B with benzene (2.5 mL, 27 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH^BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 47 mg (43%) of 2a»benzene as a white solid: mp > 250 °C; !H NMR (CDCI3, 400 MHz): 8 7.13-7.24 (m, 40H, H a , H b , and H c), 6.91 (s, 8H, H d), 6.50 (s, 8H, H e), 6.02 (d, / = 7.4 Hz, 8H, Hf), 4.88 (t, J = 7.9 Hz, 8H, H g), 4.15 (d, J = IA Hz, 8H, H h), 3.88 (br, 6H, C 6 H 6 ) , 2.66 (m, 16H, Hi), 2.50 (m, 16H, Hj). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 90), 2082 (M + ; 100), 2160 ((M«C 6H 6) +; 25). Anal. Calcd for Ci 3 8 Hn 8 024: C, 76.72; H, 5.51. Found: C, 76.59; H, 5.50. 2a»2-propanol Application of procedure B with 2-propanol (2.1 mL, 27 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 80 mg (74%) of 2a»2-propanol as a white solid: mp >250 °C; 1H NMR (CDCI3, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , and H c), 6.76 (s, 8H, H d), 6.57 (s, 8H, H e), 6.18 (d, J = 7.5 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, 109 Hg), 4.40 (d, J = 7.5 Hz, 8H, Hh), 2.64 (m, 16H, Hj), 2.45 (m, 16H, Hj), -1.91 (d, J = 1.2 Hz, IH, (CH3)2CHOH), -2.46 (m, 7H, (CH3)2CHOH). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 100), 2082 (M +; 90), 2142 ((M»(CH 3) 2CHOH)+; 70). Anal. Calcd for C i 3 5 H i 2 0 O 2 5 : C, 75.68; H, 5.65. Found: C, 75.55; H, 5.63. 2a»pyrrole Application of procedure B with pyrrole (1.9 mL, 27 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 79 mg (73%) of 2a«pyrrole as a white solid: mp > 250 °C; J H NMR (CDC13, 400 MHz): 5 7.12-7.24 (m, 40H, H a , H b , and H c), 6.86 (s, 8H, H d), 6.52 (s, 8H, H e), 6.09 (d, J = 7.4 Hz, 8H, Hf), 4.88 (t, J = 7.9 Hz, 8H, H g), 4.23 (d, J = 7.4 Hz, 8H, Hh), 4.08 (br, IH, C4H4NH), 3.41 (m, 2H, C4H4NH), 3.05 (m, 2H, C4H4NH), 2.65 (m, 16H, Hj), 2.49 (m, 16H, Hj). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 40), 2082 (M + ; 30), 2149 ((M»C 4H 4NH) +; 100). Anal. Calcd for C i 3 6 H i i 7 0 2 4 N : C, 76.00; H, 5.49; N, 0.65. Found: C, 75.85; H, 5.50; N, 0.68. 2a»tetrahydrothiophene Application of procedure B with tetrahydrothiophene (2.4 mL, 27 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 37 mg (34%) of 2a»tetra-hydrothiophene as a white solid: mp > 250 °C; 110 X H NMR (CDCI3, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , and H c), 6.77 (s, 8H, H d), 6.56 (s, 8H, H e), 6.11 (d, J = 7.6 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.71 (d, J = 7.6 Hz, 8H, H h), 2.63 (m, 16H, Hj), 2.44 (m, 16H, Hj), -0.88 (br, 2H, C 4 H 8 S), -1.08 (br, 2H, C 4 H 8 S), -1.51 (br, 4H, C 4 H 8 S). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 100), 2082 (M + ; 65), 2171 ((M«C 4H 8S) +; 20). Anal. Calcd for Ci3 6Hi2o0 2 4S: C, 75.26; H, 5.57. Found: C, 74.91; H, 5.55. 2a»l,3-dioxane Application of procedure B with 1,3-dioxane (4.9 mL, 58 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 49 mg (45%) of 2a» 1,3-dioxane as a white solid: mp > 250 °C; !H NMR (CDCI3, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , and H c), 6.74 (s, 8H, H d), 6.57 (s, 8H, H e), 6.14 (d, J = 7.6 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.55 (d, J = 7.6 Hz, 8H, H h), 2.63 (m, 16H, Hj), 2.43 (m, 16H, Hj), 1.34 (br, 2H, C 4 H 8 0 2 ) , 1.08 (br, 2H, C 4 H 8 0 2 ) , 0.62 (s, 2H, C 4 H 8 0 2 ) , -2.34 (br, 2H, C 4 H 8 0 2 ) . MS (MALDI) m/z (rel intensity): 2193 ( (M«C 4 H 8 0 2 + Na +) +; 100); Calcd for C i 3 6 H i 2 0 O 2 6 . N a + = 2193. Anal. Calcd for C i 3 6 H i 2 o 0 2 6 : C, 75.26; H, 5.57. Found: C, 75.25; H, 5.57. Ill 2a»acetamide Application of procedure B with acetamide (3.4 g, 58 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 28 mg (26%) of 2a»acetamide as a white solid: mp > 250 °C; *H NMR (CDCI3, 400 MHz): 6 7.12-7.24 (m, 40H, H a , H b , and H c), 6.75 (s, 8H, H d), 6.57 (s, 8H, H e), 6.19 (d, J = 7.4 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.36 (d, / = 7.4 Hz, 8H, Hh), 2.64 (m, 16H, Hi), 2.45 (m, 16H, Hj), 1.40 (br, 1H, CH 3 CONH a H b ) , 0.65 (br, 1H, C H 3 C O N H a H b ) , -1.55 (s, 3H, CH 3 CONH a H b ) . MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 70), 2082 (M + ; 100), 2141 ((M»CH 3CONH 2) +; 40). Anal. Calcd for C134H1 i 7 0 2 5 N : C, 75.16; H, 5.51; N, 0.65. Found: C, 74.79; H, 5.57; N, 0.48. 2a»l,3,5-trioxane Application of procedure B with 1,3,5-trioxane (5.2 g, 58 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 26 mg (24%) of 2a» 1,3,5-trioxane as a white solid: mp > 250 °C; J H NMR (CDCI3, 400 MHz): 8 7.12-7.24 (m, 40H, H a , H b , and H c), 6.74 (s, 8H, H d), 6.57 (s, 8H, H e), 6.17 (d, J = 7.4 Hz, 8H, Hf), 4.86 (t, J = 7.9 Hz, 8H, H g), 4.48 (d, J = 7.4 Hz, 8H, Hh), 2.62 (m, 16H, Hi), 2.43 (m, 16H, Hj), 1.88 (s, 6H, C 3 H 6 0 3 ) . 112 MS (MALDI) m/z (rel intensity): 2194 ((M»C 3 H 6 03 + Na +) +; 100); Calcd for C i 3 5 H i i 8 0 2 7 « N a + = 2195. Anal. Calcd for C135H118O27: C, 74.64; H, 5.47. Found: C, 74.88; H, 5.43. 2a»acetonitrile Application of procedure B with acetonitrile (3.0 mL, 58 mmol), tetrol la (102 mg, 0.100 mmol), K2CO3 (1.4 g, 10 mmol), and CH 2BrCl (65 |ll, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 37 mg (35%) of 2a#acetonitrile as a white solid: mp > 250 °C; A H NMR (CDCI3, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , and H c), 6.71 (s, 8H, H d), 6.58 (s, 8H, H e), 6.24 (d, J = 7.4 Hz, 8H, Hf), 4.87 (t, J = 7.9 Hz, 8H, H g), 4.30 (d, J = 7.4 Hz, 8H, Hh), 2.63 (m, 16H, HO, 2.43 (m, 16H, Hj), -2.41 (s, 3H, C H 3 C N ) . MS (MALDI) m/z (rel intensity): 2146 ((M«CH3CN + Na +) +; 100); Calcd for Ci3 4Hii 5024N-Na+ = 2146. Anal. Calcd for C134H115O24N: C, 75.80; H, 5.46; N, 0.66. Found: C, 75.54; H, 5.40; N, 0.60. 2a*ethanol Application of procedure B with ethanol (3.4 mL, 58 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 41 mg (38%) of 2a«ethanol as a white solid: mp > 250 °C; !H NMR (CDCI3, 400 MHz): 8 7.13-7.24 (m, 40H, H a , H b , and H c), 6.75 (s, 8H, H d), 6.56 (s, 8H, H e), 6.21 (d, J = 7.3 Hz, 8H, Hf), 4.86 (t, J = 7.8 Hz, 8H, 113 Hg), 4.26 (d, J = 7.3 Hz, 8H, Hh), 2.64 (m, 16H, Hj), 2.45 (m, 16H, Hj), 0.73 (m, 2H, CH 3 CH 2 OH), -2.53 (t, 7= 6.9 Hz, 3H, CH 3 CH 2 OH), -2.72 (br, 1H, CH3CH2OH). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 95), 2082 (M + ; 100), 2128 ((M«CH 3CH 2OH) +; 85). Anal. Calcd for C i 3 4 H i i 8 0 2 5 : C, 75.62; H, 5.59. Found: C, 75.55; H, 5.63. 2a«CH 2Cl 2 Application of procedure B with CH 2C1 2 (3.7 mL, 58 mmol), tetrol la (102 mg, 0.100 mmol), K 2C03 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 40 mg (37%) of 2a»CH 2Cl 2 as a white solid: mp > 250 °C; 1 H NMR (CDCI3, 400 MHz): S 7.11-7.24 (m, 40H, H a , H b , and H c), 6.72 (s, 8H, H d), 6.57 (s, 8H, H e), 6.23 (d, J = 7.4 Hz, 8H, Hf), 4.86 (t, J = 7.9 Hz, 8H, H g), 4.27 (d, J = 7.4 Hz, 8H, Hh), 2.63 (m, 18H, Hi and CH2C12), 2.44 (m, 16H, Hj). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 95), 2082 (M + ; 100), 2125 ((M + C(CH 3) 3) +; 35), 2167 ((M»CH 2C1 2)+; 30). Anal. Calcd for Ci33Hii 40 24Cl 2: C, 73.71; H, 5.30; CI, 3.27. Found: C, 73.81; H, 5.48; CI, 3.07. 2a«CH2BrCl Application of procedure A with CH 2BrCl (0.68 mL, 11 mmol), tetrol la (102 mg, 0.100 mmol) and K2CO3 (1.4 g, 10 mmol) in NMP (50 mL) gave, after purification from 114 carceplex 2a»NMP (4%) and recrystallization from CHCl3/EtOAc, 35 mg (31%) of 2a«CH2BrCl as a white solid: mp > 250 °C; !H NMR (CDCI3, 400 MHz): <5 7.11-7.24 (m, 40H, H a , H b , and H c), 6.72 (s, 8H, H d), 6.57 (s, 8H, H e), 6.23 (d, J = 7.4 Hz, 8H, Hf), 4.86 (t, J = 7.9 Hz, 8H, H g), 4.27 (d, J = 7.4 Hz, 8H, Hh), 2.63 (m, 18H, Hj), 2.50 (s, 2H, CH 2BrCl), 2.44 (m, 16H, Hj). MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 85), 2082 (M + ; 100), 2125 ((M + C(CH 3) 3) +; 30), 2212 ((M»CH2BrCl)+; 45). Anal. Calcd for C i 3 3 H n 4 0 2 4 B r C l : C, 72.23; H, 5.20; CI, 1.60; Br, 3.61. Found: C, 71.87; H, 5.09; CI, 1.52; Br, 3.43. 2a»CH 2 Br 2 Application of procedure A with CH 2 Br 2 (0.36 mL, 5.1 mmol), tetrol la (102 mg, 0.100 mmol) and K 2 C 0 3 (1.4 g, 10 mmol) in NMP (50 mL) gave, after purification from carceplex 2a»NMP (2%) and recrystallization from CHCl3/EtOAc, 35 mg (31%) of 2a»CH2Br2 as a white solid: mp > 250 °C; iH NMR (CDC13, 400 MHz): 5 7.12-7.24 (m, 40H, H a , H b , and H c), 6.69 (s, 8H, H d), 6.59 (s, 8H, H e), 6.22 (d, J = 7.4 Hz, 8H, Hf), 4.86 (t, J = 7.9 Hz, 8H, H g), 4.33 (d, J = 7.4 Hz, 8H, Hh), 2.64 (m, 18H, H i ) , 2.43 (m, 18H, Hj and CH 2 Br 2 ) . MS (DCI, isobutane) m/z (rel intensity): 2070 ((M - C H 2 + 2H)+; 90), 2082 (M + ; 100), 2125 ((M + C(CH 3) 3) +; 30), 2256 ((M»CH 2Br 2) +; 40). Anal. Calcd for C i 3 3 H n 4 0 2 4 B r 2 : C, 70.80; H, 5.09; Br, 7.08. Found: C, 70.93; H, 5.26; Br, 7.00. 115 2a«CH 2 I 2 Application of procedure A with CH2I2 (0.41 mL, 5.1 mmol), tetrol la (102 mg, 0.100 mmol) and K2CO3 (1.4 g, 10 mmol) in NMP (50 mL) gave, after recrystallization from CHCl 3/EtOAc, 47 mg (40%) of 2a»CH2l2 as a light pink solid: mp > 250 °C; ! H NMR (CDCI3, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , and H c), 6.63 (s, 8H, H d), 6.61 (s, 8H, H e), 6.18 (d, J = 7.3 Hz, 8H, Hf), 4.88 (t, J = 7.9 Hz, 8H, H g), 4.50 (d, J = 7.3 Hz, 8H, Hh), 2.64 (m, 18H, Hi), 2.42 (m, 16H, Hj), 1.47 (s, 2H, CH 2I 2). MS (DCI, ammonia) m/z (rel intensity): 2088 ((M - C H 2 + 2H + NH 4 ) + ; 100), 2100 ((M + NH 4 ) + ; 50), 2368 ((M«CH 2l2)+ + NH 4 ) + ; 35). Anal. Calcd for C133H1 i 40 2 4J-2: C, 67.97; H, 4.89; I, 10.8. Found: C, 68.26; H, 5.10; I, 11.12. iii. Carceplex 2a»Guest Mixtures 2a»diethyl ether Application of procedure B with diethyl ether (6.1 mL, 58 mmol), tetrol la (102 mg, 0.100 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 18 mg (17%) a mixture of carceplexes 2a»diethyl ether (96%) and 2a»CH2BrCl (4%) as a white solid: mp > 250 °C; 116 A H NMR (CDCI3, 400 MHz): <5 7.13-7.24 (m, 40H, H a , H b , and H c), 6.73 (s, 8H, H d), 6.55 (s, 8H, H e), 6.15 (d, J = 7.4 Hz, 8H, Hf), 4.88 (t, J = 7.9 Hz, 8H, H g), 4.44 (d, J = 7.4 Hz, 8H, Hh), 2.64 (m, 16H, Hi), 2.45 (m, 16H, Hj), 0.36 (q, J = 7.1 Hz, 4H, (CH 3CH 2) 20), -2.81 (t, J = 7.1 Hz, 6H, (CH 3CH 2) 20). MS (MALDI) m/z (rel intensity): 2181 ((M»C 4 H 8 0 + Na +) +; 100); Calcd for C i 3 6 H i 2 2 0 2 5 » N a + = 2179. 2a«pyran Application of procedure B with pyran (2.6 mL, 27 mmol), tetrol la (102 mg, 0.100 mmol), K2CC>3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in NMP (50 mL) gave, after recrystallization from CHCl3/EtOAc, 12 mg (11%) a mixture of carceplexes 2a«pyran (80%) and 2a«CH2BrCl (20%) as a white solid: mp > 250 °C; !H NMR (CDCI3, 400 MHz): 8 7.11-7.24 (m, 40H, H a , H b , and H c), 6.73 (s, 8H, H d), 6.57 (s, 8H, H e), 6.14 (d, / = 7.6 Hz, 8H, Hf), 4.87 (t, J = 7.8 Hz, 8H, H g), 4.54 (d, J = 7.6 Hz, 8H, Hh), 2.63 (m, 16H, Hi), 2.44 (m, 16H, Hj), -0.31 (br, 6H, C5H10O), -2.35 (br, 2H, C5H10O), -2.42 (br, 2H, C5H10O). MS (MALDI) m/z (rel intensity): 2191 ((M«C 5Hi 0O + Na +) +; 100); Calcd for C i 3 7 H i 2 2 0 2 5 « N a + = 2191. 2a»CHCl 3 Application of procedure B with CHCI3 (2.6 mL, 27 mmol), tetrol la (102 mg, 0.100 mmol), K2CC>3 (1.4 g, 10 mmol), and CH 2BrCl (65 ul, 1.0 mmol) in N-formylpiperidine (80 mL) gave, after recrystallization from CHCl3/EtOAc, 10 mg (9%) a mixture of 117 carceplexes 2a»CHCl3 (50%) 2a»CH2BrCl (25%) and 2a»ethanol (25%) as a white solid: mp > 250 °C; ! H NMR (CDCI3, 400 MHz): <5 7.11-7.24 (m, 40H, H a , H b , and H c), 6.73 (s, 8H, H d), 6.57 (s, 8H, H e), 6.16 (d, J = 7.4 Hz, 8H, H f), 4.87 (m, 8H, H g), 4.59 (d, J = 7.5 Hz, 8H, H h), 4.43 (s, 1H, CHCI3), 2.64 (m, 16H, Hj), 2.44 (m, 16H, Hj). MS (MALDI) m/z (rel intensity): 2225 ((M»CHC13 + Na +) +; 100); Calcd for Ci33Hii30 24Cl3«Na + = 2225. 2b» l , 4 -d ioxane Application of procedure A with 1,4-dioxane (0.18 mL, 2.1 mmol), tetrol lb (40 mg, 0.061 mmol), K 2 C 0 3 (400 mg, 2.9 mmol), and CH 2BrCl (40 ul, 0.61 mmol) in NMP (20 mL) gave, after recrystallization from CH2Cl2/hexanes, 32.8 mg (74%) of 2b» 1,4-dioxane as a white solid: mp > 250 °C; 118 !H NMR (CDCI3, 400 MHz): <5 6.79 (s, 8H, H d), 6.53 (s, 8H, H e), 6.17 (d, J = 7.6 Hz, 8H, Hf), 4.98 (q, J = 7.4 Hz, 8H, H g), 4.44 (d, J = 7.6 Hz, 8H, H h), 1.68 (d, / = 7.4 Hz, 24H, CH 3), -0.34 (br, 8H, 0(CH 2CH 2)20). MS (MALDI) m/z (rel intensity): 1473 ( ( M » C 4 H 8 0 2 + Na +) +; 100); Calcd for C 8 0 H 7 2 O 2 6 ' N a + = 1472. Anal. Calcd for C 8 o H 7 2 0 2 6 : C, 66.29; H, 5.01. Found: C, 66.61; H, 5.21. 2b«DMSO Application of procedure A with DMSO (0.15 mL, 2.1 mmol), tetrol lb (40 mg, 0.061 mmol), K 2 C 0 3 (400 mg, 2.9 mmol), and CH 2BrCl (40 ul, 0.61 mmol) in NMP (20 mL) gave, after recrystallization from CH2C12/Hexanes, 28.4 mg (65%) of 2b«DMSO as a white solid: mp > 250 °C; iH NMR (CDCI3, 400 MHz): 8 6.83 (s, 8H, H d), 6.53 (s, 8H, H e), 6.13 (d, J = 7.5 Hz, 8H, Hf), 4.98 (q, J = 7.4 Hz, 8H, H g), 4.47 (d, J = 7.5 Hz, 8H, H h), 1.69 (d, J = 7.4 Hz, 24H, CH 3), -1.30 (s, 6H, (CH3)2SO). MS (MALDI) m/z (rel intensity): 1463 ((M»(CH 3) 2SO + Na +) +; 100); Calcd for C 7 8 H 7 o 0 2 5 S » N a + = 1462. Anal. Calcd for C 7 8 H 7 o 0 2 5 S : C, 65.08; H, 4.90. Found: C, 65.40; H, 4.98. 2b»pyridine Application of procedure A with pyridine (0.17 mL, 2.1 mmol), tetrol lb (40 mg, 0.061 mmol), K 2 C 0 3 (400 mg, 2.9 mmol), and CH 2BrCl (40 ul, 0.61 mmol) in NMP (20 mL) gave, after recrystallization from CH2Cl2/hexanes, 23.6 mg (54%) of 2b«pyridine as a white solid: mp > 250 °C; 119 !H NMR (CDCI3, 400 MHz): S 6.98 (s, 8H, H d), 6.45 (s, 8H, H e), 6.30 (m, 1H, C 5 H 5 N), 6.01 (d, / = 7.5 Hz, 8H, Hf), 4.99 (q, J = 7.4 Hz, 8H, H g), 4.12 (d, J = 7.5 Hz, 8H, H h), 3.95 (m, 2H, C 5 H 5 N), 2.66 (m, 2H, C5H5N), 1.75 (d, J = 7.4 Hz, 24H, CH 3). MS (MALDI) m/z (rel intensity): 1463 ((M»C 5H 5N + Na +) +; 100); Calcd for C8lH 6 90 2 4N»Na + = 1463. Anal. Calcd for C 8 i H 6 9 0 2 4 N : C, 67.54; H, 4.83; N, 0.97. Found: C, 67.48; H, 4.92; N, 0.94. 2b»acetone Application of procedure A with acetone (0.16 mL, 2.1 mmol), tetrol l b (40 mg, 0.061 mmol), K 2 C 0 3 (400 mg, 2.9 mmol), and CH2J3rCl (40 ul, 0.61 mmol) in NMP (20 mL) gave, after recrystallization from CH2Cl2/hexanes, 22.4 mg (54%) of 2b»acetone as a white solid: mp > 250 °C; !H NMR (CDCI3, 400 MHz): 5 6.81 (s, 8H, H d), 6.53 (s, 8H, H e), 6.17 (d, J = 7.5 Hz, 8H, Hf), 4.98 (q, J = 7.4 Hz, 8H, H g), 4.34 (d, J = 7.5 Hz, 8H, H h), 1.68 (d, J = 7.4 Hz, 24H, CH 3), -1.67 (s, 6H, (CH 3) 2CO). MS (MALDI) m/z (rel intensity): 1443 ((M»(CH 3) 2CO + Na +) +; 100); Calcd for C79H 7 0 O 2 5»Na + = 1442. Anal. Calcd for C79H 7o0 2 5: C, 66.85; H, 4.97. Found: C, 66.67; H, 5.14. 2b»benzene Application of procedure B with benzene (2.1 mL, 23 mmol), tetrol l b (40 mg, 0.061 mmol), K 2 C Q 3 (400 mg, 2.9 mmol), and CH 2BrCl (40 ul, 0.61 mmol) in NMP (20 mL) 120 gave, after recrystallization from CH.2Cl2/hexanes, 21 mg (48%) of 2b«benzene as a white solid: mp > 250 °C; IH NMR (CDC13, 400 MHz): 8 6.98 (s, 8H, H d), 6.45 (s, 8H, H e), 6.00 (d, J = 7.4 Hz, 8H, Hf), 4.99 (q, / = 7.4 Hz, 8H, H g), 4.13 (d, J = 7.4 Hz, 8H, H h), 3.82 (s, 6H, C 6 H 6 ) , 1.75 (d, J = 7.4 Hz, 24H, CH 3). MS (MALDI) m/z (rel intensity): 1463 ((M»C 6H 6 + Na +) +; 100); Calcd for C 8 2H 7o0 24*Na + = 1462. Anal. Calcd for C82H 7o0 2 4: C, 68.42; H, 4.90. Found: C, 68.53; H, 5.05. v. Carceplex Intermediates 105-DMSO 104»DMSO 104-DMSO and 105-DMSO A mixture of tetrol l b (1.0 g, 1.5 mmol), K 2 C 0 3 (6.0 g, 43 mmol), and CH 2BrCl (1.0 mL, 15 mmol) in DMSO (125 mL) was stirred at rt for 18 h. The reaction mixture was 121 concentrated in vacuo, water (50 mL) was added, and the slurry was acidified with 2 M HC1. The slurry was extracted with CHCl3:EtOAc 2:1 (3 x 150 mL), and the combined organic extracts were washed with saturated aq. NaHCC>3 (50 mL) and brine (50 mL), and dried over anhydrous MgSC>4. Silica gel (5 g) was added to the organic solution and the solvent removed in vacuo. The silica gel-absorbed sample was dry loaded onto a silica gel gravity column (400 g) and eluted with 3% MeOH in CHCI3. The products were recrystallized from ether/acetone/hexane and dried at 110 °C (0.1 mm Hg) for 24 h affording 105-DMSO (240 mg, 22.1%), 104»DMSO (50 mg, 4.6%) and 2b»DMSO (83 mg, 7.6%) as white solids. 105»DMSO was characterized as follows: mp > 250 °C; !H NMR (CDCI3, 400 MHz): S 6.91 (s, 4H, H a), 6.88 (s, 2H, H b), 6.68 (s, 2H, H c), 6.64 (d, J = 6.4 Hz, 2H, H d or H e), 6.52 (s, 2H, Hf), 6.45 (br, 2H, H d or H e), 6.20 (d, J = 7.6 Hz, 4H, H g or Hg>), 5.97 (d, J = 7.2 Hz, 4H, H g or %), 5.70 (s, 2H, OH), 5.00 (q, J = 7.5 Hz, 4H, H h or Hh'), 4.92 (q, J = 7.5 Hz, 4H, H h or Hh'), 4.63 (br, 4H, H i or H i ' ) , 4.19 (br, 4H, H i or H i ' ) , 1.71 (d, J = 7.5 Hz, 12H, CH 3), 1.68 (d, J = 7.5 Hz, 12H, CH 3), -1.22 (s, 6H, (CH3)2SO). MS (MALDI) m/z (rel intensity): 1451 ((M»(CH 3) 2SO + Na +) +; 100); Calcd for C 7 7 H 7 o 0 2 5 S « N a + = 1450. Anal. Calcd for C 7 7 H 7 o 0 2 5 S « l H 2 0 : C, 63.98; H, 5.02. Found: C, 64.16; H, 4.92. 104»DMSO was characterized as follows: mp > 250 °C; !H NMR (CDCI3, 400 MHz): S 6.91 (s, 4H, H a), 6.69 (s, 4H, H b), 6.56 (s, 4H, H c), 6.04 (d, J = 7.2 Hz, 8H, H d), 5.74 (s, 4H, OH), 4.93 (q, J = 7.3 Hz, 8H, H e), 4.31 (br, 8H, H f), 1.69 (d, J = 7.3 Hz, 24H, CH 3), -1.15 (s, 6H, (CH3)2SO). MS (MALDI) m/z (rel intensity): 1438 ((M»(CH 3) 2SO + Na +) +; 100); Calcd for C 7 6 H 7 o 0 2 5 S « N a + = 1438. Anal. Calcd for C 7 6H 7 o0 2 5S»1.5 H 2 0 : C, 63.28; H, 5.10. Found: C, 63.25; H, 4.94. 122 102 103 monobridged 102 and A,B-bis-bridged 103 A mixture of tetrol lb (656 mg, 1.00 mmol), DBU (0.31 mL, 2.1 mmol), DMSO ( 5.0 mL, 70 mmol), and CH2I2 (0.80 mL, 10 mmol) in nitrobenzene (100 mL) was stirred at 60 °C for 2 h. The reaction mixture was concentrated in vacuo, water (50 mL) was added, and the slurry was acidified with 2 M HC1. The slurry was extracted with EtOAc (3 x 100 mL), and the combined organic extracts were washed with saturated aq. NaHC03 (50 mL) and brine (50 mL), and dried over anhydrous MgS04. Silica gel (5 g) was added to the organic solution and the solvent removed in vacuo. The silica gel-absorbed sample was dry loaded onto a silica gel gravity column (400 g) and eluted with MeOH/CHCl3 (1:9). Reverse phase tic with acetone/H20 (4:1) as eluent was used to follow the progress of the column chromatography. The products were recrystallized from benzene/ether/acetone/hexane and dried at 110 °C (0.1 mm Hg) for 24 h affording lb (120 mg, 18.3%, recovered starting material), 103 (180 mg, 27.2%) and 102 (105 mg, 15.7%) as white solids. 102 was characterized as follows: mp > 250 °C; 123 IH NMR (CDCI3, 400 MHz): S 6.98 (s, 2H, H a), 6.73 (s, 2H, H b ) , 6.72 (s, 4H, H c), 5.94 (d, J = 6.8 Hz, 4H, H e or He'), 5.65 (d, J = 7.0 Hz, 4H, H e or H e'), 5.41 (s, 2H, Hd), 5.33 (br, 6H, H h and Hh'), 4.91 (m, 8H, H f and Hf ) , 4.45 (d, J = 6.8 Hz, 4H, H g or Hg>), 4.33 (d, J = 7.0 Hz, 4H, H g or H g 0, 1.70 (m, 24H, CH 3 ) . MS (DCI, ammonia) m/z (rel intensity): 1342 ((M + NH 4 ) + ; 100). Anal. Calcd for C 7 3 H64024 '1 H 2 0 : C, 65.27; H, 4.95. Found: C, 65.28; H, 5.08. 103 was characterized as follows: mp > 250 °C; IH NMR (CDCI3, 400 MHz): 8 6.82 (s, 4H, Ha), 6.72 (s, 4H, H b), 6.62 (d, J = 6.2 Hz, 2H, H c or Hd), 6.41 (br, 2H, H c or Hd), 6.19 (d, / = 7.4 Hz, 2H, H e ' or H e"), 6.09 (d, / = 7.1 Hz, 4H, H e), 5.91 (d, J = 6.8 Hz, 2H, H e ' or H e"), 5.42 (s, 4H, OH), 5.05 (q, J = 7.4 Hz, 2H, H f or H f •), 4.89 (m, 6H, H f and ( H f or H f 0), 4.61 (br, 2H, Hg> or H g ») , 4.59 (m, 6H, H g and (Hg> or H g »)), 1.72 (d, J = 7.4 Hz, 6H, CH 3), 1.69 (d, J = 7.4 Hz, 6H, CH 3), 1.66 (d, J = 7.4 Hz, 12H, CH 3 ). MS (DCI, ammonia) m/z (rel intensity): 1354 ((M + NH 4 ) + ; 100). Anal. Calcd for C 7 4 H64024«2 H 2 0 : C, 64.72; H, 4.99. Found: C, 64.75; H, 4.68. 124 vi. Competition Experiments. The template ratios obtained directly from competition experiments are reported in Table 2.8 and Table 2.9 along with a series of cross check experiments in Table 2.10. Table 2.3 was constructed by multiplying together the template ratios of adjacent guests (or guests with very similar template abilities) starting with the poorest guest (NMP) which was arbitrarily set to 1. A sample calculation as to how Table 2.3 on page 73 was constructed follows: Template ratios from Table 2.8: DMF/NMP = 7.22/1 DMA/DMF = 2.81/1. The template ratio for DMA/NMP is calculated as follows: 7.22/1 (DMF/NMP) x 2.81/1 (DMA/DMF) = 20/1 (DMA/NMP). This same procedure was used to calculate the template ratios for 24 of the 34 guests in Table 2.3. The template ratios for the original 24 guest molecules was published in 1994.4 The template ratios of these original 24 guest molecules are listed in Table 2.8. Since our publication, ten new guest molecules have been added to Table 2.3; their template ratios are listed in Table 2.9. These new template molecules were added to Table 2.3 based on competition experiments between guests with similar template abilities. Instead of altering our original template ratios, the ten new guests were added to the table based on either a single competition experiment or an average of two or more competition experiments. A sample calculation as to how these ten new guests were added to Table 2.3 on page 73 was constructed follows: 125 Template ratios from Table 2.9: pyrazine/methyl acetate = 2.42/1 methyl acetate/1,4-dioxane = 1.80/1. 1,000,000/1 (pyrazine/NMP) x 1/2.42 (methyl acetate/pyrazine) = 425,000 methyl acetate/NMP 290,000/1 (1,4-dioxane /NMP) x 1/2.42 (methyl acetate/1,4-dioxane) = 520,000 methyl acetate/NMP The average of these two template ratios was incorporated into Table 2.3 = 470,000 methyl acetate/NMP A typical competition experiment was performed as follows: Tetrol la (508 mg, 0.500 mmol) was added to a 250 mL volumetric flask and filled to volume with NMP. Conditions A: Tetrol la was added as a stock solution in NMP (10.0 mL, 2.00 mM) to a 25 mL round bottom flask containing K2CO3 (0.20 g, 1.4 mmol). To this mixture were added 1,4-dioxane, (45 ul, 0.53 mmol), methyl acetate (42 ul, 0.53 mmol) and CH^BrCl (13 ul, 0.20 mmol), and the reaction was stirred at 60 °C for 24 h. An additional 0.20 mmol of CH2B1CI were added and the reaction was stirred for an additional 24 h at 60 °C. The reaction mixture was concentrated in vacuo, water (20 mL) was added, and the slurry was acidified with 2 M HC1. The slurry was extracted with CHCI3 (3 x 20 mL), and the combined organic extracts were washed with saturated aq. NaHC03 (20 mL) and brine (20 mL), and dried over anhydrous MgSC>4. The excess CHCI3 solvent was removed in vacuo and the residue purified by silica gel chromatography using CHCI3 as the eluent. The 126 products were recrystallized from CHCl3/EtOAc to give a mixture of carceplex 2a» 1,4-dioxane and carceplex 2a»methyl acetate (10 mg, 47%). Integration of the resolvable host peaks and guest peaks via 'H NMR gave a 1.80:1 ratio of carceplex 2a»methyl acetate: carceplex 2a# 1,4-dioxane (Table 2.8). Conditions B: Tetrol la was added as a stock solution in NMP (20.0 mL, 2.00 mM) to a 50 mL round bottom flask containing K2CO3 (0.40 g, 1.4 mmol). To this mixture were added acetamide, (340 mg, 5.8 mmol), trioxane (520 mg, 5.8 mmol) and CH^BrCl (13 ul, 0.20 mmol), and the reaction was stirred at ambient temperature for 24 h. An additional 0.20 mmol of CH^BrCl were added and the reaction was stirred for an additional 24 h at 60 °C. An additional 0.40 mmol of CH^BrCl were added and the reaction was stirred for an additional 24 h at 60 °C. The reaction mixture was concentrated in vacuo, water (20 mL) was added, and the slurry was acidified with 2 M HC1. The slurry was extracted with CHCI3 (3 x 20 mL), and the combined organic extracts were washed with saturated aq. NaHC0 3 (20 mL) and brine (20 mL), and dried over anhydrous MgS04. The CHC13 solution and the solvent removed in vacuo and purified via silica gel chromatography using CHCI3 as the eluent. The products were recrystallized from CHCl3/EtOAc to give a mixture of carceplex 2a»trioxane and carceplex 2a,acetamide (6.4 mg, 15%). Integration of the resolvable host and guest peaks via lH NMR gave a 1.64:1 ratio of carceplex 2a»trioxane: carceplex 2a»acetamide (Table 2.8). 127 Table 2.8. List of Competition Results '^•TMWBMI UtfWtUii- iWW WJWWHWiW 1 tmmmwmi y I y y y I y u llll11 y y ill Ulllllll limy lliinin yillll lill lllll 01 lllll Carceplex 2a»Guest 1 Carceplex 2a#Guest 2 Guest 1 Guest 2 pyrazine 1,4-dioxane 3.56 1.00 1,4-dioxane dimethyl sulfide 1.62 1.00 dimethyl sulfide DMSO 2.56 1.00 DMSO 1,3-dioxolane 1.84 1.00 1,3-dioxolane 2-butanone 1.02 1.00 2-butanone pyridine 1.09 1.00 pyridine dimethyl sulphone 1.86 1.00 dimethyl sulphone THF 1.56 1.00 furan THF 1.05 1.00 THF acetone 1.76 1.00 THF thiophene 1.78 1.00 acetone thiophene 1.34 1.00 acetone 2-propanol 3.50 1.00 2-propanol benzene 1.00 1.08 benzene pyrrole 2.38 1.00 2-propanol pyrrole 1.56 1.00 pyrrole tetrahydrothiophene 2.45 1.00 tetrahydrothiophene acetonitrile 5.63 1.00 tetrahydrothiophene acetamide 2.58 1.00 acetamide trioxane 1.64 1.00 trioxane DMA 4.82 1.00 trioxane ethanol 2.14 1.00 1,3-dioxane acetamide 1.33 1.00 1,3-dioxane diethyl ether 9.77 1.00 acetonitrile ethanol 1.20 1.00 ethanol DMA 3.03 1.00 DMA DMF 2.81 1.00 DMF NMP 7.22 1.00 128 Table 2.9. Additional guests added to Competition Experiments. Carceplex 2a»Guest 1 Carceplex 2a»Guest 2 Guest 1 Guest 2 methyl acetate pyrazine 1.00 2.42 methyl acetate 1,4-dioxane 1.80 1.00 ethyl methyl sulfide 1,4-dioxane 1.00 2.61 ethyl methyl sulfide DMSO 2.00 1.00 dimethyl carbonate ethyl methyl sulfide 1.00 2.05 dimethyl carbonate DMSO 1.10 1.00 1,4-thioxane benzene 3.83 1.00 1,4-thioxane THF 1.16 1.00 2,3-dihyrdofuran THF 1.22 1.00 2,3-dihyrdofuran furan 1.00 1.00 pyridazine dimethyl sulphone 1.00 2.27 pyridazine THF 1.00 1.28 1,3-dithiolane benzene 1.83 1.00 2-butanol benzene 1.10 1.00 2-butanol 2-propanol 1.86 1.00 ethyl acetate diethyl ether 2.21 1.00 diethyl ether ethanol 1.00 2.96 Cross check experiments were run to check the validity of the tabulated guest ratios based on adjacent competition experiments. A competition experiment was run starting with a 500:1 ratio of pyrazine:2-propanol as guests which gave, after adjusting for the starting ratio, an 850:1 ratio of carceplex 2a»pyrazine:carceplex 2a«2-propanol. This is in good agreement with the 670:1 ratio derived from our tabulated template ratios in Table 2.3. Likewise, a competition reaction between 2-propanol and NMP was run using 0.1% 2-propanol in NMP which gave, after adjusting for the starting ratio, a 1010:1 ratio of carceplex 2a»2-propanol:carceplex 2a«NMP. This is also in good agreement with the 1500:1 ratio derived from our tabulated template ratios in Table 2.3. Overall, the two cross check experiments yield an 850,000:1 ratio for carceplex 2a»pyrazine:carceplex 2a»NMP, which compares nicely with the 1,000,000:1 ratio from our tabulated template ratios in Table 2.3. Errors for each competition experiment are considered to be < ±20% based on integration of the *H NMR spectrum. We also used our initial mapping competition experiments as cross checks to confirm the validity of the results in Table 2.3. Cross check 129 experiments were performed, the results of which are reported in Table 2.10. All agree within a factor of 2 with template ratios in Table 2.3. Also, competition experiments starting with ratios of guests varying from 1:1 to 9:1 guest A:guest B gave the same template ratio of A:B to within 13% error. Therefore, competition experiments between guests is linearly dependent on their concentrations. Errors for individual template ratios reported in Table 2.3 are estimated to be < ±20%; this error is largely due to integration of the *H NMR spectrum. As template ratios are multiplied together to generate Table 2.3 these errors should increase substantially, but our cross check experiments with NMP:2-propanol and 2-propanokpyrazine indicate the overall error from NMP:pyrazine is approximately +15% (850,000 compared to 1,000,000). Thus, errors for the template ratios must cancel each other out when they are multiplied together, or the error in each individual template ratio is much less than 20%. Errors in the integration of a  lH NMR spectrum are dependent upon: the signal to noise ratio, spin-lattice relaxation (Tj) of the nuclei, and the limited number of data points in an actual signal. In order to minimize these error, a relaxation delay of two seconds was used whereby the sum of the acquisition time and the relaxation delay equaled ~ 4.5 seconds. The T/'s for the host signals of carceplex 2a»pyrazine in CDCI3 are ~ one second or less. Furthermore, an adequate,signal to noise ratio was obtained in the *H NMR spectrum by acquisition of at least 256 scans of the ~ 4 mM solution of the carceplex mixture in CDCI3. 130 Table 2.10. Cross Check Experiments. Carceplex 2a»Guest 1 Carceplex 2a»Guest 2 Guest 1 Guest 2 pyrazine 2-propanol 850 1.00 2-propanol NMP 1000 1.00 1,4-dioxane pyridine 15.00 1.00 1,4-dioxane pyridine 13.30 1.00 THF acetone 1.76 1.00 pyrrole ethanol 16.05 1.00 diethyl ether DMA 2.08 1.00 acetamide pyrrole 1.00 3.01 1,3-dioxane diethyl ether 9.77 1.00 1,4-dioxane DMSO 5.37 1.00 acetone 2-propanol 4.00 1.00 DMSO 1,4-dioxane 1.00 5.37 DMSO 1,3-dioxolane 1.78 1.00 DMSO 2-butanone 2.00 1.00 DMSO 2-butanone 1.88 1.00 DMSO THF 5.15 1.00 DMSO THF 7.74 1.00 DMSO thiophene 13.50 1.00 DMSO pyridine 2.53 1.00 DMSO 1,4-thioxane 5.25 1.00 fur an THF 1.12 1.00 2-butanone THF 2.89 1.00 dimethyl sulphone THF 1.78 1.00 THF acetone 1.55 1.00 thiophene 2-propanol 3.00 1.00 pyridazine benzene 3.55 1.00 Disappearance of tetrol la: Tetrol l a was added as a stock solution in NMP (20.0 mL, 2.00 mM) to a 50 mL round bottom flask containing K2CO3 (0.6 g, 2 mmol). To this mixture was added pyrazine, (170 mg, 2.1 mmol), and GhFiBrCl (12.8 ul, 0.20 mmol), and the reaction was stirred at ambient temperature for various time periods and worked up as usual. The amount of tetrol l a remaining after stopping the reaction was determined by weight. Simultaneously, this experiment was run in the absence of pyrazine as guest. The results of both series of experiments are reported in Table 2.4. 131 E . References 1. Sherman, J. C ; Cram, D. J. J. Am. Chem. Soc. 1989, 111, 4527-4528. 2. Sherman, J. C ; Knobler, C. B.; Cram, D. J. J. Am. Chem. Soc. 1991,113, 2194-2204. 3. Bryant, J. A.; Blanda, M. T.; Vincenti, M.; Cram, D. J. J. Chem. Soc, Chem. Commun. 1990, 1403-1405. 4. Chapman, R. G.; Chopra, N.; Cochien, E. D.; Sherman, J. C. J. Am. Chem. Soc. 1994, 116, 369-370. 5. Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry; 3rd ed.; Plenum: New York, 1990. p 120. 6. Reichardt, C. Solvents and Solvent Effects in Organic Chemistry; 2nd ed.; VCH: New York, 1988, see Chapter 2 and the Appendix. 7. Hildebrand, J. H.; Prausnitz, J. M.; Scott, R. L. Regular and Related Solutions; Van Nostrand Reinhold Company: New York, 1970; pp 27, 213-215. 8. Emmons, W. D.; Ferris, A. F. J. Am. Chem. Soc. 1953, 75, 2257. 9. In order to eliminate the carceplex 2a»bromochloromethane from our reaction mixtures, we explored the possibility of using a different bridging material for the formation of carceplex 2a«guest. Although both dibromomethane and diiodomethane are suitable for the bridging reaction, their template ability surpasses that of the poor guests we were attempting to isolate; therefore, they were not suitable for this purpose. Methylene ditosylate (MDT) is a bridging material that is too large to fit in the interior of a carceplex. Indeed, this bridging material was successfully used in the formation of carceplex 2a#pyrazine. 10. Sherman, J. C. Ph.D. Thesis, University of California at Los Angeles, 1988, pp 183. 11. Koltun, W. L. Biopolymers 1965, 3, 665. CPK models can be purchased from; Harvard Apparatus, Ealing Scientific Ltd. 6010 Vanden Abeele Street, St. Laurent, Quebec, H4S 1R9, 1-800-361-1905, FAX 514-335-3482. 12. Chopra, N.; Sherman, J. C. Supramol. Chem. 1995, 5, 31-37. 13. Cram, D. J.; Tanner, M. E.; Knobler, C. B. / . Am. Chem. Soc. 1991,113, 7717-7727. 14. Fraser, J. R.; Borecka, B.; Trotter, J.; Sherman, J. C. J. Org. Chem. 1995, 60, 1207-1213. 132 3 . Self-Assembling Cavitands A. Self-Assembly of Carceplex 2«Guests The templation effect in the formation of carceplex 2a»guest results from a combination of favorable noncovalent interactions between the walls of the forming carceplex and the guest as well as between adjacent bowls of the host. These favorable noncovalent interactions lead to great selectivity: subtle changes in the guest molecules cause dramatic differences in their propensity for incarceration as demonstrated by the one million-fold template effect observed for formation of carceplex 2a»guest. The formation of carceplex 2a»guest is also highly efficient, proceeding in yields as high as 87% which is far greater than the statistical yield of 60%.!>2 Furthermore, Chopra and Sherman demonstrated that the more statistically demanding formation of hemicarceplex 107»guest proceeded in as high as 57% yield, which is greater than twice the 28% statistical yield.L3 This chapter will explore the nature of these reactions in further detail to determine why they are highly guest selective and why they proceed in much greater than their statistical yields. Section i->xi discuss the characterization of complex 3b«guest, where the guest is usually pyrazine. The relevance of this complex to the formation of carceplex 2»guest is introduced as is the importance of this complex in and of itself. Section xii explores the complexation of a variety of guests in complex 3b«guest; the relative stabilities or guest-selectivity of these complexes and their relevance to carceplex 2«guest is described. Section xiii investigates two-bowl complexes that relate to complex 3b»guest. Section xiv explores the related covalently linked (i.e., one or two bridges) two-bowl complexes. Section xv presents the complexation properties of complex 3b»guest in nitrobenzene-^ 5. Section xvi presents the relative stabilities of various two-bowl complexes (both covalently linked and unlinked hosts), where the hosts are varied and the guest is held constant. The 133 absolute stability of one complex is given in section xvii. Section xviii discusses the relevance of the complexes presented in this chapter to the formation of carceplex 2»guest. Finally, a brief discussion of computations on complex 3b«guest is given in section xix, followed by a summary. Conclusions and future work are presented in Chapter 5. B. Results and Discussion i. Investigations into the Formation of Complexes of Tetrol 1 It was concluded in Chapter 2 that the guest selectivity found in the formation of carceplex 2a»guest is a reflection of the stabilization of the transition state of the GDS by the different guests. We speculate that this guest-dependent transition state stabilization might also be relevant to the better than statistical yields in formation of carceplex 2a»guest as well as hemicarceplex 107»guest. A good model for the transition state of the GDS would be the intermediate that immediately precedes the GDS. An even simpler model would be tetrol 1. That is, can two molecules of tetrol 1 form a complex by wrapping around a guest, prior to covalent bond formation and would such a complex have any bearing on the transition state of the GDS? The formation of such a complex could be stabilized by both intermolecular charged hydrogen bonding of opposing phenolic hydroxyls/phenoxides of two molecules of tetrol 1 as well as by van der Waals interactions between the bowlsvm and the template molecule. Examination of CPK models of tetrol 1 suggest that all four phenols are indeed able to form such hydrogen bonds with a second molecule of tetrol 1 (either neutral or deprotonated), and the cavity formed in such a complex has approximately the same size and shape as that of carceplex 2a»guest. Furthermore, charged hydrogen bonding between phenols and phenoxides has a literature '"' The term bowl is used to describe tetrol and its deprotonated derivatives. 134 precedent. While measuring the pKa's of phenols in NMP, Kolthoff et al. found that phenol and its monosubstituted derivatives hydrogen bond more strongly to their conjugate bases in dipolar nonhydroxylic solvents such as NMP than they do with the solvent (Scheme 3.1).4 Thus, the hydrogen bonded complex B (PhOH—~OPh) is a more dominant species than complex A (PhOH—NMP) in nonhydrolytic solvents.4-5 Extrapolation of this finding to tetrol 1, which has four accessible phenolic groups, suggests that the formation of hydrogen bonds may lead to a dimer of tetrol 1. The discovery and characterization of this and related complexes is the subject of this chapter. Scheme 3.1. Hydrogen Bonding of Phenols in Dipolar Nonhydroxylic Solvents C H 3 CH3 ^ W O H + — < ^ ^ K O H ° ^ \ ^ ) equation 1 A / \ 0 - + H 0 / - \ ^ / / \ o - - H O ^ y ^ \^ J—v — // equation 2 B 135 ii. Determination of Complex la»Guest Using Deuterium NMR NMR spectroscopy is a powerful tool for studying molecular complex formation. This is especially true when studying the binding of guest molecules in an aromatic host because substantial changes in chemical shifts (A8's) are commonly observed for the signals of the guest due to shielding or deshielding of the guest by the aryl-lined cavity. Indeed, guest molecules incarcerated in carceplex 2a»guest manifest large upfield A8 in their ! H NMR spectra of 2-4.5 ppm (see Chapter 4). Thus, if a dimer of tetrol 1 were to encapsulate a guest molecule, one would expect to see the characteristic upfield chemical shifts of the guest peaks. Initially, we used H NMR spectroscopy to explore the formation of a complex between two molecules of tetrol 1, because our reaction solvent NMP could be used in the experiments and deuterated guest molecules such as 1,4-dioxane-ds (dioxane-ds) or DMSO-^6 were readily available. Thus, the 2 H NMR spectrum of dioxane-ds (25 mM) in NMP showed no significant changes when tetrol l a (50 mM; 2 equivalents)"1 was added. Addition of excess potassium carbonate (the solution was saturated) to the NMR tube containing tetrol la and dioxane-ds also resulted in no significant changes in the spectrum. Rerunning this sample two weeks later failed to provide any evidence for complex formation, thus ruling out slow kinetic formation. Therefore, we conclude that little or no complex forms under these conditions. Disappointed by these findings, we decided to further investigate the possibility of complex formation by repeating the above experiment, except the base was substituted with DBU. DBU was chosen because it is a suitable base for carceplex formation, and its high solubility in organic solvents produces homogeneous solutions. The H NMR spectrum of tetrol l a (50 mM), DBU (500 mM), and a stoichiometric amount of dioxane-^ (25 mM) in , x High concentrations of tetrol la were used in the 2H NMR experiments to obtain adequate signal to noise ratio. 136 NMP showed a very broad peak centered at 0.2 ppm (assigned to encapsulated dioxane-Js) along with a small peak at 3.7 ppm corresponding to free dioxane-^ s- This spectrum demonstrates that a molecular complex of tetrol la and dioxane-d% does indeed form (complex 3a»dioxane-<i8) in NMP, and it is in slow exchange on the H NMR timescale at ambient temperature! The encapsulated dioxane-^ H signal is shifted 3.5 ppm from free dioxane-Js, which is consistent with the 3.96 ppm NMR A5 (in C D C I 3 ) of dioxane entrapped in carceplex 2a»dioxane. The similarity of these upfield A8's strongly suggests that dioxane-^ 8 is encapsulated between two molecules of tetrol 1 as illustrated in Scheme 3.2. Further evidence for this type of complex formation was also observed with DMSO-6?6 as guest. Under similar conditions, a new signal appears in the H NMR spectrum at -0.73 ppm, which was assigned to encapsulated D M S O - G ^ - The 3.45 ppm upfield A5 from free DMSO-^6 (2.72 ppm) is consistent with the A8 observed in carceplex 2a»DMSO. No association constants were determined for the complexes because the exact nature of the free species and the complex itself were not well described at the time. The discovery of complexes 3a»dioxane-<i8 and 3a»DMSO-<i6 are the pivotal experiment in this thesis. Molecular encapsulation of a guest molecule via self-assembly of subunits around the guest was largely unknown at the time.6 A battery of experiments were therefore designed to explore the nature of complex 3b»guest, and more importantly to investigate its significance to the formation of carceplexes in general (see section iii for change of pendent group). 137 Scheme 3.2. Encapsulation of Dioxane-^. iii. H NMR Characterization of Complex 3b»Pyrazine lH NMR spectroscopy became a very valuable tool for studying the complexes of tetrol 1 because unlike H NMR spectroscopy where only information about the deuterated guest is obtained, lH NMR provides information about both the host and the guest. We explored the potential use of CDCI3 as a solvent for studying molecular complexes of tetrol 1 with various guest molecules (Scheme 3.3). Methyl groups were chosen as the pendent group of the tetrol (as in lb) in these investigations over phenethyl groups (as in tetrol la) because it greatly simplified the *H NMR spectra.7 138 The H NMR spectrum (spectrum A, Figure 3.1) of a solution of tetrol lb and 0.50 equivalents of the best template molecule for formation of carceplex 2a»guest, pyrazine, in CDCI3 showed no change in the chemical shifts of the host or the guest signals when compared to the lH NMR spectrum of tetrol lb and pyrazine run separately. Thus, there is no complex formation between neutral tetrol lb and pyrazine in CDCI3. Addition of DBU, however, results in two sets of host and guest signals in slow exchange on the lH NMR timescale at ambient temperature as shown in Figure 3.1, spectrum B. The necessity of DBU to form this complex suggests that charged hydrogen bonds (CHB's) form between the two bowls. Most apparent in spectrum B is the appearance of a signal for encapsulated pyrazine at 4.3 ppm which is shifted 4.3 ppm upfield from free pyrazine (8.6 ppm). This 4.3 ppm upfield A8 correlates with the 4.6 ppm upfield A8 observed for incarcerated pyrazine (4.0 ppm) in carceplex 2b«pyrazine (spectrum C, Figure 3.1). Also noticeable in spectrum B is the appearance of two sets of host signals, indicating that both host and guest are in slow exchange on the NMR timescale. Furthermore, integration of the host and guest signals of complex 3b»pyrazine leads to the expected 2:1 stoichiometry for tetrol lb:pyrazine. The lK NMR data strongly suggest that the structure 139 of complex 3b»pyrazine is a charged dimer of tetrol lb interconnected by CHB's with a molecule of pyrazine encapsulated within its interior as illustrated in Scheme 3.3. The important features that drive the self-assembly of complex 3b»pyrazine are likely to be favorable van der Waals and electrostatic interactions between the interior of the dimer and pyrazine as well as favorable noncovalent interactions such as CHB's between the bowls. Several questions now arise from these results: How sensitive is the complex to the type and amount of base? Are there indeed CHB's between the bowls? What is the nature of the "free" species: is it a monomeric bowl, a complex of solvent, an empty dimer, or a combination of these? Is the complex reversible? We therefore designed a series of experiments to address these questions. 140 Figure 3.1. lH NMR of Complex 3b«Pyrazine Free JL B CHCL Free y Encapsulated f JJ Entrapped 0 Jl i—i i—i i i i i |—n-!—i i i i i i |—i i i i i i i i i | i 8 7 6 1 i—rn—r-rn—rn—i i m I I |—I—i— 5 4 P p m *H NMR spectra in CDCI3 at ambient temperature of: A, Tetrol l b (3.7 mM) and pyrazine (1.9 mM). B, Tetrol l b (3.7 mM), pyrazine (1.9 mM), and DBU (7.8 mM). In spectrum B , the host signals assigned to complex 3b*pyrazine are labeled "c" while the host signals assigned to the "free" species are labeled "f". C , Carceplex 2b«pyrazine (8.0 mM). Not shown is the signal for CH3 at -1.7 ppm for all spectra. Specific proton assignments are given in the experimental section. 141 iv. Requirements of Base in the Formation of Complex 3b»Pyrazine In an effort to evaluate the number of equivalents of DBU that are required for formation of complex 3b«pyrazine, a titration was performed on a solution of tetrol lb (4.0 mM) and pyrazine (8.0 mM) in CDCI3 with DBU (1.0 mM) and the results were followed by *H NMR spectroscopy. Integration of the resulting spectra indicated that 0.5 equivalents of DBU per phenolic hydroxyl of tetrol lb are required to reach the maximum yield of complex 3b»pyrazine as graphically displayed in Figure 3.2. We therefore conclude that two molecules of tetrol lb dimerize around the guest via the formation of O" H-0 CHB's, as depicted in Scheme 3.3. Additionally, the titration experiment suggests that the formation of complex 3b»pyrazine is highly cooperative, since the *H NMR spectra of pyrazine and tetrol lb with less than a stoichiometric amount of DBU is composed of complex 3b»pyrazine and a species that resembles tetrol lb (i.e., no intermediates are observed). Figure 3.2. Titration of Tetrol lb and Pyrazine with DBU 100.0 n # equivalents D B U (per O H of tetrol lb) 142 v. Formation of Complex 3b»CHCl3 Titration experiments of tetrol lb in neat CDCI3 with DBU were also conducted to characterize the nature of the free species (see Figure 3.1). These titrations enable us to distinguish between a dimer, where it is anticipated that 0.5 equivalents of DBU are required per hydroxyl group of tetrol lb (forming the four CHB's of complex 3b), and a monomer, where 0-4 equivalents of DBU per hydroxyl group of tetrol lb could be consumed (forming partly or fully deprotonated derivatives of tetrol lb). Two such titrations were conducted; one at high temperature and low concentration of tetrol lb (conditions that would favor a monomer) and the other at high concentration and low temperature (which would favor a charged dimer of tetrol lb). The proton para (Hp) to the hydroxyl of tetrol lb (Scheme 3.3) was used to follow the progress of the titration because it displayed the most dramatic A8. The titration curve of 2.63 mM of tetrol lb at 263 K with DBU approaches its maximum A8 when 0.5 equivalents of DBU were added per hydroxyl group of tetrol lb (Figure 3.3). Therefore, the "free" species at low temperature and high concentration is most likely a complex of two molecules of tetrol lb interconnected by charged hydrogen bonds between the bowls, since only one half of the phenolic hydroxyls are deprotonated. The observation of only one set of host signals suggests that this complex is either 100% complex 3b#chloroform or 100% empty dimer" of tetrol lb or a combination of these two in fast exchange on the *H NMR timescale. In contrast to these results, at low concentrations of tetrol lb (0.047 mM), titration with DBU at 323 K in CDCI3 exhibited no significant variation in the chemical shifts of Hp, suggesting that deprotonation of the phenolic hydroxyls of tetrol lb is not occurring to any significant extent (Figure 3.3). This titration indicates that at this temperature and concentration the "free" species is most likely a monomeric bowl. Interestingly, all spectra * The putative empty dimer could contain dissolved gases. 143 recorded during these titrations displayed only a single set of host signals. Since it is highly unlikely that both the monomer and charged dimer have coincidental lH NMR spectra, the free species observed is most likely the fast exchanging average of monomer and charged dimer. Figure 3.3. Titration of Tetrol with D B U 6.8 -, # equivalents DBU (per OH of tetrol l b ) o 0.047 mM tetrol l b at 323 K o 2.63 mM tetrol l b at 263 K Further investigation into the concentration-dependence of the dimeric species of tetrol lb in C D C I 3 was explored. Thus the lH NMR spectra of a mixture of tetrol lb:DBU (1:2.1) at 298 K were recorded over a concentration range of 0.076 mM - 3.72 mM resulting in chemical shifts of Hp ranging from 6.64 ppm - 6.46 ppm, respectively. On the other hand, the lH NMR spectra of tetrol lb alone recorded over a concentration range of 0.084 mM - 2.36 mM (at 298 K) showed no change in chemical shift of Hp (6.746 ppm). The concentration independence of tetrol lb indicates that there is no detectable charged dimer formation in the absence of base, whereas, the concentration dependence of the tetrol lb:DBU (1:2.1) mixture indicates that at higher concentration the 144 extent of dimerization increases. (The titration experiments described above indicate that the monomer has a chemical shift for Hp of ~ 6.74 ppm while the charged dimer has a chemical shift for Hp of ~ 6.43 ppm, see Figure 3.3.) Again, the appearance of only a single set of host signals strongly suggests that the charged dimer of tetrol lb and the monomer are in fast exchange on the } H NMR timescale at 298 K. The titrations experiments of the free species give no information regarding the chemical species (or lack thereof) encapsulated within this charged dimer of tetrol lb. Therefore, two experiments were designed to determine whether chloroform was encapsulated in the "free" species. The first experiment was designed to show binding of CHCI3 in complex 3b»guest by choosing a solvent that cannot act as a guest. Nitrobenzene-^ was thus chosen as the solvent due to its bulkiness, its stability towards DBU, and its ability to solubilize all the components of the reaction. The addition of CHCI3 (5.89 mM, 1.8 equivalents per tetrol) to a solution of tetrol lb (3.27 mM) and DBU (13.4 mM) at 313 K led to the formation of a new set of host and guest signals, indicating that a complex had formed in slow exchange on the lH NMR timescale. The signal at 5.38 ppm (Figure 3.4) was assigned to encapsulated CHCI3 on the basis of its integration (1 molecule of CHCI3 per two molecules of tetrol) and the 2.15 ppm upfield A8 from free CHCI3, which correlates with the 2.81 ppm (measured in CDCI3) upfield A5 observed for incarcerated CHCI3 in carceplex 2a»CHCl3. The amount of complex 3b»CHCl3 was also found to be directly proportional to the amount of CHCI3 added to the lU NMR sample. Also, decoupling the resonance for free CHCI3 (7.53 ppm) at 313 K resulted in a 50% reduction in the intensity of the peak for encapsulated CHCI3, indicating that these resonances are involved in chemical exchange.8 Furthermore, addition of excess CDCI3 to this sample causes the signal at 5.38 ppm to disappear completely due to displacement of encapsulated CHC1 3 in complex 3b»guest by the CDCI3. These experiments strongly suggest that CHCI3 is indeed encapsulated in complex 3b»guest in nitrobenzene-^ 5. 145 Figure 3.4. Formation of Complex 3b»CHCl3 in Nitrobenzene-^. encapsulated CHCI3 T 1 1 1 1 1 1 1 1 1 1 1 r Partial ! H NMR spectrum of tetrol lb (3.27 mM), DBU (13.4 mM), and CHCI3 (5.89 mM) in nitrobenzene-J5 at 313 K. Assignments: free species (4.6-5.0 ppm); Hjn of complex 3b»CHCl3 (5.20 ppm); encapsulated CHCI3 (5.38 ppm); methine of free species and complex 3b»CHCl3 (5.49 ppm); not shown is free CHCI3 (7.53 ppm). The ability of complex 3b«CHCl3 to form in nitrobenzene-ds suggests that this same complex could also form in chloroform as solvent. We thus looked for encapsulation of CHCI3 in CDCI3 by performing a solvent suppression experiment on a 50/50 v/v CHCI3/CDCI3 solution containing tetrol lb (5.0 mM) and DBU (10.5 mM).8 Performing a PI331 solvent suppression experiment8 on this sample at 298 K, however, failed to give any indication that CHCI3 was encapsulated (spectrum A, Figure 3.5). Speculating that the exchange rate of the guest in complex 3b»chloroform may be fast on the *H NMR timescale, the experiment was repeated at a lower temperature. Thus, when the PI331 solvent suppression experiment was repeated at 273 K, a new peak appeared at 4.6 ppm (spectrum B, Figure 3.5) that was assigned to encapsulated CHCI3 within complex 3b»CHCl3. This 2.64 ppm A5 shift of encapsulated CHCI3 in chloroform correlates with 146 the 2.81 ppm upfield A8 observed for incarcerated CHCI3 in carceplex 2a»CHCl3. Even though the guest signals are in slow exchange on the A H NMR timescale at 273 K in chloroform, only one set of host signals were observed. This indicates that the host signals represent either 100% complex 3b»chloroform or a fast exchanging average of complex Sb^chloroform with other species such as empty charged dimer and/or monomer. Unfortunately, the PI331 experiment does not provide accurate integration, but crude relative integration of the methine peak of tetrol lb indicates that the complex is at least 70% filled with chloroform (35% occupied with CHCI3 and 35% occupied with CDCI3). The spectrum, however, remains unchanged down to 223 K. These results strongly suggest that the free species in chloroform (see spectrum B, Figure 3.5) is predominately complex 3b*CHCl3 and possibly another as of yet unidentified species (e.g., monomer and/or empty dimer), in fast exchange. 147 Figure 3.5. Solvent Suppression of Tetrol lb complex in 50/50 v/v CHCI3/CDCI3 A I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I > I I I I I I I I I I I I I I I I I I I I M I I I I I I I I I M 5:2 5.0 4.8 4.6 p p m PI 331 solvent suppression experiment of a 50/50 v/v CHCI3/CDCI3 solution containing tetrol l b (5.0 mM) and D B U (10.5 mM). The quartet in both A and B corresponds to the methine of tetrol l b . A: temperature = 298 K; B : temperature = 273 K, the singlet at 4.6 ppm is encapsulated CHCI3. 148 vi. Charged Hydrogen Bonds in Complex 3b«Pyrazine The titration experiments discussed above indicate that half of the phenols are deprotonated to give CHB's in both complex 3b»pyrazine and complex 3b»CHCl3, yet no signal for the CHB was observed for either species in the L H NMR spectra. This is most likely due to its fast exchange on the J H NMR timescale. We therefore further explored the *H NMR of complex 3b»pyrazine, whose exchange rate is slower than 3b»CHCl3, hoping that decreasing the temperature may allow for detection of the CHB. The *H NMR spectrum of tetrol lb (5.5 mM), pyrazine (2.8 mM), and DBU (11.6 mM), at 215 K in CDCI3 (mp = 209 K) exhibited a new peak at 13.6 ppm that integrated for 8 protons per total complex. This was assigned as the fast exchanging average between the CHB of tetrol lb and the protonated DBU (DBUH+) counter ion. Various conditions were then explored to find the isolated CHB signal of the complex 3b»pyrazine. Tetrabutyl-ammonium (TBA) hydroxide was chosen as the base for complex formation to eliminate the possibility of proton exchange between complex 3b»guest and the counter ion. The resultant complex [3b»pyrazine][TBA]4 was kinetically stable as it exchanged its guest over a period of hours at ambient temperature in acetone- .^ CD3CN, CDCI3 and DMSO-d6- The ! H NMR spectrum of complex [3b«pyrazine][TBA]4 in acetone-^ at 200 K showed a new signal at 15.6 ppm (Figure 3.6) which was assigned to the CHB on the basis of its integration for 4 hydrogens per complex. The large downfield shift indicates that the CHB connecting the bowls is a strong9'10 hydrogen bond. The chemical shift of the CHB protons was highly sensitive to small amounts of water (even though the sample was run containing molecular sieves, which catalyze exchange with acetone and the condensation of acetone, generating more water). Thus, the A5 of protons versus deuterons,11'12 which is an indication of hydrogen bond strength, was not possible. The CHB is discussed further in section xi. Precise assignment of the CHB signal of complex [3b»pyrazine][TBA]4 proved to be a more daunting task than 149 anticipated: (1) Complex [3b»pyrazine][TBA]4 has a half-life for guest exchange on the order of hours at ambient temperature in acetone-^ : the spectrum reaches equilibrium when the ratio of complex [3b»pyrazine][TBA]4: complex [3b»acetone][TBA]4 is -2:1. Cooling the sample to lower temperatures greatly increased the half-life for guest exchange of this complex. (2) The CHB protons readily exchange with a deuterium from acetone-^ or residual D 2 0 present in the NMR solvent. After a period of a few hours the CHB signal would completely disappear even at 200 K; addition of a tiny amount of H 2 0 to this sample, however, regenerated this signal. Therefore, a more accurate integration for the CHB of complex [3b*pyrazine][TBA]4 was achieved by adding the solid complex [3b»pyrazine][TBA]4 salt to a lH NMR tube containing acetone-^ with 10 additional equivalents of pyrazine, and this sample was allowed to equilibrate for only ten minutes in the *H NMR spectrometer at 200 K before acquisition of the spectrum. 150 Figure 3.6. lH NMR of [complex 3b»pyrazine][TBA]4 in Acetone-J6 at 200 K I " 1 16.0 15.0 7.0 6.0 5.0 4.0 PPM 3.0 2.0 1.0 0.0 [complex 3b»pyrazine][TBA]4 2.0 mM in acetone-dg; 'H NMR assignments: 15.5 ppm CHB; Hp 6.47 ppm; H o ut 6.07 ppm; methine 4.78 ppm; encapsulated pyrazine 4.25 ppm; Hjn 4.01 ppm, 3.8 H 20; (3.40, 1.61, 1.54, and 1.30 TBA); 0.85 CH 3 . Not shown; free pyrazine at 8.6 ppm. vii. Is Complex 3b»guest pH-Switchable? Perhaps the most intriguing feature of complex 3b»guest is that it is formed by noncovalent interations. This makes it a particularly attractive target for a molecular switch. The titration experiments reported earlier clearly indicate that base is necessary for formation of complex 3b#guest. Therefore, the neutralization of the base by addition of an acid should destroy complex 3b»guest and regenerate the free guest and tetrol lb. To explore whether or not complex 3b»guest is indeed pH-switchable, trifluoroacetic acid 151 (TFA) was added to a H NMR sample of complex 3b»guest. Addition of one equivalent (3.44 mM) of TFA to a solution of tetrol lb (3.44 mM), pyrazine (1.72 mM), and DBU (7.22 mM), at ambient temperature reduced the amount of complex 3b»pyrazine by one half. Addition of a further 1.1 equivalents of TFA (7.22 mM total) to this sample resulted in disappearance of all of the pyrazine complex and gave a spectrum that consisted largely of tetrol lb and free pyrazine. Therefore, complex 3b»guest is indeed pH-switchable, which may be a useful property for controlling the release of an encapsulated guest. viii. Stability of Complex 3b»Pyrazine in DMSO. DMSO is a very strong hydrogen bonding solvent and is notorious for disrupting hydrogen bonded networks of supramolecular complexes. In fact, the stability of supramolecular complexes are often measured by the percentage of DMSO they can withstand before the complex is destroyed.13 We therefore tested the stability of complex 3b»guest by recording its *H NMR spectrum in neat DMSO-^ as the solvent. The addition of DBU (22.2 mM, 0.5 equivalents/ hydroxyl of tetrol lb) to a solution of tetrol lb (11.1 mM) in DMSO-fi?6 resulted in the appearance of a new set of host signals (assigned to complex 3b#DMSO) in addition to the original signals for tetrol lb. Both sets of host signals were broad, but addition of more DBU (178 mM) resulted in a sharpening of both sets of signals as well as an increase in the intensity of the signals of complex 3b»DMSO relative to those of tetrol lb. Furthermore, a P1331 solvent suppression experiment on a solution of tetrol lb (7.0 mM) and DBU (28 mM) in a 50/50 v/v mixture of DMSO-^VDMSO at ambient temperature resulted in a new signal at -1.26 ppm, which was assigned to encapsulated DMSO within complex 3b«DMS0.8 Crude integration of the signal for encapsulated DMSO to the signals assigned to complex 3b«DMSO gave the expected 2:1 ratio of tetrol lb:DMSO. These results encouraged us to extend this work with pyrazine (our best guest to date) in DMSO-fl^- Thus, addition of pyrazine (115 mM) 152 to a solution of tetrol l b (11 mM) and DBU (22 mM) at ambient temperature in DMS0-e?6 resulted in a new set of host and guest signals in the NMR spectrum which were in slow exchange on the A H NMR timescale. Of these signals, a singlet at 4.10 ppm ascribable to encapsulated pyrazine provided clear evidence for the formation of complex 3b«pyrazine. In addition, this spectrum indicated the presence of complex 3b»pyrazine, complex 3b»DMSO-d6, and tetrol l b in a 1.5:1.4:1 ratio, respectively. These experiments provide clear evidence that the hydrogen bonds that interconnect the two bowls of complex 3b»guest are strong enough to form even in DMSO. In contrast to these results, no evidence for formation of complex 3b»guest was observed in polar protic solvents such as C D 3 O D . In fact, complex 3b»pyrazine in CDCI3 can be eliminated by the addition of 10% C D 3 O D as indicated by the disappearance of the 5 for encapsulated pyrazine and increase in intensity of the 8 for free pyrazine in the lH NMR spectrum. ix. Electrospray Mass Spectrometry of Complex 3b«Guest Electrospray ionization is a mild method of mass spectrometry that has been traditionally used for characterization of proteins, but has recently been applied to the characterization of supramolecular complexes.14-16 Electrospray ionization mass spectrometry (ESMS)17 is based on three principles: First, a high electric field is applied to a sprayer causing the formation of small highly charged droplets by electrostatic dispersion. Second, desorption of the solvent from the droplets renders the ions into the gas phase. Finally, mass analysis of the ions occurs in the mass spectrometer.17 ESMS, therefore, does not require energy to ionize the sample but instead extracts ions from a solution of preformed ions. Proteins are particularly well suited for ESMS because they are readily protonated or deprotonated to create highly charged species in polar solvents. Unlike proteins, most supramolecular complexes are not stable in protic solvents and do not readily accept charges. Some ingenious methods have, however, been devised to 153 overcome this problem. For example, Lehn et al. incorporated a crown ether into a subunit of a neutral complex whereby complexation of a potassium ion enabled the detection of their complexes by ESMS. 1 5 Instead of chemical modification of the complex itself, Whitesides et al. used the weak ion pair of P(Ph)4Cl as a charge donor; the transfer of a chloride ion to a neutral complex in chloroform solution enabled complex detection by ESMS. 1 4 Fortunately for us, complex 3b»guest is already multi-charged; therefore, it is well suited for characterization via ESMS. ESMS of kinetically stable complex [3b»pyrazine][TBA]4 in acetonitrile resulted in the detection of a series'" of signals which included complex 3b»pyrazine (Figure 3.7 and Table 3.1).18 The signals at 462.3 m/z and 695.2 m/z were assigned to the triply and doubly deprotonated complex 3b»pyrazine, which are denoted as [(2B«P)(H+)]' 3 and [(2B»P)(H+)2l"2, respectively. The high resolution capabilities of ESMS (resolution 0.25 amu) allows the determination of the charged state of detected signals. For example, if the predominant isotope is 1 3 C (which it is in our case), signals that are 1/1, 1/2, 1/3, or 1/4X" atomic mass units (amu) apart would represent a singly, doubly, triply, or quaternary charged species, respectively. Therefore, the determination of the charge of a signal allows accurate assignment of the species it represents. For example, we observed a signal at 656.0 m/z for the complex [3b«pyrazine][TBA]4. This signal could be assigned to either the singly deprotonated tetrol lb (655.6 m/z calcd) or to doubly deprotonated dimer of tetrol lb (655.6 m/z calcd) which have the same m/z ratio. As shown in spectrum (b) in Figure 3.8, the high resolution spectrum of this region confirms that this signal corresponds to a doubly deprotonated species because the signals vary by 0.5 amu. Therefore, this peak can be correctly assigned to the doubly deprotonated dimer of Counter ion exchange from T B A to H + further complicated the spectrum. x i i The separation of signals in an isotopic cluster is based on the ratio m/z = mass/charge. For example, if a species has a charge of two the separation of signals is m/z = 1/2 or 0.5 amu (the predominate isotope is 1 3 C , [(2B«P)(H +)]" 3 = ( C 7 6 H 6 5 0 2 4 N 2 ) - 3 ; M + 1 = 80%, M + 2 = 36%,). 154 tetrol la. Figure 3.8 also shows a spectrum of a triply charged species (a) and one of a singularly charged species (c). ESMS is a powerful tool for characterizing complex 3b«pyrazine in solution and may someday be indispensable for the characterization of the next generation of supramolecular structures based on the tetrol complex. Table 3.1. Observed ionic species from the ESMS of complex [3b»pyrazine][TBA]4. m/z observed Assignment Charge(-) m/z calculated Intensity (%) 327.2 [(B)]"2 2 327.3 80 367.3 [(B.P)(H+)2]-2 2 367.4 28 462.3 [(2B.P)(H+)]"3 3 463.4 82 656.0 [(2B)(H+)2]-2 2 655.6 100 695.2 [(2B.P)(H+)2r2 2 695.7 59 776.1 [(2B)(TBA+)(H+)]"2 2 776.4 80 816.1 [(2B«P)(TBA+)(H+)r2 2 816.4 60 895.7 [(2B)(TBA+)2r2 2 897.1 29 937.1 [(2B«P)(TBA+)2r2 2 937.1 21 1551.5 [(2B)(TBA+)(H+)2r 1 1553.7 18 1633.4 [(2B.P)(TBA+)(H+)2r 1 1633.9 19 Symbols: B = (tetrol lb - 2H+); P = pyrazine; TBA = tetrabutylammonium. 155 Figure 3.7. ESMS of Complex [3b«pyrazine][TBA] 4. 100 75 CO c 0 c 50 CD > 05 0 25 0 [(2B.P)(H+)]"3 [(B)! -2 I oJuiliU [(2B)(H+)2]-1 [(2B.P)(H+)J-2 [(2B)(TBA+)(H+)]'2 [(2B»P)(TBA+)(H+)]"2 500 [(2B)(TBA+)J-2 [(2B-P)(TBA+)J-2 [(2B)(H+)J-1 [(2B)(TBA+)(H+)2]"1 [(2B.P)(H+)3]"1 [(2B«P)(TBA+)(H+)2]"1 JLdjJiJJ 1000 m/z 1500 2000 E S M S o f a 0.1 m M s o l u t i o n o f c o m p l e x [ l b # p y r a z i n e ] [ T B A ] 4 i n acetonit r i le at ambient temperature. S y m b o l s a b o v e : B = tetro l lb - 2H+); P = p y r a z i n e ; T B A = t e t r a b u t y l a m m o n i u m . Figure 3.8. Determination of Charge using ESMS. 1632.0 1634.0 m/z Determination of molecular charge by high resolution ESMS of selected mass regions for complex [3b»pyrazine][TBA]4. Signals that differ by 1/1, 1/2, or 1/3 atomic mass units (amu) have a charge of 1, 2, and 3 respectively. Symbols below: B = (tetrol lb - 2H + ) ; P = pyrazine; T B A = tetrabutylammonium. (a) m/z observed 463.2, assignment [(2B»P)(H +)]" 3 , m/z calculated 463.4, intensity 100%. (b) m/z observed 655.0, assignment [(2B)(H+)2]" 2, m/z calculated 655.6, intensity 100%. (c) m/z observed 1632.3, assignment [(2B«P)(TBA + )(H + ) 2 ]" 1 , m/z calculated 1633.9, intensity 51%. 157 x. Orientation of Pyrazine in Complex 3b«Pyrazine Both *H NMR spectroscopy and ESMS data strongly suggest that pyrazine is tightly encapsulated within the interior of complex 3b#pyrazine. Therefore, we investigated the orientation and mobility of pyrazine within complex 3b»pyrazine using a method developed in house by Fraser et al. that takes advantage of the different anisotropic environments of carceplexes that differ in their pendent groups (or feet).7 For example, the chemical shift of pyrazine in carceplex 2a»pyrazine (all pendent groups = CH2CH2Ph) and carceplex 2b»pyrazine (all pendent groups = CH3) are 4.30 ppm and 4.07 ppm, respectively, in nitrobenzene-^ . Fraser et al. synthesized asymmetric carceplex 2d»pyrazine (pendent groups = CH2CH2PI1 on bottom and = CH3 on top; schematically represented in Figure 3.9) by combination of tetrol l a and tetrol lb and separated the statistical mixture of carceplexes 2a»pyrazine, 2b»pyrazine, and 2d»pyrazine.7 The *H NMR signals of pyrazine in carceplex 2d»pyrazine consisted of two meto-split doublets (8 4.24 (7 = 1.2 Hz); 4.13 (J = 1.2 Hz)) separated by 0.11 ppm in nitrobenzene-^ at ambient temperature. The appearance of two signals for the protons of pyrazine suggests that pyrazine rotates slowly on the LH NMR timescale about the pseudo-C2 axes of the host as shown in Figure 3.9. Furthermore, the observation of meta coupling (7 = 1-2 Hz) confirms that pyrazine sits with its nitrogens at the equator as illustrated in orientation A (as opposed to orientation B ) as ortho coupling (7 = 6-7 Hz) would have been observed for orientation B (Figure 3.9). The orientation of pyrazine in carceplex 2a«pyrazine was subsequently confirmed by crystallography as described earlier in Figure 2.5. 158 Figure 3.9. Orientation of Pyrazine. (observed) (not observed) Schematic representation of asymmetric carceplex 2d»pyrazine and complex 3c»pyrazine. Pyrazine is orientated with its nitrogens at the "equator" of the complex as shown in A because meta coupling is observed for the nonequivalent hydrogens H a (d, J = 1 Hz) and Hb (d, J = 1 Hz). The vertical lines connecting the two bowls represent OCH2O bonds and CHB's for asymmetric carceplex 2d'pyrazine and complex 3c»pyrazine, respectively. Consequently, we applied Fraser's strategy to complex 3«pyrazine by recording the ! H NMR spectrum of a mixture of tetrol la (4.12 mM), tetrol lb (4.26 mM), DBU (24.0 mM) and pyrazine (21.0 mM), in nitrobenzene-^ . The *H NMR signals of the encapsulated pyrazine represented a statistical mixture of complex 3a»pyrazine, complex 3b«pyrazine, and asymmetric complex 3c»pyrazine in a 1:2:1 ratio, respectively as shown in Figure 3.10. The signals for pyrazine in asymmetric complex 3c»pyrazine consisted of two meto-split doublets at 4.31 ppm (7=1 Hz) and 4.35 ppm (7=1 Hz) in nitrobenzene-6?5 at ambient temperature.19 This confirms that pyrazine is oriented in complex 3b»pyrazine with its nitrogens at the equator and its hydrogens extending into the bowls as schematically represented by structure A in Figure 3.9. The activation energy for rotation of pyrazine about the pseudo-C2 axes in asymmetric complex 3c»pyrazine was measured by variable temperature *H NMR spectroscopy to be 18 kcal/mol which,20 agrees well with 159 the 19 kcal/mol activation barrier measured in carceplex 2d«pyrazine. In conclusion, these experiments confirm that pyrazine has the same orientation and mobility within complex 3»pyrazine and carceplex 2»pyrazine. Figure 3.10. Asymmetric Complex 3»Pyrazine. T 4.4 4.3 ppm N M R spectrum of pyrazine encapsulated in complexes 3a»pyrazine (4.36 ppm), 3c»pyrazine (d, J = 1 Hz at 4.35 and 4.31 ppm), 1 9 and 3b»pyrazine (4.29 ppm) at ambient temperature in nitrobenzene-^. 4© 3a»guest; R = R' = C H 2 C H 2 P h 3b«guest; R = R' = C H 3 3c»guest; R = C H 2 C H 2 P h , R' = CH3 xi. Crystal Structure of Complex 3b»Pyrazine To complement the solution phase characterization of complex 3b»pyrazine by H NMR and ESMS, we characterized this complex in the solid state by determining its X-ray crystal structure.21 Light yellow crystals of complex 3b»pyrazine were grown in nitrobenzene from a mixture of tetrol lb (14 mM), DBU (28 mM) and pyrazine (114 mM). Two partial views of the crystal structure of complex 3b»pyrazine that emphasize the CHB arrangement in the complex and the orientation of pyrazine are shown in Figure 3.11. Only one of the two degenerate orientations of pyrazine is displayed in Figure 3.11; the other orientation of pyrazine is perpendicular to the one shown. Also, the unit cell of this crystal 160 structure contains two crystallographically unique assemblies of complex 3b»pyrazine which differ only in the direction of the chiral twist of the upper and lower bowls with respect to each other; only one assembly is shown in Figure 3.11. The most apparent observation one can make from the crystal structure of complex 3b»pyrazine is that the pyrazine is encapsulated within the interior of two bowls, which are connected to each other by four charged hydrogen bonds. Pyrazine sits with its nitrogens along the equator and its hydrogens extending into the aromatic cavity of the bowls as previously predicted by *H NMR spectroscopy. Figure 3.11 also shows the CHB from a perspective such that the observer is in the complex looking out. This allows one to clearly see both the O-H—O hydrogen bond that connects the two arenes of the bowls and the N-H—O hydrogen bond between the phenoxide of the bowl and the protonated nitrogen of DBU. The remainder of the O-H—O hydrogen bonds and the N-H—O hydrogen bonds are equivalent to the ones shown due to the four fold symmetry of the complex. The crystal structure shows that the CHB O O distance is 2.59 A which is indicative of strong O-H—O hydrogen bonds.9 Also, the O N distance at 2.74 A is characteristic of strong N-H—O hydrogen bonds.9'10'22 The crystal structures of complex 3b«pyrazine and carceplex 2b»pyrazine show some remarkable similarities. In fact, the only major difference between the two are the noncovalent bonds that connect the bowls of complex 3b»pyrazine versus the covalent bonds that connect the bowls of carceplex 2b»pyrazine. Therefore, the remainder of the discussion of the crystal structure of complex 3b»pyrazine is as a comparison with the crystal structure of carceplex 2b,pyrazine. 161 Figure 3.11. Crystal Structure of Complex 3b»pyrazine. A view of the C H B from within the complex showing both A view of the crystal structure of complex 3b»pyrazine the O - H — O and N - H — O bonding arrangements. The showing the orientation of pyrazine. D B U ' s not shown, opposing phenol/phenoxide is shown in the foreground with D B I > H + in the background (center). Figure 3.12 shows the crystal structures of carceplex 2b*pyrazine and complex 3b«pyrazine side by side. It is evident that pyrazine displays the same orientation in both structures with the nitrogens at the equator and its hydrogens extending into the bowls. Furthermore, the bowls are twisted with respect to each other about the C4 axis by 21.0° and 21.8° for the carceplex and complex, respectively, which allows for conjugation of the aryl ethers or phenols/phenoxides with their respective aromatic rings. As mentioned in Chapter 2, the conjugation of aryl ethers likely provides 16-24 kcal/mol of stabilization energy.7'23 The twist of the bowls also maximizes van der Waals interactions between both the bowls and the guest and between the upper rims of the bowls by providing more close contacts; the twist also reduces steric strain between the opposing methylenes that line the rims of the bowls. The potential for conjugation is demonstrated in Figure 3.12 by structures A and B for carceplex 2b»pyrazine and by structures C and D for complex 3b«pyrazine. Views B and D show that both the C - O - C and the C - O - H bonds lie in the plane of the aromatic ring thus maximizing the extent of conjugation between the oxygen lone pairs and the aromatic rings. The distances between the hydrogens of pyrazine and the closest aromatic carbon atom of the top and bottom bowls are 2.78 A and 2.90 A for complex 3b»pyrazine, and 2.73 A and 2.90 A for carceplex 2b#pyrazine. These close contacts indicate that these hydrogens are close enough to form C H - T C hydrogen bonds.24 163 Figure 3.12. Crystal Structures of Carceplex 2b»Pyrazine and Complex 3b»Pyrazine. Crystal Structure of Carceplex 2b*Pyrazine Crystal Structure of Complex 3b»Pyrazine A , B; Partial views of the crystal structure of carceplex 2b*pyrazine indicating the conjugation of the OCH2O bonds into the arenes of the bowls. C, D; Partial views of the crystal structure of complex 3b»pyrazine indicating the conjugation of the O-H—O bonds into the arenes of the bowls. 164 Figure 3.13. V i e w of the Planes Connecting the A r y l Ethers of Carceplex 2b»Pyrazine and the Phenols/Phenoxides of Complex 3b»Pyrazine. Top view of carceplex 2b*pyrazine Top view of complex 3b*pyrazine Side view of carceplex 2b*pyrazine Side view of complex 3b*pyrazine Partial view of the crystal structures of carceplex 2b»pyrazine (left) and complex 3b»pyrazine (right). The planes which connect the aryl ethers of carceplex 2b»pyrazine and the phenols/phenoxides of complex 3b«pyrazine are drawn in. The distance between the planes in carceplex 2b»pyrazine is 2.23 A as calculated by the 2.36 A distance from O-to-0 and a 21.0° twist of the bowls with respect to each other (cos(21.0) x 2.36 A = 2.23 A). Similarly, the distance between the planes in complex 3b»pyrazine was calculated to be 2.43 A. Figure 3.13 shows the top and side views of the planes that connect the aryl ethers of carceplex 2b*pyrazine and the phenols/phenoxides of complex 3b«pyrazine. The top view of complex 3b»pyrazine shows that the nitrogens of pyrazine are not involved in hydrogen bonding with the O - H — O hydrogens, as the closest such N — H distance is 3.8 A. Also , the top view indicates that the nitrogens of pyrazine are involved in weak hydrogen bonds to half of the hydrogens of the intra-bowl methylenes as shown in Figure 3.13. The closest N- to-H distances for complex 3b»pyrazine are 2.32 and 2.48 A for the 165 lower and upper bowls, respectively, while the closest N-to-H distances for carceplex 2b»pyrazine are at 2.41 and 2.64 A for the lower and upper bowls, respectively. The other half of the methylene bridges (the ones perpendicular to the aromatic ring of the guest) are 3.3-3.5 A away from the plane of pyrazine indicating weak C H - T U hydrogen bonds between these hydrogens and the it system of pyrazine. Figure 3.13 also demonstrates the best view of the chiral twist of the bowls with respect to each other for each host. The distances between the planes which connect the aryl ethers of carceplex 2b«pyrazine and the phenols/phenoxides of complex 3b»pyrazine are 2.23 A and 2.43 A (respectively) indicating that the carceplex is more compressed than the complex. These differences in distances may also help explain the slightly higher activation energy barrier for rotation of pyrazine within carceplex 2b»pyrazine (19 kcal/mol) versus complex 3b»pyrazine (18 kcal/mol). The crystal structure of complex 3b»pyrazine has provided a wealth of information about the noncovalent interactions that drive its formation. Complex 3b»pyrazine is clearly stabilized by the strong CHBs that interconnect the bowls, CH-7C hydrogen bonds between the hydrogens of pyrazine and the arenes of the bowls, weak hydrogen bonds between nitrogens of pyrazine and the hydrogens of the intra-bowl methylenes as well as weak CH-7C hydrogen bonds between the hydrogens of the intra-bowl methylenes and pyrazine, favorable van der Waals interactions between pyrazine and the interior of the bowls, and conjugation of the phenol/phenoxides with their respective aromatic rings. Pyrazine's orientation in complex 3b»pyrazine with its nitrogens at the equator of the host aligns the narrower cross section of pyrazine with the narrower part of the host's cavity thereby providing the most favorable van der Waals interactions. The many structural similarities between carceplex 2b«pyrazine and complex 3b»pyrazine strongly suggest that the formation of the latter contributes to the efficiency observed in the formation of carceplexes. We wanted to explore whether complex 3b»pyrazine exhibited the same degree of guest selectivity as that determined for the 166 formation of carceplex 2a»guest. To do this we explored the encapsulation of various guest molecules within complex 3b»guest and determined the relative selectivities of these new complexes by measuring their relative stability constants. xii. Selectivity of Complex 3b»Guest Towards Successful Templates in The Formation of Carceplex 2a»Guest a Introduction Complex 3b»pyrazine and carceplex 2b»pyrazine exhibit some remarkable structural similarities as is evident by their crystal structures and  lH NMR spectra. The next obvious questions to be addressed were: (1) Is complex 3»guest selective towards the guest molecules it encapsulates?, and (2) Would the selectivity observed correlate with the selectivity determined for the formation of carceplex 2*guest; i.e., is complex 3»guest a good transition state model for the GDS in the formation of carceplex 2»guest? Previously, we showed that complex 3b«pyrazine predominated over complex 3b»CDCl3 even though there is a 10,000 fold excess of CDCI3, which suggests that there is a remarkable range of selectivity found for formation of complex 3b»guest (see section v). Furthermore, both complexes 3a»dioxane-d8 and 3a»DMSO-fl?6 were preferentially formed over complex 3b*NMP in NMP as solvent (see section ii), and complex 3b»pyrazine was preferentially formed over complex 3b»DMSO in DMSO-J6 as solvent (see section viii). Thus, we began to investigate the selective encapsulation of a number of guest molecules such as dioxane, DMSO, pyridine, acetone, and benzene in complex 3b«guest in CDCI3. Just as was observed with pyrazine, the addition of each of the other five guest molecules resulted in the formation of new host and guest signals in the presence of tetrol lb and 2.1 equivalents of DBU in CDCI3, indicating that complex 3b»guest had formed. It was, however, necessary to add increasing concentrations of 167 guest for the series pyrazine > dioxane > DMSO > pyridine > acetone = benzene to achieve a ratio of complex 3b#guest: the free species (which was mainly composed of complex 3b*CHCl3) of ~3:1. For example, it was only necessary to add a stoichiometric amount of pyrazine (2 mM) to achieve this 3:1 ratio, but it was necessary to add 500 mM acetone or benzene."1" These preliminary binding experiments indicate that complexes 3b»guest have the following relative stabilities: 3b»pyrazine > 3b»dioxane > 3b«DMSO > 3b»pyridine > 3b»acetone ~ 3b»benzene. Due to competition with complex 3b»CHCl3, we were unable to explore poorer template molecules as guest in CDCI3 as solvent. To quantify the selectivity of complex 3b»guest with these six guest molecules, we determined the relative stabilities (KTe[) y i a competition experiments, by integrating the unique host and guest signals in the lK NMR spectrum of a mixture of tetrol lb, DBU, guest 1 and guest 2 in CDCI3 (Figure 3.14). Guest 1 and guest 2 were added to the NMR sample containing tetrol lb and DBU such that a -1:1 ratio of complex 3b«guest l:complex 3b#guest 2 would be observed in the lH NMR spectrum, and such that the total amount of free species that remained represented less than 15% of the total concentration of host. The results, recorded in Table 3.2, show that complex 3b»guest is highly guest selective as discussed below. Table 3.2. Comparison of Kre\ versus Template Ratios. Template R ^ ln(TR) ln(^rei) Ratio (TR)a pyrazine 860 580 6.8 6.4 dioxane 180 71 5.2 4.3 DMSO 19 14 2.9 2.6 pyridine 14 9.5 2.6 2.2 acetone-^ 2 0.9 0.7 -0.1 benzene-^ 1 1 0.0 0.0 a TR determined at 298 K using tetrol lb. b ^ r ei 's determined at 298 K in C D C I 3 using tetrol lb. x m The guests acetone and benzene were added as acetone-and benzene-c?6 so as to not wash out the N M R signals for the host signals of complex 3b»guest. 168 Figure 3.14. ^ r e i for complex 3b»pyrazine versus complex 3b»dioxane at 298 K. CHCI3 free 1,4-dioxane free pyrazine d, P A p,d p,d,f 9.0 8 . 0 7 . 0 6 . 0 5.0 ppm 4.0 3.0 2.0 1.0 0.0 lH NMR spectrum of pyrazine (1.3 mM), dioxane (10.5 mM), DBU (5.5 mM) and tetrol lb (2.6 mM) in CDCI3 at 298 K. Signals marked p = complex 3b»pyrazine; d = complex 3b»dioxane; and f = free species. The selectivity observed for formation of complex 3b«guest correlated with the selectivity observed for formation of carceplex 2a»guest as shown by a plot of the ln(^rei) versus the ln(template ratios for carceplex 2a»guesf) for the six guest molecules in Table 3.2, which yielded a correlation factor of r 2 = 0.99 (Figure 3.15). This correlation suggests that the nature of the guest molecule imparts a similar stabilizing or destabilizing effect on the relative free energies of complex 3b»guest as it does to the relative activation energies of the transition state of the GDS in the formation of carceplex 2a»guest. This agreement implies that the interactions that govern the formation of both carceplex 2a»guest and complex 3b»guest are similar, and therefore, this complex does represents a good transition state model for the GDS in the formation of carceplex 2a»guest. 169 b. Thermodynamic Parameters for Complex 3b»Guest in CDCI3 We determined the thermodynamic parameters by measuring the temperature dependence of the relative stabilities of complex 3b»guest with the six guest molecules. An important component of calculating the relative stabilities that are reported in Table 3.3, or any stability constant, was ensuring that the sample had reached equilibrium. The amount of time necessary for equilibration of complex 3b»guest reported in Figure 3.16 depended greatly on temperature and on the guest encapsulated in complex 3b«guest. We determined the time necessary to achieve equilibrium by two different methods: (1) simply recording the A H NMR spectrum until no change was observed; or (2) Waiting five times the calculated half-life for the decomplexation rate of our best guest, pyrazine. The exchange rate of pyrazine is assumed to be the slowest of the guests,xlv and decomplexation is x i v The decomplexation rate of pyrazine from complex 3b»pyrazine was shown to be much longer than the decomplexation rates of dioxane or DMSO from complexes 3b»dioxane and 3b»DMSO, respectively (see experimental section). 170 assumed to be the slowest step in achieving equilibrium. These assumptions (1 and 2) were supported by measuring the decomplexation of other complexes 3b»guest and comparing the decomplexation rates with the time needed for the *H NMR spectra to manifest no more changes. The control experiments and the time allotted for equilibration of the lH NMR samples are described in the experimental section at the end of this chapter. A van't Hoff plot of the data (Figure 3.16) provides the relative enthalpies and entropies as well as the relative free energies for these complexes, which are given in Table 3.3; benzene was arbitrarily chosen as a reference point and set to zero for simplicity. As can be seen by the magnitude of the relative free energies of these complexes (AAG°), small changes in guest molecules are manifested by large perturbations to the energetics of the complex. For example, pyrazine and pyridine only differ by the substitution of a CH group with nitrogen, but the relative free energies of their complexes 3b»guest differ by 2.5 kcal/mol at 300 K, which represents a KTe\ of ~ 60. What factors cause this large difference in stabilities between these two complexes? It is not likely to be due to hydrogen bonding of the encapsulated guest molecule with the CHB of the complex as there is no evidence for such interactions in the crystal structure of complex 3b«pyrazine, nor due to the charge of the guest molecule itself as both pyridine and pyrazine bear the same charge. In addition to this, the strong correlation between the template ratios of carceplex 2a»guest determined in NMP (er = 32.2)25 and the KTe\s determined in CDCI3 (er = 4.8)25 suggest that solvation does not contribute significantly to the observed selectivity. (The importance of solvation is discussed in further detail in section xvic) Therefore, the source of the difference in stabilities of complex 3b#pyrazine and complex 3b«pyridine must be a result of a combination of weak noncovalent interactions. Examination of the relative enthalpic and relative entropic components of complex 3b*guest allows for a more detailed description of complex 3b»guest. 171 Figure 3.16. Graph of Rln(^rei Complex 3b»guest) versus 1/T (K). A^i's measured as a function of temperature in CDCI3 from 253 K to 323 K. The correlation factor (r2) is greater than 0.99 for all data shown. Table 3.3. Thermodynamic data for Complex 3b»Guest in CDCl3.a Guest AAH° AAS° TAA5° b AAG° kcal/mol cal/mol(K) kcal/mol kcal/mol pyrazine -0.7 10.5 3.2 -3.9 dioxane -0.7 6.3 1.9 -2.6 DMSO 7.2 29.6 8.9 -1.7 pyridine 0.8 7.2 2.2 -1.4 acetone 4.9 16.3 4.9 0.0 benzene 0.0 0.0 0.0 0.0 a Errors are estimated to be ±20%. b Temperature = 300 K. The thermodynamic data reported in Table 3.3 indicates that the relative free energies of complexes 3b»guest have a large favorable entropic component with respect to complex 3b»benzene. For example, the formation of complex 3b»DMSO and complex 172 3b«acetone are strongly favored entropically relative to benzene. This is most likely a result of the relatively small size of acetone and DMSO, which allows for greater mobility of these guests within the interior of the host, which would be favored entropically. Benzene forms the least entropically favored complex most likely because its large size restricts its mobility within the interior of the complex and perhaps distorts the bowls as well. The enthalpic component of the stabilities of the complexes is undoubtedly composed of a variety of noncovalent interactions including van der Waals, electrostatic, and 7C-7T interactions as well as hydrogen bonding such as O-H—O and CR-K. DMSO and acetone complexes are the least favored enthalpically most likely because of their size and shape which provide poor van der Waals and CH-7C interactions and no 7E-TC interactions with the interior of the complex according to CPK models. Pyrazine and dioxane complexes are the most enthalpically favored guests, which is interesting considering the similarity of their shapes. Examination of CPK models and the crystal structure of complex 3b»pyrazine reveals that (1) both guests form highly favorable van der Waals interactions due to the complementarity of the guest with the interior of the complex; (2) both guests can form CH-7C hydrogen bonds with the arenes of the bowls; and (3) both guests can form CH-X hydrogen bonds between their hetero atoms and the inward-pointing bridging acetal (OCH2O) hydrogens of the bowls. However, only pyrazine can participate in CH-7U hydrogen bonds with these inward-pointing (OCH2O) hydrogens of the bowls, and form 7t-7C interactions with the electron rich arenes of the bowls. That both pyridine and benzene complexes are less favored enthalpically compared to pyrazine could be due to the larger guests being less complementary to the interior of the host as determined by CPK models. Additionally, it could be due to the loss of one CH-X hydrogen bond between the hetero atom of the guest and the inward-pointing (OCH2O) hydrogens of the bowls in the case of pyridine, and all such bonds in the case of benzene. Complex 3b»pyridine may be less favored enthalpically compared to both complexes 173 3b#pyrazine and 3b»benzene because unlike pyrazine and benzene, pyridine has a substantial dipole moment. Complex 3b does not have a dipole moment because of its symmetry, thus it may favor guests which also do not have a dipole (pyrazine and benzene) over guests which do (pyridine). In summary, the determination of the thermodynamic components of the relative free energies of complex 3b«guest has provided a far more detailed assessment of how noncovalent interactions govern its formation. xiii. Selective Guest Binding of Other Related Complexes in CDCI3 a Introduction The high degree of guest selectivity observed in the formation of complex 3b«guest prompted us to extend these investigations to a series of related hosts which varied in the number and type of interactions that could connect the two bowls. We had seen the sensitivity of a complex to perturbations in the guest, but what about varying the host? What features of the host are essential for complexation? For example, how many CHB's are required to form stable complexes between two cavitands or bowls? If complexes form with new hosts, would they exhibit the same selectivity as observed with complex 3b»guest? The requisite cavitands 108-112 were readily available in our group (Figure 3.17). For example, monol 109, A,B-diol 110a, A,C-diol 111, and triol 112 were isolated as by-products during the synthesis of tetrol lb (see experimental). Tetraprotio 108 was synthesized as previously described in the literature.26 A more direct synthesis of A,B-diol 110a was also carried out as described in section xivb. 174 Figure 3.17. Cavitands C H 3 108 tetraprotio; R = CH 2 CH 2 Ph 109 monol; X = Y 110a A,B-diol; X = Y 111 A,C-diol; X = Z 112 triol; X = Y Z = H H, Z = OH H, Y = OH OH, Z = H b. Complexation Properties of Tetraprotio 108 Tetraprotio 108 is particularly interesting because unlike tetrol lb it cannot form CHB's between the bowls. The *H NMR spectrum of a solution of tetraprotio 108 (12 mM) and pyrazine (12 mM) in CDCI3 showed no differences in the chemical shift of any hydrogens of either the host or guest when compared to the respective *H NMR spectra of each run individually. Thus, tetraprotio 108 does not form tetraprotio-complex 113»pyrazine as illustrated in Figure 3.18, which suggests that CHB's are an essential component of complex 3b*guest. This result was not very surprising because tetrol lb also showed no evidence of complex formation in the absence of base (see spectrum A in Figure 3.1) 175 Figure 3.18 Tetraprotio-complex 113«guest. 108 tetraprotio; R = CH 2CH 2Ph tetraprotio-complex 113»guest c. Complexation Properties ofMonol 109 We therefore investigated the number and position of the CHB's that are necessary to lead to complex formation. The *H NMR spectrum of a mixture of monol 109 (9.5 mM), DBU (5.7 mM) and pyrazine (4.8 mM) in CDCI3 showed no significant difference from the spectrum in the absence of pyrazine, nor was there any significant A8 (< 0.01 ppm) for the pyrazine guest signal at both ambient temperature and at 223 K. Therefore, we concluded that monol-complex 114»pyrazine does not form in CDCI3. Interestingly, addition of DBU to a solution of monol 109 in CDCI3 does result in significant chemical shifts of the protons para to the phenolic hydroxyl (Hp, Figure 3.19). Thus at 9.5 mM monol 109, the chemical shift of H^ is 6.76 ppm in the absence of base versus 6.61 ppm in the presence of 0.6 equivalents of DBU. This 0.15 ppm A5 may be an indication of formation of a dimer of monol 109, or more likely, simply the result of deprotonation of the phenolic hydroxyl. To differentiate the two, we explored the binding properties of monol 109 in nitrobenzene-Js. The *H NMR spectrum of a mixture of monol 109 (9.2 mM), DBU (5.5 mM) and pyrazine (4.6 mM) in nitrobenzene-Js was significantly different 176 from the spectrum in the absence of DBU. We observed no signal for free pyrazine at 8.6 ppm nor was there a signal observed for encapsulated pyrazine in this spectrum (expected between 4-5 ppm). Also, there were significant differences in the chemical shifts of both the methylenes that line the upper rims of the bowls and Hp. We thus reasoned that monol-complex 114«pyrazine was indeed forming (Figure 3.19), but the decomplexation rate of pyrazine was in intermediate exchange on the *H NMR timescale. The absence of a signal for pyrazine in this  lH NMR spectrum could be due to its overlap with other signals such as residual solvent or more likely its broadening into the baseline of the spectrum. Therefore, more pyrazine was added to this sample (total 74 mM) which resulted in the appearance of a broad peak centered at 8.44 ppm (i.e., the addition of more pyrazine resulted in a better signal to noise ratio, thus allowing us to see the very broad signal for pyrazine) that is ascribable to pyrazine that is in intermediate exchange on the  lH NMR timescale between free and monol-complex 114»pyrazine. Rerunning this sample at 283 K (nitrobenzene-^ 5, m.p. 278-279 K) resulted in no significant differences changes. A magnetization transfer experiment confirmed the encapsulation of pyrazine in monol-complex 114«pyrazine, and that the guest had a similar chemical shift to complex 3b»pyrazine in nitrobenzene-J5. Recording the *H NMR spectrum of the above sample at ambient temperature while decoupling at 4.4 ppm (the suspected position of encapsulated pyrazine in monol-complex 114»pyrazine) resulted in a 3-fold reduction in intensity of the peak for pyrazine at 8.44 ppm. This confirms that the chemical shift for encapsulated pyrazine in monol-complex 114»pyrazine is approximately at 4.4 ppm and the exchange rate of pyrazine is faster than its T\. Based on this large shielding of the pyrazine guest, we suspect that the structure of monol-complex 114«pyrazine is similar to complex 3b»pyrazine as depicted in Figure 3.19. Our failure to detect monol-complex 114«pyrazine in CDCI3 is most likely due to its weak complexation properties and may also be due to stronger relative binding of CDCI3 with respect to pyrazine in monol-complex 114»guest versus complex 3b*guest. The latter explanation is more consistent with the observation of 177 small A8 for Hp of tetrol lb in CDCI3/DBU at low concentration and high temperature (see section iv). In this case monomeric tetrol lb was apparently not deprotonated to a significant extent by DBU. The base-dependence of 5 for of monol 109 indicates that some type of complex is forming. Figure 3.19. Complexes of Cavitands. 109 X = Y=Z = H 114»guest; X = Y =Z = H, H 110a X = Y = H; Z = OH 115-guest; X = Y = H, H; Z = O-H—O" 111 X = Z = H;Y = OH 116«guest; X = Z = H, H; Y = O-H—O" 112 X = Y = OH;Z = H 117a»guest; X = Y = O-H—O"; Z = H, H d. Complexation Properties of A,B-diol 110a and A,C-diollll A,B-diol HOa and A,C-diol 111 also formed complexes in the presence of pyrazine and DBU in CDCI3 as shown in Figure 3.19. The *H NMR spectrum of the A,B-diol-complex 115»guest and the A,C-diol-complex 116»guest at ambient temperature was broadened due to an intermediate exchange rate of host and guest. Therefore, these complexes were studied at 253 K where both host and guest signals were in slow exchange on the *H NMR timescale. The dimeric nature of each of these complexes was confirmed by comparison of the integrals of the bowl signals to encapsulated pyrazine, which were in a 2:1 ratio of bowls:pyrazine. Also, large upfield chemical shifts were observed for the 178 signal of the encapsulated pyrazine, thus providing clear evidence for its encapsulation in the aryl lined cavity. Moreover, the signals of encapsulated pyrazine at 4.06 ppm and 4.08 ppm at 253 K in CDCI3 for the A,B-diol-complex 115»pyrazine and A,C-diol-complex 116'pyrazine, respectively, correlate with bound pyrazine in complex 3b»pyrazine at 253 K (cf. 4.23 ppm). Therefore, A,B-diol-complex llS'pyrazine and A,C-diol-complex 116»pyrazine resemble complex 3b»guest except that there are only two CHB's interconnecting the two bowls. The ^rrei's for complexation of pyrazine, dioxane and pyridine by A,B-diol-complex 115»guest and A,C-diol-complex 116»guest at 253 K are reported in Table 3.4 along with the A"rei's determined previously for complex 3b»guest. Unfortunately, the intermediate exchange rate of the corresponding diol complexes of acetone, benzene and DMSO prevented their detailed investigation. As can be seen in Table 3.4, the same trends for selectivity are observed with both the A,B-diol-complex 115»guest and A,C-diol-complex 116»guest; i.e., pyrazine forms more stable complexes than either dioxane or pyridine. There are, however, differences observed in the magnitude of the guest selectivities in the different complexes which fall outside of our estimated 20% error. For example, the KTQ\ for A,B-diol-complex 115»dioxane to 115«pyridine differs by a factor of four from the KTe\ for complex 3b»dioxane to 3b«pyridine. Unlike complex 3b«guest and A,C-diol-complex 116«guest, A,B-diol-complex 115»guest has a dipole moment. Therefore it may favor guest molecules that also contain a dipole moment. Unlike pyrazine and dioxane, pyridine has a substantial dipole moment. The presence of such a dipole in pyridine may make it more favorable for encapsulation in A,B-diol-complex 115«pyridine relative to both complex 3b»pyridine and A,C-diol-complex 116»pyridine (see section xviib). Pyridine is disfavored more relative to dioxane and pyrazine in A,C-diol-complex 116«guest than in either complexes A,B-diol-complex 115«guest or complex 3b»guest. This is very difficult to understand. Regarding Ti-basicity of the hosts, one would expect a similar effect on the affinity toward pyrazine and pyridine as both are electron poor, but this 179 is not observed. Neither A,C-diol-complex 116»guest nor complex 3b»guest have a dipole moment, so neither appears to be more complementary to pyridine. However, the positive end of pyridine's dipole can align between two electron-rich phenolic arenes in both A,B-diol-complex 115»pyridine and complex 3b»pyridine, whereas A,C-diol-complex 116«pyridine can only offer one electron poor-arene (Figure 3.20). Perhaps computational analysis can shed light on this perplexing data (section xix). Table 3.4. Orel's of A,B-diol-complex 115«guest and A,C-diol-complex 116«guest at 253 K. Complex A,C-diol- A,B-diol-3b»guest complex complex 116«guest 115«guest pyrazine 98 460 56 dioxane 12 75 3 pyridine 1 1 1 Figure 3.20. Schematic Representation of the Dipole Interactions of A,B-diol-complex 115#pyridine, Complex Sb^pyridine, and A,C-diol-complex 116»pyridine A,B-diol-complex 115«pyridine complex 3b»pyridine A,C-diol-complex 116«pyridine In summary, A,B-diol-complex 115»guest and A,C-diol-complex 116«guest exhibit the same general trend for guest selectivity for all three guest molecules studied compared to complex 3b»guest. The differences in the magnitudes of the selectivities in these complexes is difficult to explain, but may be a result of dipolar interactions. The 180 faster exchange rates observed for the diol complexes relative to complex 3b»guest suggests that the relative free energies of the diol complexes may be lower than the free energy of complex 3b»guest. This is further addressed in section xvi of this chapter. e. Complexation Properties ofTriol 112 and Monobridged 102 Both triol 112 and monobridged 102 lead to complexes, triol-complex 117a#guest (Figure 3.19) and monobridged complex 118»guest (Figure 3.21), respectively. Again, these spectra were broad at ambient temperature due to intermediate exchange rate of the guest. Therefore, we investigated the selectivity of these complexes at 273 K where both the host and the guest were in slow exchange on the lH NMR timescale. Again, integration of the host and guest peaks in the *H NMR spectra of triol-complex 117a«guest and monobridged complex 118»guest confirms that two bowl molecules encapsulate the guest in both cases. Also, large upfield chemical shifts of encapsulated guest molecules confirms these two complexes are similar in nature to complex 3b»guest. Figure 3.21. Formation of Mono-bridged complex 118»guest. monobridged 102 monobridged complex 118»guest 181 We determined the orientation and mobility of pyrazine in triol-complex 117»guest to further compare its structure to complex 3»guest (Figure 3.22). Thus, application of Fraser's strategy7 (see section x) to asymmetric triol-complex 117b»pyrazine determined that pyrazine has the same orientation as observed in complex 3»pyrazine. Furthermore, a 17 kcal/mol energy barrier was determined (Tc = 348 K) for rotation of pyrazine about the pseudo-C2 axes of asymmetric triol-complex 117b»pyrazine,27 which correlates well with the 18 kcal/mol energy barrier for the rotation of pyrazine in asymmetric complex 3c»pyrazine (Figure 3.22). Therefore, pyrazine has the same orientation and mobility within triol-complex 117»pyrazine and complex 3»pyrazine. The formation of triol-complex 117a«guest can in principal form in three ways: (1) all three complementary phenolic hydroxyls/phenoxides align; (2) two phenolic hydroxyls/phenoxides align while the other two phenolic hydroxyls lie 180° apart on the upper and lower bowls, respectively; (3) two phenolic hydroxyls/phenoxides align while the other two phenolic hydroxyls lie 90° apart on the upper and lower bowls, respectively. The observation of a single set of host and guest peaks for triol-complex 117a»pyrazine suggests that only one of the three possible complexes forms. Furthermore, the symmetry of the triol-complex 117a»pyrazine as deduced from its *H NMR spectrum eliminates the third orientation. The remaining two possible orientations can not be differentiated based on their symmetry alone (i.e. both belong to the Civ point group), but they can be differentiated based on the number of CHB's that interconnect the two bowls. Thus, we further characterized the triol-complex 117a»guest by variable temperature *H NMR spectroscopy to examine the nature of the CHB's of this complex. The *H NMR spectrum of triol 112a (19 mM), and DBU (29 mM), at 223 K in CDC13 exhibited a broad peak at 13.1 ppm that integrated for 6 protons per total complex (see section Bvi). This was assigned as the fast exchanging average between the CHB of triol 112a and the protonated DBU (DBUH+) counter ion. This result strongly suggests that in triol-complex 117a»guest all three complementary phenolic hydroxyls/phenoxides are aligned forming 182 three CHB's. Furthermore, this orientation is also consistent with the decomplexation rate of pyrazine from this complex. Upon examining the orientation of pyrazine in asymmetric triol-complex 117b»pyrazine in nitrobenzene-Js, we found that pyrazine was in slow exchange on the *H NMR timescale up to temperature of 373 K. This is more consistent with the triol-complex 117»pyrazine containing three CHB's as complexes held together with only two CHB's would likely manifest fast guest exchange at 373 K. In summary, of the three possible ways triol-complex 117«guest can form, the one that aligns all three complementary phenolic hydroxyls/phenoxides is the most consistent with experimental evidence. Figure 3.22. Asymmetric triol-complex 117b«pyrazine. axis Schematic representation of asymmetric triol-complex 117b*pyrazine. Pyrazine is orientated with its nitrogens at the "equator" of the complex because meta coupling is observed for the nonequivalent hydrogens H a (d, J = 2 Hz) and Hb (d, J = 2 Hz) . The vertical lines connecting the two bowls represent CHB's . The coupling of H a and H b is estimated from the line width at half height (i.e., J « 7 Hz) for their corresponding signals in the ! H N M R spectrum. Equilibration to determine the Orel's for triol-complex 117a»guest only required a 20 minute period at 273 K while monobridged complex 118»guest had to be equilibrated overnight (based on the half-life for pyrazine decomplexation of 2 hours). The calculated 183 Orel's for triol-complex 117a»guest and monobridged complex 118»guest as well as those for complex 3b»guest at 273 K are tabulated in Table 3.5. These values indicate that there is a good correlation between the guest selectivity of these complexes and complex 3b»guest, as is further illustrated by the plot of ln(KTe\ of triol-complex 117a»guest) and lnt^ rei of monobridged complex 118»guest) versus ln(KK\ of complex 3b»guest) shown in Figure 3.23. It is noteworthy that whilst the overall range of selectivity for the Orel's reported in Table 3.5 are similar, the position of DMSO as guest varies considerably within this series of complexes: 3b»DMSO > 117a»DMSO > 118«DMSO, all relative to their corresponding acetone complex. In section xiib, it was shown that complex 3b»DMSO was strongly entropically favored relative to complex 3b»benzene. In other words, the stability of complex 3b»DMSO varied greatly as a function of temperature relative to complex 3b»benzene. The variance in relative stability of DMSO as guest in both its triol-complex 117a»DMSO and monobridged complex 118«DMSO compared to complex 3b»guest at constant temperature indicates that its selectivity is also host dependent. Interestingly, the A H NMR of triol-complex 117a»DMSO and monobridged complex 118»DMSO have their methylene signals broadened. Perhaps, the mobility of DMSO within these complexes is restricted and a single non-degenerate orientation of DMSO is favored. This would lower the entropy of complexation for DMSO (and therefore the free energy of complexation) in triol-complex 117a»guest and monobridged complex 118»guest (see section xiib; see also section xvi, for a discussion of methyl acetate binding in triol-complex 117a»guest). 184 Table 3.5 .£ r e i 's of Triol -complex 117a»guest and Monobridged 118«guest at 273 K. Complex 3b»guest Triol-complex 117a»guest Monobridged Complex 118«guest pyrazine 2100 2100 1300 dioxane 260 270 190 pyridine 13 28 19 DMSO 23 13 3.8 benzene 2.2 1.7 1.8 acetone 1.0 1.0 1.0 Figure 3.23. ln(KTe\ of triol-complex 117a»guest and monobridged complex 118»guest) versus ln(^Trei of complex 3b»guest). ln(Krel Complex 3b#guest) The formation of triol-complex 117a»guest has great significance toward the reaction to form hemicarceplex 107 studied by Chopra and Sherman.3 The higher than statistical yields reported (as great as 58% yield for hemicarceplex 107»pyrazine compared to a statistical yield of 28%) for the formation of hemicarceplex 107»guest clearly arises 185 from the preorganization of the two triol bowls (112) into triol-complex 117a«guest, thus aligning the complementary phenolic hydroxyls and phenoxide groups via CHB's, which promotes correct covalent bond formation to lead to hemicarceplex 107»guest. xiv. Preorganized Hosts that Exhibit Binding Under Neutral Conditions a Complexation Properties of A,B-bis-bridged 103. The complexation properties of A,B-bis-bridged 103 are unique since it has been shown to encapsulate guest molecules both in the absence of base (A,B-bis-bridged complex 103»guest) and in the presence of base (A,B-bis-bridged charged complex 119»guest) as shown in Figure 3.24. Formation of A,B-bis-bridged complex 103»guest was the first example of a tetrol-derived host molecule encapsulating a guest molecule under neutral conditions. Examination of CPK models reveals that the cavity is preorganized for binding. It is shaped like a clam shell with the two inter-bowl methylene bridges functioning like hinges which allow guest molecules easy access to the cavity. The portal of A,B-bis-bridged 103 is larger than hemicarceplex 107 (see Chapter 2, section Biii), and consequently the energies for complexation and decomplexation are lower. For example, dissolving A,B-bis-bridged 103 (2.6 mM) and pyrazine (170 mM) in C H C I 3 at ambient temperature followed by precipitation after 15 minutes with excess hexane gave (after drying at ambient temperature) A,B-bis-bridged complex 103»pyrazine in a 1:1 ratio of host:guest as determined by integration of its *H NMR spectrum. In contrast, only small guest molecules such as C H 2 C I 2 and C H 3 C N were encapsulated in hemicarceplex 107»guest at ambient temperature.28 The encapsulation of aromatic guests such as pyridine in hemicarceplex 107»guest required heat (refluxing a solution of hemicarceplex 107 in pyridine for 6 hours).28 We also examined the decomplexation of pyrazine from A,B-bis-bridged complex 103»pyrazine by following its *H NMR spectrum 186 in CDCI3 as a function of time. Thus, the half-life at ambient temperature was estimated from the initial rate for this process and was found to be 4 hours. Interestingly, equilibrium is reached after only 1/3 of the pyrazine has decomplexed, suggesting that the stability of A,B-bis-bridged complex 103#pyrazine is slightly greater than empty A,B-bis-bridged 103 or A,B-bis-bridged complex 103»CHCl3. Figure 3.24. Complexation in the Presence and Absence of Base. A,B-bis-bridged charged A,B-bis-bridged 103 A,B-bis-bridged complex 119»guest complex 103»guest We also investigated the potential encapsulation of chloroform in A,B-bis-bridged 103. Dissolving A,B-bis-bridged 103 (2.5 mM) in CHCI3 at ambient temperature followed by precipitation after 15 minutes with excess hexane resulted in a solid (after drying at ambient temperature) whose lK NMR spectrum showed no evidence for encapsulation of CHCI3 as A,B-bis-bridged complex 103*CHCl3. This suggested that either egress of CHCI3 is fast or that 103»CHCl3 is not formed at all. We confirmed that indeed the charged A,B-bis-bridged charged complex 119»CHCl3 forms by repeating this experiment in the presence of base. Thus, dissolving A,B-bis-bridged 103 (1.5 mM) and DBU (3.3 mM) in CHCI3 at ambient temperature followed by precipitation after 15 minutes with excess hexane gave (after drying at ambient temperature) [A,B-bis-bridged charged complex 119»CHCl3][DBU]i.2 as determined by integration of the *H NMR spectrum at 273 K in CDCI3. The appearance of a peak at 4.53 ppm was assigned to encapsulated CHCI3 in A,B-bis-bridged charged complex 119»CHCl3 based on its upfield A5 and its 187 integration. This signal for encapsulated CHCI3, however, disappears over a period of minutes at this temperature as the bulk CDCI3 slowly displaces it. We speculated that the rate of exchange of CHCI3 from charged A,B-bis-bridged charged complex 119»CHCl3 would be much slower than neutral A,B-bis-bridged complex 103«CHCl3. Also, the use of excess base for these exchange experiments should slow the rate of CHCI3 exchange from AB-bis-bridged charged complex 119»CHCl3. This was indeed found to be the case. Addition of solid [A,B-bis-bridged charged complex 119«CHCl3][DBU]i .2 (1.2 mM) to a solution already containing DBU (13 mM) resulted in the appearance of a signal corresponding to encapsulated CHCI3 at 298 K. The signal for encapsulated CHCI3 exchanged at a much slower rate even at this higher temperature. In fact, the signal for encapsulated CHCI3 in A,B-bis-bridged charged complex 119*CHCl3 was still present after 4 days! These experiments prove that A,B-bis-bridged charged complex 119#CHCl3 does form and its rate of decomplexation is slower when there are at least two equivalents of DBU than when there is only 1.2 equivalents of base. Our failure to observe A,B-bis-bridged complex 103«CHCl3 was probably due to kinetic instability, most likely due to its lack of stabilizing CHB's. b. Synthesis and Binding Properties of A,B-bis-bridged Tetraprotio 123a We thus decided to synthesize A,B-bis-bridged tetraprotio 123a to determine if other A,B-bis-bridged compounds could also encapsulate guest molecules. In particular, what if the hydroxyl groups are absent? Since isolation of A,B-diol 110a as a by-product in the synthesis of tetrol lb is extremely inefficient (see experimental), we began synthesizing A,B-diol 110a via a more direct route. We followed a procedure that was similar to one developed by Reinhoudt and coworkers.29-30 for the synthesis of A,B-diol 110b (Scheme 3.4). A,B-dibromo-tris-bridged bowl 121a was synthesized in 67% yield 188 by treatment of tetra-bromo-tris-bridged bowl 120a31 with n-butyllithium followed by quenching with H2O. The resultant A,B-dibromo-tris-bridged bowl 121a was then bridged with bromochloromethane to give A,B-dibromo-bowl 122a in 95% yield. Treatment of this bowl (122a) with n-butyllithium followed by addition of trimethyl borate and finally oxidation and hydrolysis with H 20 2/NaOH of the intermediate borate esters gave A,B-diol 110a in 40% yield. Finally, tetraprotio 123a was obtained in 61% yield from the reaction of A,B-diol 110a, bromochloromethane, potassium carbonate, and methyl acetate (as guest) in NMP under standard conditions. Previously, Reinhoudt and coworkers,29-30 and Cram and Robbins32 reported the synthesis of tetraprotio 123b and tetraprotio 123c, which were also synthesized from their corresponding A,B-diol 110b and A,B-diol 110c, respectively, but neither reported any binding experiments for these compounds. Scheme 3.4. Synthesis and Complexation of Tetraprotio 123a. 123a«guest 123a, R = CH 3 110a, R = CH 3 R= CH 3 123b,R= (CH2)i0CH3 110b, R = (CH 2)i 0CH 3 123c, R = CH2CH2Ph 110c, R = CH2CH2Ph 189 The H NMR spectrum of tetraprotio 123a was extremely broad in C D C I 3 at ambient temperature as shown in spectrum A in Figure 3.25. This broadening was not simply due to aggregation of tetraprotio 123a because its lH NMR spectrum was concentration independent. Also, saturating this sample with N 2 gas to displace any residual 0 2 gas had no effect on the lH NMR spectrum; therefore, the broadening is not due to encapsulation of 0 2 in the cavity. We assumed that the broadening of the J H NMR spectrum was due to the dynamics of the host at this temperature and therefore cooled the sample to 223 K which resulted in a well resolved l H NMR spectrum as shown in spectrum B in Figure 3.25. Furthermore, addition of pyrazine to this sample led to the formation of a new set of host and guest peaks which were in slow exchange on the lH NMR timescale at ambient temperature as shown in spectrum C in Figure 3.25. The formation of tetraprotio 123a»pyrazine confirms that the structure of the host is indeed the "C" isomer shown in Scheme 3.4 and not the "Z" isomer. 190 Figure 3.25. H NMR Characterization of tetraprotio 123a. A i i i i i i i i i 9.0 8.0 7.0 6.0 5.0 PP™ 4.0 3.0 2.0 1.0 free i i i i i i i i i 9.0 8.0 7.0 6.0 5.0 PPm 4.0 3.0 2.0 1.0 J H NMR in CDC13 of: A; tetraprotio 123a (1.6 mM) at 296 K. B; tetraprotio 123a (1.6 mM) at 223 K. C; tetraprotio 123a (6.5 mM) and pyrazine (32 mM) at 296 K. Specific proton assignments are given in the experimental section. 191 c. Krei 'sfor AB-bis-bridged 103»Guest and Tetraprotio 123a*Guest The guest K^i's for both A,B-bis-bridged complex 103»guest and tetraprotio 123a»guest at 298 K in CDCI3 were determined (Table 3.6). Again, there was reasonably good correlation (r > 0.92) for the guest ATrei's of A,B-bis-bridged complex 103»guest and tetraprotio 123a»guest and complex 3b»guest as can be seen by the plots of ln(Kre\ tetraprotio 123a»guest) and \n(KTe\ A,B-bis-bridged complex 103»guesf) versus ln(KTe\ complex 3b»guest) in Figure 3.26. These similarities in complexation suggests that the noncovalent forces which govern the formation of these three complexes are similar. These studies show that preorganization of the host by a hinge eliminates the need for CHB's, or any hydroxyls at all, for complexation to occur. The relative stability could not be determined for A,B-bis-bridged complex 103»DMSO because of extreme broadening in the lH NMR spectra. Unfortunately, the extremely slow exchange rate of guests in A,B-bis-bridged charged complex 119»guest (see page 187) prevented our investigation of its guest selectivity. Table 3.6. Orel's of tetraprotio 123a«guest and A,B-bis-bridged complex 103»guest at 298 K. complex tetraprotio A,B-bis-bridged 3b«guest 123a»guest complex 103»guest pyrazine 680 1500 1800 dioxane 80 88 160 pyridine- 11 16 45 DMSO- d6 15 2.6 benzene- d(, 1.1 1.1 2.6 acetone-dg 1.0 1.0 1.0 192 Figure 3.26. Graphs of ln(KTe\ tetraprotio 123a»guest) and ln(i^rei A,B-bis-bridged complex 103»guest) versus ln(^Trei Complex 3b»guest). d- Kap for AB-bis-bridged 103'Guest We also approximated the stabilities for AB-bis-bridged 103»guest with five guest molecules by treating the free species in the *H NMR spectra as empty AB-bis-bridged 103. This allowed for calculation of apparent stability constants (Kap) which are reported in Table 3.7 along with the Orel's determined previously. We call this an apparent stability constant because the free species of AB-bis-bridged 103 in CDCI3 most likely contains AB-bis-bridged 103»CDCl3. We found a good correlation (r2 = 0.99) between the Orel's and ^ap determined for AB-bis-bridged 103»guest. Furthermore, the A'ap for AB-bis-bridged 103«pyrazine was concentration independent over a concentration range of 0.15-1.5 mM (X a p = 5100 + 150 M _ 1). In order to determine the absolute stability constant for AB-bis-bridged 103»guest the nature of the free species would have to be further characterized and any binding of solvent to form AB-bis-bridged 103»CDCl3 would have 193 to be quantified. The absolute stability constant for a related AB-bis-bridged host is discussed in section xviib and compared with the value reported here. Table 3.7. Kap of AB-bis-bridged 103«guest at 298 K. _ _ _ _ ^ (M-1) bridged 103»guest £ap(acetone) pyrazine 1800 2000 5100 dioxane 160 120 290 pyridine-^5 45 56 140 benzene- de 2.6 2.2 5.6 acetone-^ 1.0 1.0 2.5 e. Other Hinged Compounds. The neutral binding properties of the A,B -bis-bridged compounds such as tetraprotio 123a»guest or A,B -bis-bridged complex 103»guest prompted us to re-investigate the binding properties of monobridged compounds such as monobridged 102. We also synthesized another monobridged compound by bridging monol 109 with diiodomethane in NMP to give hexaprotio 124 in 81% yield. The binding properties of these two compounds were examined in CDCI3 under neutral conditions. The addition of our best guest (pyrazine) to either of these compounds in CDCI3 failed to demonstrate any significant changes in the signals in the *H NMR spectra of either the host or guest. Therefore, these compounds do not form monobridged complex 102«guest or hexaprotio complex 124*guest (respectively) due to their inability to preorganize as illustrated in Figure 3.27. Examination of the CPK models of both monobridged 102 and hexaprotio 124 shows that the bowls have a large degree of mobility about the intermolecular methylene bridge that connects the two bowls. The favorable interactions of the guest with the interior of these hosts is not sufficient to compensate for the loss of mobility about this bridge (i.e., it is entropically disfavored), nor are neutral hydrogen bonds. In the case of monobridged 102, however, the use of DBU leads to the formation of CHB's between the 194 bowls to yield monobridged complex 118»guest; the CHB's allow the complex to overcome the entropic cost of its formation. Figure 3.27. hexaprotio 124, X = H hexaprotio complex 124»guest, X = H monobridged 102, X = OH monobridged complex 102»guest, X = OH The selectivity observed in the complexation studies performed in this chapter all showed the same basic trend. For example, pyrazine is the most favored guest relative to the others investigated for each complex studied. The similarities in the relative stabilities of these complexes with various guests are not too surprising considering the structural similarities of the cavities created in each complex. Nevertheless, there are striking differences in the hosts. The question then arises: do the relative stabilities of the hosts vary with the same guest molecule? As explained in section xvi, this investigation required the use of a solvent other than CDCI3. Therefore, we began to investigate the formation of the tetrol-related complexes in nitrobenzene-^ . 195 xv. Nitrobenzene-d5 as a Solvent for Complexation Nitrobenzene-^ is an excellent solvent for studying the complexation properties of cavitands. Like toluene-d% and l,l,2,2-tetrachloroethane-<i2, nitrobenzene-^ cannot be encapsulated in complex 3b»guest or other related complexes because of its large size (as determined by examination of CPK models), but unlike toluene-ds and 1,1,2,2-tetrachloroethane-<i2> nitrobenzene-Js is stable to DBU and can solubilize all the components of complex 3b»guest.xv Therefore, nitrobenzene-ds does not compete with suitable guest molecules for encapsulation in complex 3b»guest or other related complexes. This noncompetitive property of nitrobenzene-ds differentiates it from CDCI3 which was previously shown to form complex 3b»CHCl3 and A,B-bis-bridged charged complex 119»CHC13. We wanted to address the following questions by using nitrobenzene-^: (1) Is the formation of complex 3b»guest selective in this solvent and if so does the observed selectivity correlate with that found in CDCI3? (2) Can a larger range of guest molecules be encapsulated in complex 3b»guest in this noncompetitive solvent? (3) Are the thermodynamic parameters for complex 3b»guest the same as those determined in CDCI3; i.e., how important is solvation? (4) What are the relative stabilities of complexes where the guest is constant, but the host is varied? Are there large changes in entropy or enthalpy as one varies the number or type of interactions that connect the two bowls of the complex together? (5) Finally, can an absolute stability constant be determined? We began to address these questions by repeating similar complexation experiments to those performed on complex 3b»guest in CDCI3 to ensure that this complex has the same properties in the two solvents. Previously in this chapter, we reported that CHCI3 can be encapsulated in complex 3b»guest in nitrobenzene-Js (Figure 3.4, page 146). We therefore investigated the l,l,2,2-Tetrachloroethane-d2 reacts with DBU and complex 3b»guest is not soluble in toluene-dg. 196 encapsulation of other guest molecules in this solvent. The following guest molecules were found to form complex 3b»guest in nitrobenzene-^ : pyrazine, methyl acetate, dioxane, DMSO, pyridine, benzene, and acetone. The noncompetitive nature of nitrobenzene-ds required only slightly greater than stoichiometric amounts of these guests to lead to -100% complex 3b»guest. In contrast, the formation of complex 3b»acetone in the competitive solvent CDCI3 required - 200 equivalents of acetone per tetrol lb to furnish a ~ 75% yield of complex 3b»acetone. The relative stability constants of complex 3b»guest at 298 K for these seven guest molecules in nitrobenzene-ds are reported in Table 3.8 along with the relative stability constants for six of these guest molecules previously determined in CDCI3 at the same temperature. The relative stabilities of complex 3b»guest in these two different solvents correlate fairly well (r2 = 0.94) as demonstrated by a plot of ln(KTe\ in CDCI3) versus ln(KTe\ in nitrobenzene-i/5). (Figure 3.28). This correlation suggests that the noncovalent interations that govern the relative stabilities of complex 3b»guest is not strongly solvent dependent. There are some differences observed for the Kre\s reported in Table 3.8, but these differences are not significant enough for detailed interpretation. It should be noted, the time required to reach equilibrium at 298 K for the Kre\ in CDCI3 and nitrobenzene-^ vary considerably. In CDCI3 equilibration is achieved on the order of minutes while in nitrobenzene-ds this process takes several hours (half-life =1.5 hours for the out rate of pyrazine at 296 K). 197 Table 3.8. Orel's of Complex 3b»guest in Nitrobenzene-d5 and CDCI3 at 298 K. KrQ\ of complex KTe\ of complex ln(ATrei) of complex ln(ATrei) of Guest3 3b»guest in 3b»guest in 3b»guest in complex 3b»guest nitrobenzene-ds CDCI3 nitrobenzene-^  in CDCI3 pyrazine 250 580 5.5 6.4 methyl acetate 140 — — — dioxane 94 71 4.5 4.3 DMSO 35 14 3.6 2.6 pyridine 7.3 9.5 2.0 2.2 acetone-^ 1.7 0.9 0.5 -0.1 benzene-^ 1 1 0.0 0.0 a The Orel's determined in nitrobenzene-^ for complex 3b»acetone and complex 3b»benzene were done using non-deuterated acetone and benzene as guests. Figure 3.28. Graph of ln(^Trei of Complex 3b«guest in CDCI3) versus ln(^Trei of Complex 3b*guest in Nitrobenzene-ds) at 298 K. -1.0 ln(Kr ei in Nitrobenzene) The relative stability constants for complex 3b»guest in nitrobenzene-Js were measured as a function of temperature as shown Figure 3.29. The resulting thermodynamic data for complex 3b«guest in nitrobenzene-ds are reported in Table 3.9 198 along with the thermodynamic data previously determined in CDCI3. Again, for simplicity benzene was assigned a value of zero. Table 3.9. Thermodynamic data for Complex 3b#Guest in Nitrobenzene-<i5 and CDCl3.a Guest AAH° TAA5° b AAG° AAH° TAAS°b AAG° kcal/mol kcal/mol kcal/mol kcal/mol kcal/mol kcal/mol C 6 D 5 N 0 2 C 6 D 5 N 0 2 C 6 D 5 N 0 2 CDCI3 CDCI3 CDCI3 pyrazine -2.3 0.9 -3.3 -0.7 3.2 -3.9 methyl acetate -2.2 0.7 -2.9 — — — dioxane -3.5 -0.8 -2.7 -0.7 1.9 -2.6 DMSO 0.2 2.3 -2.1 7.2 8.9 -1.7 pyridine -0.9 0.3 -1.2 0.8 2.2 -1.4 acetone 0.5 0.8 -0.3 4.9 4.9 0.0 benzene 0.0 0.0 0.0 0.0 0.0 0.0 a Errors are estimated to be ±20%. b Temperature = 300 K. Figure 3.29. Graph of ln(A'j-ei) Versus 1/T (K) in Nitrobenzene-J5. 4 . 0 -r 0 . 0 -I 1 1 1 1 1 1 0 . 0 0 2 9 0 . 0 0 3 0 0 . 0 0 3 1 0 . 0 0 3 2 0 . 0 0 3 3 0 . 0 0 3 4 0 . 0 0 3 5 1/T ( K ) °Rln(pyrazine/MeOAc); r 2 = 0.96 • Rln(DMSO/pyridine); r2 = 0.94 • Rln(MeOAc/dioxane); r 2 = 0.96 • Rln(pyridine/acetone); r2 = 0.95 xRln(dioxane/DMSO); r 2 = 0.99 A Rln(acetone/benzene); r 2 = 0.95 199 The thermodynamic data in Table 3.9 allow a comparison to be made of the relative enthalpic and entropic components for complexation of six different guest molecules in complex 3b»guest in nitrobenzene-d$ and CDCI3. The values for relative free energies (AAG°) for complex 3b»guest are very similar in both nitrobenzene-ds and CDCI3 which is expected considering the Abel's for complex 3b»guest were measured at close to standard temperature (Figure 3.28). There are differences observed for the values of TAAS° and AAH° for complex 3b»guest in these two solvents, but the same trends still remain. For example, guest molecules that are highly complementary to the interior of the complex 3b»guest (as determined by CPK models) such as dioxane and pyrazine are enthalpically favored relative to benzene for complexation in complex 3b»guest in both nitrobenzene-d$ and CDC13. Also, guest molecules that are not as complementary to the interior of complex 3b»guest due to their small size (as determined by CPK models) are enthalpically disfavored and entropically favored relative to benzene for complexation in complex 3b»guest in both nitrobenzene-ds and CDCI3. Also, large guest molecules that may have restricted motion in the interior of complex 3b*guest (as determined by CPK models) such as benzene and dioxane are entropically unfavorable relative to the other guest molecules studied in both solvents. For further discussion on the factors that make up the entropic and enthalpic components of complex 3b»guest see section xiib. Thus, although the values for TAAS° and AAH° for complex 3b«guest in these two solvents vary, the same relative trends are observed as discussed above with only a few exceptions. Complex 3b»pyridine is favored enthalpically relative to complex 3b*benzene in nitrobenzene-Js while complex 3b»benzene is favored enthalpically relative to complex 3b»pyridine in CDCI3. This is most likely due to the different solvation of pyridine and benzene in CDCI3 and nitrobenzene- .^ These two solvents differ greatly in there polarities (nitrobenzene, er = 34.8; CDCI3, 8r = 4.8)25 and thus it is not unexpected that the solvation energies of various guests would change. Further analysis of the solvation energies of these guest molecules in CDC13 and nitrobenzene-^ may shed light on these data. 200 In addition to the six guest molecules discussed above (pyrazine, dioxane, DMSO, pyridine, acetone and benzene), the relative thermodynamic data for complex 3b»methyl acetate was determined in nitrobenzene-^ . Examination of a CPK model of complex 3b»methyl acetate suggests that the guest is highly complementary to the interior of the host and thus should form favorable van der Waals interactions. The methyl groups of the guest in complex 3b»methyl acetate extend deep into the aromatic cavity and the carbonyl oxygen is positioned such that it can interact with the methylenes that line the equator of the complex. Thus, it is not surprising that complex 3b«methyl acetate is enthalpically favored relative to complex 3b»benzene. Furthermore, methyl acetate is a particularly useful guest molecule because it forms highly stable complexes with the hosts discussed in this chapter, and its *H NMR spectra lend themselves toward studying the relative stabilities of complexes using different hosts. xvi. Relative Stabilities of Complexes'Methyl Acetate a Introduction The remarkable similarities between the guest selectivity for each of the complexes prompted us to investigate the relative stabilities of the complexes. For example, would the number of CHB's in a complex lead to more stable complexes: e.g., complex 3b«guest > triol-complex 117a»guest > A,B-diol-complex 115«guest? Would the substitution of a covalent bond (OCH2O) for a CHB lead to more stable complexes: e.g., monobridged complex HS^guest > complex 3b*guest? 201 b. Determination of the Relative Stability Constants of Complexes'Methyl Acetate Using methyl acetate as guest, we addressed the relative stabilities of various complexes. Methyl acetate was chosen as the guest molecule for these complexation studies for the following reasons: (1) Relative to complex 3b«benzene it formed the second strongest complex (second only to pyrazine). (2) The signals for encapsulated methyl acetate in complex 3b»methyl acetate appeared in the high field region (open window) in the lH NMR spectra. (3) The lH NMR signals for encapsulated methyl acetate vary considerably among the series of complexes studied as shown Table 3.35; this facilitated integration of the encapsulated guest signals since overlap was minimal. Two complexes were competed for a limited amount of methyl acetate (less than a stoichiometric amount of methyl acetate for the yield-limiting complex). Integration of the signals for encapsulated methyl acetate in the lH NMR spectra of each of the respective complexes yielded, after correcting for the starting ratio of the hosts, the relative ratio of their stability constants for the complexation of methyl acetate (Figure 3.30). For example, a mixture of tetrol lb, triol 112, DBU and methyl acetate gave signals in the upfield region of the lH NMR that correspond to complex 3b»methyl acetate and triol-complex llTa^methyl acetate as well as signals that were assigned to the asymmetric complex of (tetrol lb)(triol 112)»methyl acetate (spectrum A, in Figure 3.30). (Asymmetric complex (tetrol lb)(triol 112)»methyl acetate is illustrated in Figure 3.31.) Integration of the signals in spectrum A gave a 3.2:1.2:1 ratio of complexes 3b»methyl acetate: 117a»methyl acetate:(tetrol lb)(triol 112)»methyl acetate. The relevance of this ratio is discussed later in this section. The formation of asymmetric complex (tetrol lb)(triol 112)»methyl acetate has relevance to the formation of hemicarceplex 100, (see Chapter 1 section Eviii). Note, the appearance of two signals for each methyl group of methyl acetate in asymmetric (tetrol 202 lb)(triol 112)»methyl acetate is due to the two different orientations this guest can adopt in this asymmetric host (Figure 3.31). The relative stability of complex 3b»methyl acetate and monobridged complex 118»methyl acetate at 313 K are shown in spectrum B in Figure 3.30. This lH NMR spectrum is much simpler than spectra A because no mixed complex was observed. The relative stability constants determined for various complexes are reported in Table 3.10. Figure 3.30. *H NMR Spectra of the Relative Stability of Complexes»methyl acetate in Nitrobenzene-Js. ttw*w#F ^ ^ ^ ^ i w *\%JWW,% 0.5 0.0 -0.5 ppm -1.0 -1.5 -2.0 B r T T n i i i i i i I 0.5 0.0 -0.5 ppm -1.0 -1.5 -2.0 LH NMR in nitrobenzene-rf5 for A, tetrol lb (2.17 mM), triol 112 (2.17 mM), DBU (3.9 mM) and methyl acetate (0.81 mM) at 298 K; complex 3b«methyl acetate (0.07, -1.47 ppm); complex 3b«methyl acetate (-0.09, -1.62 ppm); asymmetric complex (0.01, -0.01, -1.53, -1.56 ppm). B, tetrol lb (1.03 mM), monobridged 102 (1.54 mM), DBU (5.9 mM) and methyl acetate (0.38 mM) at 313 K; complex 3b«methyl acetate (0.04, -1.51 ppm); monobridged complex 118*methyl acetate (-0.09, -1.62 ppm); 203 Figure 3.31. Schematic Illustration of the two Orientations of Methyl Acetate in Asymmetric Complex (Tetrol lb)(Triol 112)»Methyl Acetate. In addition, we wanted to compare the relative stability of A,B-bis-bridged complex 103»methyl acetate and A,B-bis-bridged charged complex 119,methyl acetate to examine the effect of the CHB's (see page 187). We planned to measure the relative stabilities of these two complexes separately by competing each with a common neutral binder such as tetraprotio 123a»methyl acetate. Unfortunately, the insolubility of tetraprotio 123a in nitrobenzene-ds prevented us from using it in these experiments. Therefore, we synthesized the analogue tetra-OMe 125 in 71% yield from A,B-bis-bridged 103 by methylating A,B-bis-bridged 103 with dimethyl sulfate (Figure 3.32). Tetra-OMe 125 was also instrumental in the determination of absolute stability constants (see section xvii). 204 Figure 3.32. Synthesis of tetra-OMe 125. A,B-bis-bridged 103 tetra-OMe 125 tetra-OMe 125»guest Table 3.10. Abel's of Various Complexes'Methyl Acetate in Nitrobenzene-<i5. complex'methyl acetate Kref A,B -bis-bridged charged complex 19b 119#methyl acetate complex Sbnnethyl acetate 18 triol-complex 117a»methyl acetate 6.7 tetra-OMe 125»methyl acetate 5.9 (tetrol lb)(triol 112)»methyl acetate 5.6 monobridged complex 118»methyl acetate 4.1 A,B -bis-bridged complex 103»methy 1 3.9 acetate A,C-diol-complex 116»methyl acetate 1.0 A,B-diol-complex 115»methyl acetate 1.0 a KK\s determined at 298 K in nitrobenzene-^ ;b Relative stability constant of A,B-bis-bridged charged complex 119,methyl acetate to complex 3b»methyl acetate was determined at 333 K. The host ^rei's of the various complexes of methyl acetate in nitrobenzene-Js are given in Table 3.10 and indicate that there are no great differences in the relative stabilities of these complexes because they are within a factor of 20 of one another. The differences that are observed in the relative stabilities do, however, show some predicted trends. For instance, the number of CHB's is important to the overall stability of the methyl acetate 205 complex as shown by the following trend: complex 3b»methyl acetate > triol-complex 117a»methyl acetate > A,C-diol-complex 116»methyl acetate = A,B-diol-complex 115»methyl acetate. Also, the addition of CHB's to neutral A,B-bis-bridged complex 103«methyl acetate results in the formation of a stronger complex (i.e., AB-bis-bridged charged complex 119«guest). Interestingly, the three complexes which have three CHB's (asymmetric (tetrol lb)(triol 112)»methyl acetate, triol-complex 117a«methyl acetate, and monobridged complex 118«methyl acetate) all have very similar stabilities (Table 3.10). This suggests that the presence of the CHB's in these three complexes is a major component in their overall stabilities. c. Thermodynamic Parameters for Complexes'Methyl Acetate The small differences in selectivity for these complexes was somewhat surprising considering the structural differences between them. We therefore decided to explore these differences in further detail by determining the thermodynamic data for these complexes. Their great differences in molecularity during complexation (e.g., seven molecules for complex 3b»guest versus only two for A,B-bis-bridged complex 103»guest) create the possibility of substantial entropic effects. The temperature dependence of the relative stability constants for this series of complexes of methyl acetate are presented in Figure 3.33 and in Table 3.11. The relative stability constants of both A,B-diol-complex 115»methyl acetate and A,C-diol-complex 116»methyl acetate could not be studied because at temperatures greater than 298 K the signals in the *H NMR spectra for methyl acetate were broadened due to the intermediate rate of exchange of methyl acetate in these complexes. In addition, the very slow rate of exchange of methyl acetate in A,B-bis-bridged charged complex 119«methyl acetate prevented an investigation of the temperature dependence of its relative stability constants due to excessively long equilibration times. (Equilibration of A,B-bis-bridged charged complex 119»methyl acetate with other 206 complexes took 4 days at 333 K and other hosts have intermediate exchange rates above this temperature.) Figure 3.33. Graph of Rln(Complex 1'methyl acetate / Complex 2»methyl acetate) versus 1/T (K) in Nitrobenzene-d5. A Rln(triol complex 117a*methyl acetate/monobridged complex 118«methyl acetate); r 2 = 0.89 o Rln(AB-bis-bridged complex 103»methyl acetate/tetra-OMe 125»methyl acetate); r 2 = 0.86 + Rln(tetra-OMe 125*methyl acetate/monobridged complex 118*methyl acetate); r 2 = 0.94 • Rln(complex 3b«methyl acetate/triol complex 117a«methyl acetate); r 2 = 0.98 • Rln(complex 3b»methyl acetate/monobridged complex 118*methyl acetate); r 2 = 0.94 Table 3.11. Thermodynamic Data for KTe\s of Various Complexes»Methyl Acetate in Nitrobenzene- .^ complex triol-complex A,B-bis-bridged tetra-OMe monobridged 3b»methyl 117a«methyl complex 125»methyl complex acetate3 acetate 103»methyl acetate 118»methyl acetate acetate AAH° kcal/mol 1.8 -1.0 2.7 3.7 0.0 AA5° cal/mol(K) 8.8 -2.3 9.8 12.3 0.0 TAA5° kcal/molb 2.6 -0.7 3.0 3.7 0.0 AAG° kcal/mol -0.9 -0.3 -0.2 0.0 0.0 •ma a Errors are estimated to be ±30% mainly due to integration of the ! H N M R spectra. b Temperature = 300 K 207 The thermodynamic data for the relative stabilities of the five different complexes of methyl acetate studied (Table 3.11) were calculated from the van't Hoff graph of Rln(complex 1 "methyl acetate/complex 2»methyl acetate) versus 1/T (K), Figure 3.33. The resulting thermodynamic data from this graph were calculated relative to monobridged complex 118»methyl acetate. The most obvious trend observed in Table 3.11 is the small changes in free energy that differentiate these complexes of methyl acetate. The relative free energies for these five complexes are all within 1 kcal/mol of each other. As expected, the bimolecular complexes A,B-bis-bridged complex 103«methyl acetate and tetra-OMe 125«methyl acetate are more entropically favored than monobridged complex 118»methyl acetate (molecularity = five). Unexpectedly, complex 3b*methyl acetate (molecularity = seven) is entropically favored over monobridged complex HS^methyl acetate (molecularity = five). This may be due to the entropic cost of restricting the motion of the inter-bowl methylene bridge (OCH2O) in the monobridged complex 102 being greater than the entropic cost of bringing two molecules together in the complex 3b»guest. Also statistically speaking, complex 3b«methyl acetate has four different ways of forming, that is rotation of the upper and lower bowls with respect to each other by 90°, 180°, 270° and 360°. This is more entropically favorable compared to monobridged complex 118»methyl acetate which only has one way in which it can form. Triol-complex 117a»methyl acetate is the most entropically unfavorable complex of the five complexes studied. This is not surprising because its formation is statistically demanding and it has a molecularity of six. Monobridged complex 118«methyl acetate is the second most entropically unfavorable complex investigated. Like triol-complex 117a»methyl acetate, the formation of monobridged complex 118»methyl acetate is more statistically demanding and it has a molecularity of five. Triol-complex 117a,methyl acetate is the most favored enthalpically compared to the other four complexes of methyl acetate investigated. One feature that differentiates triol-complex 117a»methyl acetate from the other four studied is that it has a larger portal. The 208 portal of the triol-complex llTa^methyl acetate may allow more room for the carbonyl of methyl acetate, thus reducing steric strain and creating more favorable van der Waals interactions compared to other complexes (see section xiiie for discussion of the relative binding of guests in triol-complex 117a»guest). (The effect of the guest (methyl acetate) on the thermodynamic data reported in Table 3.11 is largely unknown but can be determined by repeating these experiments in the presence of other guest molecules.) Monobridged complex 118*methyl acetate is enthalpically more stable than complex 3b»methyl acetate. This may be due to a tighter complex in monobridged complex 118»methyl acetate versus complex 3b»methyl acetate due to the covalent linkage of monobridged complex 118»methyl acetate; note the inter-bowl distances in carceplex 2b»pyrazine are shorter than in complex 3b»pyrazine (see section xi). The bis-bridged complexes AB-bis-bridged 103»methyl and tetra-OMe 125»methyl acetate are enthalpically less favored relative to complex 3b»methyl acetate, triol-complex 117a»methyl acetate, and monobridged complex 118»methyl acetate possibly due to the lack of stabilizing CHB's in the bis-bridged complexes. Of these two bis-bridged complexes, AB-bis-bridged 103»methyl acetate is favored enthalpically over tetra-OMe 125«methyl acetate perhaps because it can form neutral hydrogen bonds between the phenolic hydroxyls of the upper and lower bowls and potentially form hydrogen bonds between the phenolic hydroxyls to the carbonyl of methyl acetate. Furthermore, as determined by examination of CPK models, the OMe groups of tetra-OMe 125»methyl acetate cannot be in conjugation with the aromatic rings of the bowls due to steric interactions. This alone should not effect the stability of tetra-OMe 125»methyl acetate complex relative to the others studied because it occurs in both empty tetra-OMe 125 and tetra-OMe 125»methyl acetate, but these steric interactions of the OMe groups may constrain the bowls to be slightly apart, which may effect this host's ability to bind methyl acetate by causing poor van der Waals interactions between the interior of the host and the guest relative to the other complexes studied. 209 The determination of the thermodynamic parameters for these complexes of methyl acetate have provided further insight into the properties that determine the stabilities of these assemblies. Such data further our knowledge about the elusive noncovalent interactions that govern the formation of both natural and nonnatural assemblies. xvii. Absolute Stabilities of Complex tetra-OMe 125»Guest a. The Nature of the Free Species Only relative stability constants and one example of an apparent stability constant have been reported so far in this thesis. We cannot measure absolute stability constants in CDCI3 because we know that the free species is not simply an empty dimer. If the free species is a monomer, then we must take into account the concentration of DBU and DBUH + , which complicates matters immensely. If we assume the free species is all complex 3b»CDCl3 (likely to be true as it is at least 70% complex 3b«CDCl3) then we can get a KrQ\ for any guest:CDCl3 (see section xviib). Then a determination of the absolute stability constant for complex 3b»CDCl3 in CDCI3 would require either (1) generation and characterization of a monomer species and determination of DBU/DBUH"1" concentrations (the equation of the absolute stability constant would include four deprotonations, which would complicate the simple affinity between the host and the guest) or (2) generation and characterization of an empty dimer. Neither is possible. We were unable to determine the nature of the free species of tetrol lb and DBU in nitrobenzene-J5. This is because the *H NMR spectrum of a solution of tetrol lb and DBU in nitrobenzene-Js is very broad as shown in Figure 3.34. Saturation of this sample with nitrogen does not sharpen the lH NMR spectrum, indicating that the broadness of the lH NMR is not simply due to the binding of oxygen. Either dilution or increasing the temperature does, however, result in a better resolved lH NMR spectrum, suggesting that 210 the free species is an aggregate. Thus, if we were to calculate an absolute stability constant for complex 3b#guest in nitrobenzene, we would have to determine the stability constant Ks based on the dynamic processes occurring in equation 1: (n+2)(l)+Guest+(m+4)(DBU) ^((l-guesfl)(DBU)4) + ((l)n(DBU)m) {equation 1} where 1 is tetrol 1; ((l»guest»l)(DBU)4) is complex 3»guest; and ((l)n(DBU)m) is the free species which is an aggregate (the aggregate may or may not contain guest molecules). Ks = [(l.guesfl)(DBU)4][(l)n(DBU)m] / [l] ( n + 2 )[DBU] ( m + 4lGuest] f l ree We do not know the concentration of the aggregate ((l)n(DBU)m), DBU, or free tetrol (1); therefore this equation cannot be solved. If we assume the free species is an empty dimer or treat it as such, it allows a calculation of an apparent stability constant (i.e., equation 1 simplifies to equation 2): ((M)(DBU)4) + Guest ^ ( ( l » g u e s f l ) ( D B U ) 4 ) {equation 2} where 1 is tetrol 1; ((l»guest»l)(DBU)4) is complex 3»guest; and ((1#1)(DBU)4) is an empty dimer. ^a P = [(l-guesfl)(DBU)4] / [((M)(DBU)4)][Guest]free The concentrations of complex 3»guest, empty dimer and free guest can be determined from integration of the lH NMR spectrum, again assuming all the free host is an empty dimer. Thus, the Kap for complex 3b»NMP in nitrobenzene-ds at 313 K was determined to be 280 M" 1 . Thus, not knowing the exact nature of the free species of tetrol lb/DBU in nitrobenzene-ds prevents the direct determination of an absolute stability constant; if assumptions are made, an apparent stability constant can be calculated but this is not totally 211 satisfactory. We therefore explored other complexes that may provide an absolute stability constant. Figure 3.34. Tetrol lb and DBU in Nitrobenzene-^ at Ambient Temperature. i i i i i i i i i i 9.0 8.0 7.0 6.0 5.0 ppm 4.0 3.0 2.0 1.0 0.0 • H NMR of tetrol lb (2.89 mM) and DBU (6.07 mM) in nitrobenzene-^  at ambient temperature. The large signals at 8.11, 7.67 and 7.50 ppm are due to the solvent. The signals for DBUH+ are at 3.57, 3.32, 3.15, 1.83, 1.73, and 1.36 ppm. The rest of the signals are due to tetrol lb and/or its deprotonated derivatives/aggregates. b. Determination of an Absolute Stability Constant Using Tetra-OMe 125 Tetra-OMe 125 (structure on page 205) forms a neutral complex, and appears to be essentially empty when "free" in nitrobenzene-Js (see below); thus tetra-OMe 125 is an ideal host for determining the absolute stability constant. This complex is much simpler than complex 3b»guest because there are no counterions to be concerned with. A large variety of guest molecules can be encapsulated to form complex tetra-OMe 125»guest including: pyrazine, methyl acetate, 1,4-dioxane, pyridine, DMSO, acetone, benzene, 1,3-dioxane, DMA, CHC13, NMP, morpholine, CH 2 I 2 , CH 2 Br 2 , CH 2BrCl, and CH 2C1 2. 212 Before measuring the value for the absolute stability constant for tetra-OMe 125»guest, we explored the nature of the empty host at 333 K.X V 1 The lH NMR signals for the methylene protons that line the interior of the bowls of tetra-OMe 125 (1.15 mM) were slightly broad in nitrobenzene-Js at 333 K. Saturating this sample with O2 gas further broadened the signals for these hydrogens as shown in lH NMR spectrum A in Figure 3.35. This demonstrates that the broadening observed in the *H NMR spectra of tetra-OMe 125 was due to binding of triplet oxygen whose paramagnetic character causes substantial broadening of nearby hydrogens. Displacing the dissolved O2 gas in this sample by saturation with N2 gas resulted in considerable sharpening of the these inside methylene hydrogens as shown in spectrum B in Figure 3.35. Therefore, the free species for tetra-OMe 125 can contain some bound O2, N2, or presumably, other gases that are present in the nitrobenzene-Js solvent. These species are treated as being an empty host. In order to determine the absolute stability constant for tetra-OMe 125»guest, we had to use a guest that showed weak complexation (otherwise, no free species would be observed). NMP was chosen as the guest because it exhibits weak binding, and had host 1 1 H NMR signals that were unique from free tetra-OMe 125. The H NMR spectrum of a mixture of tetra-OMe 125 (1.36 mM) and NMP (3.26 mM) in nitrobenzene-d5 at 333 K showed a ratio of -1:1 (by integration) for tetra-OMe 125:tetra-OMe 125»NMP (spectrum C in Figure 3.35). The complexity of this *H NMR spectrum is due to the top/bottom asymmetry of tetra-OMe 125#NMP induced by the large NMP, which rotates slowly about the host's C2 axes on the *H NMR timescale. We used the integration of one of the inside methylene signals (Hin) of tetra-OMe 125-NMP (3.74 ppm) and tetra-OMe 125 (3.58 ppm) to determine the stability constant. From this spectral data, a stability constant (ZQ 3 3 of 410 + 40 M"1 was calculated. As a control experiment, we determined the Ks of tetra-x v i We performed our measurements at 333 K because spectra of tetra-OMe 116»guest were better resolved, equilibrium was reached faster and tetra-OMe 116 had greater solubility. 213 OMe 125«NMP on a sample that was degassed and sealed under dynamic vacuum. This gave a Ks = 390 M ' 1 which is well within the experimental error, thereby suggesting that the binding of residual 0 2 by tetra-OMe 125 does not significantly effect its absolute stability constant. The absolute stability constant of tetra-OMe 125«CDCl3 was calculated to be 560 ± 50 M* 1 . The ratio of £ s ' s determined for tetra-OMe 125*CDC13 and tetra-OMe 125-NMP (560/410 = 1.36) correlated with the Kre\ (1.40) determined for these two complexes. We thus used the absolute value of the stability constant for tetra-OMe 125»NMP and the relative stability constants determined for other guests (Table 3.12, third column) to calculate the absolute stability constants for the series of guest molecules encapsulated in tetra-OMe 125»guest (Table 3.12, fourth column). Table 3.12. Kre\ and A:s for tetra-OMe 125»guest at 333 K in Nitrobenzene-J5. guest Template KTef at 333 K Ksa at 333 K /<:relaat298 K #reiaat313 K Ratios tetra-OMe tetra-OMe complex complex Carceplex 125 125 M" 1 3b»guest 3b»guest 2a»guest pyrazine 1000000 1500 630000 35000 18000 methyl acetate 520000 860 360000 20000 10000 1,4-dioxane 290000 690 280000 13000 6100 DMSO 70000 170 68000 5000 3000 pyridine 34000 950 390000 1000 560 acetone 6700 29 12000 230 150 benzene 2400 52 22000 140 85 1,3-dioxane 200 23 9500 51 27 DMA 20 29 11000 6.8 4.5 NMP 1 1.0 410 1.0 1.0 CHCI3 1.4 560 4.8 3.1 morpholine 340 140000 3900 CH 2C1 2 5.6 2300 CH2J3rCl 25 10000 CH 2 Br 2 72 30000 CH9I2 950 390000 a Errors are estimated to ± 20%. 214 Figure 3.35. The effect of 0 2 and N 2 gases on the lR NMR Spectra of tetra-OMe 125 in Nitrobenzene-d5. i : 1 1 1 1 1 1 i 1 1 i 9.0 8.0 7.0 6.0 5.0 PP"i 4.0 3.0 2.0 1.0 0.0 -1.0 -2.0 L i H ;, B i — 9.0 8.0 ~ ~ i — 7.0 5.0 ppm 4.0 ~~i i 3.0 2.0 1.0 0.0 ~~1— -1.0 I — 9.0 I — 7.0 encapsulated NMP 5.0 ppm 4.0 f — 1.0 — I — 0.0 ' H NMR spectra of tetra-OMe 125 in nitrobenzene-ds at 333 K. A, tetra-OMe 125 (1.2 mM) saturated with O 2 gas, H j n signals are broad (4.8, 4.5, and 3.6 ppm). B, tetra-OMe 125 (1.2 mM) saturated with N 2 gas, H j n signals appear as doublets at 4.81, 4.49, and 3.58 ppm. C, tetra-OMe 125 (1.36 mM) and NMP (3.3 mM), sample was degassed and sealed under dynamic vacuum. The H ; n signal for empty tetra-OMe 125 at 3.55 ppm and the signal for tetra-OMe 125*NMP at 3.75 ppm were used to determine KS; the inset shows an expansion of the region of the ' H NMR used to for determination of the KS. 215 The relative selectivity of tetra-OMe 125»guest correlated reasonably well (r2 = 0.89) with the template ratios for the formation of carceplex 2a«guest. This again indicates that the noncovalent interactions that govern the formation of tetra-OMe 125»guest are similar to the noncovalent interactions that govern the selectivity in the transition state of the GDS for the formation of carceplex 2a»guest. The overall range of selectivity observed from the guest Orel's of tetra-OMe 125»NMP to tetra-OMe 125»pyrazine is only 1500 when compared to the million fold range in template ratios observed for the formation of carceplex 2a»guest (Table 3.12). The guest KK\'s for complex 3b»guest for this same range of guest molecules at 313 K and 298 K were also determined (Orel's of complex 3b»NMP to complex 3b«pyrazine are 18,000 and 35,000 at 313 K and 298 K, respectively).xv" Complex 3b«guest is more than ten times more selective than tetra-OMe 125«guest; how much of this is due to the lower temperature used in these experiments is unknown. c. Interesting Features of the Guest Binding in tetra-OMe 125*guest In Chapter 2, we showed that the four dihalomethane guests (CH2I2, CH2Br2, CH2BrCl , and CH2CI2) led to the formation of carceplex 2a»guest and that CHCI3 led to a mixture of carceplex 2a ,CHCl3 and carceplex 2a , CH2BrCl, but template ratios for these guests were not determined. Therefore, we determined the selectivity observed for these five guest molecules by measuring the relative stability constants of tetra-OMe 125»guest of each of these guests. As shown in Table 3.12, the relative stability of tetra-OMe 125»CHCl3 is by far the worst of these five guests, while the relative stability of the four x v i i The relative stability constants for complex 3b»guest at temperatures above 313 K were not measured for D M A , 1,3-dioxane, N M P and CHCI3 due to the intermediate exchange rate of these guests from complex 3b»guest. 216 dihalomethane guests decreases with decreasing size: i.e., CH2I2 > CH2Br2 > CH^BrCl > CH2CI2 > CHCI3. Furthermore, the relative stability of tetra-OMe 125»CH2l2 was the second largest (equal to tetra-OMe 125»pyridine) of all the guests studied, indicating that it forms highly favorable noncovalent interactions with the interior of this host. The greater relative stability of tetra-OMe 125«pyridine compared to the other host systems studied may reflect the reduced steric interactions of the CH's of pyridine at the equator of this complex. Furthermore, tetra-OMe 125 and pyridine are complementary to each other because each has a dipole moment. Pyridine may orientate in tetra-OMe 125»pyridine such that their dipoles cancel, thus providing favorable electrostatic interactions relative to the other guest molecules studied see (section iv). Surprisingly, we found that morpholine can also be encapsulated as tetra-OMe 125»morpholine, and it forms an even more stable complex than tetra-OMe 125»DMSO. This is the first example of complexation of a guest molecule that was not found to lead to the formation of carceplex 2a»guest, (see Table 2.2). Furthermore, morpholine was also encapsulated as complex 3b»morpholine, (Table 3.12). These results, contradicts our finding that complex 3b«guest is a good model for the transition state for formation of carceplexes because carceplex 2a»morpholine was never isolated. Possible explanations for this are: (1) Morpholine is a sterically demanding guest and may be too large to allow for efficient bridge formation subsequent to the GDS; thus polymerization effectively competes against carceplex formation. (2) Morpholine itself could act as a nucleophile from within a carceplex intermediate and attack the chloromethyl ether (ArOCH2-Cl) formed during formation of the methylene bridge (OCH2O); this would leave a free phenolic hydroxyl which could lead to the formation of polymer. (3) Carceplex 2a«morpholine was lost during work up due to formation of its R.2NH2+ salt which would have remained on the silica gel during column chromatography. 217 d. Calculation of the Absolute Stability Constants for Other Complexes*Methyl Acetate The determination of the absolute stability of tetra-OMe 125»methyl acetate (Table 3.12) and the relative stability of various host molecules (Table 3.10) indirectly allow a calculation of the apparent stabilities of a series of complexes'methyl acetate. The Kaps were determined by multiplying Ks (tetra-OMe 125,methyl acetate) x ^ r e i (complexes»methyl acetate/tetra-OMe 125»methyl acetate). For example, the Kap for complex 3b»methyl acetate is equal to Ks 360,000 M _ 1 (tetra-OMe 125»methyl acetate) x Krei 2.1 (complex 3b*methyl acetate/tetra-OMe 125»methyl acetate) = 750,000 M" 1 (Table 3.13) . We performed this calculation on all the complexes for which we had data at 333 K. The results are reported in Table 3.13 (333 K was the temperature used for calculation of the absolute stability constant of tetra-OMe 125»NMP). Again, we call these values Kap's because the calculations treat the free species as empty complex. Alternatively, the Kap's can be calculated from the data in Table 3.12, also assuming the free species is an empty dimer. For example, one can use the Kap calculated for complex 3b»NMP in nitrobenzene-d$ at 313 K to calculate Kap for complex 3b»methyl acetate; thus Kap 280 M" 1 (complex 3b«NMP see section Bxviia) x Kxe\ 10,000 (complex 3b»methyl acetate to complex 3b»NMP, Table 3.12) = 2,800,000 M" 1. The Kap for complex 3b«methyl acetate calculated in this manner is four times greater than the Kap determined above (750,000 M"1) but this difference is most likely due to the difference in temperatures used for these experiments (313 K versus 333 K). 218 Table 3.13. iv a p ' s of Various Complexes»Methyl Acetate in Nitrobenzene-d5 complex^ methyl acetate tetra-OMe 125»methyl acetate complex 3b#methyl acetate A,B -bis-bridged charged complex 119»methyl acetate monobridged complex 118»methyl acetate A,B-bis-bridged complex 103»methyl acetate 360000 750000 830000 190000 470000 aKap determined by multiplying Ks (tetra-OMe 125»methyl acetate, Table 3.12) x Kre\ (complexes 'methyl acetate/tetra-OMe 125»methyl acetate); values at 333 K in nitrobenzene-d$. xviii. Energetics for Formation of Carceplex 2a»guest The strong correlation (r2 = 0.99 for ln(KTe\) versus ln(TR's)) of template ratios for carceplex 2a«guest and relative stability constants for complex 3b»guest (Table 3.14) suggests that the favorable noncovalent interactions that govern the formation of complex 3»guest play an integral role in the stabilization of the transition state for formation of carceplex 2»guest (i.e., complex 3b»guest is an excellent transition state model for the GDS in the formation of carceplex 2a#guest). For example, the many favorable noncovalent interactions of complex 3b»pyrazine are likely present in the transition state for formation of the second O C H 2 O bridge (formation of the second O C H 2 O bridge was found to be the GDS, see Chapter 2, section Biig). In addition, pyrazine is likely to provide optimum geometry for the phenoxide to attach the chloromethyl ether thereby allowing efficient formation of the O C H 2 O bridge. Other guest molecules such as NMP, may provide fewer favorable noncovalent interactions and more unfavorable interactions in the transition state of the GDS, thus leading to different product ratios for these two guests. 219 Table 3.14. Template Ratios and KK\'s for A,B-bis-bridged 103 and Tetrol l b Determined at 298 K. Guest KTe\ Complex Template Ratios Template Ratios Kre\ A,B-bis-3b»guest tetrol lb A,J3-bis-bridged bridged complex 103 103«guest pyrazine 580 860 40 1800 1,4-dioxane 71 180 5 160 DMSO 14 19 17 pyridine 9.5 14 19 45 acetone 0.90 2 3 3 benzene L0 1 1 1 One final note about the mechanism of carceplex formation is worth noting. The formation of the second bridge was determined to be the GDS for the formation of carceplex 2a»guest (see Chapter 2, section Biig) because competition experiments performed with carceplex intermediates formed subsequent to this step such as A,B-bis-bridged 103 gave vastly different template ratios than was determined using tetrol la (Table 3.14). We suggest that the poor correlation (r2 = 0.7) for the template ratios determined for the formation of carceplex 2a»guest from A,B-bis-bridged 103 versus tetrol lb is a result of guest exchange being slower than formation of the third OCH2O bridged. Indeed, A,B-bis-bridged charged complex 119»guest was shown to exchange its guest very slowly, which made determination of guest Abel's for this species impractical (see section xiva). The guest ^Trei's determined for neutral A,B-bis-bridged complex 103»guest (this is not the best model because it does not have CHB's) do correlate with the template ratios determined for formation of carceplex 2a»guest from tetrol lb, which suggests again that poor correlation of the template ratios from A,B-bis-bridged 103 versus starting from tetrol lb is due to slow guest exchange with A,B-bis-bridged charged complex 119»guest. 220 xix. Computations Carceplex 2a»guest and complex 3b»guest are excellent systems for studying the importance of noncovalent interactions between molecules because small changes in the guest results in large changes in selectivity. These assemblies are attractive models for computational chemists because they provide a relatively simple model to test the force fields used in calculations. We began a collaborative project in 1994 with Houk's group at UCLA. They were interested in determining the origin of the stability of complex 3b»guest as well as the template ratios for formation of carceplex 2a*guest. Houk and coworkers had previously performed theoretical investigations on the egress of guest molecules from hemicarceplexes and therefore were familiar with this field.34"36 Houk and coworkers calculated complexation energies of complex 3b»guestxvl" and carceplex 2b«guest in the gas phase (A£ , a(g)) using Macromodel 4.5,37>38 The absolute free energy of solvation of the guests in chloroform (AG(S)G) were calculated using BOSS 3.539 using OPLS 4 0 parameters. Thus, the free energy of complexation in solution was calculated from the equation (AGa(s) ~ A£a(g) - AG(S)G + constant), where AGa(S) is energy of complex 3b«guest in CDCI3; AEa(g) is the free energy of complex 3b«guest in the gas phase (AGa(g) is approximated by the calculated A£ a( g)); AG(S) is the free energy of solvation of the guest in CDCI3; and the constant represents the free energy of solvation of tetrol lb and complex 3b»guest which were assumed to be approximately constant. The correlation between the calculated AGa(S) for complex 3b»guest and the template ratios for formation of carceplex 2a»guest is illustrated in a logarithmic plot shown in Figure 3.36. The template ratios of carceplex 2a»guest were used for this comparison instead of the ^rei's of complex 3b»guest because more experimental data was available for this system and the template ratios strongly correlated with the KK\s. As can be seen in Figure 3.36, "" Calculations were performed on a tetrol 1 dimer, no CHB's. 221 the theoretical calculations predict the general trend observed for the template ratios for carceplex formation but there is considerable scatter of the data (r2 = 0.61). The scatter in the data may be due to the assumptions given above or inaccuracies in the force fields. This represents the first calculations on complex 3b«guest, and we anticipate that future experimental data will help improve upon the parameters used in these theoretical calculations. Figure 3.36. Correlation Between log(Template Ratios For Carceplex 2a»Guest) and log(Theoretical Calculated Energies of Complex 3b»Guest). 25" o I 20-'5b a o 'S X <0 JJ. 15-0 o U •a 3 U 1 0 -o o o o o 8 o o o o o 0 I f V o 6 m o^ o* O * • . O , 6 o o o o o o 6 « . JO. 0 0 ° Q ^ OH log (Template Ratio) • with calculated solvation energy correction o without calculated solvation energy correction 222 C. Summary This chapter presented the discovery and characterization of a new family of switchable self-assembling structures. These assemblies are excellent models for studying the importance of noncovalent interactions between molecules, because small changes in guests result in large differences in the relative free energy of binding of these complexes. Characterization of complex 3b»guest, the prototypical model for these complexes, included extensive lH NMR spectroscopy as well as 2 H NMR spectroscopy, ESMS, and X-ray crystallography. Characterization of complex 3b»guest and related assemblies revealed that the driving forces that govern their formation include CHB's between the bowls, van der Waals interactions between both the guest and the interior of the complex as well as between the upper rims of the bowls themselves, CH-7C interactions between both the hydrogens of the guest with the arenes of the bowls and the hydrogens of intra-bowl methylene bridges with the guest, CH-X hydrogen bonding between the hydrogens of the intra-bowl methylene bridges of the bowls to the guest, conjugation of the OHO and OCH2O bonds into their respective aromatic rings, and n-n interactions between the arenes of the bowls and the guest. The preorganization of A,B-bis-bridged hosts such as tetra-OMe 125»guest allowed complex formation in the absence of base. The simplicity of neutral binding hosts such as tetra-OMe 125»guest also allowed the determination of absolute stability constants in the noncompetitive solvent nitrobenzene-Js. Thus, the Ks for tetra-OMe 125»NMP and tetra-OMe 125«CHC13 were 410 ± 40 M" 1 and 560 ± 40 M" 1 , respectively at 333 K. The determination of the stability constant for tetra-OMe 125#NMP and the relative stabilities determined for other guest molecules in tetra-OMe 125#guest allow the calculation of the absolute stability constants for the entire series of guest molecules studied. The absolute stability of tetra-OMe 125»pyrazine was thus calculated to be 630,000 M" 1. 223 Evaluation of the enthalpic and entropic components of the free energy of these complexes allows a more in-depth analysis of the noncovalent interactions that drive their formation. Relative thermodynamic data was determined for complex 3b«guest with six or more guest molecules in both CDCI3 and nitrobenzene-J5. Guest molecules with the best apparent van der Waals interactions were generally enthalpically favored relative to those guests that provide poor van der Waals interactions; in terms of entropy, small guest molecules that could have substantial mobility within the interior of the complex were found to be favored relative to larger sterically demanding guests. In contrast to the large range in selectivity observed for various guest molecules within each host system, we found only a small range in selectivity when the host was varied, suggesting that the bonding interactions between the bowls of these assemblies (both number and type of bonds) has only a small impact on the relative affinities of the host toward a guest. The relative thermodynamic data determined for the complexes of methyl acetate revealed that the bimolecular complexes tetra-OMe 125»methyl acetate and A,B-bis-bridged complex 103»methyl acetate are more entropically favored than complexes with greater molecularity. Triol-complex 117a»methyl acetate had the lowest entropy of complexation possibly due to the statistical demands of its formation. Triol-complex 117a«methyl acetate was also the most enthalpically favored complex studied which, in addition to its three CHB's, may be due to the presence of its portal, which may reduce steric interactions of the carbonyl oxygen and thus provide even more favorable van der Waals interactions between methyl acetate and the interior of the host. The formation of these assemblies has great importance to the formation of carceplexes and hemicarceplexes as the relative free energy of binding in complex 3b»guest and triol-complex 117a»guest correlate with the template ratios determined for the formation of carceplex 2a»guest and hemicarceplex 107»guest, respectively. This suggests that these complexes are excellent models for the transition state of the GDS for 224 these respective reactions. Also, the discovery of the triol-complex 117a»guest helped to explain the much greater than statistical yields obtained for hemicarceplex 107»guest. With respect to solvation, the correlation of the relative stability constants of complex 3b»guest in CDCI3 and nitrobenzene-ds with the template ratios determined for the formation of carceplex 2a«guest in NMP suggests that solvation is not very important in terms of the free energy of the complexes. The relative thermodynamic data determined in CDCI3 for complex 3b»guest differed slightly from that determined in nitrobenzene-^ 5; thus the enthalpy and entropy of solvation of various guest molecules in these two solvents must differ. Assemblies such as complex 3b»guest and carceplex 2a«guest are excellent systems for theoretical analysis due to their high sensitivity to small perturbations. The energies calculated for complex 3b»guest reproduced the trend observed for the template ratios determined for the formation of carceplex 2a»guest. Re-evaluation of the force field parameters should improve these correlations thereby honing the forcefields used. The insight gained from the computations will also aid in the development of self-assembling structures. Thus, a combination of experiments and computations should increase our knowledge of noncovalent interactions, which are instrumental to the formation of so many natural and nonnatural assemblies. Future directions and further implications of this work are presented in Chapter 5. 225 D. Experimental i. General All reactions involving air sensitive or moisture sensitive reagents were conducted under nitrogen atmosphere. THF was distilled over sodium/benzophenone. All commercially available chemicals were used without further purification unless otherwise stated. CDCI3, acetone-^ and DMSO-afo NMR solvents were stored over crushed 4A molecular sieves. *H NMR spectra were recorded on a Bruker WH-400 spectrometer in CDCI3 at ambient temperature [(22 + 2) °C] using the residual CHCI3 as a reference (8 = 7.24 ppm) unless noted otherwise. *H NMR spectra recorded on the Bruker WH-400 spectrometer in nitrobenzene-i/5 were referenced to the residual nitrobenzene (8 = 8.11, 7.67, and 7.50 ppm). At temperatures other than ambient temperature, the *H NMR samples were equilibrated in the spectrometer for 20 minutes prior to their acquisition unless noted otherwise. H NMR spectra were recorded on a Varian XL-300 spectrometer. ESMS were recorded on a SCJEX API 300 triple quadrupole mass spectrometer in negative mode. X-ray crystallography measurements were made on a Rigaku AFC6S diffractometer using graphite monochromated Cu-Ka radiation machine and the data analysis was performed using the program package teXsan.41-42 LSIMS and DCI mass spectrometry as well as elemental analyses and column chromatography were performed as described in Chapter 2. 226 ii. Synthesized Compounds Monol 109, A,B-diol 110a, A,C-diol 111, and triol 112 were isolated as by-products in the formation of tetrol lb. Yields were not optimized for any of these compounds as sufficient quantities were already available from previous preparations of tetrol lb. Insufficient amounts of A,B-diol 110a were obtained by this method; a direct synthesis of A,B-diol is given later in this section. Tetrol lb A suspension of tetrabromo-bowl 125 (3.10 g, 3.41 mmol) in dry THF (500 mL) was warmed until the tetrabromide dissolved. The solution was cooled to -78 °C and n-butyllithium (18.2 mL of a 1.5 M solution in hexanes, 27.3 mmol) was added. After 1 min, B(OMe)3 (3.5 mL, 15.0 mmol) was added and the solution was allowed to warm to ambient temperature over 3 h. The reaction mixture was re-cooled to -78 °C, 1.5 M NaOH-15% H2O2 (90 mL) was added and the reaction mixture was again allowed to warm to ambient temperature over 3 h. Na2S205 (15 g, 79 mmol) was carefully added, the THF was removed in vacuo and the yellow solid in the residual water was filtered and washed with water. The filtrate was acidified with 10% aqueous HC1 and extracted with ethyl acetate (3 x 60 mL). The combined organic extracts were washed with brine (50 mL), dried with MgSO*4 and concentrated in vacuo. The two solids were added to an 1 L Erlenmeyer flask with 500 mL of CHCI3 and refluxed. To this suspension DMSO (2.4 mL, 34 mmol) was added and the suspension was cooled to 0 °C and filtered after 1 h. The resulting white solid (95% tetrol lb and 5% triol 112) was then dissolved in THF (200 mL) and dry loaded onto a silica gel gravity column that was eluted with ethyl acetate:hexanes (4:1), affording tetrol lb which was dried at 110 °C (0.1 mm Hg) for 24 h (1.0 g, 45%). The CHCI3 filtrate was also dry loaded onto a silica gel gravity column 227 which was eluted with ethyl acetate:hexanes (1:1), affording the following by-products which were recrystallized from ethyl acetate/hexanes and dried at 110 °C (0.1 mm Hg) for 24 h: monol 109 (230 mg, 11%), A,C-diol 111 (30 mg, 1.4%), a mixture of A,C-diol 111 and A,B-diol 110 (235 mg, 11%), A,B-diol 110 (20 mg, 1%), and triol 112 (417 mg, 21%). For all alcohols, mp > 250 °C; monol 109 triol 112 Monol 109 was characterized as follows: ! H N M R (CDC1 3 , 400 MHz): 8 7.22 (s, 3H, H a and H b ) , 6.76 (s, IH, H c), 6.47 (s, 2H, Hd), 6.46 (s, IH, H e), 5.84 (d, J = 6.9 Hz, 2H, H f or H f ) , 5.74 (d, J = 7.1 Hz, 2H, H f or Hf), 5.29 (s, IH, OH), 4.94 (m, 4H, H g and Hg>), 4.44 (d, J = 6.9 Hz, 2H, H h or Hh'), 4.43 (d, J = 7.0 Hz, 2H, H h or Hh'), 1.74 (m, 12H, CH 3 ) . M S ( L S I M S + , Thioglycerol) m/z (rel intensity): 608 (M + ; 100). Anal. Calcd for C36H32O9: C, 71.04; H, 5.30. Found: C, 70.79; H, 5.23. Triol 112 was characterized as follows: IH N M R (CDCI3, 400 MHz): 8 7.21 (s, IH, H a), 6.75 (s, 3H, H b and H c), 6.49 (s, IH, H d), 5.95 (d, / = 7.1 Hz, 2H, H e or He>), 5.85 (d, J = 7.2 Hz, 2H, H e or H e 0, 5.29 (s, 3H, OH), 4.92 (m, 4H, H f and Hf), 4.44 (d, J = 7.1 Hz, 2H, H g or H g'), 4.43 (d, J = 7.2 Hz, 2H, H g or Hg>), 1.72 (m, 12H, CH 3). 228 MS (LSIMS+, Thioglycerol) m/z (rel intensity): 640 (M+; 100). Anal. Calcd for C73H64O2W /2 H 2 0: C, 66.56; H, 5.12. Found: C, 66.71; H, 4.99. H' 1 1 --H A,C-diol 111 C H 3 A,B-diol 110 A,C-diol 111 was characterized as follows: !H NMR (CDCI3, 400 MHz): 8 7.22 (s, 2H, Ha), 6.75 (s, 2H, H b), 6.49 (s, 2H, H c), 5.84 (d, J = 7.0 Hz, 4H, Hd), 5.27 (s, 2H, OH), 4.93 (q, J = 7.4 Hz, 4H, H e), 4.44 (d, J = 7.0 Hz, 4H, Hf), 1.73 (d, J = 7.4 Hz, 12H, CH 3). MS (LSIMS+, Thioglycerol) m/z (rel intensity): 624 (M+; 100). Anal. Calcd for C3 6H320io'l/2 H 2 Q: C, 68.24; H, 5.25. Found: C, 68.61; H, 5.26. A,B-diol 110 was characterized as follows: !H NMR (CDCI3, 400 MHz): 8 7.21 (s, 2H, H a), 6.76 (s, 2H, H b), 6.47 (s, 2H, H c), 5.94 (d, J = 6.8 Hz, 1H, Hd> or Hd>>), 5.84 (d, J = 7.0 Hz, 2H, H d), 5.74 (d, J = 7.1 Hz, 1H, Hd> or H d »), 5.37 (s, 2H, OH), 4.93 (m, 4H, H e , He>, and He>>), 4.44 (m, 4H, H f , Hf and Hf >), 1.73 (m, 12H, CH 3). (LSIMS+, Thioglycerol) m/z (rel intensity): 624 (M+; 100). Anal. Calcd for C36H 3 2Oio»l/2 H 2 0: C, 68.24; H, 5.25. Found: C, 68.11; H, 5.36. 229 A,B-dibromo-tris-bridged bowl 121a A solution of tetrabromo-tris-bridged bowl 120a31 (2.0 g, 2.23 mmol) in dry THF (300 mL) was cooled to -78 °C and n-butyllittiium (4.2 mL of a 1.5 M solution in hexanes, 6.3 mmol) was added. After 1 min, the reaction mixture was quenched with excess H 2 O and the solution was allowed to warm to ambient temperature over 1 h. The solvent was removed in vacuo and the resulting solid was dissolved in C H 2 C I 2 (200 mL) and washed with 2M HC1 (25 mL), saturated aqueous NaHC03 (25 mL) and brine (25 mL) and dried over MgSC»4. The crude product was purified by silica gel gravity column (eluted with C H 2 C I 2 ) , affording A,B-dibromo-tris-bridged bowl 121a as a white solid, which was recrystallized from CH2Cl2/hexane and dried at 110 °C (0.1 mm Hg) for 24 h (1.1 g, 67%): mp > 250 °C; A H NMR (CDCI3, 400 MHz): 8 7.26 (s, 2H, H a or H b), 7.18 (s, 2H, H a or H b), 7.07 (s, 2H, OH), 6.48 (s, 2H, H c), 5.86 (d, / = 7.2 Hz, 2H, H d), 5.70 (d, J = 7.2 Hz, IH, HH.0. 5.00 (q, J = 7.4 Hz, 2H, H e), 4.90 (q, J = 7.4 Hz, IH, H e - or H e"), 4.64 (q, J = 7.2 Hz, IH, He> or H e"), 4.42 (d, J = 7.2 Hz, 2H, Hf), 4.35 (d, J = 7.2 Hz, IH, H f ) , 1.77 (d, J = 7.4, Hz 6H, CH 3), 1.76 (d, J = 7.2 Hz, 3H, CH 3), 1.71 (d, J = 7.4 Hz, 3H, CH 3). (LSIMS+, NOBA) m/z (rel intensity): 738 (M + ; 100). Anal. Calcd for C 35H3o08Br2: C, 56.93; H, 4.09. Found: C, 57.00; H, 4.14. 230 A,B-dibromo-bowl 122a A mixture of A,B-dibromo-tris-bridged bowl 121a (400 mg, 0.542 mmol), K2CO3 (1.0 g, 7.2 mmol), and CH 2BrCl (1.4 mL, 22 mmol) in NMP (30 mL) was stirred at 60 °C for 2 d. The reaction mixture was concentrated in vacuo, water (50 mL) was added, and the slurry was acidified with 2 M HC1. The slurry was extracted with CHCI3 (3 x 50 mL), and the combined organic solutions were washed with saturated aq. NaHC03 (30 mL) and brine (30 mL), and dried over anhydrous MgSO*4. Silica gel (0.5 g) was added to the CHCI3 solution and the solvent was removed in vacuo. The silica gel absorbed sample was purified by dry loading onto a silica gel gravity column (20 g) and eluted with CHCI3, affording A,B-dibromo-bowl 122a as a white solid which was recrystallized from CH2Cl2/hexane and dried at 110 °C (0.1 mm Hg) for 24 h (404 mg, 95%): mp > 250 °C; ! H NMR (CDCI3, 400 MHz): 5 7.20 (s, 2H, H a or H b), 7.18 (s, 2H, H a or H b), 6.49 (s, 2H, H c), 5.94 (d, J = 7.3 Hz, 1H, H d or H d »), 5.84 (d, J = 7.2 Hz, 2H, Hd'), 5.74 (d, J = 7.2 Hz, 1H, H d or H d -), 5.00 (m, 4H, H e , He> and H e »), 4.45 (d, J = 7.2 Hz, 1H, H f or H f»), 4.40 (d, J = 7.2 Hz, 2H, H f ) , 4.37 (d, J = 7.3 Hz, 1H, H f or H f 0, 1-74 (m, 12H, CH 3). (LSIMS + , NOBA) m/z (rel intensity): 750 (M + ; 100). Anal. Calcd for C 3 6 H 3 o 0 8 B r 2 : C, 57.62; H, 4.03. Found: C, 57.96; H, 3.94. 231 A,B-diol 110 A solution of A,B-dibromo-bowl 122a (500 mg, 0.666 mmol) in dry THF (50 mL) was cooled to -78 °C and n-butyllithium (2.08 mL of a 1.5 M solution in hexanes, 3.33 mmol) was added. After 1 min, B(OMe)3 (0.454 mL, 4.00 mmol) was added and the solution was allowed to warm to ambient temperature over 2 h. The reaction mixture was cooled again to -78 °C, 1.5 M NaOH-15% H2O2 (24 mL) was added and the reaction mixture was again allowed to warm to ambient temperature over 2 h. N a 2 S 2 0 s (15 g, 79 mmol) was carefully added to quench the excess H2O2 followed by H 2 O (100 mL), and removal of the THF in vacuo, furnishing a yellow solid which was filtered and washed with water. This material was then dissolved in C H C I 3 and dry loaded onto a silica gel gravity column that was eluted with ethyl acetate:hexanes (1:1), affording A,B-diol 110 as a white solid, which was recrystallized from CH2Cl2/hexane and dried at 110 °C (0.1 mm Hg) for 24 h (165 mg, 40%): This material was identical (by *H NMR) to that obtained as a by-product in the synthesis of tetrol lb. 232 Tetraprotio 123a A mixture of A,B-bis-diol 110 (0.106 g, 0.168 mmol,), K 2 C 0 3 (1.0 g, 7.23 mmol), methyl acetate, (2.5 mL, 31.4 mmol) and CH 2BrCl (0.11 mL, 1.7 mmol) in NMP (50 mL) were stirred at rt for 24 h. An additional 1.7 mmol of CH 2BrCl were added and the reaction was stirred for an additional 48 h at 60 °C. The reaction mixture was concentrated in vacuo, water (50 mL) was added, and the slurry was acidified with 2 M HC1. The slurry was extracted with CHCI3 (3 x 40 mL), and the combined organic solutions were washed with saturated aq. NaHCC>3 (30 mL) and brine (30 mL), and dried over anhydrous MgSC>4. Silica gel (0.5 g) was added to the CHCI3 solution and the solvent removed in vacuo. The silica gel absorbed sample was purified by dry loading onto a silica gel gravity column (20 g) and eluted with (CH2Cl)2:CCl4 (3:1) affording tetraprotio 123a as a white solid which was recrystallized from CH2Cl2/hexane and dried at 110 °C (0.1 mm Hg) for 24 h (65 mg, 61% of tetraprotio 123a): mp > 250 °C; X H NMR (CDCI3, 400 MHz at 223K): 8 7.15 (s, 4H, H a), 6.80 (s, 4H, H b), 6.61 (d, J = 6.1 Hz, 2H, H c or H d), 6.40 (d, J = 6.1 Hz, 2H, H c or H d), 6.36 (s, 4H, H e), 6.13 (d, J = 7.6 Hz, 2H, H f or Hf 0, 5.97 (br, 4H, Hf), 5.79 (d, / = 6.6 Hz, 2H, Hf or H f 0, 5.00 (q, J = 7.4 Hz, 2H, H g - or H g ») , 4.88 (q, J = 7.3, Hz 2H, 233 Hg- or H g"), 4.81 (q,7 = 7.1 Hz, 4H, H g), 4.54 (d, 7=7.6 Hz, 2H, H h ' orH h"), 4.22 (m, 6H, H h and (Hh« or Hh-)), 1.73 (m, 12H, C H 3 ) , 1.65 (d, 7 = 7.1, 12H, CH 3 ) . MS (DCI, ammonia) m/z (rel intensity): 1291 ((M + NH 4 ) + ; 100). Anal. Calcd for C 7 4 H 6 4 0 2 ( r l H 2 0 : C, 68.83; H, 5.15. Found: C, 68.92; H, 5.10. Hexaprotio 124 A mixture of monol 109 (0.078 g, 0.128 mmol,), K 2 C 0 3 (1.5 g, 11 mmol), and CH 2 I 2 (0.26 mL, 3.2 mmol) in NMP (50 mL) was stirred at 60 °C for 24 h. An additional 3.2 mmol of CH 2 I 2 were added and the reaction was stirred for an additional 24 h at 60 °C. The reaction mixture was concentrated in vacuo, water (50 mL) was added and the slurry was acidified with 2 M HC1. The slurry was extracted with CH 2C1 2 (3 x 40 mL), and the combined organic solutions were washed with saturated aq. NaHC0 3 (30 mL) and brine (30 mL), and dried over anhydrous MgS0 4. Silica gel (0.5 g) was added to the CH 2C1 2 solution and the solvent removed in vacuo. The silica gel absorbed sample was purified by dry loading onto a silica gel gravity column (15 g) and eluted with ethyl acetate/hexanes (1:1) affording hexaprotio 124 as a white solid which was recrystallized from ethyl acetate/CH2Cl2/hexane and dried at 210 °C (0.1 mm Hg) for 24 h (64 mg, 81% of hexaprotio 124): mp > 250 °C; 234 ! H NMR (CDCI3, 400 MHz): 8 7.21 (s, 2H, H a), 7.20 (s, 4H, H b), 6.99 (s, 2H, H c), 6.47 (s, 2H, Hd), 6.44 (s, 4H, H e), 5.72 (d, J = 7.0 Hz, 4H, H f or Hf), 5.50 (d, J = 7.3 Hz, 4H, H f or H f ) , 5.39 (s, 2H, H g), 4.91 (m, 8H, H h and Hh'), 4.43 (d, J = 7.3 Hz, 4H, Hj or Hj-), 4.32 (d, J = 7.0 Hz, 4H, Hj or Hj>), 1.74 (m, 24H, CH 3). (LSIMS + , NOBA) m/z (rel intensity): 1228 (M + ; 100). Anal. Calcd for C 7 3H640i 8'lH20: C, 70.30; H, 5.33. Found: C, 70.60; H, 5.17. Tetra-OMe 125 A mixture of A,B-bis-bridged 103 (0.053 g, 0.040 mmol,), K 2 C 0 3 (1.0 g, 7.2 mmol), and S02(OCH3)2 (0.15 mL, 1.6 mmol) in acetone (40 mL) was refluxed for 16 h. Diethyl amine (1.0 mL, 10 mmol) was added and the reaction mixture was stirred for lh to quench any residual SO(OCH3)2. The reaction mixture was concentrated in vacuo, water (20 mL) was added, and the slurry was acidified with 2 M HC1. The slurry was extracted with CHC13 (3 x 50 mL), and the combined organic extracts were washed with saturated aq. NaHC0 3 (30 mL) and brine (30 mL), and dried over anhydrous MgS04. The CHC13 solution was concentrated in vacuo and purified by chromatography on a silica gel gravity column (20 g), and eluted with CHCl3/hexanes/ethyl acetate (60:20:1), affording tetra-OMe 235 125 as a white solid which was recrystallized from CEkCVhexane and dried at 210 °C (0.1 mm Hg) for 48 h (41 mg, 74% of tetra-OMe 125): mp > 250 °C; A H NMR (CDC13, 400 MHz): S 6.89 (s, 4H, H a or H b), 6.79 (s, 4H, H a or H b), 6.69 (d, J = 6.1 Hz, 2H, H c or Hd), 6.39 (br, 2H, H c or H d), 6.16 (d, J = 1A Hz, 2H, H e - or He"), 6.03 (d, J = 7.5 Hz, 4H, H e), 5.77 (d, J = 7.5 Hz, 2H, H e - or H e"), 5.05 (q, J = 7.4 Hz, 2H, H f or H f >), 4.97 (q, J = 1A Hz, 2H, Hf or Hf ) , 4.87 (q, J = 7.4 Hz, 4H, Hf), 4.59 (br, 2H, H g - or H g - ) , 4.26 (d, J = 7.5 Hz, 2H, Hg- or H g"), 4.26 (d, J = 7.5 Hz, 4H, H g), 3.84 (s, 12H, OCH3), 1.72 (d, J = 7.4 Hz, 12H, CH 3), 1.64 (d, J = 7.4 Hz, 12H, CH 3). (LSIMS + , Thioglycerol) m/z (rel intensity): 1393 (M + ; 100). Anal. Calcd for C 78H7 2024'H 20: C, 66.38; H, 5.28. Found: C, 66.39; H, 5.17. iii. Crystal Structure of Complex 3b«pyrazine21 Crystals were grown by dissolving tetrol l b (0.11 mmoles), DBU (0.22 mmoles), and excess pyrazine (0.91 mmoles), in nitrobenzene (8.0 mL) and letting the mixture stand in a closed vial for 30 days. A yellow prism-shaped crystal of dimensions 0.3 x 0.5 x 0.5 mm was mounted in a glass capillary. The asymmetric unit consists of 1/4 of two crystallographically independent pyrazine complexes (1/4 x 2 x 2bowls»pyrazine), two DBU»H + ions, one nitrobenzene and two water molecules. The pyrazine guest molecule sits on a four-fold axis and is thus disordered. Crystal data: tetragonal P4cc (no. 103); a = b = 22.661(3) A, c = 30.209(2) A; V = 15512(2) A 3 ; Formula unit = [C 9Hi7N 2]4 +[C76H64N 20 24] 4- • 2C 6 H 5 N0 2 • 4H20 with Z = 4; Racemic; F.W. = 2320.61; F(000) = 4936; li(Cu-Ka) = 6.0 cm"1; ^(Cu-Ka) = 1.54178 A; p c a i c = 0.99 g/cm3; t = 21°C; 9337 reflections collected (20m a x = 130.0°). Final full-matrix least-squares refinement13 (on F 2 ) based on 4426 unique reflections converged with 791 variable parameters and 928 restraints. Non-hydrogen atoms were refined anisotropically, 236 hydrogen atom positions were calculated. R(F) = 0.1130 for 2018 F 0 > 4a(F0); GooF(F2) = 1.047; A p m a x = 0.44 e7A3, Apmin = -0.36. All measurements were made on a Rigaku AFC6S diffractometer and the data analysis was performed using the program package teXsan.41 iv. General procedure for determining the t\/2 for decomplexation of pyrazine from complexes complex 3b»pyrazine. The lH NMR spectrum of a mixture of tetrol lb (2.89 mM), DBU (6.1 mM), and pyrazine (1.45 mM) was recorded at 273 K. Excess dioxane-Js (145 mM) was added to this sample and spectra were recorded as a function of time. The decrease in the concentration of complex 3b»pyrazine was measured by integration of the unique host signals. A plot of ln [complex 3b«pyrazine] versus time gives the approximate rate constant (&d = 1.2 x 10 min ) for the decomplexation of pyrazine from this complex (Figure 3.37). Thus, the decomplexation was first order in complex 3b»pyrazine, with a t\/2 of 59 min. Only the initial rate is used (first five data points) to estimate &d because as equilibrium of this sample approaches the plot becomes nonlinear. Figure 3 . 3 7 . Plot of ln[complex 3b»pyrazine] versus time at 273 K, r = 0.99. 0 . 8 T 237 In order to establish equilibrium for relative binding studies, samples were r equilibrated at the appropriate temperature for at least five times the half-life for the decomplexation of pyrazine (from the slowest complex if two hosts were in competition). This method for determining the time necessary for equilibration was found to correlate with simply recording the lH NMR of the samples, repeatedly until no change was observed. Pyrazine was shown to have the slowest half-life for decomplexation from complex 3b»pyrazine (fi/2 of 21 h.) relative to complex 3b» 1,4-dioxane (fi/2 of 4.2 h.) and complex 3b«DMSO (t\/2 of 1 h.) at 253 K. Pyrazine (five equivalents) was used for displacing 1,4-dioxane and DMSO in complexes 3b» 1,4-dioxane and 3b»DMSO, respectively. Pyrazine was assumed to have the slowest decomplexation in the other host systems studied. v. Complexation Experiments a. General procedure for determination ofKrei 'sfor complex 3b*guest in CDCI3: Tetrol l b (19.0 mg, 0.0289 mmoles) was weighed into a 10.0 mL volumetric flask. To this, 2.1 equivalents of DBU (9.3 |iL, 0.059 mmoles) were added and the flask was filled to volume with CDCI3. 4A molecular sieves (ca. 0.3 mg) and 500 jaL of the above stock solution were added to lH NMR tubes to create samples 1-6 (Table 3.15): Stock solutions of the guests were added to each sample as summarized in Table 3.15. lH NMR samples were prepared such that complex 3b»guest 1: complex 3b#guest 2 were -1:1 and the total amount of free species was less than 15% of the total host (Table 3.15). The KK\s were determined by integration of the unique signals (Hp, H o u t , Hjn and guest) of the host and/or guest in each complex as shown in Table 3.16. 238 Table 3.15. Quantities of Host and Guest Added to A H NMR Samples at 298 K CDCI3. Sample # Guest 1 (Gl) Guest 2 (G2) equivalents G1:G2 C D C I 3 Added (uL) 1 pyrazine3 (11.5 uL, 0.723 umol) 1,4-dioxane (9.85 uL, 5.79 umol) 0.5:4 28.7 2 1,4-dioxane (9.85 uL, 5.79 umol) DMSOc (10.3 uL, 28.9 umol) 4:20 33.5 3 DMSOc (6.66 uL, 18.8 umol) pyridine (2.34 uL, 28.9 umol) 13:20 38.7 4 pyridine (2.34 uL, 28.9 umol) acetone-fife (21.3 uL, 289 umol) 20:200 24.1 5 acetone-d6 (21.3 uL, 289 umol) benzene-c?6 (25.6 uL, 289 umol) 200:200 3.1 6 Stock, blank 50.0 a Pyrazine was added as a stock solution (63.1 mM) in CDCI3. b 1,4-dioxane was added as a stock solution (587 mM) in CDCI3. c D M S O was added as a stock solution (2.82 mM) in CDC1 3 . 4© complex 3b»guest 239 Table 3.16. Integration results for Orel's3 for Complexes 3b»guest at 298 K in CDCI3. Guest 1: Guest 2b (G1:G2) equiv. per tetrol l b * H NMR integration complex 3b»guest 1 ! H NMR integration complex 3b»guest 2 ^ r e l G1:G2 -Krel Gl:benzene pyrazine: 1,4-dioxane (0.5:4) 1.02 H p , H o u t , H j n 1.00 H p , H 0 u t , H j n 8.20 580 pyrazine: 1,4-dioxane (0.5:4)c 1.00 H p , H o u t , H j n 1.02 H p , H0ut> H j n 7.84 560 l,4-dioxane:DMSO (4:20) 1.06 H p , H 0 u t 1.00 H p , H o u t 5.32 71 DMSO:pyridine (13:20) 1.00 H p , H 0 ui> H i n 1.07 H p , H 0 u t , H i n 1.43 14 pyridine:acetone-c?6 (20:200) 1.05 H p , H 0 u t , H i n 1.00 H p , H o u t , H j n 10.5 9.5 acetone-^benzene--d6 (200:200) 0.90 H p , H o u t , H j n 1.00 H p , H o u t , H i n 0.90 0.9 benzene-J6 1 a Kre\ = starting ratio (G2:G1) x integration ratio (G1:G2); errors are estimated to be 20%. b The IH NMR signals integrated for complex 3b»guest are indicated by H p , H o u t , H;„ and/or guest. The IH NMR assignments for complex 3b»guest are tabulated in Table 3.7. If the signals for the free species or a guest peak overlapped with the host signals of complex 3b»guest the host signals were subtracted from the total system. c Spectrum rerun after 24 hours. 240 Table 3.17. H NMR assignments for complexes 3b»guest in C D C I 3 . complex 3b»guesta H0ut methine Hin C H 3 guest signals pyrazine 6.66 6.00 4.90 4.01 1.69 4.30 (s, 4H) 1,4-dioxane 6.50 6.16 4.89 4.27 1.62 0.03 (br, 8H) DMSO 6.51 6.12 4.87 4.27 1.63 -1.02 (s, 6H) pyridine 6.66 5.98 4.89 • 3.99 1.69 6.27 (m, 1H), 4.30 (m, 2H), 3.00 (m, 2H), acetone-^ 6.47 6.14 4.87 4.09 1.61 benzene-^ 6 6.74 6.09 4.98 4.12 1.77 Stock Solution 6.48 6.14 4.89 4.40 1.63 a ! H N M R spectra were all well resolved: H p (s, 8H) H o u t (d, J = 7.2-7.7 Hz, 8H) methine (m, 8H) H i n (d, J = 7.4-7.7 Hz, 8H) C H 3 (d, J = 7.4-7.9 24H). The signals for the DBUH+ in the stock solution are at: 3.44 (m, 2H) 3.26 (m, 6H) 2.67 (m, 2H) 1.85 (m, 2H) 1.70 (m, 2H) 1.56 (m, 2H). The relative stability constants for complex 3b»guest at 283 K, 313 K and 323 K in CDCI3 were determined in the same manner as described above. In all cases equilibration was achieved within twenty minutes. The moderate exchange rate of complex 3b»benzene and complex 3b»acetone at 323 K prevented the determination of KTC\ for these complexes at this temperature. Complex 3b»DMSO and complex 3b#pyridine were broadened at 323 K but were still in slow exchange on the *H NMR timescale. The Orel's determined at 253 K, and 273 K were run as above except the *H NMR samples were equilibrated in a constant temperature bath over 48 hours and 16 hours, respectively, prior to recording their *H NMR spectra. b. Determination ofKrei 'sfor A, C-diol-complex 116»guest and A,B-diol-complex 115»guest A,C-diol 111 (10.8 mg, 0.0173 mmoles) was weighed into a 5.00 mL volumetric flask. To this, 1.0 equivalents of DBU (2.6 uL, 0.017 mmoles) were added and the flask was filled to volume with CDCI3. *H NMR samples were prepared by adding 500 uL of 241 this 3.46 mM stock solution to *H NMR tubes containing molecular sieves. The equivalents of guests added to the *H NMR samples are indicated in Table 3.18; pyrazine was added as a stock solution. The *H NMR tubes were equilibrated (out rate of pyrazine is too fast to determine by *H NMR spectroscopy at 253 K) for 20 minutes in the spectrometer at 253 K prior to data acquisition. Table 3.18. ^r ei's A,C-diol-complexes 116-guest (3.44 mM) at 253 K in CDCI3. Guest 1-Guest 2a ! H NMR integration IH NMR integration Kre\ i£rel b (G1:G2) A,C-diol-complex A,C-diol-complex G1:G2 Gl:pyridine equivalents per H6«guest 1 116«guest 2 ' A,C-diol 111 pyrazine: 1,4- 1.00 dioxane-ds (1:8) H p , H o u t , H i n , guest 1.32 Hout» H j n 6.06 450 l,4-dioxane-c?8: pyridine-d5 (8:200) pyridine-^ 3.00 Hin 1.00 Hin 75.0 75 1 a The IH N M R signals integrated for A,C-diol-complex 116*guest are labeled in Figure 3.38. Specific peak assignments are given in Table 3.19. b Errors are estimated to be ± 20%. Figure 3.38. A,C-diol-complex 116»guest and A,J3-diol-complex 115»guest 2© CH 3 CH 3 CH 3 CH 3 H . | I , H H j .c: CH 3 CH 3 CH 3 CH 3 I I , H H I I , H A,C-diol-complex 116»guest 2© 242 Table 3.19. H NMR Assignments for A,C-diol-complexes 116«guest at 253K in CDCI3. A,C-diol- pyrazine 1,4-dioxane-dg pyridine-^ Stock Solution complex 116»guest H p 7.32 (s, 4H) 7.12 (s, 4H) 7.28 (s, 4H) 7.17 (s, 4H) 6.58 (s, 4H). 6.42 (s, 4H). 6.62 (s, 4H). 6.43 (s, 4H). Haryl 6.41 (s, 4H) 6.42 (s, 4H) 5.97 (s, 4H) 6.49 (s, 4H) Hout 5.94 (d, 4H) 6.12 (br, 4H), 5.70 (br, 8H) 5.85 (d,/ = 5.77 (d, 4H) 5.84 (br, 4H) 6.8 Hz, 8H) Hin 4.19 (d, 4H) 4.33 (br, 4H) 4.09 (br, 4H) 4.42 (d, J = 3.19 (d, 4H) 4.15 (br, 4H) 3.76 (br, 4H) 6.8 Hz, 8H) guest signals 4.08 (s, 4H) Hout (d, J = 6.7-7.4 Hz) H i n (d, J = 6.8-7.4 Hz). The signals for the DBUH+ in the stock solution are: 3.36 (m, 8H) 2.70 (m, 2H) 1.86 (m, 2H) 1.68 (m, 4H). A,B-diol 110 (4.3 mg, 0.0069 mmoles) was weighed into a 2.00 mL volumetric flask. To this, 1.0 equivalents of DBU (1.0 uL, 0.0069 mmoles) were added and the flask was filled to volume with CDCI3. ! H NMR samples were prepared by adding 0.50 mL of this 3.4 mM stock solution to *H NMR tubes containing molecular sieves. The equivalents of guests added are indicated in Table 3.20; pyrazine was added as a stock solution. The *H NMR tubes were equilibrated for 20 minutes (out rate of pyrazine is too fast to determine by *H NMR spectroscopy at 253 K) in the spectrometer at 253 K prior to data acquisition. 243 Table 3.20. tfrei's for A,B-diol-complexes 115»guest (3.44 mM) at 253 K in CDC13. Guest l:Guest2 (G1:G2) equivalents per A,B-diol 110 i H NMR a integration A,B-diol-complex 115»guest 1 i H NMR a integration A,B-diol-complex 115»guest 2 G1:G2 ^relD Gl:pyridine -d5 pyrazine: 1,4-dioxane-ds (1:25) 1,4-dioxane-dg-.pyridine-ds (60:200) pyridine-^ 1.00 Haryl, H o u t , guest 3.33 H0ut 1.33 Hout 1.00 Hout* H j n 18.8 3.0 56 3.0 1.0 a The IH N M R signals integrated for A,B-diol-complex 115»guest are labeled in Figure 3.38 and assigned in Table 3.21. b Errors are estimated to be ± 20%. Table 3.21. ! H NMR Assignments for A,B-diol-complexes 115»guest at 253 K in CDCI3. A,B-diol pyrazine 1,4-dioxane-dg pyridine-Js Stock Solution Complex 115*guest Hp, Haryl 7.36 (s, 4H) overlap with 7.24 (s, 4H) 7.18 (s, 4H) 6.59 (s, 2H) free species 6.41 (s, 4H) 6.48 (s, 4H) 6.56 (s, 2H) 6.28 (s, 2H) overlap with 6.29 (s, 2H) 6.48 (s, 4H) 6.25 (s, 2H) free species 6.23 (s, 2H) H o u t 6.05 (d,2H) 6.17 (d, 2H) 6.00 (d, 2H) 5.86 (d, J = 6.9 Hz, 2H) 5.95 (d,2H) 6.10 (d,2H) 5.91 (d, 2H) 5.79 (d, J = 7.0 Hz, 4H) 5.79 (m,2H) 5.86 (m, 4H) 5.66 (m, 2H) 5.73 (d, J = 7.1 Hz, 2H) 5.63 (m, 2H) 5.60 (m, 2H) H i n 4.39 (m, 2H) 4.33 (m, 2H) 4.30 (m, 2H) 4.57 (d, J = 7.1 Hz, 2H) 4.26 (m,2H) 4.26 (br, 4H) 3.86 (br, 4H) 4.44 (d, J = 7.0 Hz, 4H) 3.68 (d, 4H) 4.21 (d, 2H) 3.74 (d, 2H) 4.33 (d, / = 6.9 Hz, 2H) 3.53 (d,2H) 4.11 (d,2H) guest signals 4.06 (s, 4H) H o u t (d, J = 6.8-7.4 Hz); H i n (d, / = 6.8-7.4 Hz). The signals for the DBUH+ in the stock solution are: 3.34 (m, 8H) 2.37 (m, 2H) 1.92 (m, 2H) 1.68 (m, 4H). 244 c. Determination ofKre[ 'sfor triol-complex 117a*guest and monobridged complex 118*guest Triol 112 (10.2 mg, 0.0159 mmoles) was weighed into a 5.00 mL volumetric flask. To this, 1.55 equivalents of DBU (3.7 J I L , 0.25 mmoles) were added and the flask was filled to volume with CDCI3. NMR samples were prepared by adding 500 |iL of this 3.18 mM stock solution to J H NMR tubes containing molecular sieves. The equivalents of guests added to the lH NMR samples are indicated in Table 3.22; pyrazine and 1,4-dioxane were added as a stock solution. The lH NMR tubes were equilibrated in the spectrometer for 20 minutes at 273 K prior to data acquisition (out rate of pyrazine is too fast to determine by lH NMR spectroscopy at 273 K). 245 Table 3.22. Integration results for ^ rei's of Triol-complexes 117a»guest at 273 K in CDCI3. Guest l:Guest2 iff NMR integration3 1H NMR integration3 ^ r e l Zi:reib (Gl:G2)equiv.per triol-complex triol-complex G1:G2 Gl:benzene triol 112 H7a»guest 1 117a»guest 2 pyrazine: 1,4- 1.00 1.00 8.00 2150 dioxane(0.5:4) Hp, H a r y i , H o u t , H o u t , Hj n H in 1,4-dioxane: 1.95 1.00 9.75 270 pyridine-^ (4:20) H o ut, Hi n Hp, H a r y i , H o ut. Hin pyridine-fife: 1.00 1.88 2.13 27 DMSO-fife (20:80) H i n H i n DMSO-J6: 1.55 1.00 7.75 13 benzene-fife H o u t H o u t (40:200) benzene-^: 1.00 1.00 1.67 1.7 acetone-fife Hi n Hj n (120:200) acetone-«fe 1 pyridine-^: 1.55 1.00 15.5 benzene-fife(20:200) H o u t , Hj n H o u t , Hj n 1,4-dioxane: 4.00 1.00 20.0 DMSO-fife(8:40) H o u t H o u t a The !H NMR signals integrated for triol-complex 117a»guest are indicated by H p , H^i, H0ut> Hi n and/or guest. The proton assignments for triol-complexes 117a»guest are tabulated in Table 3.23. b Errors are estimated to be + 20%. 246 Table 3.23. X H NMR Assignments for Triol-complexes 117a»guest at 273 K in CDCI3. triol-complex Hp, H a r y i , H o u t Hi n guest signals 117a*guest pyrazine 7.32 (s, 4H), 6.60 (s, 12H) 6.35 (s, 4H) 6.05 (d, 2H) 6.02 (d, 2H) 5.97 (d, 2H) 5.76 (d, 2H) 4.09 (d, 2H) 3.96 (d, 2H) 3.85 (d, 2H) 2H overlapping 4.18 (s, 4H) 1,4-dioxane overlapping 6.23 (d, 2H) 6.16 (d, 2H) 6.12 (d, 2H) 5.85 (d, 2H) 4.42 (d, 2H) 4.28 (d, 2H) 4.24 (d, 2H) 4.15 (d, 2H) 0.08 (br, 8H) DMSO-^6 overlapping peaks overlapping peaks 3.95 (br, 4H) 3.88 (br, 4H) pyridine-ds 6.63 (s, 4H) 6.58 (s, 4H) 6.56 (s, 4H) 6.54 (s, 4H) 4H overlapping 6.06 (d, 2H) 5.98 (d, 2H) 5.90 (d, 2H) 5.74 (d, 2H) 4.49 (d, 2H) 4.06 (d, 2H) 3.86 (d, 2H) 2H overlapping acetone-fife overlapping peaks overlapping peaks 4.01 (br, 8H) benzene-fife Stock Solution 7.15 (s, 4H), 6.48 (s, 4H), 6.44 (s, 12H) 6.15 (br, 4H) 5.93 (br, 4H) 4.44 (br, 8H) H o u t (d, J = 6.8-7.7 Hz); H i n (d, J = 6.8-7.7 Hz, 8H). The signals for the DBUH+ in the stock solution are: 3.44 (m, 2H), 3.26 (m, 6H), 2.67 (m, 2H), 1.85 (m, 2H), 1.70 (m, 2H), 1.56 (m, 2H). 247 triol-complex 117a«guest monobridged complex 117a»guest Monobridged 102 (10.2 mg, 0.00770 mmoles) was weighed into a 5.00 mL volumetric flask. To this, 3.1 equivalents of DBU (3.6 uL, 0.24 mmoles) were added and the flask was filled to volume with CDCI3. lH NMR samples were prepared by adding 500 uL of this 1.54 mM stock solution to ' H NMR tubes containing molecular sieves. The equivalents of guests added to each lH NMR sample are indicated in Table 3.24; pyrazine and 1,4-dioxane were added as a stock solution. The lH NMR tubes were equilibrated in an ice bath overnight prior to data acquisition at 273K (half-life for decomplexation of pyrazine (kd) from monobridged complex 118»pyrazine at 273 K is 130 minutes). 248 Table 3.24. Integration results for ATrei's of Monobridged complexes 118«guest at 273 K in CDCI3. Guest 1: Guest 2 (G1:G2) equiv. per monobridge 102 IH NMR integrations monobridged-complex 118»guest 1 i H NMR integration3 monobridged-complex 118»guest 2 Krel G1:G2 ^rel b Gl:benzene pyrazine: 1,4-dioxane(l:8) 1.00 HaryU Hb, H o u t 1.15 Haryl, Hb, H o u t 6.96 1300 1,4-dioxane: pyridine-^ (8/80) 1.00 Haryb Hb, H o u t 1.00 Haryl, Hb, H o u t 10.0 190 pyridine-^ :DMSO-d6 (40/160) 1.22 H0ut 1.00 H0ut 4.88 19 :DMSO-J6: benzene-6?6 (160:200) 1.00 Haryl, Hb 1.16 Haryb Hb 2.16 3.8 benzene-cfe: acetone-dg (240:400) acetone-d?6 1.06 Haryl, Hb, H j n 1.00 Haryb Hb, H j n 1.77 1.8 1 pyridine-ds: benzene-4(40:400) 1.00 Hin 1.05 H i n 9.52 a The I H NMR signals integrated for monobridged-complexes 118»guest are indicated by Haryl, Hb, H o u t , H ; n and/or guest. The proton assignments for monobridged-complexes 118»guest are tabulated in Table 3.25. b Errors are estimated to be ± 20%. 249 Table 3.25. *H NMR Assignments of Monobridged complexes 118«guest at 273 K in CDCI3. monobridged- H a ryi, Hb complex 118*guest H out H in guest signals pyrazine 1,4-dioxane DMSO-4 pyridine-ds acetone-^ 6 benzene-dg Stock Solution 6.97 (s, 2H) 6.66 (s, 2H) 6.60 (s, 2H) 6.57 (s, 2H) 6.40 (s, 2H) 6.81 (s, 2H) 6.58 (s, 2H) 6.47 (s, 2H) 6.43 (s, 2H) 6.34 (s, 2H) 6.78 (s, 2H) 6.52 (s, 2H) 6.44 (s, 2H) overlapping peaks 6.92 (s, 2H) 6.68 (s, 2H) 6.61 (s, 2H) 6.56 (s, 2H) 6.30 (s, 2H) 6.77 (s, 2H) 6.52 (s, 2H) 6.41 (s, 2H) overlapping peaks 6.98 (s, 2H) 6.77 (s, 2H) 6.71 (s, 2H) 6.61 (s, 2H) 6.32 (s, 2H) 6.80 (s, 2H) 6.59 (s, 2H) 6.44 (br, 4H) 6.38 (br, 2H) 6.08 (d, 2H) 6.04 (d, 2H) 6.00 (d, 4H) 6.18 (b, 8H) 6.10 (br, 8H), 6.07 (d, 2H) 6.05 (d, 2H) 5.97 (d, 2H) 5.95 (d, 2H) 6.14 (br, 8H) 6.14 (d, 2H) 6.11 (d, 2H) 6.00 (d, 4H) 6.17 (d, 4H), 6.12 (br, 4H) 4.15 (d, 6H) 3.82 (d, 2H) 4.40 (d, 2H) 4.32 (d, 2H) 4.22 (m, 4H) 4.08 (br, 2H) 4.14 (br, 2H) 4.32 (br, 2H) 4.51 (br, 2H) 4.56 (d, 2H) 4.22 (d, 2H) 3.92 (d, 2H) 2H overlapping 3.98 (d, 4H) broad overlapping peaks 4.60 (d, 2H) 4.48 (d, 2H) 3.93 (d, 2H) 3.63 (d, 2H) 4.52 (br, 2H) 4.47 (br, 4H) 4.38 (br, 2H) 4.19 (s, 4H) 0.09 (br, 8H) Hout (d, / = 6.8-7.7 Hz); H i n (d, J = 6.8-7.7 Hz). The signals for the DBUH+ in the stock solution are: 3.44 (m, 2H), 3.26 (m, 6H), 2.67 (m, 2H), 1.85 (m, 2H), 1.70 (m, 2H), 1.56 (m, 2H). 250 d. Determination ofKrei 'sfor A,B-bis-bridged complex 103»guest and tetraprotio 123a*guest A,B-bis-bridged 103 (10.2 mg, 0.00763 mmoles) was weighed into a 5.00 mL volumetric flask and the flask was filled to volume with CDCI3. *H NMR samples were prepared by adding 500 uL of this 1.53 mM stock solution to *H NMR tubes containing molecular sieves. The equivalents of guests added are indicated in Table 3.24; pyrazine and 1,4-dioxane were added as a stock solution. The *H NMR tubes were equilibrated for 48 hours at 298 K prior to data acquisition (half-life for decomplexation of pyrazine (kd) from A,B-bis-bridged complex 103»guest is 10 hours at 296 K). Table 3.26. Integration results for Orel's A,B-bis-bridged complex 103»guest at 298 K in CDCI3. Guest l:Guest2 1H NMR integration3 1H NMR integration3 ^ r e l ^ r e i b (Gl:G2)equiv. per A,B-bis-bridged A,B-bis-bridged G 1 : G 2 GLbenzene A,B-bis-bridged c o m p l e x complex 103«guest 1 103«guest 2 pyrazine: 1,4- 1.41 1.00 11.3 1800 dioxane(1.5:12) Hp, OH, H o u t Up, OH, H o u t 1,4-dioxane: 1.05 1.00 3.52 160 pyridine-^ (12:40) Wp, H i n Up, H i n pyridine- 2.36 1.00 17.7 45 d^/benzcne-d^: Hj n Hj n (40:300) benzene-^: 1.00 1.18 2.55 2.6 acetone-fife H o u t , Hj n H}n (200:600) acetone-i/6 1 a The lH NMR signals integrated for A,B-bis-bridged complex 103*guest are indicated by H p , Hb, H o u t , OH, H;n and/or guest. The proton assignments for A,B-bis-bridged complexes 103«guest are tabulated in Table 3.27. b Errors are estimated to be ± 20%. 251 Table 3.27. *H NMR Assignments for A,B-bis-bridged complexes 103»guest at 298 K in CDCI3. H„ 1% f W OH H A,B-bis-bridged complex 103»guest in pyrazine 6.98 (s, 4H) guest signal 6.91 (s, 4H) 4.25 (s, 4H) 1,4-dioxane 6.81 (s, 4H) guest signal 6.73 (s, 4H) 0.04 (s, 8H) pyridine-^ 6.97 (s, 4H) 6.89 (s, 4H) acetone-d6 benzene-^ Stock Solution 6.83 (s, 4H) 6.71 (s, 4H) 7.04 (s, 4H) 6.97 (s, 4H) 6.82 (s, 4H) 6.72 (s, 4H) 6.50 (d, 2H) 6.37 (d, 2H) 6.62 (d, 2H) 6.46 (d, 2H) 6.54 (d, 2H) 6.37 (d, 2H) 6.66 (d, 2H) 6.38 (br, 2H) 6.58 (d, 2H) 6.45 (d, 2H) 6.62 (d, 2H) 6.41 (br, 2H) 6.16 (d, 2H) 5.51 (s, 4H) 5.30 (d, 2H) 6.01 (d, 2H) 4.83 (d, 2H) 5.76 (d, 4H) 3.30 (d, 4H) 6.18 (d, 2H) 5.72 (s, 4H) 4.42 (d, 2H) 6.11 (d, 2H) 4.29 (m, 6H) 5.95 (d, 4H) 6.17 (d, 2H) 7.95 (br, 4H) 5.50 (d, 2H) 6.07 (d, 2H) 4.54 (d, 2H) 5.77 (d, 4H) 2.90 (d, 4H) 6.06 (d, 2H) 6.60 (s, 4H) 4.27 (d, 2H) 5.96 (d, 2H) 4.23 (d, 2H) 5.86 (d, 4H) 3.81 (d, 4H) 6.23 (d, 2H) 5.44 (s, 4H) 4.63 (d, 2H) 5.97 (d, 2H) 4.38 (d, 2H) 5.84 (d, 4H) 3.63 (d, 4H) 6.19 (d, 2H) 5.44 (s, 4H) 5.05 (br, 2H) 6.09 (d, 2H) 4.29 (m, 6H) 5.91 (d, 4H) H b (d, J = 6.1-6.7 Hz) H o u t (d, J = 6.8-7.4 Hz) H i n (d, J = 6.8-7.4 Hz, 8H). 252 tetraprotio 123a«guest A,B -bis-bridged complex 103«guest Tetraprotio 123a (10.5 mg, 0.00825 mmoles) was weighed into a 5.00 mL volumetric flask and the flask was filled to volume with CDCI3. lH NMR samples were prepared by adding 500 p:L of this 1.65 mM stock solution to NMR tubes containing molecular sieves. The equivalents of guests added are indicated in Table 3.28; pyrazine, 1,4-dioxane and pyridine were added as a stock solution. Equilibration of the KTe\ for tetraprotio 123a»pyrazine to tetraprotio 123a« 1,4-dioxane is established in less than one hour as determined by running the lH NMR spectrum at different time intervals. Nevertheless, lK NMR tubes were equilibrated for 6 hours at 298 K prior to data acquisition. 253 Table 3.28. Integration results for KK\s for tetraprotio 123a»guest at 298 K in CDCI3. Guest 1: Guest 2 (G1:G2) equiv. per tetraprotio 123a 1H N M R integration3 tetraprotio 123a«guest 1 1H N M R integration3 tetraprotio 123a»guest 2 KKl G1:G2 *rel b Gl:benzene pyrazine: 1,4- 1.04 1.00 16.7 1500 dioxane(l:16) Hp, OH, H o u t Hp, OH, H o u t 1,4-dioxane: 1.09 1.00 5.47 88 pyridine (8/40) Hp, H i n Hp, H j n pyridine:DMSO-c?6: 1.00 1.06 6.06 16 (14:90) Hg Hin DMSO- 1.00 1.16 2.39 2.6 dfi/benzene-cfe: Hin Hin (90:250) benzene-fi?6: 1.00 1.45 1.10 1.1. acetone-^ 6 Hout> H i n Hin (250:400) acetone-c?6 1 a The !H NMR signals integrated for tetraprotio 123a»guest are indicated by Hp, Hb, H o u t , OH, Hj n and/or guest. The proton assignments for tetraprotio 123a»guest are tabulated in Table 3.29. D Errors are estimated to be + 20%. 254 Table 3.29. ! H NMR Assignments of tetraprotio 123a»guest at 298 K in CDC13. tetraprotio Hp H b Haryi H o u t 123a *guest pyrazine guest signal 4.09 (s, 4H) 7.39 (s, 4H) 6.52 (d, 2H) 6.29 (s, 4H) 6.98 (s, 4H) 6.36 (d, 2H) 6.14 (d, 2H) 5.91 (d, 2H) 5.57 (d, 4H) H 5.33 (d, 2H) 4.92 (d, 2H) 2.85 (br, 4H) 1,4-dioxane guest signal -0.17 (s, 8H) 7.21 (s, 4H) 6.63 (d, 2H) 6.40 (s, 4H) 6.82 (s, 4H) 6.45 (d, 2H) 6.17 (d, 2H) 5.97 (d, 2H) 5.80 (d, 4H) 4.40 (d, 2H) 4.33 (m, 6H) pyridine guest signal 5.97 (m, IH) 4.23 (m, 2H) 2.85 (m, 2H) 7.33 (s, 4H) 6.54 (d, 2H) 6.28 (s, 4H) 6.98 (s, 4H) 6.36 (d, 2H) 6.16 (d, 2H) 5.92 (d, 2H) 5.52 (br, 4H) 5.26 (d, 2H) 4.52 (d, 2H) 3.02 (d, 4H) DMSO-^6 7.20 (s, 4H) 6.65 (d, 2H) 6.37 (s, 4H) 6.85 (s, 4H) 6.37 (b, 2H) 6.17 (d, 2H) 5.89 (d, 4H) 5.68 (d, 2H) 4.78 (br, 2H) 4.66 (d, 2H) 4.27 (br, 4H) acetone-4 7.25 (s, 4H) 6.67 (d, 2H) 6.37 (s, 4H) 6.90 (s, 4H) 6.37 (br, 2H) 6.27 (br, 2H) 5.90 (d, 4H) 5.79 (d, 2H) 3.87 (d, 4H) over lapping peaks benzene-4 7.08 (br, 4H) 6.53 (br, 2H) 6.43 (br, 6.92 (br, 4H) 6.32 (br, 2H) 4H) 6.18 (br, 2H) 5.82 (br, 2H) 5.60 (br, 4H) 4.66 (br, 2H) 4.44 (br, 2H) 3.51 (br, 4H) Stock Solution 7.21 (s, 4H) 6.87 (s, 4H) broad 6.40 (s, 4H) 5.81 (br, 4H) 6.12 (br, 2H) broad, 2H broad, 8H Stock (223 K) Solution13 7.15 (s, 4H) 6.61 (d, 2H) 6.36 (s, 4H) 6.80 (s,4H) 6.40 (d,2H) 6.13 (d, 2H) 5.97 (d, 4H) 5.79 (d, 4H) 4.54 (d, 2H) 4.22 (m, 6H) a H b (d, J = 6.1-6.7 Hz), H o u t (d, J = 6.1-7.6 Hz), H i n (d, J = 6.8-7.6 Hz, 8H). b The ! H NMR spectrum for tetraprotio 123a in CDC13 at 298 K is very broad (line above); cooling to 223 K resulted in a well resolved spectra. 255 vi. Complexation Experiments in Nitrobenzene-^ a General procedure for determination ofKre[ 'sfor complex 3b»guest in nitrobenzene-6.5: Tetrol l b (15.9 mg, 0.0218 mmoles) was weighed into a 5.00 mL volumetric flask. To this, 2.1 equivalents of DBU (7.6 (iL, 0.046 mmoles) were added and the flask was filled to volume with nitrobenzene- .^ 4A molecular sieves (ca. 0.3 mg) and 450 |iL of the above stock solution were added to each lH NMR tube (Table 3.30); all guests were added from stock solutions. lJi NMR samples were prepared such that complex 3b»guest 1: complex 3b«guest 2 were -1:1 and the total amount of free species was less than 1% of the total host (Table 3.15). The KTe\s were determined by integration of the unique signals of the host (e.g., Up, H o u t , H m ) and/or guest in each complex as shown in Table 3.30. For determination of the relative stability constants for NMP, CHCI3, DMA, and 1,3-dioxane, 4.1 equivalents of DBU were used. 256 Table 3.30. Integration results for Orel's f ° r Complexes 3b*guest at 298 K in nitrobenzene-0I5. Guest l:Guest2 1H NMR integrationa i H NMR integration3 K ^ KTelb (Gl:G2)equiv.per complex 3b.guest 1 c o m p l e x 3b»guest 2 G 1 : G 2 Glrbenzene tetrol lb pyrazine: methyl acetate (0.6:1.2) methyl acetate: 1.4-dioxane (0.6:1.2) l,4-dioxane:DMSO (0.6:1.2) DMSO:pyridine (0.6:2.4) pyridine: acetone (1.2:3) acetone :benzene (0.6:0.6) benzene: 1,3-dioxane (0.6:1.8) l,3-dioxane:DMA (0.6:2.4) DMA:CHC13 (0.6:1.8) CHC13:NMP (1.8:3.6) NMP 1.00 Hout, guest 1.00 H i n , guest 1.34 Hout, Hin, guest 1.21 Hout, Hjn, guest 1.76 Hout, Hjn, guest 1.65 Hp, Hout, guest 1.00 Hp, Hout, guest 1.89 t 1.00 Hout, H i n , guest 2.42 Hout, H j n 1.11 1.81 35000 H 0ut, guest 1.36 1.47 20000 Hjn, guest 1.00 2.68 13000 Hout, H i n , guest 1.00 4.84 H 0 ut, Hin, guest 1.00 4.40 Hout, Hin, guest 1.00 1.65 Hp, H o u t 1.09 2.75 Hp, H o u t , H j n , guest 1.00 7.56 guest 2.14 1.40 Hout, H i n , guest 1.00 4.84 Hout, Hin 5000 1000 230 140 51 6.8 4.8 DMSO:morpholine (0.6:0.6) 1.28 Hin 1.00 Hin 1.28 3900 a The ! H N M R signals integrated for complex 3b»guest are indicated by H p , H o u t , H i n and/or guest. The ! H N M R assignments for complexes 3b»guest are tabulated in Table 3.31. 257 Table 3.31. *H NMR assignments for complex 3b»guest in nitrobenzene-^ at 298 K. complex 3b»guesta H0ut Hin guest signals pyrazine 7.26 (s, 8H) 6.80 (d, 8H) 4.82 (d, 8H) 5.04 (s, 4H) methyl 7.06 (s, 4H) 6.94 (m, 4.87 (d, 8H) 0.04 (s, 3H), -1.50 (s, acetate 7.05 (s, 4H) 8H) 3H), 1,4- 7.10 (s, 8H) 6.97 (d, 8H) 5.08 (d, 8H) 0.79 (br, 8H) dioxane DMSOb 7.12 (s, 8H) 6.88 (d, 8H) 5.05 (d, 8H) -0.29 (s, 6H) pyridine 7.27 (s, 8H) 6.79 (d, 8H) 4.79 (d, 8H) 7.01 (m, IH), 5.04 (m, 2H), 3.72 (m, 2H), acetone 7.10 (s, 8H) 6.95 (d, 8H) 4.91 (d, 8H) -0.67 (s, 6H) benzene 7.26 (s, 8H) 6.74 (d, 8H) 4.81 (m, 8H) 4.75 (s, 6H) 1,3-dioxane 7.10 (s, 8H) 6.93 (d, 8H) 5.18 (d, 8H) -1.28 br, 2H (OCH 2CH 2CH 20) other signals overlapping DMA 0 7.10 (s, 8H) 6.85 (d, 8H) 5.15 (m, 8H) -0.56 (s, 3H), -1.51 (s, 3H), other C H 3 overlapping CHC13 C 7.08 (s, 8H) 6.95 (d, 8H) 5.23 (d, 8H) 5.41 (s, IH) NMPC 7.09 (s, 4H) 6.74 (d, 8H) 5.02 (br, 8H) -0.72 (br, IH), -0.89 (br, 7.06 (s, 4H) 4H), -1.22 (br, IH), -1.36 (br, IH), one C H 2 overlapping morpholine 7.10 (s, 8H) 6.91 (d, 8H) 5.16 (m, 8H) (br, 2H), 0.78 (br, 2H), 0.63 (br, 2H), 0.26 (br, IH), -0.11 (br, 2H), -0.20 (br, 2H) Stock 7.08 (br, 16H) Up and H o u t 4.66 (br, Solution 8H) a H o u t (d, J = 6.8-7.7 stock solution are at: 1.51 (m, 4H). b lH N M R spectrum c ! H N M R spectrum Hz, 8H), H i n (d, J = 6.8-7.7 Hz, 8H). The signals for the DBUH+ in the 3.47 (m, 2H), 3.23 (m, 4H), 2.72 (m, 2H), 1.78 (m, 2H), 1.67 (m, 2H), recorded at 333 K. recorded at 313 K. The relative stability constants for complex 3b«guest at 303 K, 303 K , 313 K, 323 K and 333 K in CDC13 were performed in the same manner as described above. Samples were equilibrated for 2 days in a constant temperature bath prior to data 258 acquisition. The moderate exchange rate of complex 3b» 1,3-dioxane, complex 3b»DMA, complex 3b«CHCl3 and complex 3b»NMP at temperatures above 313 K prevented the determination of their Kre\ above 313 K. Equilibration of complex 3b« 1,3-dioxane, complex 3b»DMA, complex 3b»CHCl3 and complex 3b»NMP was achieved within 20 minutes at 313 K; thus, these four samples were only equilibrated in the spectrometer prior to data acquisition at 313 K. Complex Sb^morpholine at 298 K has five signals for the guest due to slow inter-conversion of its chair conformation when encapsulated within this host (see Chapter 4). b. Determination of K.rei for tetra-OMe 125*guest in Nitrobenzene-ds at 333 K Tetra-OMe 125»guest (11.6 mg, 0.00832 mmoles) was weighed into a 5.00 mL volumetric flask and the flask was filled to volume with nitrobenzene-6?5. lH NMR samples were prepared by adding 450 oL of this 1.66 mM stock solution to lH NMR tubes containing molecular sieves. The equivalents of guests added are indicated in Table 3.28; all guests were added as a stock solution. Equilibration for tetra-OMe 125»pyrazine to tetra-OMe 125* 1,4-dioxane is established in less than one hour at 333 K as determined by rerunning the lH NMR spectrum after 24 hours. Therefore, lH NMR tubes were equilibrated for 1 hour in a constant temperature bath at 333 K prior to data acquisition. 259 Table 3.32. Integration results for ^rrei's for tetra-OMe 125»guest at 333 K in Nitrobenzene- .^ Guest 1 .Guest 2 I H NMR integration3 I H NMR integration3 ^ r e l ^ r e i b (G1:G2) equiv. per tetra-OMe tetra-OMe G 1 : G 2 Glrbenzene tetra-OMe 125 i25*guest 1 125«guest 2 pyrazine:methyl acetate (1.2:2.4) 1.00 H b , H o u t , H i n , guest 1.12 H p , H b , H i n , guest 1.79 1500 methyl acetate: pyridine (1.2:2.4) 1.00 H p , H o u t , guest 2.22 H p , H i n , guest 1.10 860 pyridine: 1,4-dioxane (1.2:1.2) 1.39 H b , H 0 u t , H j n , guest 1.00 H b , H 0 u t , H i n , guest 1.39 950 1,4-dioxane: DMSO (1.2:3.6) 1.39 H i n , guest 1.00 H i n , guest 4.16 680 DMSO:acetone (1.2:6) 1.12 H i n , guest 1.00 H i n , guest 5.60 160 acetone: benzene (1.2:1.2) 1.00 H b , H i n , guest 1.79 H b , H o u t , H j n , guest 1.79 29 benzene:1,3-dioxane (1.2:3) 1.00 H b , H 0 u t , H i n 1.10 H b , H i n , guest 2.28 52 1,3-dioxane: DMA (1.2:1.2) 1.00 guest 1.19 guest 1.19 23 D M A : C H C 1 3 (1.2:24) 1.00 H i n 1.02 H i n 19.6 29 C H C 1 3 : NMP (3:3) 1.39 H j n 1.00 H i n 1.39 1.4 NMP l.C C H 2 C I 2 : benzene (12:1) 1.28 H b , H o u t , H i n 1.00 H b , H i n , guest 9.39 5.6 260 Guest l:Guest2 (G1:G2) equiv. per tetra-OMe 125 1H NMR integration3 tetra-OMe 125»guest 1 1H NMR integration3 tetra-OMe 125»guest 2 G1:G2 * r e l b Gl:benzene CH2BrCl: benzene 1.42 1.00 2.12 25 (3:1) Hb, H o u t , H; n Hb, guest CH2Br2: benzene 1.37 1.00 1.37 72 (1.2:1.2) Hb. H o u t , Hi n H b , guest CH2I2: pyridine 1.00 1.00 1.00 950 (1.2:1.2) Hb. H o u t , Hin, Hb, H o ut, H^, guest morpholine: DMSO 1.00 1.39 2.08 340 (1.8:1.2) guest guest a The 1H N M R signals integrated for tetra-OMe 125»guest are indicated by Hp, Hb, H o u t , OMe, H i n and/or guest. The proton assignments for tetra-OMe 125»guest are tabulated in Table 3.33. b Errors are estimated to be ± 20%. tetra-OMe 125»guest 261 Table 3.33. ! H NMR Assignments for tetra-OMe 125»guest at 298 K in Nitrobenzene-J5. tetra-OMea 125«guest H b H out OMe H; pyrazine 7.53 guest signal 7.36 4.41 (s, 4H) pyridine guest 7.53 signal 7.34 6.36 (m, IH) 4.37 (m, 2H) 3.30 (m, 2H) methyl acetate 7.32 guest signal 7.17 -0.54 (s, 3H) -2.09 (s, 3H) 1,4-dioxane b guest signal 0.26 (s, 8H) DMSO b guest signal -0.82 (s, 8H) acetone guest signal -1.24 (s, 6H) benzene guest signal 4.25 (s, 6H) 1,3-dioxane guest signal 0.96 (s, 2H) -1.76 (br, 2H) DMA b guest signal 1.47 (s, 3H) -1.14 (s, 3H) -1.99 (s, 3H) 7.31 7.16 7.33 7.16 7.35 7.18 7.35 7.34 7.35 7.20 7.33 7.14 s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H s, 4H 6.63 6.38 6.66 6.40 6.74 6.43 6.71 6.47 6.70 6.47 6.74 6.44 6.65 6.44 6.78 6.44 6.70 6.48 d, 2H) 6.24 d, 2H) 6.06 5.76 d, 2H) 6.28 d, 2H) 6.14 5.69 d, 2H) 6.27 d, 2H) 6.10 6.01 d, 2H) 6.23 d, 2H) 6.13 5.94 d, 2H) 6.21 d, 2H) 6.10 5.94 d, 2H) 6.29 d, 2H) 6.09 6.02 d, 2H) 6.28 d, 2H) 6.14 5.69 d, 2H) 6.18 d, 2H) 6.12 5.95 d, 2H) d, 2H) 6.14 6.08 5.94 d, 2H) d, 2H) d, 4H) d, 2H) d, 2H) d, 4H) d, 2H) d, 4H) d, 2H) d, 2H) d, 4H) d, 2H) d, 2H) d, 4H) d, 2H) d, 2H) d, 4H) d, 2H) d, 2H) d, 2H) d, 4H) d, 2H) d, 4H) d, 2H) d, 2H) d, 4H) d, 2H) 3.91 (s, 12H) 5.58 (d, 2H) 5.06 (d, 2H) 2.75 (br, 4H) 3.93 (s, 12H) 3.96 (s, 12H) 3.98 (s, 12H) 3.98 (s, 12H) 3.97 (s, 12H) 3.93 (s, 12H) 3.98 (s, 12H) 5.84 (d, 2H) 4.82 (m, 2H) 2.80 (br, 4H) 4.73 (d, 2H) 4.37 (d, 2H) 3.98 (d, 4H) 4.57 (d, 2H) 4.51 (d, 2H) 4.39 (d, 4H) 4.3l(d, 2H) 4.2 l(d, 4H) overlap 4.98 (d, 2H) 4.46 (d, 2H) 3.85 (d, 4H) 4.96 (d, 2H 4.78 (d, 2H) 3.17 (d, 4H) 4.72 (d, 2H) 4.41 (d, 4H) 4.29 (d, 2H) 3.99 (s, 12H) 5.90 (d, 2H) 3.92 (d, 2H) 3.77 (d, 4H) 262 tetra-OMea HP H b H ° u t 0 M e H i n 125*guest CHCI3 b 7 , 2 8 ( s ' 4 H ) 6 > 7 2 ( d ' 2 H ) 6 - 2 4 ( d ' 2 H ) 3 - 9 7 ( s ' 1 2 H ) 4 ' 7 4 ( d ' 2 H ) guest signal 4.93 (s, 1H) overlaps methine 7.14 (s,4H) 6.44 (d,2H) 6.12 (d, 4H) 4.50 (d, 2H) 5.91 (d, 2H) 4.34 (d, 4H) NMP b euest 7.34 (s,4H) 6.72 (d, 2H) overlap free 4.00 (s, 6H) 4.50 (d, 2H) signal 7.18 (s,4H) 6.43 (d, 2H) species 3.98 (s, 3H) 3.74 (d, 4H) 1.2 (br, 2H) 3 - 9 3 (s' 3 H ) o v e r l a P -1.33 (m, 5H) -1.54 (t, 2H) CH2C12 guest 7.32 (s, 4H) 6.79 (d, 2H) signal 2.90 7.17 (s, 4H) 6.41 (br, 2H) (s, 2H) CH 2BrCl 7.31 (s,4H) 6.79 (d, 2H) guest signal 7.16 (s, 4H) 6.40 (br, 2H) 2.89 (s, 2H) CH 2 Br 2 guest 7.30 (s, 4H) 6.79 (br, 2H) signal 2.75 7.15 (s, 4H) 6.45 (br, 2H) (s, 2H) CH2I2 guest 7.27 (s, 4H) 6.81 (d, 2H) signal 1.76 7.09 (s, 4H) 6.44 (d, 2H) (s, 2H) morpholine b 7.30 (s, 4H) 6.71 (d, 2H) guest signal 7.16 (s, 4H) 6.49 (d, 2H) 0.18 (br, 4H) -0.09 (br, 4H) -0.58 (br, 4H) Stock 7.39 (s, 4H) 6.22 (d, 2H) 6.34 (d, 2H) 3.90 (s, 12H) 4.85 (m, 2H) Solution b 7.29 (s,4H) 6.12 (d, 2H) 6.01 (d, 4H) 4.50 (d, 2H) 5.87 (d, 4H) 3.58 (d, 4H) a !H NMR spectra for tetra-OMe 125#guest in nitrobenzene-Js at ambient temperature: H D (d, J = 5.0-6.7 Hz), H o u t (d, J = 6.8-7.6 Hz), H i n (d, J = 6.8-7.6 Hz). b ] H NMR spectra for tetra-OMe 125»guest in nitrobenzene-d5 at 333 K: H b (d, J = 5.0-6.7 Hz), H o u t (d, J = 6.8-7.6 Hz), H i n (d, J = 6.8-7.6 Hz). The *H NMR assignments for the guest signal of tetra-OMe 125»CHCl3 was based on a magnetization transfer experiment. The lH NMR assignments of tetra-OMe 125#DMA was based on a homonuclear decoupling experiment. 6.30 (br, 2H) 6.16 (d, 4H) 6.02 (d, 2H) 3.96 (s, 12H) 4.41 (br, 2H) 4.23 (d, 4H) 4.13 (d, 2H) 6.29 (br, 2H) 6.15 (d, 4H) 6.01 (d, 2H) 3.97 (s, 12H) 4.43 (br, 2H) 4.22 (m, 6H) 6.31 (br, 2H) 6.15 (d, 4H) 6.00 (d, 2H) 3.97 (s, 12H) 4.48 (br, 2H) 4.24 (m, 6H) 6.32 (d, 2H) 6.11 (d, 4H) 5.97 (d, 2H) 3.98 (s, 12H) 4.65 (d, 2H) 4.50 (d, 4H) 4.23 (d, 2H) 6.23 (d, 2H) 6.09 (d, 4H) 5.95 (d, 2H) 3.99 (s, 12H) 4.67(d, 2H) 4.60(d, 4H) 4.38(d, 2H) 263 vii. Determination of the relative stability constants for complexes'methyl acetate. Stock solutions of tetrol lb, triol 112, A,C-diol 111, A,B-diol 110, monobridged 102, and AB-bis-bridged 103 were prepared with 0.55 equivalents of DBU per phenolic hydroxyl in nitrobenzene-J5. Stock solutions of tetra-OMe 125 and AB-bis-bridged 103 were prepared without DBU. Each of two hosts were added as a stock solution to a lH NMR tube containing ca. 0.3 mg of crushed 4A molecular sieves (the total volume was ~ 450 uL). To this sample, methyl acetate was added as a stock solution (0.75 equivalents per yield limiting-complex). A new NMR cap was placed on the tube which was then wrapped with parafilm. The mole ratio of hosts and methyl acetate added as well as the integration results from the *H NMR spectra are indicated in Table 3.34. The ! H NMR samples were equilibrated for two days at each temperature with the exception of the Abel's for AB-bis-bridged charged complex 119»guest which required four days at 333 K to reach equilibrium. The time necessary to achieve equilibrium was determined by rerunning the *H NMR samples until they remained constant. 264 Table 3.34. Integration results for ATrei's for complex 1 "methyl acetate/ complex 2»methyl acetate in Nitrobenzene-^ 5 at 296 K. Host 1/Host 2 a [Host 1] Host 1 Host 2 MeOAc Integration Host /[Host mmoles mmoles |J,L added Host 1/ v c 2]mM (mL) (mL) [126 mM] Host 2 r d A,B-bis-bridged charged 1.91/ 0.732 0.487 1.45 3.31 0.91 complex 119b/ tetrol lb 4.87 (0.383) (0.100) tetrol Vol triol 112 3.66/ 1.06 1.06 3.16 2.61 2.61 5.31 (0.290) (0.200) tetrol lb/ monobridged 4.87/ 0.487 0.731 1.45 1.08 3.25 102 1.96 (0.100) (0.373) tetra-OMe 125/ A,B-bis- 1.97/ 0.690 0.460 2.74 1 0.67 bridged 103 2.13 (0.350) (0.216) tetra-OMe 125/ 1.83/ 0.490 0.490 2.92 0.95 0.95 monobridged complex 1.96 (0.268) (0.250) 118 monobridged 118/ triol 1.96/ 0.663 1.33 3.95 0.62 0.62 112 5.31 (0.338) (0.250) triol 112/ A,C-diol 111 5.31/ 0.467 1.40 1.39 6.7 6.7 4.00 (0.088) (0.350) A,B-diol 110/ A,C-diol 4.00/ 1.00 1.00 2.98 1.0 1.0 111 4.00 (0.250) (0.250) a Integration results are from integration of the lH NMR signals for encapsulated methyl acetate in the separate complexes. J H NMR spectral assignments for methyl acetate are reported in Table 3.35. b ! H NMR spectrum recorded at 333 K in nitrobenzene-ds. c Errors are estimated to be 30%. 265 Table 3.35 lH NMR Chemical Shifts for Methyl Acetate Encapsulated in Various Complexes in Nitrobenzene-J5 at Ambient Temperature. CH3CO2CH3 CH3CO2CH3 CH3CO2CH3 CH3CO2CH3 8, ppm 8, ppm -A8a -A8 tetra-OMe 125«methyl -0.54 -2.09 4.13 4~06 acetate A,B-bis-bridged -0.47 -2.05 4.06 4.01 103»methyl acetate A,B-bis-bridged charged -0.36 -1.88 3.95 3.85 complex 119#methyl acetate A,C-diol-complex -0.29 -1.84 3.88 3.80 116»methyl acetate A,B-diol-complex -0.25 -1.78 3.84 3.75 115»methyl acetate monobridged complex -0.14 -1.68 3.73 3.65 118»methyl acetate triol-complex -0.09 -1.62 3.68 3.58 117»methyl acetate complex 3b»methyl 0.07 -1.48 3.52 3.44 acetate free methyl acetate 3.59 1.97 -AS = 8 free methyl acetate - 8 encapsulated methyl acetate. 266 E . References 1. Chapman, R. G.; Chopra, N.; Cochien, E. D.; Sherman, J. C. J. Am. Chern. Soc. 1994,116, 369-370. 2. In the forming carceplex 2»guest, the formation of the first inter-bowl bridge cannot form "incorrectly". The formation of the second inter-bowl bridged can form incorrectly two ways, both producing a Z shaped product. The second bridge can form correctly three ways producing either AB-bis-bridged 103 (two ways) or AC-bis-bridged 104 (one way), both of which only can form carceplex 2»guest. Therefore, the statistical yield for formation of carceplex 2»guest is 3/5 = 60%.). 3. Chopra, N.; Sherman, J. C. Supramol. Chern. 1995, 5, 31-37. 4. Kolthoff, I. M.; Chantooni, M. K.; Bhownmik, S.O.J. Am. Chern. Soc. 1968, 90, 23-28. 5. Bordwell, F., G.; Branca, J. C ; Hughes, D. L.; Olmstead, W. N. / . Org. Chern. 1980,45,3305-3313. 6. Branda, N.; Wyler, R.; Rebek, J. J. Science 1994, 263, 1267-1268. 7. Fraser, J. R.; Borecka, B.; Trotter, J.; Sherman, J. C. / . Org. Chern. 1995, 60, 1207-1213. x 8. Derome, A. E. Modern NMR Techniques for Chemistry Research; Pergamon: New York, 1987; Vol. 6. For P1331 solvent suppression experiment see pp 172-176. For magnetization transfer see pp 188-190, 215. A pre-saturation experiment is inappropriate for suppressing resonances that are involved in chemical exchange. For example in the commonly used watemul experiment (waternul.au on the WH-400), the resonance being suppressed is irradiated to eliminate the population difference across its transition (equal populations of a and 8 states) immediately prior to acquisition of the spectrum. This type of experiment is inappropriate for suppressing resonances that are involved in chemical exchange, because the chemical species whose resonance is being suppressed can transfer from one environment to the other while the population of its a and 6 states are still equilibrated resulting in suppression of both resonances. Of course, the rate of exchange and rate of relaxation of the nucleus back to equilibrium will determine if this does occur. For example in our case, the use of a watemul could produce an incorrect negative result for encapsulation of CHCI3. Thus the solvent suppression experiment known as a P1331 was used which does not use decoupling to reduce the intensity of the signal one wishes to suppress. A P1331 solvent suppression experiment is performed by placing the transmitter frequency exactly to the frequency of the peak being suppressed. The with of the pulse used is divided into eight portions, i.e., two l/8th portions and two 3/8th portions. The sequence: (K/2X1/8) x — T — (7t/2x3/8)-x — T — (7t/2x3/8)x — T — (7t/2xl/8)-x acquire. The magnetization of the peak being suppressed is returned to the z-axis resulting in no signal detection for the resonance being suppressed. Other resonances that precess at frequencies that differ from that of the peak being suppressed have some of their magnetization 267 remaining in the xy plane and thus result in a signal. The delay is chosen in order to maximize the intensity of the region of interest. 9. Hibbert, F.; Emsley, J. Adv. Phys. Org. Chem. 1990, 26, 255-379. 10. Emsley, J. Chem. Soc. Rev. 1980, 9, 91-124. 11. Cleland, W. W. Biochemistry 1992, 31, 317-319. 12. Smirnov, S. N.; Golubev, N. S.; Denisov, G. S.; Benedict, H.; Schah-Mohammedi, P.; Limbach, H.-H. J. Am. Chem. Soc. 1996,118, 4094-4101. 13. Wyler, R.; de Mendoza, J.; Rebek, J. Jr. Angew. Chem. Int. Ed. Engl. 1993, 32, 1699-1701. 14. Cheng, X.; Gao, Q.; Smith, R. D.; Simanek, E. E.; Mammem, M.; Whitesides, G. M. Rapid Commun. Mass Spectrom. 1995, 9, 312-316. 15. Russel, K. C ; Leize, E.; Dorsselaer, A. V.; Lehn, J.-M. Angew. Chem. Int. Ed. Engl. 1995, 34, 209-213. 16. Saf, R.; Mirtl, C ; Hummel, K. Tetrahedron Lett. 1994, 35, 6653-6656. 17. Mann, M.; Wilm, M. Trends Biochem. Sci. 1995, 20, 219-224. 18. We warmly thank Don Douglas and Bruce Collings for performing the electrospray mass spectrometry. 19. Resolution enhancement of these signals indicated they were coupled, J = 1 Hz. Enhancement was done by performing a gausine multiplication (GB = 0.5) of the fid and setting line broadening to -1. 20. The activation barrier for rotation of pyrazine about the pseudo-C"2 axes of asymmetric complex 3c«pyrazine was calculated to be 18.3 kcal/mol based on a coalescence temperature (Tc) of 353 K and separation of the signals (A5HZ) of 14.3 Hz using the following equation: AGC+ = RTJ22.96 + ln(Tc/A5Hz)] where AGC+ is the activation barrier in kcal/mol; T c is the temperature of coalescence, and A8HZ is the separation of the signals in Hz. See: Abraham, R. J.; Fisher, J.; Loftus, P. Introduction to NMR Spectroscopy; Wiley: New York, 1990; pp 194-7. 21. We warmly thank Gunnar Olovsson and James Trotter of the University of British Columbia X-ray crystallography laboratory for solving this crystal structure. 22. Novak, A. Struct. Bonding 1974,18, 177-216. 23. Spellmeyer, D. C ; Grootenhuis, P. D. J.; Miller, M. D.; Kuyper, L. F.; Kollman, P. A. J. Phys. Chem. 1990, 94, 4483-4491. 24. Nishio, M.; Umezawa, Y.; Hirota, M.; Takeuchi, Y. Tetrahedron 1995, 51, 8665-8701. 25. Reichardt, C. Solvents and Solvent Effects in Organic Chemistry; 2nd ed.; VCH: New York, 1988. 268 26. Gibb, B. C.; Chapman, R. G.; Sherman, J. C. J. Org. Chern. 1996, 61, 1505-1509. 27. The activation barrier for rotation of pyrazine about the pseudo-C2 axes of asymmetric triol-complex 117b»guest was calculated to be 17.4 kcal/mol based on a coalescence temperature (Tc) of 348 K and separation of the signals (ASH z) of 41 Hz. 28. Cram, D. J.; Tanner, M. E.; Knobler, C. B. J. Am. Chern. Soc. 1991,113, 7717-7727. 29. Timmerman, P.; van Mook, M. G. A.; Verboom, W.; van Hummel, G. J.; Harkema, S.; Reinhoudt, D. N. Tetrahedron Lett. 1992, 33, 3377-3380. 30. Timmerman, P.; Boerrigter, H.; Verboom, W.; van Hummel, G. J.; Harkema, S.; Reinhoudt, D. N. / . Inclusion Phenom. Mol. Recognit. Chern. 1994,19, 167-191. 31. Cram, D. J.; Karbach, S.; Kim, Y. H.; Baczynskyj, L.; Marti, K.; Sampson, R. M.; Kalleymeyn, G. W. J. Am. Chern. Soc. 1988,110, 2554-2560. 32. Robbins, T. A.; Cram, D. J. J. Chern. Soc, Chern. Common. 1995, 1515-1516. 33. The calculated ^ fs based on spectrum C and three other spectra was 410 ± 40 M" 1 as calculated by the following equation: Ks = [tetra-OMe 125«NMP]/([tetra-OMe] [NMP]free) 34. Nakamura, K.; Houk, K. N. J. Am. Chern. Soc. 1995, 117, 1853-1854. 35. Houk, K. N.; Nakamura, K.; Sheu, C ; Keating, A. E. Science 1996, 273, 627-629. 36. Sheu, C ; Houk, K. N. J. Am. Chern. Soc. 1996,118, 8056-8070. 37. Still, W. C. Macromodel; Version 4.5 ed.; Still, W. C , Ed.; Department of Chemistry, Columbia University: New York, NY, 1994. 38. Nakamura, K.; Sheu, C ; Constable, A. E.; Houk, K. N.; Chapman, R. G.; Sherman, J. C. 7. Am. Chern. Soc. 1997, (in press). 39. Jorgensen, W. L. BOSS; Version 3.5 ed.; Jorgensen, W. L., Ed.; Department of Chemistry, Yale University: New Haven, Connecticut, 1995. 40. Jorgensen, W. L.; Tirado-Rives, J. J. Am. Chern. Soc 1988,110, 1657-1666. 41. Corporation, M. S. teXsan. Single Crystal Structure Analysis Software; Version 1.7. MSC ed.; Corporation, M. S., Ed.: 3200 Research Forest Drive, The Woodlands, TX 77381, USA., 1995. 42. Sheldrick, G. M. SHELXL-93. Program for the Refinement of Crystal Structures.; Sheldrick, G. M., Ed.: University of Gottingen, Germany, 1993. 269 4. Mobility of Host and Guest in Carceplex 2*Guest and Related Compounds A. Orientation of Guest Molecules in Carceplex 2»Guests and Related Compounds i. Orientation of Guest Molecules in Carceplex 2»Guests Guest molecules encapsulated in the interior of two-bowl host compounds such as carceplex 2*guest or complex 3»guest generally manifest upfield shifts (A8) in their *H NMR spectra of 2-4.5 ppm (A8 is the 8 of free guest minus the 8 of the encapsulated guest). Here we present a method for the prediction of the orientation of guest molecules with respect to the interior of various two-bowl host compounds based on the AS values of their encapsulated guest. This method draws from the orientation of pyrazine in carceplex 2b«pyrazine1 and DMA in carceplex 2a»DMA2 as determined from their X-ray crystal structures (Figure 4.1, Figure 2.4, Figure 2.6 and Figure 3.11). These orientations are compared to the AS values for the incarcerated guest molecules (Table 4.1). The six protons of incarcerated DMA (CH3 (b) and CH3 (c)) and all the protons of incarcerated pyrazine are positioned deep in the interior of the bowls as determined by their crystal structures (Figure 4.1) and all have AS values greater than 4.4 ppm (Table 4.1). In contrast to this, the CH3 (a) group of DMA that is located at the equator of carceplex 2a»DMA has a AS of 1.98 ppm (Table 4.1). Thus, there are substantial differences for AS values in the *H NMR spectra for signals of incarcerated guest molecules whose protons are positioned deep within the interior of the bowls versus those protons positioned near the equator of the host. Thus, it is reasonable to use AS values to predict the orientation of other guest molecules. 270 By analogy with DMA and the analysis of the A8's from Table 4.1, one can predict that both butanone and ethylmethyl sulfide are oriented within carceplex 2a»guest with then-methyls sticking into the interior of the bowls, while their midsections are along the equator of the host. In addition, ethylmethyl sulfide and butanone rotate slowly about the C 2 axes of the host on the lH NMR timescale,3 as the top and bottom halves of the host exhibits nonequivalent sets of protons (see section Bi and Chapter 3, section Bx). Figure 4.1. Schematic Representation of the Orientation of DMA in Carceplex 2a»DMA and Pyrazine in Carceplex 2b#Pyrazine as Determined by X-ray Crystallography. C H 3 C H 3 C H 3 C H 3 R R ' 3 C H 3 C H 3 U N 3 R = CH 2 CH 2 Ph 271 Table 4.1. 8 (ppm) of Guest Molecules in Carceplexes 2a»Guest at Ambient Temperature in CDCI3. Carceplex 2a*guest 5 bound (ppm) 5 free (ppm)' A5 ppm pyrazine pyrazine11 DMA O A. H 3 C " ^ N ' C C H 3 'CH3 H a H b H c 4.07 3.99 1.04 -1.46 -2.40 8.58 8.58 3.02 2.94 2.09 4.51 4.59 1.98 4.40 4.49 2-butanone O H 3 b C H 3 c H a H b H r -0.05 -2.36 -3.43 2.41 2.10 1.01 2.46 4.46 4.44 ethylmethylsulfide b 3 pyridine H. H H a H b H c H a H b H c -0.10 -2.29 -3.23 6.34 4.02 2.73 2.46 2.05 1.21 7.68 8.62 7.30 2.56 4.34 4.44 1.34 4.60 4.57 pyridazine H b H a H b 4.64 4.39 9.22 7.49 4.58 3.10 NMP c N- - C H 3 H a H b H c Hd 1.46 -1.79 -1.79 -2.13 3.25 2.70 2.22 1.89 1.79 4.49 4.01 4.02 I 8 of free guest determined in CDCI3 at ambient temperature. I I carceplex 2b»pyrazine. 272 For carceplex 2a,pyridine, the large AS values for Hb and H c of pyridine suggest that they are positioned deep in the bowls while the small A5 value for H a suggests it is positioned on or near the equator of the host (Table 4.1). Pyridazine may be oriented in carceplex 2a»pyridazine such that its N-N bond is bisected by the equator of the host and its protons extend into the bowls. Examination of the CPK model of this orientation indicates that H a extends deeper into the bowls than Hb, thus accounting for the larger A8 value of H a (Table 4.1). The AS values for the protons for incarcerated NMP in carceplex 2a»NMP suggest that Hb, H c and H<j are positioned deep in the interior of the bowls while H a is positioned near the equator of the host (Table 4.1). Thus, NMP and DMA have similar orientations in the carceplex (both have their carbonyl oxygen positioned at the equator of the host) which is not surprising considering the similarities in the shapes of these two molecules. Furthermore, Cram et al. have recently determined by X-ray crystallography that NMP is indeed orientated in the manner described above in a hemicarceplex that differs from carceplex 2a»NMP by the substitution of its four OCH2O inter-bowl bridges with four 0(CH2)30 bridges.4 In summary, the orientation of guest molecules incarcerated in carceplex 2a»guest can be predicted based on the A8 values of signals of the free and incarcerated guest molecule. The A8 values for the guest signals of the forty carceplexes 2a»guest described in Chapter 2 are tabulated in Table 4.9. ii. Sensitivity of the *H NMR 8 of the Host Signals of Carceplex 2a#Guests to the Nature of the Incarcerated Guest Molecule The host signals in the *H NMR spectra of carceplex 2a»guest were also observed to be sensitive to the nature of the incarcerated guest molecules. For example, the 8 of Hj n (Figure 4.2) in the 33 carceplexes 2a»guest with non-aromatic guest molecules ranged from 4.26 - 4.86 ppm while the 8 of H m in the seven carceplexes 2a»guest with aromatic guests molecules ranged from 4.10 - 4.26 ppm (Table 4.8). The chemical shifts of three other 273 host protons of carceplex 2a»guest exhibited a similar trend as summarized in Table 4.2. The chemical shifts of the pendent group protons (CH2CH2Ph) or the methine protons of carceplex 2a»guest were not sensitive (AS < 0.06 ppm) to the nature of the guest molecule incarcerated. Cram et al. has recently reported similar effects of the guest on the host signals in the *H NMR spectra of various hemicarceplexes.5 The sensitivity of the host protons of carceplex 2a»guest to the nature of the guest molecule was instrumental in determining accurate product ratios from our competition experiments since it provided unique host signals allowing for accurate integration of the carceplex mixture (see Chapter 2, section Bib). Table 4.2. *H NMR A8 (ppm) of Host Signals for Carceplexes 2a«guest in CDCI3 at Ambient Temperature. Carceplexes Hp Hb H o u t H i n 2a»Guest (ppm) (ppm) (ppm) (ppm) All Guests Rangea A8 6.93 - 6.63 0.30 6.61 - 6.47 0.14 6.24 - 6.02 0.22 4.86 - 4.10 0.76 All Aromatic Guests Range A8 6.93 - 6.84 0.09 6.52 - 6.47 0.05 6.12 - 6.02 0.10 4.26 -4.10 0.76 All non-Aromatic Guests Rangea AS 6.78 - 6.63 0.15 6.61 - 6.54 0.07 6.24 - 6.08 0.16 4.86 - 4.26 0.60 a Excludes the ! H N M R signals for carceplex 2a»NMP which was anomalous: Hp (6.74, 6.71), H b (6.32), H o u t (6.03), and H i n (4.68). 274 Figure 4.2. Carceplex 2»Guest Carceplex 2a»guest R = CH^CH^Ph Carceplex 2b*guest R = CH3 iii. Sensitivity of the H NMR 5 of Host and Guest Signals of Carceplex 2«-Guest to a Change of Pendent Group Changing the pendent group of carceplex 2,guest resulted in modest A5 values of both host and guest signals as shown in Table 4.3. The magnitude of the A5 values for a particular host signal were very similar among the series of carceplexes investigated (Table 4.3). For example, the A5 value for Hj n was 0.02 ppm for each pair of the six carceplexes 2"guest studied. What is interesting about this comparison is that the 5 of the incarcerated guest molecule also changed when the pendent group of carceplex 2»guest changed; this feature was instrumental for the determination the orientation of pyrazine in asymmetric carceplex 2d>pyrazine1 and subsequently its orientation in asymmetric complex 3c»pyrazine (see Chapter 3, section Bx). 275 Table 4.3. Comparison of the 8 (ppm) in the H NMR Spectra of Carceplexes 2a»guest versus Carceplexes 2b»guest in CDCI3 at ambient temperate. Description H b H 0 ut Hm Hin Guest Signals 2a«pyrazine 6.93 6.47 6.03 4.90 4.26 4.07 2b»pyrazine 7.00 6.44 6.01 5.01 4.24 3.99 A8 (2a -2b) -0.07 0.03 0.02 -0.11 0.02 0.08 2a«1,4- 6.72 6.57 6.19 4.87 4.46 -0.28 dioxane 2b«l,4- 6.79 6.53 6.17 4.98 4.44 -0.34 dioxane AS (2a -2b) -0.07 0.04 0.02 -0.11 0.02 0.06 2a»DMSO 6.76 6.57 6.16 4.88 4.49 -1.24 2b«DMSO 6.83 6.53 6.13 4.98 4.47 -1.30 AS (2a -2b) -0.07 0.04 0.03 -0.10 0.02 0.06 2a»pyridine 6.91 6.49 6.03 4.89 4.14 6.34 4.02 2.73 2b»pyridine 6.98 6.45 6.01 4.99 4.12 6.30 3.95 2.66 A8 (2a -2b) -0.07 0.04 0.02 -0.10 0.02 0.04 0.07 0.07 2a#acetone 6.74 6.57 6.19 4.87 4.36 -1.63 2b«acetone 6.81 6.53 6.17 4.98 4.34 -1.67 A8 (2a -2b) -0.07 0.04 0.02 -0.11 0.02 0.04 2a»benzene 6.91 6.50 6.02 4.88 4.15 3.88 2b»benzene 6.98 6.47 6.00 4.99 4.13 3.82 AS (2a -2b) -0.07 0.03 0.02 -0.11 0.02 0.06 Host protons are labeled in Figure 4.2. AS is the lH NMR 8 of carceplex 2a»guest that have phenethyl pendent groups minus ! H NMR 8 of carceplex 2b»guest that have methyl pendent groups. 276 iv. Orientation of Guest Molecules in Other Two-Bowl Complexes#Guest Guest molecules encapsulated in other two-bowl complexes are predicted to have the same orientation in the interior of their respective complexes as that predicted for carceplex 2»guest due to the similarity of the A 5 values. For a comparison of the 8 of a guest molecule in a series of two-bowl complexes see Table 3.35. The 8 of the host for A,B-bis-bridged complexes»guest did provide further insight to the favored orientations of aromatic guest molecules with respect to the pseudo-C4 axis of these hosts. For example, tetraprotio 123a»guest ( C 2 V ) has three nonequivalent H i n and Hout protons which give rise to three doublets in a 2:2:4 ratio, whereas all the H ; n and H o u t protons of carceplex 2a»guest (D^) are equivalent; these protons are labeled in a schematic representation of the bottom bowl of these hosts in Figure 4.3. The 8 of the H i n and H o u t protons of four representative complexes of tetraprotio 123a»guest are listed in Table 4.4; a complete list of the 8 values for all the complexes of tetraprotio 123a*guest studied are given in Table 3.29. Dramatic differences are observed for the AS of H j n of tetraprotio 123a*guest when aromatic guests are encapsulated versus non-aromatic guests. Remarkably, the AS of H j n for tetraprotio 123a*pyrazine is 2.48 ppm ( A 8 = (Hj n( a) or Hin(b)) - Hi n(C))! Similarly, the A 8 values for tetraprotio 123a«pyridine and tetraprotio 123a»benzene are 2.24 ppm and 1.15 ppm, respectively. In contrast to this, the AS value of tetraprotio 123a»dioxane is merely 0.07 ppm. Other non-aromatic guest molecules encapsulated in tetraprotio 123a»guest also had small AS values relative to those observed for aromatic guest molecules. We suggest that the differences observed in A 8 of H j n of tetraprotio 123a«guest for aromatic guests versus non-aromatic guests is due to the preferred orientations of aromatic guests in cavity of the host as discussed below. 277 Figure 4.3. Schematic Representation of the Lower-Bowl of tetraprotio 123a»guest and carceplex 2»guest. Schematic Representation of the Lower-Bowl of tetraprotio 123a»guest Schematic Representation of the Lower-Bowl of carceplex 2»guest H out(b) out(a) out(c) H, out out out R = Upper Bowl of tetraprotio 123a, see R = Upper Bowl of Carceplex 2»guest; see Figure 4.6 for entire structure. Figure 4.2 for entire structure. Table 4.4. 'H NMR Assignments for tetraprotio 123a»guest at 298 K in CDCI3. Guest Proton 8 Proton 8 pyrazine H0ut( a) or Hout(b) 6.14 Hin(a) or Hin(b) 5.33 H0ut( a) or Ho ut(b) 5.91 Hin(a) or Hi n ( b) 4.92 Hout(c) 5.57 Hin(c) 2.85 pyridine H0ut( a) or Hout(b) 6.16 Hin(a) or Hin(b) 5.26 H0ut( a) or Hout(b) 5.92 Hin(a) or Hin(b) 4.52 H0ut(c) 5.52 Hin(c) 3.02 benzene H0ut( a) or Hout(b) 6.27 Hin(a) or Hin(b) 4.66 H0ut( a) or Hout(b) 5.90 Hin(a) or Hin(b) 4.44 Hout(c) 5.79 Hin(c) 3.51 1,4-dioxane H0ut( a) or Hout(b) 6.17 Hin(a) or Hin(b) 4.40 H0ut( a) or H o u t(b) 5.97 Hin(a) or Hin(b) 4.33 Hout(c) 5.80 Hin(c) 4.33 278 The large A8 values of Hi n of tetraprotio ^ Sa^aromatic guest) is consistent with the aromatic guest being oriented with its aromatic ring in the plane that bisects each bowl and contains the Hin(a)'s and H in(b)'s. For example, pyrazine would preferentially sit with one of its nitrogens directed between the protio groups and the other directed between the OCH2O inter-bowl bridges of tetraprotio 123a»pyrazine as illustrated in Figure 4.4. The aromatic ring current of the guest in this orientation would therefore shield Hjn(c) and deshield both Hin(a) and Hjn(b), thereby accounting for the disparity in the 5 of these protons. The observation of a single peak for pyrazine encapsulated in tetraprotio 123a»pyrazine suggests that the rotation of pyrazine about the pseudo-C4 axis of the host is rapid on the lH NMR timescale.3 Similar orientations are also predicted for pyridine and benzene based on their A8 values. Non-aromatic guest molecules such as dioxane, however, are unable to induce large shifts in the 8 of these protons because they lack an aromatic ring current. Therefore, the orientation of dioxane in tetraprotio 123a»dioxane cannot be predicted based on its *H NMR spectra. These same trends for the H o u t protons of the tetraprotio 123a were also observed but the magnitude of the AS values are much smaller than Hjn, presumably because this proton is more distant from the guest (Table 4.4). Similarly, aromatic guest molecules encapsulated in other A,B-bis-bridged hosts (tetra-OMe 125«guest and A,B-bis-bridged 103«guest) have similar orientations based on their similar AS values to those of tetraprotio 123a»guest. lH NMR spectroscopy of carceplex 2a«guest and A,B-bis-bridged complexes»guest have provided a wealth of information about the orientation of the guest molecule in the interior of these hosts. Variable temperature *H NMR spectroscopy was employed to further probe the guest orientation and to investigate the mobility of these guest molecules in these host guest systems. 279 Figure 4.4. Schematic Representation of the Lower-Bowl of tetraprotio 123a»guest Illustrating the Favored Orientation of Pyrazine. Favored Orientation of Pyrazine Disfavored Orientation of Pyrazine B. Variable Temperature 1 H NMR Spectroscopy of Two-Bowl Complexes»Guest i. Introduction Cram et al. explored the variable temperature *H NMR behavior of carceplexes 2a«DMA, 2a«DMF, and 2a»DMSO as discussed in Chapter 1, section Eiv. They found that DMA did not rotate about the C 2 axes of the host on the lU NMR timescale even at temperatures up to 448 K. The % NMR spectra of carceplex 2a»DMF did not exhibit any asymmetry even at temperatures down to 235 K suggesting that DMF rotates rapidly about the C 2 axes of the host.2 In contrast, the lH NMR spectra for carceplex 2a»DMSO changed over the temperature range of the experiments: at 255 K the *H NMR spectrum had two signals for the DMSO guest as well as two signals for the Hi n proton of the host; at 298 K each of these protons manifested one signal. Cram et al. suggested "that the rotation of DMSO within the shell is constrained enough on the lH NMR timescale to relegate one methyl group and the sulfur oxygen to one hemisphere and the other methyl 280 group to the other hemisphere."2 Although this process was evidently not well understood, the energy barrier for the process was calculated as 12.7 kcal/mol and 13.6 kcal/mol from the coalescence temperatures of the guest signal and the Hj n signal, respectively. They subsequently explored the "energy barrier for rotation of DMSO about the Ci axes" for a number of different hosts as discussed in section Biv. The orientation of DMSO in asymmetric carceplex 2d»DMSO is also discussed in section Biv. Our group has also determined that the energy barrier for rotation of pyrazine in asymmetric carceplex 2d«pyrazine,1 asymmetric complex 3c»pyrazine, and asymmetric triol-complex 117d»pyrazine were 19, 18, and 17 kcal/mol, respectively (see Chapter 3). Thus, there was substantial information available concerning the rotation of guest molecules about the C 2 axes of carceplex 2#guest. Cram et al. also explored the variable temperature *H NMR of triol 112 based hemicarceplexes 107»DMA, 107»DMF, and 107-DMSO (Figure 4.9 on page 291, Scheme 2.4). Upon cooling to -53 °C, the lH NMR spectra of all three hosts increased in complexity. They suggested that this loss of symmetry was due to the reduced motion of the guest relative to the host.6 Furthermore, the guest signals for hemicarceplexes 107«DMA and 107»DMF split out into two sets of signals with unequal populations. This was interpreted as being due to two different guest orientations whose energy barrier for interconversion was 13 kcal/mol. (Reinhoudt et al. has studied other carceplex systems that displayed preferred orientation of their guest molecule with unequal populations.7) In contrast to this, the guest signals of hemicarceplexes 107«DMSO did not split out. Like Cram et al., we repeatedly observed increased complexity of the *H NMR spectra of various two-bowl complexes»guest as the temperature was lowered. This observation was not unique to any single complex nor was it unique to a particular guest molecule but, it occurred with various complexes and guests. We speculated that the cause of this phenomenon had to have a similar origin because of the large number of two-bowl 281 complexes«guest that it affected. We therefore re-examined the crystal structures that were known for these and related compounds, hoping to find a common denominator. The shell of the host for complex 3b»pyrazine (see Figure 3.11), carceplex Zb'pyrazine,1 carceplex 2a»DMA2 and hemicarceplex 107»DMF6 all have a chiral twist, due to the rotation of the upper bowl with respect to the lower bowl, of 13-21° as determined by their crystal structures (see Chapter 3, section Bxi). This chiral twist presumably adds stability to these structures by allowing the aryl ethers and the phenolic hydroxyls/phenoxides to conjugate with their respective aromatic rings; it also would relieve strain between the methylenes (Hjn and H o u t ) that line the upper rims of the bowls. Would the host of other complexes, such as tetraprotio 123a«guest, also have a chiral twist and if so, would the rate of interconversion be too fast to observe by *H NMR spectroscopy? Could the freezing out of such a conformer explain the increased complexity observed for the *H NMR spectra of various two-bowl complexes«guest at lower temperature? The remainder of this chapter explores the dynamic *H NMR properties of various host guest complexes in order to address these questions. ii. Variable Temperature *H NMR Experiments with Tetraprotio 123a»Pyrazine Tetraprotio 123a»pyrazine is an excellent host-guest system to begin our investigation because we had a good idea about the orientation of pyrazine with respect to the interior of this host (see section Aiv). Here we examine the variable temperature *H NMR of tetraprotio 123a»pyrazine in detail, in order to explore the molecular motions of both host and guest. The *H NMR spectra of tetraprotio 123a«pyrazine at 298 K shows the characteristic 2:2:4 ratio for protons Hjn(a): Hjn(b):Hin(C) of the host (Figure 4.5). Upon cooling, the spectrum shows substantial loss of symmetry. For example, the broad signal 282 for Hin(C) at 298 K (2.83 ppm) eventually splits into two broad doublets at 218 K (6 = 3.20 ppm and 2.06 ppm)! . Furthermore, the *H NMR spectrum of tetraprotio 123a»pyrazine at 218 K versus 298 K indicates that the following host signals have split into two signals: Hb(a). Hb(b)> Hp(a), H^b), Hout(c), Hm( c), and Crfyc). The remaining protons of the host and those of pyrazine showed no apparent temperature dependence. To account for the increased complexity of the *H NMR spectra of tetraprotio 123a»pyrazine at 218 K, we propose that the chiral twisting of the upper and lower bowls of tetraprotio 123a»pyrazine are interconverting slowly on the *H NMR timescale (i.e., right-handed helical twist interconverts to a left-handed helical twist) to give a pair of isomers we will call "twistomers". If this were the case, the symmetry of tetraprotio 123a»pyrazine would be reduced from C 2 v (symmetry elements: E, C 2 , o"v(xz), o~v(yz),) to C 2 (symmetry elements: E, C 2 ) . Thus, all protons of tetraprotio 123a»pyrazine that were formally equivalent due only to the presence of a mirror plane of symmetry (fjv(xz), Ov(yz)) are now nonequivalent, while those protons that were equivalent due to the presence of a C 2 axis of symmetry remain equivalent. Based on the predicted loss of symmetry, the following protons are expected to split into two signals: Haryl, Hb(a), Hb(b), Hp(a), Hp(b), Hout(c), Hm(c), and CH3( C ) while the following signals are predicted to remain equivalent: the signal for pyrazine, Hi„(a), Hi„(b), H o u t( a), Hout(b) Hm(a)> Hm(b), CH3(a) and CH3(b), (Figure 4.6). The *H NMR spectrum of tetraprotio 123a«pyrazine at 218 K is consistent with slow interconversion of the twistomers on the *H NMR timescale except the signal for Haryl did not split out as anticipated, but this may be due to coincidental 5 of the Haryi's. The energy barrier determined for interconversion of the twistomers was 11.5 ± 0.1 kcal/mol based on the coalescence temperatures of five separate protons of tetraprotio 123a»pyrazine (Table 4.5).8 Similarly, the variable temperature *H NMR of complexes A,B-bis-bridged 103»pyrazine and tetra-OMe 125»pyrazine gave energy barriers for 283 Figure 4.5. Variable Temperature 400 MHz lH NMR Spectra of tetraprotio llSa-'pyrazine in CDCI3. 2 9 8 K ** encapsulated pyrazine H in(c) X 2 7 3 K ** u Lul ULUOAOI H in(c) 2 5 8 K H in(c) J L 2 3 8 K H in(c) 2 1 8 K 8.0 7.0 6.0 5.0 ppm 4.0 3.0 2.0 ! H NMR spectra of tetraprotio 123a (6.3 mM) and pyrazine (25 mM) in CDCI3. ! H NMR assignments at 298 K (ppm): Hp (7.39, 6.98), H b (6.52, 6.32), H a r y i (6.29), H o u t (6.14, 5.91, 5.57), H m (5.20, 5.11, 4.72), pyrazine (4.09) and CH 3 (1.80, 1.66); not shown is the signal for free pyrazine at 8.6 ppm. * Signal for residual water. ** Signal for residual CHCI3. 284 twistomer interconversion of 11.8 and 12.1 kcal/mol, respectively.8 In summary, slow interconversion of the twistomers on the lH NMR timescale is consistent with the lH NMR spectrum of tetraprotio 123a»pyrazine at 218 K. Furthermore, the energy barrier for twistomer rotation in A,B-bis-bridged complexes of pyrazine have very similar energy barriers. Therefore, we further explored whether other two-bowl complexes»guest also demonstrated this phenomenon. Figure 4.6. Tetraprotio 123a»pyrazine. Schematic representation of the lower- tetraprotio 123a»pyrazine bowl of tetraprotio 123a»pyrazine 285 Table 4.5. Activation Barrier for the Interconversion of Twistomers of tetraprotio 123a»pyrazine in CDCI3. Proton of tetraprotio A5 (Hz) T c (K) AG* (kcal/mol) 123a»pyrazine1 Hp(a) 13 223 11.5 Hp(b) 27 238 11.9 Hout(c) 209 248 11.4 Hm (c) 63 238 11.5 Hin(c) 456 258 11.5 Average 11.5 ±0 .14 1 The protons of tetraprotio 123a»pyrazine are labeled in Figure 4.6. iii. Chiral Carceplex The above observations prompted us to explore the dynamic *H NMR of carceplex 2»guest to determine whether twistomers of these carceplexes could also be observed by *H NMR spectroscopy, and if so what would the energy barrier be for their interconversion. Carceplex 2»guest and complex 3b»guest are unique from the other complexes studied in this thesis due to their higher symmetry (point group ZX#,). If the interconversion of the twistomers became slow on the *H NMR timescale, all planes of symmetry would be lost and the symmetry of carceplex 2«guest and complex 3»guest would be reduced to the D4 point group. Amazingly, all the protons of carceplex 2»guest and complex 3»guest still remain equivalent in this D4 point group due to the presence of a C4 axis and 4 C2 axes of symmetry. Thus, one would not expect to see twistomers for carceplex 2»guest and complex 3»guest based on the symmetry of the host alone. For example, the *H NMR spectrum of complex 3b»pyrazine at 223 K showed no changes from the spectrum at 298 K (e.g., all Hi n protons are equivalent). The above argument assumes that the guest molecule does not, itself, induce asymmetry. Therefore, what if the guest molecule did induce asymmetry: could twistomers be observed for these systems? What if the protons of the guest were diastereotopic (see section iv)? Would a chiral guest molecule encapsulated in the interior of carceplex 2»guest lead to diastereomeric 286 twistomers? We therefore synthesized carceplex 2b#(/?)-(-)-2-butanol to address these questions. The variable temperature *H NMR spectra of carceplex 2b»(i?)-(-)-2-butanol are shown in Figure 4.7. The top and bottom asymmetry of the host is apparent (e.g., 2 doublets for Hj n at 4.43 and 4.34 ppm). This asymmetry suggests that 2-butanol rotates slowly about the C 2 axes of the host in the temperature range studied (223 K - 323 K). The ! H NMR spectra of carceplex 2b«(/?)-(-)-2-butanol at 223 K indicates that indeed the twistomers are frozen out on the *H NMR timescale. The resulting two diastereomeric twistomers are schematically illustrated in Figure 4.8. The hydroxyl proton of the guest in the two diastereomeric twistomers has the largest A§ value of 65 Hz (one position at -0.91 ppm and the other at -1.07 ppm, which overlaps with one of the protons of the methylene of 2-butanol), which is not surprising because hydroxyls are notoriously sensitive to their chemical environment. Also, the terminal methyl groups of the guest, and the Hj n of the host each split into two signals (one for each diastereomeric twistomer). Also, as expected the inter-bowl methylene bridges (OCH2O) of the host at 6.54 ppm were broadened and form what appears to be overlapping AB quartets. Relative integration of the guest signals of the diastereomeric twistomer at 223 K gives a 1:1.2 ratio which does not fall much outside of the 10% expected experimental error for integration; thus, the diastereomers are of very similar energy. The energy barrier for interconversion of the twistomers of carceplex 2b»(i?)-(-)-2-butanol was calculated to be 12.6 + 0.1 kcal/mol based on the coalescence temperature of five host and guest signals (Table 4.6).8 Therefore, the twistomers of carceplex 2a*guest can be observed if one chooses the appropriate guest molecule. Furthermore, the energy barrier is similar to that for tetraprotio 123a»pyrazine. 287 Figure 4.7. Variable Temperature 400 MHz lH NMR Spectra of Carceplex 2b»(/c)-(-)-2-butanol in CDCI3. 293 K 7.0 6.0 ppm 5.0 4.0 -1.0 -2.0ppm -3.0 -4.0 J H NMR assignments at 293 K: Up 6.80, 6.78 ; H b 6.54; H o ut 6.14; H m 4.99; Hi n 4.43, 4.34; CH3CHOHCHxHyCH3 -0.97; CH 3CHOHCH xH yCH 3 -1.37; CH 3CHOHCH 2CH 3 -3.33; CH3CHOHCH2CH3 -3.54. Not shown: CH 3CHOHCH 2CH 3 0.82; and the pendent methyl group (1.69 ppm): 288 Figure 4.8. Schematic Representation of the Diastereomeric Twistomers of Carcepli 2b-(/e)-(-)-2-butanol Table 4.6. Activation Barrier for the Interconversion of Twistomers of Carceplex 2b»(/?)-(-)-2-butanol in CDCI3. carceplex 2b»(/c)-(-)-2-butanol A8 (Hz) TC(K) AG* (kcaymol) (guest) OH 65 263 12.7 (guest) CHOHCH3 19 248 12.6 (guest) CH2CH3 18 248 12.6 Hin 22 248 12.5 Hin 14 243 12.5 Average 12.6 ± 0.06 289 iv. Twistomers with DMSO as Guest. The mobility of DMSO within the interior of carceplex 2a»DMSO and in other related systems has been extensively studied by Cram et al.2>9 They found that the energy barrier for rotation of DMSO about the C 2 axes of carceplex 2a»DMSO was 12.7 kcal/mol and 13.6 kcal/mol as determined from the coalescence temperatures of the guest signal and the Hj n signal, respectively. Later, they used this activation energy for DMSO rotation (energy barrier was calculated from the coalescence temperature of the guest DMSO only) as a measure of the internal size of the cavity in a number of related host systems shown in Figure 4.9.9 Their data are shown in Table 4.7, and a schematic of the host systems studied is shown in Figure 4.9. Table 4.7. Free energies of Activation for Rotation of DMSO about the C2 axes of the Host.9 Host Solvent AG* (kcaymol)1 TC(K) 2a»DMSO CDCI3 13.6 (12.7)11 275 107«DMSO (CD 2) 40 9.8 193 126-DMSO (CD 2) 40 12.1 261 127«DMSO (CD 2) 40 12.6 272 128»DMSO (CD 2) 40 12.1 258 128»DMSO CDCI3 12.1 258 128»DMSO CD 2C1 2 12.1 258 128-DMSO CDC1 2CDC1 2 12.1 258 129-DMSO (CD 2) 40 12.1 261 130«DMSO (CD?,)40 13.0 283 I Energy barrier determined by the coalescence temperature of the guest. I I Energy barrier determined by the coalescence temperature of the H; n proton of the host. 290 Figure 4.9. Schematic Representation of Host Systems Investigated by Cram et al. R = CH2CH2Ph; line connecting the bowls R = CH2CH2Ph; line connecting the bowls represents a OCH2O bridge represents a OCH2O bridge 107-DMSO X = H 129»DMSO X = OMe 126»DMSO X = OMe 130»DMSO X = OEt 127«DMSO X = OEt 128»DMSO X = OCH2Ph Cram et al. found the energy barrier for the process they were observing was -12 kcal/mol with the exception of hemicarceplex 107»DMSO which had an uncharacteristically low energy barrier (Table 4.7). We therefore suggest that the freezing out of twistomers of the hosts on the H NMR timescale provides a more complete explanation for Cram's data. The application of the twistomer theory to complexes of DMSO suggests that the enantiotopic protons of the two methyl groups of DMSO's become diastereotopic (non-equivalent) when the twistomers interconvert slowly on the ! H NMR timescale because they are in a chiral environment.3 Thus, regardless of the rate of rotation of DMSO about the C 2 axes of the host, one should observe two signals for DMSO, (i.e., one signal for each unique methyl group) once the twistomers have frozen out on the *H NMR timescale. Similarly, the enantiotopic methyl groups of DMSO were demonstrated to be non-equivalent (diastereotopic) in both a chiral solvent10 and in the presence of a chiral shift reagent.11 If the twistomers are interconverting rapidly on the H NMR timescale, then in 291 effect DMSO is not in a chiral environment and the two methyl groups are equivalent (enantiotopic). Cram et al. did suggest that: "the rotation of the northern and the southern hemispheres of carceplex 2a»DMA, as determined by its crystal structure, probably extends to carceplex 2a»DMSO as well. If so, the transition occurring at 255 K may reflect a freezing out of equilibrations between the directions of rotations of the northern relative to the southern hemispheres, which may occur once the guests' motions are constrained."2 Thus, Cram did suggest twistomers may be responsible for the asymmetry observed in the lH NMR spectrum of carceplex 2a»DMSO at 255 K, but in order to explain this asymmetry with the twistomer theory, there is no need to invoke additional forms of asymmetry (e.g., constrain the motion of the guest). To substantiate this theory, we synthesized asymmetric carceplex 2d»DMSO (Figure 4.10) in order to specifically examine the rotation of DMSO about the C 2 axes of the host. We had previously used asymmetric hosts to examine the rotation of pyrazine about the C2-axes of other complexes of pyrazine (see Chapter 3, section Bx and Bxiiie). As shown in Figure 4.11, the *H NMR spectrum of asymmetric carceplex 2d»DMSO in CDCI3 at 293 K indicates that DMSO rotates rapidly about the C 2 axes of the host and the interconversion of the twistomers is fast, both with respect to the lH NMR timescale, due to the observation of a single guest peak for DMSO at -1.27 ppm. There are, however, two sets of signals for the host at this temperature because the top and bottom bowls differ in their attached pendent group. As the temperature is decreased the signals for the DMSO guest broaden and eventually split out into four separate singlets at 223 K (as a comparison, the guest peaks for symmetric carceplex 2a»DMSO split out into two singlets). We suggest that these four singlets of DMSO are due to the slow interconversion of the two twistomers of asymmetric carceplex 2d»DMSO and in addition to this, DMSO is rotating slowly about the pseudo-C2 axes of the host (both with respect to the lH NMR timescale), which results in two diastereomeric twistomers. The splitting out of the Hj n signals of the host is only due to the slow interconversion of the twistomers at 223 K (i.e., 292 because the top and bottom bowls differ in their pendent group, the host signals are independent of the rotation of the guest about the pseudo-C2 axes. The energy barrier for interconversion of the twistomers as calculated from the coalescence temperature (Tc = 258 K) of Hj n is 12.8 kcal/mol. We were unable to distinguish which pair of the guest signals corresponded to a single molecule of DMSO. Nevertheless, the energy barrier calculated for both rotation of DMSO with respect to the C2 axis of the host and twistomer interconversion based on the coalescence of any sets of guest signals gave energy barriers of 12.8 - 13.1 kcal/mol. Thus, the two processes have the same energy barrier (within experimental error). The energy barrier for interconversion of the twistomers of carceplex 2d»DMSO (12.8 kcal/mol) correlates with carceplex 2a«DMS0 2 (12.7 and 13.6 kcal/mol) and carceplex 2d»(R)-(-)-2-butanol (12.6 kcal/mol). Figure 4.10. Schematic Representation of the Two Diastereomers of Carceplex 2d»DMSO. lines interconnecting the two bowls represent O C H 2 O bridges 293 For carceplex 2d«DMSO, if the interconversion of the twistomers and one methyl group and the sulfur oxygen of DMSO was constrained to one hemisphere and the other methyl group to the other hemisphere (i.e., DMSO sits asymmetrically in the cavity) as suggested by Cram, one should observe a total of eight signals for DMSO. This was not observed. If the twistomers interconverted rapidly on the *H NMR timescale and DMSO was to rotate slowly about the pseudo-C2 axes of the host with respect to the NMR timescale and DMSO sat asymmetrically in the cavity, then one would observe four signals for DMSO which is also consistent with the observed data. Thus, either the slow interconversion of twistomers, or DMSO sitting asymmetrically in the cavity are possible explanations for the observed *H NMR spectra of carceplex 2d»DMSO, but not both. We suggest that indeed the twistomer theory provides an elaboration on the nature of the asymmetry to which Cram referred, as it was shown to occur for carceplex 2a»(R)-(-)-2-butanol and three A,B-bis-bridged complexes of pyrazine. We also determined the energy barrier for interconversion of the twistomers of complex 3b»DMSO in CDC13 to be 12.7 (Tc = 245 K) and 13.5 (Tc = 260 K) kcai/mol based on the coalescence of Hj n proton of the host and the guest signal, respectively. The similarity of this value to that of carceplex 2a»DMS02 (12.7 and 13.6 kcal/mol) suggests that the interconversion of twistomers is not sensitive to the nature of the linkages that interconnects the two bowls. 294 Figure 4.11. Variable Temperature 400 MHz A H NMR of Carceplex 2d»DMSO in CDCI3. 1 1 Partial H N M R spectra of carceplex 2d>DMSO in CDCI3; H N M R spectral assignments: H m 4.98, 4.87; H i n 4.47; (CH 3 ) 2 SO -1 .27. Specific ! H N M R spectral assignments for all of carceplex 2d»DMSO can be found in the experimental section. 295 The energy barrier determined by Cram et al. for the dynamic process occurring with hemicarceplex 107«DMSO is substantially lower than the other host systems studied (Table 4.7).9 Furthermore, Cram et al. had shown previously that a dynamic process occurring with hemicarceplex 107»DMF had an energy barrier of 13.0 kcal/mol.12 Again, we suggest that these dynamic processes are interconversion of twistomers as the energies are consistent with those reported above. We determined the twistomer energy barrier of hemicarceplex 107»pyrazine in CDCI3 to be 12.9 ± 0.3 kcal/mol based on the coalescence temperature of the two signals for H o u t (Tc = 260 K and 265 K). This value correlates with the energy barrier determined for hemicarceplex 107«DMF (13.0 kcal/mol).12 Thus, the low energy barrier determined by Cram et al. for the dynamic process occurring with hemicarceplex 107»DMSO differs from that of other hemicarceplexes 107»guest and is anomalous.9 v. Dynamic lH NMR of Carceplex 2a«l,4-Thioxane Unique *H NMR spectral properties were observed for several carceplexes 2a»guest whose guests contained six-membered rings. Often, interconversion of conformers was slow on the *H NMR timescale. Thus, separate signals could be observed for the axial and equatorial protons. The variable temperature lH NMR spectra of carceplex 2a» 1,4-thioxane serves as a good example of this phenomenon. Unfortunately, the lH NMR spectra of carceplex 2a» 1,4-thioxane are further complicated by the presence of carceplex 2a» 1,4-dioxane (see Chapter 2, section Bid). At 293 K, the lH NMR spectrum carceplex 2a» 1,4-thioxane has four signals for the encapsulated 1,4-thioxane representing the two sets (C(HD)2S and C(Ha)20) of axial and equatorial protons of its chair conformation which are interconverting slowly with respect to the *H NMR timescale (Figure 4.12). The activation barrier for interconverting the chair conformation of 1,4-thioxane inside carceplex 2a» 1,4-thioxane was thus 296 determined to be 16.3 and 16.5 kcal/mol based on the coalescence temperature of the C(Ha)20 protons (Tc = 333 K) and C(Hb)2S protons (Tc = 323 K), respectively (Figure 4.13). The limited internal size of the interior of the host presumably creates further steric strain in the transition state of the converting chair conformation of 1,4-thioxane, thereby resulting in a higher energy barriers for this process in carceplex 2a»guest than observed free in solution/"1 The diastereotopic nature of the axial and equatorial protons of the chair conformation of 1,4-thioxane also allowed the determination of the energy barrier for the interconversion of twistomers of carceplex 2a» 1,4-thioxane. Thus, at 223 K all eight protons of 1,4-thioxane are non-equivalent because of the slow interconversion of the twistomers and slow conformational inversion of the chair of 1,4-thioxane; actually in the lH NMR spectra only six separate signals are observed presumably because of the coincidental overlap of two pairs of hydrogens. The energy barrier determined for twistomer interconversion was 10.7 kcal/mol based on the coalescence temperature (Tc = 243 K) of both C(HbHD')S protons. This energy barrier is smaller than the twistomer energy barriers observed for carceplex 2a»2-butanol (12.6 kcal/mol) and carceplex 2d«DMSO (12.8 kcal/mol). This may be due to the steric bulk of 1,4-thioxane which may hold the bowls apart, thus raising the ground state closer to the transition state. Also note, there are substantial differences in the A8 values for the protons of 1,4-thioxane at 223 K. We suggest that the protons with the largest A5 are the equatorial protons which extend deeper into the bowls (as determined by examination of CPK models). X1X Little is known about the conformational analysis of 1,4-thioxane; the energy barrier for interconversion of the chair conformation of thiane is 11.7 kcal/mol. The energy barrier for 1,4-thioxane should be similar or slightly lower. ^ 297 Figure 4.12. Variable Temperature 400 MHz lR NMR Spectra of Carceplex 2a«l thioxane in CDCI3. 273 K Hi 263 K 253 K 243 K 233 K 223 K 215 K 1 1 1 1 r 0.5 0.0 -0.5 -1.0 -1.5 ppm * Carceplex 2a» 1,4-dioxane impurity in the sample. -2.0 ~ l 1 -2.5 -3.0 298 Figure 4.13. Interconversion of the Chair Conformations of 1,4-Thioxane. vi. Dynamic H NMR of Carceplex 2b»l,4-Dioxane and Complex 3b»l,4-Dioxane We explored the dynamic behavior of both complex 3b» 1,4-dioxane and carceplex 2b» 1,4-dioxane in order to compare their energy barriers for interconversion of twistomers and chair conformations. Unlike carceplex 2b» 1,4-thioxane, the chair conformation of both complex 3b» 1,4-dioxane and carceplex 2b» 1,4-dioxane are interconverting rapidly on the proton lH NMR timescale at 298 K. The *H NMR spectra of these two compounds at 223 K is consistent with both the chair and twistomer conformations being slow on the % NMR timescale. The energy barrier for twistomer interconversion of carceplex 2b» 1,4-dioxane was determined to be 11.6 kcal/mol based on T c = 243 K while the energy barrier for interconversion of the chair conformation was determined to be 11.3 kcal/mol based on T c = 253 K. Similarly, the energy barrier for interconversion of the twistomers of complex 3b» 1,4-dioxane was determined to be 11.7 kcal/mol based on T c = 243 K while the energy barrier for interconversion of its chair conformation was determined to be 11.3 kcal/mol 299 based on T c = 253 K. The similarities in the energy barriers for interconversion of the chair conformation of dioxane in both the carceplex and the complex suggests that the cavity sizes are similar. Also, the energy barrier for interconversion inside the carceplex and complex are slightly larger than that determined in solution (9.7 kcal/mol, T c = 179.4 K, A8 Hz = 3.3).14 The similarities of the energy barrier for interconversion of the twistomers suggests that this process is not very sensitive to the nature of the linkages interconnecting the two bowls (see section iv). The similarities of the energy barrier for ring-flipping and for twistomer interconversion is likely to be coincidental. 300 Figure 4.14. Variable Temperature 400 MHz *H NMR Spectra of Carceplex 2b» 1,4-dioxane and Complex 3b»l,4-dioxaxe in CDCI3. 303 K 263 K MiimiinMiineniMiMi/ 253 K ^HJL 243 K 223 K 1 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 0.7 0.5 0.3 0.1 -0.1 -0.3 -0.5 -0.7 -0.9 -1.1 ppm Guest region of Complex 3b» 1,4-dioxane r\ Ha1 Ha 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 0.7 0.5 0.3 0.1 -0.1 -0.3 -0.5 -0.7 -0.9 -1.1 -1.3 ppm Guest region of Carceplex 2b» 1,4-dioxane T ^ \ \ H a ' H a H a H a ' 0 O I H H 301 vii. Summary The H NMR spectral properties of both host and guest for carceplex 2a»guest and complexes such as tetraprotio 123a»guest allow one to predict the most probable orientation of the guest molecules with respect to the interior of the host. There is a preferred orientation of aromatic guest molecules in their A,B -bis-bridged complexes as predicted by the substantially different 8 of the three non-equivalent Hj n protons. The dynamic lH NMR spectra of carceplex 2a»guest and other related complexes suggest that both host and guest molecules have unique conformational mobilities. For example, energy barriers of 10 -13 kcal/mol were found for interconversion of the twistomers of the host while the conformational interconversion of chair conformations of 1,4-dioxane and 1,4-thioxane were 11.3 and 16.5 kcal/mol in their respective carceplexes 2a»guest. Conformational interconversion of the chair conformation of guest for carceplexes 2a» 1,3-dioxane, 2a»pyran, 2a» 1,3-dithiolane, 2a»tetrahydrothiophene as well as complex 3b,morpholine were also slow at ambient temperature. The twistomer theory further clarifies the dynamic processes that are occurring for two-bowl hosts systems with DMSO as guest. The diastereomeric twistomers of carceplex 2a»(R)-(-)-2-butanol have very similar free energies as determined by integration of their unique guest signals in the NMR spectra. Therefore, the chirality of encapsulated guest molecules does not cause one twistomer to preferentially form over another to a significant extent for this particular guest. 302 C. Experimental For general experimental conditions and the synthesis of other carceplexes 2«guest see Chapter 2. Variable temperature *H NMR spectroscopy was performed on a Bruker WH-400 in CDCI3 using the residual *H (7.24) as a reference unless otherwise noted. i. Synthesis of Carceplex 2»Guest Carceplexes 2b«Guest 2b»(R)-(-)-2-butanol A mixture of tetrol lb (133 mg, 0.20 mmol), (R)-(-)-2-butanol (0.50 mL, 5.4 mmol), K 2 C 0 3 (1.4 g, 10 mmol), and CH 2BrCl (27 ul, 0.42 mmol) in NMP (50 mL) was stirred at rt for 24 h. An additional 0.42 mmol of CH2B1CI were added and the reaction was stirred at 60 °C for an additional 24 h. A further 0.42 mmol of CH2BrCl were added and the reaction was stirred at 60 °C for an additional 24 h. The reaction mixture was concentrated in vacuo, water (50 mL) was added, and the slurry was acidified with 2 M 303 HC1. The slurry was extracted with CHCI3 (3 x 60 mL), and the combined organic extracts were washed with saturated aq. NaHCC>3 (30 mL) and brine (30 mL), and dried over anhydrous MgS04. Silica gel (0.5 g) was added to the CHCI3 solution and the solvent was removed in vacuo. The silica gel-absorbed sample was dry loaded onto a silica gel gravity column (20 g) and eluted with CHCl3:hexanes (3:1), affording lb»(R)-(-)-2-butanol as a white solid which was recrystallized from CH.2Cl2/hexanes and dried at 110 °C (0.1 mm Hg) for 24 h (60 mg, 41%) as a white solid: mp > 250 °C; ! H NMR (CDCI3, 400 MHz): 6 6.80 (s, 4H, H d or Hd>), 6.78 (s, 4H, H d or Hd>), 6.54 (s, 8H, H e), 6.14 (d, J = 7.4 Hz, 8H, Hf and Hf), 4.99 (q, J = 7.4 Hz, 8H, H g and Hg>), 4.43 (d, J = 7.4 Hz, 4H, H h or Hh'), 4.34 (d, J = 7.4 Hz, 4H, H h or Hh'), 1-69 (d, J = 7.4 Hz, 24H, CH 3), ), 0.82 (br, 1H, C H 3 C H O H C H 2 C H 3 ) , -0.97 (br, 2H, CH 3 CHOHCH x H y CH 3 ) , -1.37 (br, 1H, CH 3 CHOHCH x H y CH 3 ) , -3.33 (d, J = 6.1 Hz, 3H, CH3CHOHCH2CH3), -3.54 (t, / = 7.4 Hz, 3H, CH3CHOHCH2CH3). MS (MALDI) m/z (rel intensity): 1458 ((M»CH 3CHOHCH 2CH 3 + Na +) +; 100); Calcd for C 8oH 7 4025'Na+ = 1458. Anal. Calcd for C80H74O25: C, 66.94; H, 5.20. Found: C, 66.62; H, 5.23. 304 2d»DMSO A mixture of tetrol la (102 mg, 0.10 mmol), tetrol lb (66 mg, 0.10 mmol), DMSO (1.0 mL, 14.1 mmol), K 2 C 0 3 (1.4 g, 10.0 mmol), and CH 2BrCl (130 jil, 2.0 mmol) in NMP (50 mL) was stirred at 60 °C for 24 h. An additional 1.0 mmol of CH 2BrCl were added and the reaction was stirred for an additional 24 h at 60 °C. The reaction mixture was concentrated in vacuo, water (50 mL) was added, and the slurry was acidified with 2 M HC1. The slurry was extracted with CHCI3 (3 x 60 mL), and the combined organic extracts were washed with saturated aq. NaHC03 (30 mL) and brine (30 mL), and dried over anhydrous MgS04. Silica gel (2 g) was added to the CHCI3 solution and the solvent removed in vacuo. The silica gel-absorbed sample was dry loaded onto a silica gel gravity column (200 g) and eluted with (CH2Cl)2:CCl4 (3:1). The purified products were recrystallized from CH2Cl2/hexanes and dried at 110 °C (0.1 mm Hg) for 24 h affording 305 2a»DMSO of (20.8 mg, 9.6%), 2b»DMSO of (14.2 mg, 9.8%) and 2c»DMSO of (33.1 mg, 18.3%) as white solids. 2c«DMSO was characterized as follows: mp > 250 °C; J H NMR (CDC13, 400 MHz): d 7.11-7.24 (m, 20H, H a , H b , and H c), 6.84 (s, 4H, H d or Hd>), 6.75 (s, 4H, H d or Hd>), 6.55 (s, 8H, H e), 6.15 (d, / = 7.5 Hz, 8H, Hf and Hf), 4.98 (q, J = 7.4 Hz, 4H, Hg'), 4.87 (t, / = 7.9 Hz, 8H, H g), 4.47 (m, 8H, Hh and Hh'), 2.63 (m, 8H, Hj), 2.43 (m, 8H, Hj), 1.70 (d, J = 7.4 Hz, 12H, CH 3), -1.27 (s, 6H, (CH3)2SO). MS (MALDI) m/z (rel intensity): 1823 ((M»(CH 3) 2SO + Na +) +; 100); Calcd for Ci06H9 4 O 2 5 S»Na + = 1823. Anal. Calcd for C106H94O25S: C, 70.73; H, 5.26. Found: C, 70.43; H, 5.25. ii. Supplementary Tables Table 4.8. 6 (ppm) of Host Signals in the H NMR Spectra of Carceplex 2a»Guest in CDCI3 at Ambient Temperature. Protons Description H, H b Hout H m Hin 2a»pyrazine 6.93 6.47 6.03 4.90 4.26 2a» 1,4-dioxane 6.72 6.57 6.19 4.87 4.46 2a»dimethyl sulfide 6.73 6.57 6.21 4.87 4.47 2a»dimethylcarbonate 6.67 6.54 6.16 4.92 4.38 2a«DMS0 2 6.76 6.57 6.16 4.88 4.49 2a»l,3-dioxolane 6.75 6.56 6.19 4.87 4.37 2a,pyridine 6.91 6.49 6.03 4.89 4.14 2a#dimethyl sulfone 6.76 6.57 6.14 4.88 4.49 2a»thioxane 6.73 6.57 6.16 4.87 4.80 2a#2,3-dihydofuran 6.78 6.55 6.16 4.87 4.34 2a»furan 6.84 6.52 6.12 4.88 4.16 2a»tetrahydrofuran 6.76 6.56 6.16 4.87 4.40 2a»pyridazine 6.91 6.49 6.03 4.89 4.10 2a»acetone 6.74 6.57 6.19 4.87 4.36 2a»thiophene 6.88 6.51 6.05 4.88 4.26 2a» 1,3-dithiolane 6.78 6.56 6.11 m,4.86 m, 4.86 2a«benzene 6.91 6.50 6.02 4.88 4.15 2a»2-propanol 6.76 6.57 6.18 4.87 4.40 306 Protons Description UP H b H o u t H m H i n 2a«pyrrole 6.86 6.52 6.09 4.89 4.23 2a«tetrahydrothiophene 6.77 6.56 6.11 4.87 4.71 2a» 1,3-dioxane 6.74 6.57 6.14 4.87 4.55 2a»acetamide 6.75 6.57 6.19 4.87 4.36 2a»trioxane 6.74 6.57 6.17 4.86 4.48 2a»acetonitrile 6.71 6.58 6.24 4.87 4.30 2a#ethanol 6.75 6.56 6.21 4.86 4.26 2a»diethylether 6.73 6.55 6.15 4.88 4.44 2a»DMF2 6.78 6.55 6.12 4.89 4.50 2a»dichloromethane 6.72 6.57 6.23 4.86 4.27 2a,bromochloromethane 6.71 6.58 6.23 4.86 4.30 2a»dibromomethane 6.69 6.59 6.22 4.86 4.33 2a»diiodomethane 6.63 6.61 6.18 4.88 4.50 2a»CHCl3 6.73 6.57 6.16 4.87 4.59 2a»pyran 6.73 6.57 6.14 4.87 4.54 2a»methyl acetate 6.71, 6.69 6.57 m, 6.19 4.89 4.35, 4.31 2a»ethylmethylsulfide 6.69 6.59 6.19, 6.17 m, 4.89 4.59, 4.39 2a»2-butanone 6.71, 6.70 6.58 6.16 m, 4.88 m, 4.40 2a»(±)-2-butanol 6.73, 6.71 6.58 6.16 4.89 4.46, 4.37 2a»ethylacetate 6.71, 6.67 6.58 m, 6.13 4.90 4.54, 4.51 2a«DMA2 6.74 6.51 6.08 4.91 4.64 2a«NMP 6.74, 6.71 br, 6.32 m, 6.03 m, 4.85 m, 4.68 Host protons are labeled in Figure 4.2. 307 Table 4.9. 8 (ppm) of Guest Signals in the *H NMR Spectra of Carceplex 2a*guest in CDCI3 at Ambient Temperature. Guest 5 bound (ppm) free (ppm) AS ppm pyrazine 1,4-dioxane methylacetate O H 3 / C H 3 O a H a H b 4.07 -0.28 -0.88 -2.40 8.58 3.68 3.67 2.05 4.51 3.96 4.55 4.45 dimethyl sulfide ethylmethylsulfide H 3 C ^ S ^ C c H 3 H a H b H , -1.24 -0.96 -2.29 -3.23 2.61 2.46 2.05 1.21 3.85 3.42 4.34 4.44 dimethylcarbonate DMSO 2 -0.76 -1.24 3.66 2.61 4.42 3.85 H a H b 0.65 0.30 4.80 3.78 4.15 3.48 2-butanone O H 3 C b pyridine H •CH3 c H c H a H b H c H a H b H , -0.05 -2.36 -3.43 6.34 4.02 2.73 2.41 2.10 1.01 7.68 8.62 7.30 2.46 4.46 4.44 1.34 4.60 4.57 dimethyl sulfone -0.89 2.95 3.84 308 Guest 8 bound (ppm) '5 free (ppm) A8 ppm H a Ha' Hb H b ' -0.01 -0.17 -1.55 -1.60 3.88 3.88 2.60 2.60 3.89 4.05 4.15 4.20 2,3-dihydofuran Hb furan Q H b H a H b H c Hd H a H b 1.90 1.19 0.29 -0.25 3.09 3.06 6.29 4.91 4.26 2.56 7.48 6.43 4.39 3.72 3.97 2.81 4.39 3.37 tetrahydrofuran H a H b -1.26 -0.30 1.85 3.75 3.11 4.05 pyridazine N H b H a H b 4.64 4.39 9.22 7.49 4.58 3.10 acetone -1.63 2.17 3.70 thiophene H a H b H a H b 3.73 3.73 7.37 7.15 3.64 3.42 H a and H a-H b H b ' -0.59 -0.93 -1.05 3.85 3.12 4.44 4.05 4.17 309 Guest 8 bound (ppm) »8 free (ppm) AS ppm H a H b H c Hd H e Hf 0.91 -1.32 -0.93 -0.93 -3.27 -3.47 3.65 1.40 1.40 1.94 1.11 0.86 2.72 2.72 2.33 2.87 4.38 4.33 benzene 3.88 7.36 3.48 2-propanol OH H C H, -1.91 1.35 -2.46 2.14 3.94 1.13 4.05 2.59 3.56 pyrrole H b H a H b 4.08 3.41 3.05 8.00 6.68 6.22 3.92 3.27 3.17 tetrahydrothiophene H b H a Ha' Hb and Hb' H a H a ' Hb and Hb1 H c and H c ' -0.88 -1.08 -1.51 1.34 1.08 0.62 -2.32 2.75 1.86 3.80 3.80 4.74 1.67 3.63 3.83 3.37 2.46 2.72 4.12 3.99 acetamide O H 3 Mb H a H b CH, 1.40 0.65 -1.55 5.76 5.98 1.95 4.36 5.33 3.50 trioxane acetonitrile 1.88 -2.41 5.12 2.00 3.24 4.41 310 Guest 8 bound (ppm) »S free (ppm) A5 ppm CH 3 CH 2 OH C H 2 0.73 3.72 2.99 CH 3 -2.53 1.24 3.77 OH -2.72 3.76 6.48 ethylacetate H a 0.58 4.12 3.54 O H b -2.50 2.04 4.54 H c -2.63 1.25 3.88 C H 2 0.36 3.48 3.12 CH 3 -2.81 1.20 4.01 H a 1.04 3.02 1.98 H b -1.46 2.94 4.40 H c -2.40 2.09 4.49 H a 4.28 7.99 3.71 H b -0.04 2.94 2.98 H c -1.02 2.86 3.88 NMP H a 1.46 3.25 t 1.79 H b -1.79 2.70 s . 4.49 H c -1.79 2.22 1 4.01 Hd -2.13 1.89 m 4.02 dichloromethane 2 63" ^ - 2 7 bromochloromethane 2.54 5.14 2.60 dibromomethane 2.39 4.69 2.30 diiodomethane 1.47 3.85 2.38 CHC13 4.43 7.24 2.81 311 Guest 8 bound (ppm) '5 free (ppm) A8 ppm pyran H a H a ' H b H b ' H c and H c ' -0.31(4H) -0.31(2H) -2.35(2H) -2.42(2H) 3.52 1.51 1.51 1.51 3.83 1.82 3.86 3.93 I 8 (ppm) for free guests in CDCI3 at ambient temperature. I I Guest signal overlaps the signal for the CH2 of the pendent group of the host. 312 D. References 1. Fraser, J. R.; Borecka, B.; Trotter, J.; Sherman, J. C. J. Org. Chem. 1995, 60, 1207-1213. 2. Sherman, J. C ; Knobler, C. B.; Cram, D. J. J. Am. Chem. Soc. 1991,113, 2194-2204. 3. Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry of Organic Compounds; John Wiley and Sons: New York, 1994, pp 54, 635, 641. For a discussion of prostereoisomerism see chapter 4. 4. Helgeson, R. C ; Paek, K.; Knobler, C. B.; Maverick, E. F.; Cram, D. J. J. Am. Chem. Soc. 1996,118, 5590-5604. 5. Yoon, J.; Sheu, C ; Houk, K. N.; Knobler, C. B.; Cram, D. J. / . Org. Chem. 1996, 61, 9323-9329. 6. Cram, D. J.; Tanner, M. E.; Knobler, C. B. / . Am. Chem. Soc. 1991,113, 1111-1121. 7. Timmerman, P.; Verboom, W.; van Veggel, F. C. J. M.; van Duynhoven, J. P. M.; Reinhoudt, D. N. Angew. Chem. Int. Ed. Engl. 1994, 33, 2345-2348. 8. Abraham, R. J.; Fisher, J.; Loftus, P. Introduction to NMR Spectroscopy; Wiley: New York, 1990, pp 194-7. Activation barriers were calculated using the following equation: A G C ^ = RTC[22.96 + ln(Tc/A8HZ)] where AG C ^ is the activation barrier in kcal/mol; T c is the temperature of coalescence, and A 8 H z is the separation of the signals in Hz. 9. Kurdistani, S. K.; Robbins, T. A.; Cram, D. J. J. Chem. Soc, Chem. Commun. 1995, 1259-1260. 10. Kainosho, M.; Ajisaka, K.; Pirkle, W. H.; Beare, S. D. J. Am. Chem. Soc. 1972, 94, 5924-5926. 11. Goering, H. L.; Eikenberry, J. N.; Koermer, G. S.; Lattimer, C. J. J. Am. Chem. Soc. 1974, 96, 1493-1501. 12. Cram, D. J.; Tanner, M. E.; Knobler, C. B. / . Am. Chem. Soc. 1991,113, 11X1-1121. 13. Riddell, F. G. The Conformational Analysis of Heterocyclic Compounds; Academic : New York, 1980, pp 104-118. 14. Vergamini, P. J.; Vahrenkamp, H.; Dahl, L. F. J. Am. Chem. Soc. 1971, 93, 6329-6330. 313 5 Conclusions and Future Applications A. Carceplex 2a»Guest The work presented in this thesis began as an investigation of the driving forces for formation of carceplex 2a»guest. This investigation led to the discovery of a new family of self-assembled structures whose prototypical member is complex 3b»guest. A wealth of information about noncovalent interactions has been gleaned from the encapsulation of various molecules in these and related assemblies. The goal of this chapter is to place this new found knowledge from Chapters 2 and 3 in perspective, in terms of how it relates to the field of supramolecular chemistry, and also to provide suggestions for applications of these new systems in the design of larger self-assembling structures. The reaction to form carceplex 2a»guest is an excellent example of a templated reaction because, not only is a suitable template a necessity for formation of carceplex 2a»guest but the reaction is very sensitive to the nature of the template used in its synthesis. The million-fold range in template ratios determined for the formation of carceplex 2a»guest and the diversity of guest molecules that can be incarcerated in carceplex 2a»guest make it an excellent system for studying noncovalent interactions between molecules. The template ratios determined for formation of carceplex 2a»guest are a result of different rates of reaction of the forming carceplex at the GDS; this step was determined to be the formation of the second OCH2O bridged (either A,B-bis or A,C-bis). The rate of formation of this bridge is 106 times faster in the presence of pyrazine relative to NMP. The discovery of complex 3b»guest has great relevance to the formation of carceplex 2a»guest. It serves as an excellent model for the transition state of the GDS as demonstrated by the good correlation between the template ratios for the formation of carceplex 2a«guest and the relative stability constants for the formation of complex 314 3b»guest. The Orel's of other related complexes, which vary in the number and type of bonding interactions that interconnect the two bowls, show the same trends in their guest selectivity as was observed for complex 3b»guest. Therefore, the type and number of interactions that interconnect the two bowls does not have a significant impact on the guest selectivity observed in the formation of these complexes. Furthermore, the relative stabilities determined for these various hosts suggest that they have similar affinities for encapsulation of a given guest molecule. Thus, the driving forces for the formation of these nine complexes are dominated by the host-guest interactions, which are fairly uniform over the range of hosts explored. These host-guest interactions include favorable van der Waals and electrostatic interactions, CH-TC hydrogen bonds from the guest to the arenes of the bowls, X - H hydrogen bonds between the hetero atom of the guest to the methylene hydrogens that line the upper rims of the bowls, and K-TZ interactions between the aromatic guest molecules and the arenes of the bowls. These favorable noncovalent interactions between host and guest are inextricably linked to the selectivity observed for formation of carceplex 2a«guest. Carceplex 2a*guest and complex 3b«guest are excellent systems for studying noncovalent interactions between molecules because small changes in guest molecules result in large changes in selectivity. Thus, these assemblies are attractive models for computational chemists to develop better force fields. Such computations on complex 3b»guest have regenerated the trends observed for formation of carceplex 2a«guest and complex 3b»guest.1 Further comparisons of computations and experimental data on systems such as complex 3b»guest hold great promise for improvement of both sciences. Rebek et al. have played a major role in developing the sub-field of supramolecular chemistry that is concerned with the reversible encapsulation of guest molecules by two or more concave hosts.2"4 They reported their first example of encapsulation of a guest molecule within a molecular capsule in 19935 and have since reported a number of related systems. Sanders et al. and Aoyama et al. have reported the selective encapsulation of 315 substituted pyranosides in a steroid based capsule6 and in a resorcinarene-based capsule,7 respectively. More recently, Kim et al. have reported the encapsulation of molecules such as THF in a cucurbituril assembly. In time we will see ever increasing examples of this type of encapsulation with greater size and greater complexity. Ultimately, these self-assembled capsules may be used as miniature reaction vessels for the combination of two or more molecules in their interior. Other such future directions are discussed in the following sections. Complex 3b»guest has many unique features that differentiate it from other systems known in the literature. Complex 3b«guest represents one of the very few examples of self-assembly where two or more concave host molecules reversibly wrap around a guest molecule in solution. The important features of complex 3b»guest are: (1) It is highly selective towards the guest molecule it encapsulates as demonstrated by the 35,000 fold range in its relative stabilities in nitrobenzene-Js. (2) It does not appear to form in the absence of a suitable guest molecule. (3) Its charged character and high degree of crystallinity allow for its detailed characterization by ESMS and X-ray crystallography. (4) It is reversible and can be switched on and off by adjustment of the pH of the solution. This last property may render complex 3b»guest suitable for applications such as drug delivery. B . Larger Self-Assembling Structures i. Introduction The structural integrity and selectivity of complex 3b»guest make it an excellent model from which to design even larger and more sophisticated assemblies. How far can one extend this prototypical system? Can more than two bowls be joined via guest 316 encapsulation? Can polymeric assemblies be generated? Can the properties of polymeric assemblies be controlled by the choice of guest molecule? Will multiple guest molecules in an assembly be able to communicate with one another? Besides being intellectually stimulating, such futuristic assemblies may have interesting properties and/or further our understanding of the noncovalent interactions that govern their formation. ii. Polymeric Side-to-Side Assemblies Tetra-bromo-tris-bridged bowl 120a8 holds great promise for construction of polymeric side-to-side assemblies. Tetra-bromo-tris-bridged bowl 120a could be doubly bridged to another molecule of tetra-bromo-tris-bridged bowl 120a via long rigid spacers (e.g., OCH2-C^C-CH20) and the bromides could be converted to phenols to give either cis or trans bis-cavitand 131 (Figure 5.1). The two resulting stereoisomers (cis and trans) may be separable by chromatography, but their complexation properties in the presence of base and a suitable guest should be substantially different. For example, the complex of the trans isomer will most likely lead to polymeric assemblies that would resemble a structure of a staircase (Figure 5.1) while the cis isomer may form a polymeric assembly or a dimer about two guest molecules. The aggregation properties of polymeric assemblies based on bis-cavitand 131, both cis and/or trans, may be controlled by the choice of guest molecule used (i.e. strong binding guests should lead to larger aggregates) and also may be controlled by the addition of a chain terminating cap such as tetrol lb which would prevent further aggregation at either end of the polymeric assembly. The noncovalent assemblies may also be bridged with bromochloromethane to form covalently bridged bis- and polymeric carceplexes. Both the noncovalent and covalent assemblies may be further studied by X-ray crystallography which could further explore the communication properties of encapsulated guest molecules. 317 iii. Polymeric Tail-to-Tail Assemblies In a similar vein, tail-to-tail assemblies 132 may be constructed via interconnecting the pendent groups of two separate molecules of tetrol 1 (Figure 5.2). The addition of DBU and a suitable guest to tail-to-tail assembly 132 should lead to polymeric rods similar to those of the side-to-side assemblies discussed above. Again, the aggregation properties may be controlled by the addition of a chain terminating bowl such as tetrol lb (Figure 5.2). These polymeric assemblies may be covalently linked to form unique polymeric rods that may have interesting properties. For example, CPK models suggest that the tiny portals at the lower end of the bowls of 132 are aligned. Could guest molecules communicate through these tiny portals in these polymeric assemblies? 318 iv. Other Assemblies The construction of even more ambitious side-to-side self-assembling structures can also be envisioned such as the dimerization of five-bowl assembly 133 (Figure 5.3). Five-bowl assembly 133 could be synthesized by tetra-alkylating tetrol lb with a linker such as l,4-dichloro-2-butyne followed by reaction with tris protected tetrol lb and subsequent deprotection. The formation of the dimer (133»133«(guest)5) is the one of the smallest self-assembled structures that five-bowl assembly 133 can form and even it would weigh in excess of 10 kDaltons. The sophistication of and complexity of such an assembly is approaching the complexity of natural assemblies. 319 Figure 5.3. Five-bowl Assembly 133. five-bowl assembly 133 133»133»(guest)5 v. Monolayers Another interesting application of complex 3*guest would be in the formation of monolayers on a solid surface, as shown schematically in Figure 5.4. These monolayers have potential to be used as sensors. They also may form switchable monolayers where species that can be linked via the pendent groups of the incoming tetrol could be adsorbed/desorbed from the surface.9 Figure 5.4. Formation of Self-assembled Monolayers of Tetrol 1. 320 vi. Incarceration of Reactive Intermediates in Carceplex 2b»guest The formation of carceplex 2a»guest was demonstrated to form only in the presence of a suitable template molecule. What if the molecule one desired to incarcerate is not a suitable guest (e.g., a-pyrone) or it was not stable to the conditions used for formation of carceplex 2a»guest? This section proposes a different procedure for entrapment of such problematic guest molecules by drawing on the chemistry of hemicarceplexes.10 The property of hemicarceplexes which makes them appealing to us is their ability to expel one guest and incorporate another by heating them in a solvent that is too large for the interior of the hemicarcerand (hemicarcerand is a hemicarceplex without an encapsulated guest molecule). Ultimately, we propose to construct a hemicarceplex whose portal can be subsequently sown up after guest exchange to give the corresponding carceplex. The formation of such a versatile hemicarceplex could be synthesized from tetra-bromo-tris-bridged bowl 120a as described below.8 Protection of the phenolic hydroxyls of tetra-bromo-tris-bridged bowl 120a and subsequent conversion of the bromides to phenols may give tetrol 135 (Figure 5.5). Shell closure of tetrol 135 with and an appropriate guest molecule may lead to the formation of hemicarceplex 136»guest. There are three possible stereoisomers of hemicarceplex 136»guest, all of which should be separable by silica gel chromatography. As determined by CPK models, the portal of the stereoisomer of hemicarceplex 136»guest that aligns the benzyl ethers of the upper and lower bowls is sufficiently large to allow for slot-shaped guests to escape. After guest exchange and deprotection, the portal may be sewed closed by bridging to give carceplex 2b»guest 2 (Figure 5.5). Alternatively, deprotection could be done prior to guest exchange. This procedure may allow for incarceration of photolabile guest molecules such as a-pyrone, which could be subsequently photolysed to give cyclobutadiene. Although this fascinating chemistry has already been performed inside 321 hemicarceplex 107'guest,11 the advantage of our system is the permanent incarceration of this reactive species. For example, the permanent incarceration of cyclobutadiene in carceplex 2a«cyclobutadiene may allow for the determination of its crystal structure which may lead to further insight into the nature of highly antiaromatic molecules. Alternatively, this procedure may be used to form the first-ever carcerand (a carcerand is a carceplex without a entrapped guest molecule). Figure 5.5. An Alternative Route to Carceplex 2a*Guest. 120a, R = H 135, R = CH 2Ph hemicarceplex 136»guest 134, R = CH 2Ph R = CH 2Ph It Y carceplex 2b»guest hemicarceplex 136«guest R = CH 2Ph 322 C. References 1. Nakamura, K.; Sheu, C ; Constable, A. E.; Houk, K. N.; Chapman, R. G.; Sherman, J. C. J. Am. Chern. Soc. 1997, (in press). 2. Rebek, J., Jr. Chern. Rev. 1996, 96, 255-263. 3. Hamann, B. C ; Shimizu, K. D.; Rebek, J., Jr. Angew. Chern. Int. Ed. Engl. 1996,55, 1326-1329. 4. Kang, J.; Rebek, J. Jr. Nature 1996, 382, 239-241. 5. Wyler, R.; de Mendoza, J.; Rebek, J., Jr. Angew. Chern. Int. Ed. Engl. 1993, 32, 1699-1701. 6. Bonar-Law, R. P.; Sanders, J. K. M. J. Am. Chern. Soc. 1995,117, 259-271. 7. Kikuchi, Y.; Tanaka, Y.; Sutarto, S.; Kobayashi, K.; Toi, H.; Aoyama, Y. J. Am. Chern. Soc. 1992,114, 10302-10306. 8. Cram, D. J.; Karbach, S.; Kim, Y. H.; Baczynskyj, L.; Marti, K.; Sampson, R. M.; Kalleymeyn, G. W. J. Am. Chern. Soc. 1988,110, 2554-2560. 9. Huisman, B.-H.; Rudkevich, D. M.; van Veggel, F. C. J. M.; Reinhoudt, D. N. J. Am. Chern. Soc. 1996,118, 3523-3524. 10. Cram, D. J.; Cram, J. M. Container Compounds and Their Guests; The Royal Society of Chemistry: Cambridge, 1994; Vol. No. 4. 11. Cram, D. J.; Tanner, M. E.; Thomas, R. Angew. Chern. Int. Ed. Engl. 1991, 30, 1024-1027. 323 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059500/manifest

Comment

Related Items