Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigations into the formation of [4] cavitand derived hosts that encapsulate multiple guest molecules Makeiff, Darren A. 2003

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

Item Metadata

Download

Media
831-ubc_2003-853888.pdf [ 33.18MB ]
Metadata
JSON: 831-1.0061215.json
JSON-LD: 831-1.0061215-ld.json
RDF/XML (Pretty): 831-1.0061215-rdf.xml
RDF/JSON: 831-1.0061215-rdf.json
Turtle: 831-1.0061215-turtle.txt
N-Triples: 831-1.0061215-rdf-ntriples.txt
Original Record: 831-1.0061215-source.json
Full Text
831-1.0061215-fulltext.txt
Citation
831-1.0061215.ris

Full Text

Investigations into the Formation of [4]Cavitand Derived Hosts that Encapsulate Multiple Guest Molecules by Darren A . M a k e i f f B.Sc, University College of the Cariboo, 1996 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Chemistry) We accept this thesis as conforming tf^mTrtsguired standard THE UNIVERSITY OF BRITISH C O L U M B I A January, 2003 © Darren A. Makeiff, 2003 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. The University of British Columbia Vancouver, Canada Department DE-6 (2/88) Abstract The main goals of the thesis work is to investigate the formation and applications of a class of host-guest molecules known as carceplexes and hemicarceplexes. Carceplexes are large, globe-shaped molecules that permanently entrap smaller guest molecules within their inner confines. Hemicarceplexes are similar to carceplexes, but differ in that hemicarceplexes possess larger pores though which guest egress is possible with sufficient heating. Template effects in the formation carceplexes/hemicarceplexes were examined. One study involves investigations into the role of single-molecule guests as templates in the formation of the "two-bowl" hemicarceplex 25»guest. A second study investigates the role of single and multiple-molecule guests as templates in the formation of a larger "three-bowl" carceplex 56«guests. Complexation of various guests that were determined to be suitable templates to transition state models were also examined to help elucidate the driving forces in the formation of 56*guests. Restricted conformational and orientational mobility of several incarcerated guests, as well as the reversible of complexation of water to 56«guests are described. 25«guest S6*guests The application of carceplex 56«guests in the generation and stabilization of the simple enol, acetophenone enol (139), as a guest in 56 was also studied. Enol 139 was successfully generated and is remarkably stable as an incarcerated guest in the absence of water, even at high i i temperatures. Ketonization of the entrapped enol is facilitated by water, and the observed rate of reaction is significantly slower than for the free enol 139 in solution. The mechanism of ketonization of the entrapped enol 139 is also reported. OH 139 A new "six-bowl" carceplex, which entraps seven DMSO guest molecules (149«(DMSO)7) was also synthesized and characterized. Dynamic behavior for both the host shell and its entrapped DMSO guests was observed in organic solutions by *H N M R spectroscopy. 149«(DMSO)7 iii Abstract The main goals of the thesis work is to investigate the formation and applications of a class of host-guest molecules known as carceplexes and hemicarceplexes. Carceplexes are large, globe-shaped molecules that permanently entrap smaller guest molecules within their inner confines. Hemicarceplexes are similar to carceplexes, but differ in that hemicarceplexes possess larger pores though which guest egress is possible with sufficient heating. Template effects in the formation carceplexes/hemicarceplexes were examined. One study involves investigations into the role of single-molecule guests as templates in the formation of the "two-bowl" hemicarceplex 25«guest. A second study investigates the role of single and multiple-molecule guests as templates in the formation of a larger "three-bowl" carceplex 56»guests. Complexation of various guests that were determined to be suitable templates to transition state models were also examined to help elucidate the driving forces in the formation of 56»guests. Restricted conformational and orientational mobility of several incarcerated guests, as well as the reversible of complexation of water to 56«guests are described. The application of carceplex 56»guests in the generation and stabilization of the simple enol, acetophenone enol (139), as a guest in 56 was also studied. Enol 139 was successfully generated and is remarkably stable as an incarcerated guest in the absence of water, even at high 25*guest 56»guests temperatures. Ketonization of the entrapped enol is facilitated by water, and the observed rate of reaction is significantly slower than for the free enol 139 in solution. The mechanism of ketonization of the entrapped enol 139 is also reported. OH 139 A new "six-bowl" carceplex, which entraps seven DMSO guest molecules (149»(DMSO)7) was also synthesized and characterized. Dynamic behavior for both the host shell and its entrapped DMSO guests was observed in organic solutions by ] H N M R spectroscopy. 149»(DMSO)7 Table of Contents Abstract i i Table of Contents iv List of Charts x List of Schemes xi List of Figures xiii List of Tables xvi List of Abbreviations xix Acknowledgements xxii 1. The Formation of Molecular Containers and Their Applications 1 1.1 Definitions, Concepts and A Selected Historical Account of "Molecules Within Molecules" 1 1.1.1 Supramolecular Chemistry 1 1.1.2 Template Effects 2 1.1.3 The Evolution of Carceplexes and Hemicarceplexes From "Cyclophane-Type" Hosts 3 1.1.4 The Inner Phase of "Molecules Within Molecules" 6 1.2 Temptation In the Formation of Carceplexes and Hemicarceplexes 8 1.2.1 Acetal-bridged Carceplexes and Hemicarceplexes 8 1.2.1.1 The First Soluble Carceplex and Hemicarceplex 8 1.2.1.2 Template Ratios In The Formation of Acetal-Bridged Carceplex and Hemicarceplexes 9 1.2.1.3 [4]Cavitand-Based Self-Assembling Capsules 10 1.2.1.4 Mechanism of Formation for Acetal-Bridged Carceplex 10a«Guest .... 13 1.2.2 Synthesis of a Benzylthia-Bridged Carceplex 14 1.2.3 Hemicarceplexes With Four Slotted Portals 16 1.2.4 Calix[4]arene-[4]Cavitand Hybrid Carceplexes 19 1.2.5 Metal-Ligand Bridged Carceplexes 21 1.3 Container Molecules That Entrap Multiple Guests 22 1.3.1 Benzyl Thia-Bridged Carceplex 24«(Guest)2 and Trismethylene-Bridged Hemicarceplex l l»(Guest) 2 (Revisited) 22 1.3.2 Larger Carceplexes Derived From Cyclic Arrays of [4]Cavitands 24 1.3.3 Synthesis of Bis(carceplexes) and Bis(complexes) 25 1.3.4 Bis-Bridged Hemicarcerand Dimers 26 1.3.5 Synthesis of a Trimer Carceplex Containing Three D M F Molecules 28 1.3.6 The First C5-Symmetric Disulfide-Bridged Carceplex 29 1.3.7 Rebek's "Sports-balls" Capsules 31 1.3.8 Cylindrical-Shaped Capsules 35 1.3.9 Molecular Paneling • 37 1.3.10 Hexameric Resorcin[4]arene Capsules 41 1.4 The Inner Phase Reactivity of Hemicarceplexes 44 1.4.1 Acid/Base reactions 44 1.4.2 Oxidation/Reduction Reactions 46 1.4.3 Alkylation Reactions 47 1.4.4 Generation and Stabilization of Cyclobutadiene 49 1.4.5 Generation and Stabilization of ortho-Benzyne 50 1.4.6 Generation and Stabilization of Cycloheptatriene 51 1.5 Conclusions 55 1.6 References 57 2. Template Effects In the Formation of a Tetramethylene-bridged Hemicarceplex 67 2.1 Introduction 67 2.2 Results and Discussion 69 2.2.1 Reaction Conditions/Synthesis of Hemicarceplexes 25»guest 69 2.2.2 Unsuccessful templates 70 2.2.3 Successful Templates 72 2.2.4 Guest Orientation and Mobility 73 2.2.5 Determination of Template Ratios 75 2.2.5.1 Competition experiments 75 2.2.5.2 Control Experiments 78 2.2.5.3 General Trends in Templating Abilities 78 2.3 Conclusions 81 2.4 Experimental Section 85 2.4.1 General Experimental 85 2.4.2 Synthesis of Hemicarceplexes 25»guest 86 2.4.2.1 General Templating Procedure for the Synthesis of 25»guest 86 v 2.4.2.2 General Guest Exchange Procedure for the Synthesis of 25*guest • 88 2.4.3 Competition experiments 103 2.4.4 Template Ratios 104 2.4.5 Control experiments 107 2.5 References 109 3. Investigation into the Formation of a Carceplex Derived From Three Cavitand Subunits I l l 3.1 Introduction H I 3.2 Improved Synthesis of the A,C-Trimer 42 113 3.3 Synthesis and Characterization of Trimer Carceplexes 56»Guests 114 3.3.1 Carceplexes From Neat Solvent 114 3.3.2 Trimer Carceplex Water Complexes 116 3.3.2.1 Characterization of Complexes 56«[Guests»(H20)>,] 116 3.3.2.2 Trimer Carceplex Complexes 56«[(DMA) 2«(H 20)>,] 118 3.3.2.3 Effects of Acid and Base in the Complexation of Water 121 3.3.3 Single-Molecule Guest Carceplexes 123 3.3.3.1 Synthesis and Characterization 123 3.3.3.2 Trimer Carceplex 56»trimethyl-1,3,5-benzenetricarboxylate 127 3.3.4 Carceplexes 56»(NFP«Guest) 131 3.3.4.1 Carceplexe 56»(NFP»DMSO) 131 3.3.4.2 Carceplexes 56«(NFP»Guests) (Guest = Aryl Ketones) 134 3.4 Templation Studies 134 3.4.1 Single-Molecule Guest Competitions 134 3.4.2 Template Ratios for 56«(NFP«Guest) 137 3.4.2.1 Competitions Between Two-Molecule Pairs in the Formation of56»(NFP«Guest) 137 3.4.2.2 Single-Molecule Templates Versus NFP»Guest 139 3.4.3 Other TR22s, 77?23s, and 77?33s 140 3.4.4 r / ? i 3 s ( a s w e l l a s r ^ i 2 s a n d r i ? 2 3 s ) 141 3.4.4.1 Competitions Between Tris-Acetylene 116, (DMSO) 3 , and (DMF) 3 141 3.4.4.2 Temperature Effects on 77? 13, TRn, and 77*23 143 3.5 Complexation of Single-Molecule Guests 146 3.5.1 Trimer Complexes 42«Guest and 43«Guest 146 3.5.1.1 Potential Transition State Models for the GDS 146 3.5.2.2 Structure, Binding, and Dynamic Features of Trimer and Trimer Cavitands and Their Complexes 155 vi 3.5.2.3 The Stabilities of Trimer Cavitand Complexes 114»Guest, 125»Guest, and 116«Guest 161 3.5.2.4 Complexation/Decomplexation Rates for Complexes 42»116 andl25»116 164 3.5.2.5 meta-Xy\y\ Capped Trimer Cavitand 127 165 3.6 Complexation of Multiple-Molecule Guests To Trimer 42 168 3.7 Summary 171 3.8 Experimental Section 174 3.8.1 General Experimental 174 3.8.2 M A L D I Mass Spectrometry 174 3.8.3 Synthesis and Characterization 175 3.8.3.1 A,C-Trimer Derivatives 175 3.8.3.2 Single-Guest Trimer Carceplexes (56«guest and 115»guest) 182 3.8.3.3 Two-Guest Carceplexes (56«(guest)x and 115«(guest)x, x = 2) 190 3.8.3.4 Two-Guest Carceplexes (56«(NFP-guest)) 194 3.8.3.5 Three-Guest Carceplexes (56»(guest)x and HS^guest)*, x = 3) 198 3.8.4 N M R spectroscopy 202 3.8.4.1 General 202 3.8.4.2 Coalescence temperature (r c) measurements 203 3.8.4.3 ID E X S Y Experiments 204 3.8.5 Template Ratios 211 3.8.5.1 Competition Experiments 211 3.8.5.2 Solvent Competitions 214 3.8.5.3 Temperature Dependence of the Template Ratios 77?i3, TR\2, and TR2i 215 3.8.6 Host-Guest Complexes with Trimer Derivatives 218 3.8.6.1 General Complexation Experiments 218 3.8.6.2 A,C-Trimers 41 and 42 219 3.8.6.3 Trimer Cavitand Complexes 225 3.9 References 228 4. The Generation, Stabilization, and Ketonization of Acetophenone Enol as a Guest in Trimer Carceplex 56»Guests 234 4.1 Introduction 234 4.1.1 The Importance of Enols 234 4.1.2 Simple Enols 235 4.1.3 The Norrish Type II Photocleavage of Aryl Ketones 236 4.1.4 Okazaki's Endohedral Enol 237 4.1.5 Objectives of This Chapter 238 4.2 Synthesis and Characterization of Trimer Carceplexes with Aryl Ketone Guests 239 4.3 Photolysis of Carceplexes 56*121 and 56*123 240 4.3.1 Photolysis of Carceplex 56-121 240 4.3.2 Photolysis of Carceplex 56-123 244 4.4 The Stability of Acetophenone Enol in the Inner Phase of 56-(139»140) 246 4.4.1 Objectives for the Ketonization of 56-(139»140) 246 4.4.2 Ketonization of Acetophenone Enol in Carceplex 56»(139«140) 247 4.4.3 Ketonization in Nitrobenzene 250 4.4.3.1 The Rate of Ketonization with H 2 0 250 4.4.3.2 The Rate of Ketonization with D 2 0 251 4.4.3.3 H/D Exchange of the Enol Hydroxyl Proton by D 2 0 in Nitrobenzene 252 4.4.3.4 The Formation of Protio Acetophenone in Carceplex 56«(142»140) with D 2 0 in Nitrobenzene at 100 °C 253 4.4.4 The Accepted Mechanism for the "Uncatalyzed" Ketonization of Acetophenone Enol in Aqueous Solution 255 4.4.5 The Mechanism for the Ketonization of Acetophenone Enol in 56»(139«140) by Water 257 4.4.6 Ketonization in Benzene 259 4.4.6.1 Ketonization by H 2 0 259 4.4.6.2 The Effect of Acid 260 4.4.6.3 The Effect of Base 261 4.5 Attempts To Generate Other Enols Inside Carceplex 56»Guests 262 4.6 Summary 263 4.7 Experimental • 265 4.7.1 General 265 4.7.2 Synthesis and Characterization 266 4.7.3 Ketonization Experiments 269 • 4.8 References 278 viii 5. A Hexa-Cavitand Derived Carceplex That Entraps Seven Guest Molecules 282 5.1 Introduction • • 282 5.2 Synthesis of A,B-Trimer 152 284 5.3 Complexation Experiments Involving A,B-Trimer 152 286 5.4 The Formation of 149»(DMSO)7 288 5.5 N M R Spectroscopy of Carceplex 149«(DMSO)7 291 5.5.1 Conformation of 149«(DMSO)7 in Dry CDC1 3 291 5.5.2 Host dynamics 298 5.5.2.1 Host dynamics in Dry CDCI3 298 5.5.2.2 Host dynamics in Dry CD2CI2 302 5.5.3 Hydration of 149«(DMSO)7 in CDCI3 304 5.5.4 Summary of the Observed Conformational Interconversion for 149«(DMSO)7 307 5.5.5 Hydration of 149«(DMSO)7 in CD 2 C1 2 308 5.5.6 Dehydration of 149«(DMSO)7 in CDC1 3 and CD 2 C1 2 311 5.5.7 ' H N M R Guest Mobility Studies in CDCI3 and CD 2 C1 3 312 5.6 Bound DMSO in 149«(DMSO)7 Compared to Other Carceplexes in CDCI3 319 5.7 Entrapment of Guests Other than DMSO as Carceplex 149«guests 320 5.8 Summary 326 5.9 Outlook 330 5.10 Experimental 333 5.10.1 General Experimental 333 5.10.2 Synthesis and Characterization 333 5.10.3 H 2 0 ' H N M R titration experiments 337 5.10.4 2D ROESY Versus NOESY for 149«(DMSO)7 in Dry CDC1 3 338 5.10.5 ID E X S Y experiments 338 5.11 References 343 ix List of Charts Chart 2.1 Unsuitable guests for the templated formation of hemicarceplex 25«guest 71 Chart 3.1 Single molecule guests 125 Chart 3.2 Guests that did not form complexes with 41 148 x List of Schemes Scheme 1.1 Metal ion templated formation of 18-crown-6 3 Scheme 1.2 Synthesis of an acetal-linked carceplex and hemicarceplex 8 Scheme 1.3 Formation of charged dimeric tetrol complexes 11 Scheme 1.4 Mechanism for the guest-templated formation of acetal-bridged carceplex 10a»guest • 13 Scheme 1.5 Synthesis of benzylthia-bridged carceplex 24»guest 15 Scheme 1.6 Synthesis of Reinhoudt's carceplex 38»guest 19 Scheme 1.7 Formation of metal-ligand bridged carceplex 40«guest 22 Scheme 1.8 Synthesis of cyclic trimer and tetrameric [4]cavitand arrays 25 Scheme 1.9 Synthesis of bis(carceplex) 46»(guest)2 and bis(complex) 47»(guest)2 26 Scheme 1.10 Schematic representation of the synthesis of bis-hemicarceplex dimers 27 Scheme 1.11 Synthesis of trimer carceplex 56«(DMF)3 28 Scheme 1.12 Formation of disulfide-bridged carceplex 58«guests 29 Scheme 1.13 Formation of Rebek's "Tennis" ball 32 Scheme 1.14 Displacement of solvent from the interior of capsule 622 by a single-molecule guest 33 Scheme 1.15 Diels-Alder reaction within dimer capsule 622 33 Scheme 1.16 Turnover cycle for the Diels-Alder reaction catalyzed by Rebek's "softball" 34 Scheme 1.17 Synthesis of molecular paneled hosts 38 Scheme 1.18 Formation of hydrophobic dimers as guests in hosts 75a/76a 39 Scheme 1.19 Oxidation/reduction of guests entrapped in 25«guest 47 Scheme 1.20 Inner phase reactions conducted within hemicarcerand 11a 50 Scheme 1.21 Generation of ort/zo-benzyne (104) in hemicarcerand 25 51 Scheme 1.22 Generation and stabilization 106a»cycloheptatrienes 52 Scheme 1.23 Enantiomerization of cycloheptatriene 53 Scheme 1.24 Intermolecular inner phase reactions with 106a«110 54 Scheme 1.25 Proposed mechanism of intermolecular inner phase reaction between 0 2 with incarcerated cycloheptatriene 55 Scheme 2.1 Capsule 12a«guest as a transition state model in the formation of carceplexes/hemicarceplexes 68 Scheme 2.2 Templated synthesis of hemicarceplex 25»guest 70 Scheme 2.3 Proposed mechanisms in the formation of 25»guest and 10a«guest 82 Scheme 3.1 Synthesis of trimer carceplex 56»guest from tetrol 8a 112 Scheme 3.2 Conformational interconversion processes for 117 129 Scheme 3.3 DMSO mobility in 56»(NFP«DMSO) 133 Scheme 3.4 Speculated mechanism and intermediates in the formation of 56«guest 147 Scheme 3.5 Conformational equilibria for trimer cavitand derivatives 156 Scheme 4.1 Norrish II photocleavage/cyclization of butyrophenone (121) 237 Scheme 4.2 Biradical intermediate geometries and their chemical consequences 237 Scheme 4.3 Synthesis of trimer carceplex 56»guests 239 xi Scheme 4.4 Proposed mechanisms for the "uncatalyzed" ketonization of acetophenone enol by water 256 Scheme 4.5 Ketonization of 56«(139»140) with water 258 Scheme 4.6 Photohydration of 116 263 Scheme 5.1 Synthesis of A,B-trimers 285 Scheme 5.2 A,B-trimer 152 coupling reaction 289 Scheme 5.3 Interconversion between degenerate flattened conformations (A-D) for 149»(DMSO) 7 299 Scheme 5.4 Schematic representation of the fluxional behavior of 149»(DMSO)7 in solution 308 List of Figures Figure 1.1 Generic structures of various neutral aromatic organic hosts 4 Figure 1.2 Classification of inner molecular reactions 7 Figure 1.3 Self-assembling capsules derived from [4]cavitands 12 Figure 1.4 Hemicarceplexes 25-27«guest 15 Figure 1.5 weto-Xylyl-bridged hemicarceplexes 28-33 and precursor [4]cavitands 34-36 .... 17 Figure 1.6 Monomers for Rebek's "Sports balls" 31 Figure 1.7 Cylindrical capsules from tetraimide-bridged cavitand dimers 36 Figure 1.8 Fujita's coordination nanobowl 80 41 Figure 1.9 Hexameric rescorin[4]arene capsules 42 Figure 1.10 Orientations of aryl phenol guests and transition state of alkylation through the equatorial portals of 25-guest 48 Figure 1.11 Hemicarcerands used in cycloheptatriene studies 52 Figure 2.1 Predicted guest orientations of substituted benzene and aliphatic alcohol in 25»guest 74 Figure 3.1 Expanded regions of ' H N M R spectra (400 MHz) of 56«(DMA)2 119 Figure 3.2 Proton peak labels for trimer carceplex 56«guest 119 Figure 3.3 ! H N M R spectra (500 MHz, H20-saturated C 6 D 6 ) of the acetyl methyl protons of 56«[(DMA)2«(H20)^] 123 Figure 3.4 M M 2 minimized space filling models of carceplexes 56»guest 126 Figure 3.5 ' H N M R spectra (500 MHz) of 56*117 128 Figure 3.6 Symmetry of the host imposed by the guest in 56«117 130 Figure 3.7 1 H N M R spectra (400 MHz, CD 2C1 2) of 56-(NFP«DMSO) in CD 2 C1 2 at various temperatures 132 Figure 3.8 M M 2 minimized structures of trimer carceplexes 56«(NFP«aryl ketone) 138 Figure 3.9 *H N M R (nitrobenzene-d5, 300 K) spectra of A,C-trimers 42 and 43 and complexes with tris-acetylene 116 150 Figure 3.10 A,C-Trimers 42 and 43 150 Figure 3.11 Plot of ln(^ s) against \n(TR\0 153 Figure 3.12 Trimer cavitands 114,125 and 126 158 Figure 3.13 ' H N M R (400 MHz, nitrobenzene-^, 300 K) spectra of trimer cavitand derivatives 159 Figure 3.14 *H N M R spectra of trimer cavitand 125 161 Figure 3.15 ' H N M R (500 MHz, nitrobenzene-^, 300 K) spectra of trimer cavitand 125 and complexes 162 Figure 3.16 Bis-capped trimer cavitands 166 Figure 3.17 Expanded regions of *H N M R spectra of complex 127»116 in nitrobenzene-c?5 at various temperatures 167 Figure 3.18 Guest mobility in complex 127«116 169 Figure 3.19 ' H N M R spectra (400 MHz, toluene-^, 300 K) of trimer 42 with DMSO 169 Figure 3.20 ' H N M R spectra (400 MHz, toluene-<i8) of trimer 42 with DMSO at various temperatures 170 Figure 3.21 Proton labels and structure of 127 181 Figure 3.22 ] H N M R spectra of 127 and its complexes 182 xiii Figure 3.23 Plot of X(lnA)X _ 1 matrix elements (a\2 and a2\) versus mixing time (V) for complex 42»116 (nitrobenzene-^, 330 K) 208 Figure 3.24 Plot of X(lnA)X _ 1 matrix elements (a\2 and a2\) vs. mixing time for complex 125*116 (nitrobenzene-J5, 330 K) 209 Figure 3.25 Plot of X(lnA)X _ 1 matrix elements (an and 021) versus mixing time for56»117 (CD 2 Cl 2 ,267K) 210 Figure 3.26 Plot of ln(77?i3) versus 1/7/for 116/(DMSO)3 competitions 216 Figure 3.27 Plot of ln(77?i2) versus 1/7/for 116/(NFP«DMSO) competitions 217 Figure 3.28 Plot of ln(77?23) versus l/T for (NFP»DMSO)/(DMSO) 3 competitions 217 Figure 4.1 Okazaki's endohedral cavitand derivatives 238 Figure 4.2 ! H N M R spectra of 56*121 and 56»(139*140)/56*141 241 Figure 4.3 Intra and intermolecular NOEs observed between 139 and 140 in carceplex 56»(139»140) 243 Figure 4.4 M M 2 minimized structures of carceplexes 56«121 and 56*123 246 Figure 4.5 *H N M R spectra (400 MHz, CDCI3, 300 K) of carceplexes 56«(139«140)/56»141 after various stages of tautomerization 249 Figure 4.6 Arrhenius plot (-ln(Xe) versus time) for the ketonization of 56»(139»140) in D 2 0 saturated nitrobenzene at 100 °C 252 Figure 4.7 Disappearance of the enol hydroxyl proton with time in the presence of D 2 0 253 Figure 4.8 Expanded region of *H N M R spectra (CDC13) of carceplex of 56»(139«140) after heating at 100 °C under various conditions 254 Figure 4.9 Arrhenius plots (-ln(Xe) versus time) for the ketonization of 56«(139«140) with H 2 0 in nitrobenzene at 100 °C 273 Figure 4.10 Arrhenius plots (-ln(%e) versus time) for the ketonization of 56»(139»140) with D 2 0 in nitrobenzene at 100 °C 275 Figure 4.11 Arrhenius plot for the H/D exchange of the enol hydroxyl proton in 56»(139*140) in D 2 0 saturated nitrobenzene at 100 °C 276 Figure 4.12 *H N M R spectra (400 MHz) before and after photolysis in H 2 O saturated C 6 D 6 277 Figure 5.1 *H N M R spectra (400 MHz, 300 K) of A,B-trimer 152 286 Figure 5.2 Hydrogen-bonded capsules from A,B-trimer 152 287 Figure 5.3 lH N M R (500 MHz, nitrobenzene-^, 300 K) spectra of A,B-trimer 142 288 Figure 5.4 M A L D I MS spectrum of carceplex 149«(DMSO)7 290 Figure 5.5 *H N M R spectrum (400 MHz, dry CDC1 3, 300 K) of 149»(DMSO)7 292 Figure 5.6 M M 2 minimized space filling representations of the C 3 v conformation of 149»(DMSO)7 293 Figure 5.7 Schematic representation of the flat conformation of 149«(DMSO)7 (top view down the C 3 v axis) dry CDC1 3 solution 295 Figure 5.8 2D ROESY spectrum of 149»(DMSO)7 at 250 K in dry CDC1 3 296 Figure 5.9 2D ROESY spectrum of 149»(DMSO)7 at 300 K in dry CDC1 3 297 Figure 5.10 Sections of *H N M R spectra (500 MHz) of 149«(DMSO)7 in pyridine-^ 300 Figure 5.11 M M 2 minimized space filling representations of the Oh-symmetric conformation of 149«(DMSO)7 301 xiv Figure 5.12 Sections of ' H N M R (400 MHz, dry CD 2C1 2) spectra of 149-(DMSO)7 at different temperatures -303 Figure 5.13 Sections of *H N M R spectra (400 MHz, CDC1 3 , 300 K) in the presence of 149»(DMSO)7 at varying ratios of H20:149«(DMSO)7 305 Figure 5.14 Sections of ! H N M R spectra of 149«(DMSO)7 (400 MHz) in H20-sarurated CDCI3 at different temperatures 306 Figure 5.15 *H N M R spectra (400 MHz, CD 2 C1 2 , 300 K) of 149*(DMSO)7 309 Figure 5.16 DMSO methyl proton signals for 149«(DMSO)7 in ! H N M R spectra (400 MHz, CD 2 C1 2 , 300 K) at different H 2 0 concentrations 310 Figure 5.17 [ H N M R spectra of 149»(DMSO)7 solutions after extended drying (>48 h) 312 Figure 5.18 Schematic representation of proposed DMSO methyl environments (a, b, and c) in 149-(DMSO)7 314 Figure 5.19 Sections of ' H N M R spectra (400 MHz, CDCI3) of 149«(DMSO)7 at different temperatures and conditions 315 Figure 5.20 Sections of ' H N M R spectra (500 MHz) of 149«(DMSO)7 in CD 2 C1 2 at various temperatures 316 Figure 5.21 Full ' H N M R spectra (500 MHz, H20-saturated CD 2C1 2) of 149»(DMSO)7 at various temperatures 318 Figure 5.22 M A L D I MS spectra of 149«[(DMSO)^-(guest)>,] mixtures (guest = D M A , DMF) 321 Figure 5.23 ' H N M R spectra (400 MHz, sieve-dried CDCI3, 300 K) of carceplex 149-[(DMSO)x»guests] 322 Figure 5.24 ID N M R spectra (400 MHz, sieve-dried CDCI3) of 149«(DMSO)7 at 300 K 339 Figure 5.25 Plot of rate (ay) versus mixing time (in s) for 149»(DMSO)7 in sieve-dried CDCI3 at 300 K 341 xv List of Tables Table 1.1 Selected template ratios for the formation of acetal-bridged carceplex 10a»guest and hemicarceplex lla»guest 10 Table 1.2 Templating ability and yields for the formation of Reinhoudt's carceplex 38«guest 20 Table 1.3 Energy barriers for amide methyl interconversion for free and incarcerated DMF in various sized carceplexes in nitrobenzene-Js at 77 °C 30 Table 2.1 Template ratios in the formation of 25»guest 77 Table 2.2 Conditions for the synthesis of hemicarceplexes 25»guest using procedure B 89 Table 2.3 ! H N M R chemical shifts for free and bound guests of hemicarceplex 25«guest in CDCI3 at ambient temperature 99 Table 2.4 Competition experiment results 105 Table 2.5 Crosscheck experiments results 106 Table 2.6 Control experiment results 108 Table 3.1 ] H N M R chemical shift data for bound H 2 0 in several carceplexes 56«[guests«(H20)>,] in CDC1 3 at 300 K 117 Table 3.2 ' H N M R chemical shifts (CDC13) for various 1,3,5-trisubstituted benzene derivative guests in 56«guest 124 Table 3.3 Template ratios (TR\\, unitless) in the formation of 56»guest from single-molecule templates 136 Table 3.4 Template ratios (TR22) for 56«(NFP»guest) 138 Table 3.5 Template ratios (77?i2 (G/NFP»GB)) for single-molecules versus NFP»aryl ketone (NFP«GB) at 70 °C 140 Table 3.6 TRs (7Y?22, 77?32, 77?33) for multiple-molecule templates in the formation of 56«guests 141 Table 3.7 Template ratios (77?i3) for (DMSO) 3 and (DMF) 3 against tris-acetylene 116 142 Table 3.8 Dependence of the template ratios (77?i3, TRX2 and 77?23) on temperature (7) 144 Table 3.9 Thermodynamic/kinetic values for template ratios for one versus, three, one versus, two, and two vs. three-molecule templates 145 Table 3.10 Stability constants (Ks) for complexes 42*116 and 43«116 in various deuterated solvents 152 Table 3.11 Stability constants (Kss) for 42-guest (nitrobenzene-d5, 300 K) 153 Table 3.12 ' H N M R chemical shift data for interbowl acetal protons of A,C-trimer derivatives in different solvents at 300 K 157 Table 3.13 Stability constants for trimer cavitand complexes and various guests (nitrobenzene-^, 300 K) 163 Table 3.14 Stability (Ks) and rate (kc, kd) constants for complexes 42»116 and 125»116 (nitrobenzene-^, 330 K) 164 Table 3.15 Stability constants for trimer derivative complexes and template ratios (TR\\) for carceplex 56«guests with various guests 172 Table 3.16 Chemical shift differences (A5 = 5b0und-8free) for protons of various guests in 56»(guest)x 200 Table 3.17 *H N M R chemical shifts (CDCI3) for aryl ketone guests in 56»(NFP»guest) 201 xvi Table 3.18 Additional ID E X S Y data for complex 42*116 (nitrobenzene-d5, 330 K) 206 Table 3.19 Additional ID E X S Y data for complex 125*116 (nitrobenzene-c/5, 330 K) 208 Table 3.20 Additional ID E X S Y data for trimer carceplex 56*117 (CD 2C1 2 , 267 K) 210 Table 3.21 Single-guest competition experiments at room temperature 211 Table 3.22 Single-guest competition experiments at 70 °C 212 Table 3.23 Competition results for 56*(NFP*guest) at 70 °C 212 Table 3.24 Single- versus Two-molecule template competition results at 70 °C 213 Table 3.25 M A L D I mass spectrometric data on carceplex product mixtures in different binary solvent mixtures at room temperature 214 Table 3.26 *H N M R chemical shifts (ppm) of 42 in various deuterated solvents at 300 K 219 Table 3.27 ' H N M R chemical shifts (ppm) for selected protons of 42*116 in various deuterated solvents at 300 K 220 Table 3.28 ' H N M R chemical shifts (ppm) of complex 43*116 in various deuterated solvents at 300 K 220 Table 3.29 *H N M R chemical shifts of selected protons of complexes 42*guest (nitrobenzene-ds, 300 K) 221 Table 3.30 Additional data for Ks measurements for complexes 42*116 and 43*116 222 Table 3.31 Additional data for Ks measurements for complexes 42*guest in nitrobenzene-^ 223 Table 3.32 Additional data for relative stability measurements (KK{) of 42*guest in nitrobenzene-^ 224 Table 3.33 ' H N M R chemical shifts (ppm) of trimer cavitand 125 and complex 125*116 in (nitrobenzene-ci5, 300 K) 225 Table 3.34 ' H N M R chemical shifts (ppm) of trimer cavitand 126 and complex 126*116 in nitrobenzene-^ 226 Table 3.35 ' H N M R chemical shifts (ppm) of trimer cavitand 114 and complex 114*116 in nitrobenzene-iis 226 Table 3.36 Additional data for Ks measurements for complexes 114*116 ,125*116 and 126*116 in nitrobenzene-^ at different temperatures (T) 227 Table 4.1 Chemical shift data (CDC1 3, 300 K) for guest protons of butyrophenone (121) and hexanophenone (123) 245 Table 4.2 The results of ketonization experiments under various conditions 260 Table 4.3 2D N O E S Y data for carceplexes 56*(139*140) and 56*141 268 Table 4.4 Additional data for single time (t) ketonization experiments with H 2 0 in nitrobenzene at 100 °C 270 Table 4.5 Additional data for multiple time (0 ketonization experiments with H 2 0 in nitrobenzene at 100 °C 272 Table 4 .6 Additional data for ketonization experiments with D 2 0 in nitrobenzene at 100 °C 274 Table 4.7 Additional data for H/D exchange of the enol hydroxyl proton in 56*(139*140) with D 2 0 in nitrobenzene at 100 °C 276 Table 5.1 Chemical shift differences (A6) between free (8free) and bound (5b0Und) DMSO for various carceplexes and hemicarceplexes in CDCI3 (300 K) 319 Table 5.2 ' H N M R chemical shift data for DMF and D M A in various environments 324 Table 5.3 * H N M R chemical shift data for 149»[(DMSO)x»(DMA)^] and 149»[(DMSOWDMF)J 325 Table 5.4 Template ratios for 149«[(DMSO)Jt«(DMF)J,] and 149»[(DMSO)x»(DMA)>,] 326 Table 5.5 Relative integration values from ' H N M R spectra of 149«(DMSO)7 in H 2 O and D20-saturated CDCI3 336 Table 5.6 ID E X S Y integration intensities for 149«(DMSO)7 sieve-dried CDC1 3 at 300 K 340 Table 5.7 X( lnA)X _ 1 matrix elements (ay') calculated at various mixing times (tm) 340 Table 5.8 Rate constants measured between Hn/Hi2 /Hi3 in dry C D C I 3 at 300 K 342 List of Abbreviations atm - atmospheres P - overall binding constant cald - calculated CIDNP - Chemically Induced Dynamic Nuclear Polarization COSY - Correlation Spectroscopy C P K - Cory-Pauling-Koltun (molecular models) CS2CO3 - cesium carbonate A8 - change in chemical shift (in ppm) A 8 m a x - maximum change in chemical shift (in ppm) 5 - chemical shift (in ppm) G - guest AG* - free energy of activation &GCX - free energy of activation from coalescence Av - frequency difference in Hz d - deuterium (i.e., nitrobenzene-ds) d - day(s) D B U - l,8-diazabicyclo[5.4.0]undec-7-ene DCC - dicyclohexylcarbodiimide D C U - dicyclohexylurea DHB - 2,5-dihydroxybenzoic acid D M A - AyV-dimethylacetimide D M F - A^iV-dimethylformamide xix DMI D M S O D M P U E X S Y equiv. h H/D H M Q C IR J KBr KI AG* GDS ^cat ^obs fcobs (H/D) •^app KKl Ks M M A L D I MS min mlz 1,3-dimethyl-2-imidazolidinone dimethyl sulfoxide 1,3-dimethyl-3,4,5,6-tetrahydro-2( l//)-pyrimidinonone EXchange SpectroscopY equivalents hour(s) hydrogen/deuterium (exchange) Heteronuclear Multiple Quantum filtered Correlation infrared (spectroscopy) coupling constant potassium bromide potassium iodide Gibbs free energy of activation guest determining step rate constant for a "catalyzed" reaction observed rate constant observed rate constant for H/D exchange rate constant for an "uncatalyzed" reaction apparent association constant relative stability constant stability constant parent mass (mass spectra) or molar, moles per liter (concentration) matrix assisted laser desorption ionization mass spectrometry minute(s) mass to charge ratio xx nuclear magnetic resonance N-methylpyrrolidinone Af-formylpiperidine nuclear magnetic resonance (spectroscopy) Nuclear Overhauser Effect Spectroscopy Palladium (0) over carbon parts per million correlation factor Rotating-frame Overhauser Effect via correlation Spectroscopy room temperature second(s) 2,2,2-trichloroacetic acid tetrahydrofuran 2,2,2-trifluoroacetic acid half-life mixing time coalescence temperature trifluoroacetic acid thin layer chromatography tetramethylsilane mixing time template ratio template ratio, x-molecule template versus ^ -molecule template ultra violet (spectroscopy) volume per volume Acknowledgements I would like to thank the following individuals: M y supervisor, Professor John Sherman, for his guidance, insightful discussions, constructive criticisms, and endless patience over the years. Dr. Nick Burlinson and the N M R facility staff for their expertise and advice in using the N M R instruments. Past and present members of the Sherman group for their regular participation in group meetings. I would also like to thank my family and loving girlfriend, and her family for their endless support and patience throughout my years as a graduate student 1. The Formation of Molecular Containers and Their Applications 1.1 Definitions, Concepts, and A Selected Historical Account of "Molecules Within Molecules" 1.1.1 Supramolecular Chemistry Supramolecular chemistry is a rapidly progressing field in chemical research,1 which involves the chemistry beyond the molecule, where non-covalent bonding and spatial fit between 2 3 molecular entities that form a specific host-guest complex are of paramount importance. ' Consequently, the emergent properties of the assembled complex may differ greatly from those exhibited by in its individual components.2 The host is an organic molecule or ion whose binding sites converge in the complex about a smaller guest molecule(s) or ion(s) possessing complementary divergent binding surfaces.2 Molecular recognition is fundamental to the formation of these assemblies where multiple (two or more) carefully designed components are self-assembled in a spontaneous manner through the manipulation of weak noncovalent interactions.2 Although the growth of this field has only spanned the latter half of the last century,1 supramolecular systems in nature have existed and evolved for many millions of years. D N A / R N A , enzymes, receptors, antibodies, membranes, carriers, and channels are a few examples. Recent developments within the field of supramolecular chemistry have been geared towards attempts to unravel the complexity displayed within these natural systems. In order to gain a full understanding of how these sophisticated biological processes work, carefully devised, yet simpler models can help facilitate the study of the specific individual interactions 1 involved. Hence, it must be stressed that the primary aim in studying supramolecular systems is not necessarily to mimic or compete with a particular natural system, but to use nature as an inspiration to gain a thorough understanding of the fundamental principles involved in designing and synthesizing similar systems that possess the same subtle recognition properties. 1.1.2 Template Effects Many important synthetic supramolecular systems possess rigid or flexible (dynamic), two or three dimensional macrocyclic structures. Several fundamental strategies have been developed to efficiently synthesize these types of molecules. Among these include template effects.4'5 In (supramolecular) chemistry, a template is defined as a chemical entity that organizes an assembly of atoms within a specific region of space such that the interacting groups become fixed in some fashion that ultimately controls the overall topology of the product.6 In the extreme, a single product is formed from a reactant or reactants that have the potential to assemble in a variety of different ways. Two fundamental aspects are involved in templation. First, the template must bind to a particular substrate containing the interacting groups to form some type of complex. Second, a reaction must occur with the functional groups in the formed complex.6 The reaction may be reversible or irreversible, giving rise to either thermodynamic or kinetic template effects, respectively.6 For thermodynamic template effects, the more stable complex forms between template (guest) and substrate (host).6 Upon complex formation, the reaction equilibrium is shifted towards the products and the yield is increased.6 For kinetic template effects, the template effectively lowers the energy of the transition state of a reaction to favor a certain reaction path (product).6 Template effects can also act positively by increasing the product yield, 2 or negatively by inhibiting formation of the product. Finally, the template involved in a reaction may be incorporated into the product by covalent bonding or noncovalent interactions, or it may be released after the product is formed. 1.1.3 The Evolution of Carceplexes and Hemicarceplexes From "Cyclophane-Type" Hosts This section will discuss a brief history of the evolution of synthetic molecular hosts often labeled "molecules within molecules". Only selected examples of hosts will be presented, which are most relevant to hosts that are studied in this thesis. Crown ethers are polycyclic ethers (i.e., 1, Scheme 1.1) that bind various sized alkali metal ions within the center of the macrocycle to form highly structured complexes, and are one of the earliest studied and most classically renowned synthetic supramolecular hosts.7'8 Affinity for particular cations is largely dependent on the complementarity between the polycyclic ether cavity and the size of the cation. The first template effects reported were introduced in the metal ion templated synthesis of crown ethers such as 18-crown-6 (1) from the corresponding diol and ditosylate reactants in the presence of a suitably sized metal ion template (Scheme 1.1, M = K ). Scheme 1.1 Metal ion templated formation of 18-crown-6. ;> - -vtO - C-?C k ^ 6 - T s C X ^ J k ^ O - k Q T s 3 The host-guest concept was first introduced with crown ethers in mind, 2 and was extended to many other existing synthetic and natural systems. For example, shortly after crown ethers, host-guest systems were reported between small neutral molecules bound in the cavities of belt-shaped synthetic hosts such as cyclophanes (Figure l . l ) . 9 Cyclophanes are ortho (i.e., compound 7), meta (compounds 3-6), and/?ara-bridged (i.e., structure 2) aromatic macrocycles that possess nonpolar cavities that can bind various neutral aromatic guests utilizing interactions such hydrogen bonding, pi-pi stacking, van der Waals, and hydrophobic effects.9 Larger concave surfaces later appeared, such as the ort/zo-cyclophanes called cyclotriveratrylenes (CTVs), and more importantly the slightly more voluminous and dynamic /weta-cyclophanes, called calixarenes (4).10 These hosts can be appropriately functionalized and can bind varieties of neutral and charged guests in both polar and nonpolar environments. Coupling of two C T V components, yielded the crytophanes (7)," which were the first hosts to fully enclose smaller guests such as Xe, C H 3 , C H 3 C H 3 , C H 2 C H 2 , CH2CI2, and C H C I 3 within their spherical cavities.12 4 6 Figure 1.1 Generic structures of various neutral aromatic organic hosts. Larger hosts were later reported by Cram, 1 1 derived from meta-cyclophanes, also known as [4]calixarenes or resorcin[4]arenes (5), 1 3 which are available via the acid-catalyzed condensation reaction between resorcinols and aldehydes.13 The rigidity of these building blocks could be improved by bridging the adjacent phenols of resorcin[4]arenes (5), to give birth to the [4] cavitands (6), which are bowl-shaped molecules that possess rigid, enforced concave cavities." A wide variety of [4]cavitands with different functional groups (A, Z of 6 in Figure 1.1) are available.11 Intermolecular bridging of functionalized [4]cavitands by reaction with suitable linker groups furnished new families of novel molecular containers called carceplexes and hemicarceplexes (see section 1.2).14 Cram defined carceplexes as globe-shaped molecules that permanently entrap smaller guest molecules within their confines.11 Consequently, the incarcerated guest cannot escape without breaking covalent bonds. Hemicarceplexes are closely related to carceplexes, except they contain portals through which the entrapped guest can exit, given the appropriate conditions. Carcerands and hemicarcerands are the corresponding empty hosts that contain no entrapped guests. The lifetimes of entrapped guests range from infinity for carceplexes to minutes for hemicarceplexes.11 Closely related to carceplexes and hemicarceplexes are reversibly forming capsules, which are defined by Rebek as "receptors with enclosed cavities, formed by the reversible noncovalent interaction of two or more, not necessarily identical, subunits".15 Capsules have well-defined structures in solution, and show binding capabilities that are absent for the individual components alone. Guest exchange rates for capsules are intermediate to cryptophanes and hemicarceplexes, ranging from milliseconds to minutes. The presence of a suitable template always appears to be a necessary requirement for the formation of carceplexes, while only sometimes for the formation of hemicarceplexes and capsules. 5 1.1.4 The Inner Phase of "Molecules Within Molecules" Initially, container compounds such as carceplexes, hemicarceplexes, and capsules were only used to recognize smaller complementary guests, but as an advanced understanding was attained, more sophisticated properties began to emerge. The inner phase of these containers is believed to consist of a new phase of matter somewhere in between the liquid and gas phase.16 Probably the most impressive applications of these "molecules within molecules" has been in their ability to act as miniature reaction chambers and in the generation and stabilization of reactive species within their confines.1 7'1 8 Four general classes of chemical transformations have been observed within the inner phase of molecular containers (i.e., capsules and hemicarceplexes): (1) intermolecular, (2) intramolecular, (3) mother-daughter molecule reactions, and (4) innermolecular inner phase reactions as depicted in Figure 1.2. Intermolecular reactions involve reactions between an incarcerated guest with a reactant from the bulk phase. The reactant may completely enter the host cavity or may interact with the entrapped guest through a sufficiently large portal in the host shell, without completely entering the host. A n intramolecular inner phase reaction requires the transformation from the original incarcerated guest into another different guest, or cleavage into separate noncovalently linked species. For the latter, the resulting products are of smaller size, which may easily depart from the interior of the host reaction vessel and exit to the bulk phase. Usually these types of reactions occur thermally or photolytically.18 In mother molecule-daughter molecule inner phase reactions, a covalent bond between host and guest is broken. The guest fragment is initially free inside the host and may or may not egress to the external environment. Finally, in innermolecular inner phase reactions, the noncovalently bound guest 1 R reacts with the host. This typically involves the irreversible formation of covalent linkages. 6 Inner Phase Reactions tt ^ <^ Mother molecule- _ , . Intermolecular Intramolecular Daughter molecule l n n e r m o l e c u l a r • Reactant O Product Figure 1.2 Classification of inner molecular reactions. This chapter will present a variety of different recently reported molecular containers that entrap neutral guest molecules, which will be discussed as follows. Section 1.2 will discuss work involving template effects in the formation of carceplexes and hemicarceplexes by single-molecule guests, to serve as a background to the work that is presented in Chapters 2 and 3 of this thesis. Section 1.3 will discuss examples of larger molecular containers (carceplexes, hemicarceplexes, and capsules) that entrap multiple guest molecules. These will serve as background for the work presented in both Chapters 3 and 5. Finally, section 1.4 deals with the use of hemicarceplexes in the generation, stabilization, and characterization of highly reactive intermediates such as cyclobutadiene, ortho-benzyne, and 2,4,6-cycloheptatriene. The material in section 1.4 is relevant to the work that is presented in Chapter 4. 7 1.2 Templation In the Formation of Carceplexes and Hemicarceplexes 1.2.1 Acetal-Bridged Carceplexes and Hemicarceplexes 1.2.1.1 The First Soluble Carceplex and Hemicarceplex Cram and Sherman reported the preparation of the first soluble carceplex, 10a«guest, from a shell closure reaction between two bowl-shaped tetrol molecules (8a), base, and four molecules of bromochloromethane under conditions of high dilution in dipolar, aprotic solvents (Scheme 1.2).1 9 Since Cram and Sherman's initial report product yields as high as 87 % have been achieved for formation of carceplex 10a»guest, which are remarkable for a reaction that joins seven molecules and makes eight new covalent bonds. The presence of a suitable template (guest) is required in the formation of 10a«guest as no carceplex or carcerand products were isolated from reactions conducted in solvents that are too big for the interior of carcerand 10a.19 However, 10a»guest did form when the solvent was doped with a suitable template, which suggests that a template is required.19 Scheme 1.2 Synthesis of an acetal-bridged carceplex and hemicarceplex. 2 10«guest, A = OCH 20 11 "guest, A = H, H Tris-bridged hemicarceplex lla«guest also forms under similar reaction conditions from two triol (9a) bowls. 2 1 The shell of hemicarcerand 11a has a single modest-sized hole, through which the trapped guest can escape with sufficient time and heat. 1.2.1.2 Template Ratios In The Formation of Acetal-Bridged Carceplex and Hemicarceplexes Further investigation by the Sherman group led to the discovery of a one million-fold range in selectivity for various small molecules found to be suitable guests (Table 1.1) in the formation of 10a«guest. Template ratios (77?s) were calculated from direct competition experiments between pairs of guests through measurement of the carceplex product ratios obtained from integration of bound guest ' H N M R signals in the product mixtures. Just as product ratios reflect relative rates of rate determining steps in irreversible reactions, TRs reflect the relative rates of the guest determining step (GDS) for each of the suitable templates. The GDS refers to the step occurring along the reaction pathway during which the guest becomes permanently entrapped within the forming host.20 Therefore, the template effect observed is kinetic in origin. Unlike most template effects, the carceplex/hemicarceplex product is "tagged" with the template. Therefore, large template ratios can be measured with good precision without the need for the determination of the rates of each individual step in the reaction. Often in template studies, yields are compared, which do not provide quantitative measurement of template ratios, and the range in template effects measured by yields is small. Table 1.1 shows that pyrazine is the best template and is a million times better than the poorest measured template, JV-methylpyrrolidinone (NMP), for 10a«guest. The template ratios 9 in the formation of lla»guest measured under identical conditions correlated well with those for 10a«guest, suggesting that the similar driving forces are at play. Table 1.1 Selected template ratios for the formation of acetal-bridged carceplex 10a»guest and 22 hemicarceplex lla«guest. Guest 10a»guest lla«guest pyrazine 1000000 170000 methyl acetate 470000 -1,4-dioxane 290000 52000 DMSO 70000 6200 acetone 6700 620 thiophene 5800 -± 2-butanol 2800 -benzene 2400 -pyrrole 1000 360 1,3,5-trioxane 100 10 DMA 20 2 DMF 7 -NMP 1 1 1.2.1.3 [4]Cavitand-Based Self-Assembling Capsules Further investigation into the driving forces responsible in the formation of carceplex 10a»guest led to the discovery of a complex (12b»guest, Scheme 1.3) that forms in solution from two tetrol molecules (8b) in the presence of base. In complex 12b»guest, the guest is reversibly encapsulated between two tetrol bowls that are held together by four charged hydrogen bonds (CHBs). 2 3 Measurement of relative stabilities of complexes formed with selected guests from the series in Table 1.1 showed that complex 12b»guest expresses the same guest-selectivity as carceplex 10a»guest. Thus, complex 12b»guest serves as a good transition state model for the GDS in the formation of carceplex 10a»guest.23 Consistent with these results was the finding that triol 9b forms a reversible capsule (13b«guest) with like guest-selectivity to 12b«guest, and that the template effect in forming the corresponding tris-bridged hemicarceplex lla»guest, proceeds with like guest-selectivity to carceplex 10a»guest.22 Non-covalent interactions between the forming host and suitable templates promote the most stable alignment of the triol bowl precursors, and thus, the templates facilitate the formation of hemicarceplex lla«guest. 2 2 Scheme 1.3 Formation of charged dimeric tetrol complexes. a R - C H 2 C H 2 C 6 H 5 12«guest, A = O-H-O, n = 4 bR = C H 3 13*guest, A = H, H, « = 3 Striking similarities are manifested in the ' H N M R spectra of 10a»pyrazine2 0'2 4 and 12b«pyrazine,23'25 and X-ray crystal structures of carceplex 10b«pyrazine24 and complex 12b«pyrazine. These similarities, along with the guest selectivities observed for each indicate that the same interactions are at play in driving the thermodynamic formation of 12b«guest and the kinetic formation of 10a»guest.25 Favorable interactions include: charged hydrogen bonds between the bowls, favorable van der Waals contacts, C H - p i interactions, C H - X (X = O) hydrogen bonding, conjugation of O-H—O" and O C H 2 O bonds to their respective aromatic 11 rings, and pi-pi interactions.19"25 The general trend observed experimentally was also 18 19 20 21 Figure 1.3 Self-assembling capsules derived from [4]cavitands. In addition to 12b and 13b«guest, singly and doubly covalently-bridged intermediates (Figure 1.3) were also synthesized, which form charged hydrogen-bonded capsules 14-21 that also reversibly encapsulate guests with the same relative guest affinity. 7 The relative stabilities of complexes formed with several guests were found to mirror the template effect observed in the formation of carceplex 10a«guest.27 Thus, each of these complexes, along with complex 12b«guest, are valid transition state models for the formation of carceplex 10a»guest. Note that 14»guest, 18«guest, and 20«guest do not have CHBs; thus these complexes are neutral. 12 1.2.1.4 Mechanism of Formation for Acetal-Bridged Carceplex 10a«Guest The first step in the formation of carceplex 10a»guest is the reversible formation of complex 12b»guest. Installation of the first acetal bridge to form a mono-bridged intermediate then concurs, with guests still in rapid exchange.20b The GDS then occurs during the formation of a second bridge at any position to form doubly-bridged intermediates, where guest exchange ceases under the reaction conditions. The ability of a particular guest in aligning the phenoxides of each hemisphere of the mono-bridged intermediate ultimately determines the rate 9(1K at which the second bridge is installed. Hence, the more stable complexes (i.e., those containing the best template molecule) form the second bridge the fastest, thereby ensnaring more of the preferred guest under the given reaction conditions. The relative rates of the GDSs are largely dictated by ground state effects, where the more stable complexes are formed in higher concentration. In other words, the rate constants for forming the second bridge are likely to be very similar for the guests studied. Installations of the third and fourth bridges then ensue, providing the completed carceplex product. (Incidentally, the formation of the fourth bridge is the rate determining step, but not the GDS.) Scheme 1.4 Mechanism for the guest-templated formation of acetal-bridged carceplex 10a«guest. GDS 12«guest 10»guest 13 1.2.2 Synthesis of a Benzylthia-Bridged Carceplex Carceplexes have also been synthesized using templates from [4] cavitands with functionalities other than phenols. Cram synthesized benzylthia-bridged carceplex 24»guest in a shell-closure reaction between tetra-benzyl chloride cavitand 22 and tetra-benzylthiol cavitand 23 in the solvents 2-butanone, 3-pentanone, ethanol:benzene (1:2), dimethyl formamide (DMF), MeOH:benzene (2:1), and acetonitrile:benzene (2:1), yielding carceplexes 24«2-butanone, 24«3-pentanone, 24«ethanol, 24«DMF, 24»(MeOH)2, and 24»(acetonitrile)2, respectively (Scheme 1.5).28 Preliminary experiments conducted by the Sherman group suggest that template effects are also evident in the formation of carceplex 24»guest, where the range in selectivity observed was calculated to span two million-fold. 2 9 Templating abilities for similar guests used in the templation studies in the formation of both 10a«guest and 24«guest were completely different. Thus, different driving forces are at play in the formation of a carceplex (24«guest) from reactants that cannot form (hydrogen-bonded) preorganized intermediates than for a system (12b»guest) that can. 14 Scheme 1.5 Synthesis of benzylthia-bridged carceplex 24»guest. R R R R 23 24«guest Chapter two of this thesis also discusses published work 3 0 involving investigations into the role of preorganized intermediates (i.e., complexes shown in Schemes 1.3 and Figure 1.3) in hemicarceplex formation. A templation study into the formation of hemicarceplex 25»guest 3 1 (Figure 1.4) from a shell-closure reaction analogous to the reaction involved the formation of 10a»guest, in which two cavitand subunits (8a) are bridged with a longer (tetramethylene) spacers is discussed. 15 1.2.3 Hemicarceplexes With Four Slotted Portals Using the same shell closure approach for the synthesis of carceplex 10a»guest (and hemicarceplex lla«guest), a large number of hemicarceplexes have been formed that incorporate a wide variety of different sized spacers. 1 1 , 1 4 , 3 2 Thus, fairly large-ring portals have been created in the hemicarceplex shell, which allow entrapped guest molecules to escape into the external medium. Few template studies exist on these hemicarceplexes, as most have been prepared by Cram to investigate complexation/decomplexation behaviors and to investigate novel reactivity of the entrapped species (see section 1.4). In turn, the larger empty spaces created by these hosts, have allowed for the entrapment of guests significantly larger in size and with greater variety in shape. For example, template effects have been observed in separate shell-closure reactions forming 26»guest and 27»guest (Figure 1.4).33 Yields of 26*guest and 27«guest were increased from 20 and 18 %, to 51 and 27 %, respectively, upon addition of 1,2-dimethoxybenzene to the reactants. This is the first of two reported examples where the formation of a hemicarceplex is templated, but the product is not "tagged" with the template, as it escapes after the host is formed. Template effects in the formation of weto-xylyl bridged hemicarceplexes (Figure 1.5) have also been reported.34 The robustness of 28«guest prompted Cram and coworkers to investigate how binding ability is changed upon altering the cavity size of the host. This was accomplished through various shell closure reactions using [4]cavitands that differ in their interhemispheric "spanners": methylene (34), ethylene (35), and propylene (36) (Figure 1.5) were explored.34 Hemicarceplexes having like (28-30) and unlike (31-33) northern and southern hemispheres were prepared by two different strategies, each revealing striking differences in yields. 3 4 These methods were analogous to the methods used to form carceplex 10a»guest ("2 + 16 4" addition of two tetrol bowls and four linker molecules), and carceplex 24»guest ("1 + 1" addition of tetrol with tetrachloride bowl). 34 35 36 R - r u a X - OH b x = OQTc' Figure 1.5 meta-Xylyl-bridged hemicarceplexes 28-33 and precursor [4]cavitands 34-36. Applying the "2 + 4" conditions, 28»NMP was synthesized in 50 % yield from tetrol 34a (Figure 1.5) and linker in N M P . 3 4 In contrast, the "1 + 1" conditions produced only a 2.2 % yield of 31»NMP from tetrol 34a plus tetrachloride 34b. 3 4 These results suggest that N M P acts as a template, where two tetrol bowls can pre-associate to form complex 12»NMP (R = C 5 H 1 1 ) . 17 In the first reaction, N M P is a "positive" template, as the two preorganized tetrol molecules react with the linker to form bridges leading to the desired product in high yield. However, N M P is a "negative" template in the second reaction since the formation of the desired product is inhibited as the effective concentration of free tetrol 34a available for reaction with the tetrachloride bowl 34b is reduced in forming a dimer ((34a)2). In addition, installation of the first bridge would lead to an intermediate that is not preorganized (as inter-cavitand hydrogen bonds would be unable to form), and, thus would be poised for oligomerization/polymerization. N M P is also a "negative" template for the syntheses of hemicarceplexes 31«guest and 32»guest. Reaction of 34b and 35a gave a much higher product yield of 31«guest (21 %), than the corresponding "reverse" reaction involving 34a and 35b (2 %). 3 4 Similarly, 32»guest is formed in 1.8 % yield from 34b and 36a, and ~0 % from 34a and 36b.34 These results are consistent with 34a promoting a complex like 12-NMP whereas 35a and 36a are precluded from such. As no complex akin to 12»guest forms, tetrol molecules are available for the "1 + 1" reaction, but are not preorganized to facilitate the "2 + 4" reaction. Lack of complex formation for 35a and 36a also explains the differences in the yields of 29»guest and 30»guest, between reactions involving tetrol bowl plus tetrachloride bowl (43 % and 6 %, respectively) and those involving tetrol bowl plus linker (8 % and 0 %, respectively). Template effects appear to be absent in three out of four of these reactions. However, addition of 36a to 36b only gives the product 30»guest when 1,2,3-trimethoxybenzene is present in the reaction at 5 % (w/w) of the solvent.34 1,2,3-Trimethoxybenzene apparently templates the formation 30«guest, which escapes during the work-up, as only the free host 30 was isolated.34 This is the second example where such a template/escape process occurs in the formation of hemicarceplexes (vide infra). Support of this notion is gleaned from the isolation of 32« 1,2,3-18 trimethoxybenzene from the reaction of 34b and 36a. 32»guest did not form in the absence of 1,2,3-trimethoxybenzene.34 1.2.4 Cali\|4|arenc-|4|C avitand Hybrid Carceplexes Reinhoudt and coworkers have prepared asymmetric carceplexes composed of both calix[4]arenes and [4]cavitands.35 The synthesis of 38»guest involved intramolecular cyclization of ettfio-coupled compound 37, where the tert-butyldimethylsilyl groups are removed in situ with CsF, followed by displacement of the chlorides with the phenoxides of the resorcin[4]arene (Scheme 1.6).35 Suitable guests/templates reported are DMF, D M A , N M P , l,5-dimethyl-2-pyrrolidinone, DMSO, ethyl methyl sulfoxide, thiolane-1-oxide, 3-sulfolene, and 2-butanone.36 Scheme 1.6 Synthesis of Reinhoudt's carceplex 38«guest. 37, R = TBDMS 38«guest The relative templating abilities of several guests were investigated through guest competitions reactions 1,5-dimethyl-2-pyrrolidinone solvent.36 The results are displayed in 19 Table 1.2, where D M A is the best template in the formation of 38»guest. The relative templating abilities of the guests in Table 1.2 were described as comparable to the association strengths between the calix[4]- and resorcin[4]arene cavity of the host and the guest.36 D M A is a superior template as it provides the best solvation of the transition state during shell closure. Stabilization of the transition state, thus promoting formation of the resulting product, is believed to arise partly from hydrogen-bonding between the N.//protons of the host and the incarcerated guests.36 Slow association between the host and guests that are poorer templates would lead to the formation of intermolecularly coupled products, and/or decomposition of the chloroacetamido groups of precursor 37. 3 6 Table 1.2 Templating ability and yields for the formation of Reinhoudt's carceplex 38«guest. G « e « T " £ « J « Yield <%)> D M A 100 27 DMSO 63 16c DMF 27 13c 2-butanone 27 16 a D M A is set at 100. isolated carceplex when only one guest is used during doped inclusion. °Yield of deuterated guests. Diastereomeric carceplexes 38«guest have been reported, each differing in the orientation of the contained guest, which provided means for the discovery of a new form of stereoisomerism called carceroisomerism.37 Carceroisomers are carceplexes that contain the same guest, but differ in the orientation of the guest within the host cavity. Generally there is an appreciable energy barrier associated with their interconversion. Application of carceroisomers as molecular switches has been suggested.36'37 20 1.2.5 Metal-Ligand Bridged Carceplexes Carceplexes traditionally consist of neutral organic molecules imprisoned by the covalently-linked, neutral host carcerands. Dalcanale and coworkers at the University of Parma, have expanded the scope of carceplexes by characterizing carceplex 40«guest, synthesized through the metal-induced self-assembly of two molecules of tetra-cyano cavitand 39 connected with four Pd 1 1 or Pt11 square-planar linkages, where the incarcerated guest is a triflate anion, CF3SO3" (Scheme 8). 3 8 The structure of 40«guest was confirmed through extensive characterization involving N M R ( 'H, 1 3 C , 3 1 P , 1 9 F) and FT-IR spectroscopies, electrospray-MS, and vapor-phase osmometry. Carceplex 40b»guest was also discovered to disassemble by ligand exchange upon addition of eight equivalents of triethylamine to give two molecules of precursor bowl 39 and four PtL(NEt3)2(OTf>2 (L = l,3-bis(diphenylphosphino)propane). Subsequently, 40»guest reforms upon addition of eight equivalents of triflic acid (CF3SO3H). 3 8 Ligand exchange controls the reversible cage formation.38 As no empty cage 40 was ever observed as a product, CF3SO3" is required as a template. Dalcanale and coworkers have also synthesized other similar derivatives to 40»guest from tetrakis(benzylnitrile) [4] cavitands.39 Several other metal-ligand bridged hemicarcerands have also been recently reported, which incorporate the transition metals C o 4 0 and Fe. 4 1 21 Scheme 1.7 Formation of metal-ligand brid ged carceplex 40«guest. R 7 CF3SO3 N C a M = Pd, R = C 1 1 H 2 ; b M = Pt, R = C 1 1 H 2 c M = Pd,R = C 6 H 1 3 R R 40«guest The next section (1.3) will discuss the formation of large host molecules that can entrap two or more guest molecules either each in separate chambers, or within the same chamber 1.3 Container Molecules That Entrap Multiple Guests 1.3.1 Benzyl Thia-Bridged Carceplex 24«(Guest)2 and Trismethylene-Bridged Hemicarceplex l l»(Guest) 2 (Revisited) In section 1.2, carceplexes 24»(MeOH)2 and hemicarceplexes 24»(CH3CN)2 and lla«(CH3CN)2 were mentioned to have been synthesized by the Cram group. 2 1 ' 2 8 Carceplex 24»(MeOH)2 and hemicarceplex 24»(CH3CN)2 were prepared in templated shell-closure reactions, while hemicarceplex lla»(CH3CN)2 was formed by complexation to hemicarcerand 11a in CHCl3:CH3CN (5:2).21 Guest orientations are predicted to be similar in 24»(MeOH)2 and 24«(CH3CN)2 by ' H N M R chemical shift data, where the methyl groups are directed into opposite hemispheres (bowls) with the polar OH and C N functionalities near the equator.28 In 22 contrast, the two guests in l la«(CH 3 CN) 2 are predicted to be oriented with the methyls near the 21 equatorial region of the host and the C N groups in opposing bowls (below). Carceplex 24»(MeOH)2 was isolated as a pure and stable substance. No loss of guest was 28 observed after prolonged heating of 24*(MeOH)2 in solution (110 °C, 5 d., toluene-cis). Noncovalent (van der Waals) interactions between the host and the two guests and hydrogen-bonding between the two guests both contribute to the overall stability of this complex. With the hydroxyls near the equator (vide supra), inter-guest hydrogen-bonding can occur between the two MeOH guests.28 As a dimer, the two MeOH guests can shift to and from the northern and southern hemispheres, with each of hydroxyl protons hydrogen bonding to the inward directed sulfur lone pair of the host's benzyl thiol linkers (below).28 In contrast to 24*(MeOH), hemicarceplexes 24*(CH 3CN)2 and l la»(CH 3 CN)2 are not as kinetically stable. Hemicarceplex 24«(CH3CN)2 was isolated as an inseparable mixture with carceplex 24»CH3CN. At 353 K , decomplexation of 24»(CH3CN)2 to 24»CH3CN occurred with a half-life of -26 hours. Thermally induced kinetic motions of the mobile C H 3 C N guests along the polar axes of the host were explained to provide the driving force for guest expulsion via guest collisions, producing a "billiard-ball" effect (below).28 Hemicarceplex l la«(CH 3 CN)2 is even less stable than 24«(MeOH) or 24«(CH3CN)2, which readily loses one guest even in the S N = C - C H 3 H 3 C - C = N fS| H 3 C - O - - - H - O - C H 3 H S H 2 C - a - C H 2 N 23 solid state.21 A decomplexation half-life of 26 min was measured for l l a«(CH 3 CN) 2 in solution at 22 °C. 2 1 b+o-H 3 C-C=N • • ——— H 3 C-C=N N = C - C H 3 N E C - C H 3 8- &+• "billiard-ball" effect 1.3.2 Larger Carceplexes Derived From Cyclic Arrays of [4]Cavitands Our group has recently used a new approach in synthesizing new host molecules, which is different than the conventional method developed by Cram for two-bowl carceplexes. We have reported several new carceplexes from precursors consisting of multiple covalently linked [4]cavitands.42'43 Cyclic trimer 42 (5 %) and tetramer 44 (15 %) are synthesized from the cyclo-oligomerization of A,C-diol 41 with base and BrCE^Cl (Scheme 1.8) in D M F 4 2 a The free hydroxyls of trimer 43 and tetramer 45 were obtained by hydrogenolysis.42a Cyclic trimer 43 possesses a fairly large cavity for a synthetic host, and is conformationally very rigid. Whereas, cyclic tetramer 45 has an even larger interior, its structure is far more flexible at room temperature. 24 Scheme 1.8 Synthesis of cyclic trimer and tetrameric [4]cavitand arrays. ? H O H K 2 C 0 3 PH ^ 3 BnBr rjz. OR' DMF K 2 C O 3 ( f W / / * " BrCH,CI k. 42 R' = C H 2 C 6 H 5 43 R' = H H 2 Pd/C DMF 8a 41 R' = CH2CgHg p or . O R * F T O , ,OR' R.0, V 44 R' = C H 2 C 6 H 5 45 R' = H H 2 Pd/C 1.3.3 Synthesis of Bis (carceplexes) and Bis( complexes) The addition of appropriate sized caps to the upper and lower rims of cyclic tetramer 45 could potentially render a carceplex with a cavity of unprecedented size. However, due to its flexible structure, tetramer 45 can fold in upon itself, forming a carceplex with two separate chambers (Scheme 1.9). Subjecting tetramer 45 to optimal carceplex forming conditions found for 10«pyrazine gave bis(carceplex) 46»(pyrazines)2 in 74 % yield. 4 2 3 In addition, pairs of adjacent covalently attached bowls of tetramer 45 cooperatively clamp around suitable guest molecules in the presence of base to form bis-capsules, 47»(guest)2. 4 2 b Both homo and hetero bis-capsules were formed with suitable templates such as DMS, 1,4-dioxane, pyrazine, acetone-fife, DMSO, methyl acetate, benzene-^, THF, and CDC1 3 4 2 b *H N M R chemical shift data for hetero bis-capsules suggests that guest communication is likely in the form of conformational changes, where the ability of the two bowls defining each capsule to clamp down about a certain guest depends on how far the bowls of the adjacent capsule clamp down about their guest.42b This type of interbowl communication could be important in the development of molecular switching devices.42 46*(guest)2 1.3.4 Bis-Bridged Hemicarcerand Dimers Bis-hemicarceplexes have also been prepared with similar topology to 46«(guesf)2 and 47«(guest)2. 4 5 Hemicarcerand diol (48) was bridged with different spacers of appropriate lengths to synthesize 52«(guesf)2 and 55«(guest)2, in light of the previously demonstrated ability of 25»guest in promoting through-shell reactions (see sections 1.4.1-1.4.3) between incarcerated guests and bulk phase reactants. It was hoped that similar interactions may occur between designable guests bound in the adjacent capsules of structures 52 and 55. 4 5 26 Scheme 1.10 Schematic representation of the synthesis of bis-hemicarceplex dimers. 53 .CHCI3 5 4 . C H C I 3 5 5 . ( D M A ) 2 # = C H C I 3 Two different strategies were employed to synthesize bis-hemicarceplexes 52»(guesf)2 and 55»(guest)2. In the first, 52«(DMA) 2 was synthesized in a single pot by bridging two diols (48) with a durene spacer in either N M P or D M A . 4 5 The second dimeric hemicarceplex prototype, 55»(DMA)2, was also synthesized in several steps by first separately preparing 53«CHC13 and 54«CHC13 by bridging diol 48 with dibenzyl bromides 50 and 51, respectively.45 Hemicarceplexes 53»CHC13 and 54«CHC13 were then efficiently coupled in D M A at 55 °C to give 55«(DMA) 2 (82 % yield). 4 5 The advantage with the latter method for 55«(DMA) 2 is that hetero-guest hemicarceplexes dimers could be prepared, while avoiding the inevitable formation of homo-guest hemicarceplex dimers.45 Although no guest communication between 27 hemicarceplex chambers was reported, this work does represent a method for synthesizing robust polymers whose core of guests is potentially conducting, proximate shells are electron permeable, and pendant (R) groups are insulating.4 5'4 6 1.3.5 Synthesis of a Trimer Carceplex Containing Three D M F Molecules Our group has also recently reported large novel hosts composed of multiple linked [4]cavitands that entrap multiple small molecule guests within the same chamber.43 Trimer carceplex 56«(DMF) 3 has been synthesized in 37 % yield, from trimer 43 in a reaction with 2,4,6-tris(bromomethyl)mesitylene under basic conditions (Scheme 1.11).43 The entrapment of multiple guests in carceplex 56«guests provides an excellent opportunity to study how clusters of molecules can act as a single template. This begs various questions. How well do multiple guests compete with a large single guest in carceplex formation? Can a large single template molecule form trimer carceplex 56»guest? When does a template effect become a solvent effect? Attempts to address these questions involving further studies into the formation of carceplex 56»guests is presented in Chapter 3 of this thesis. Scheme 1.11 Synthesis of trimer carceplex 56»(DMF)3. 56.(DMF)3 28 1.3.6 The First Cs-Symmetric Disulfide-Bridged Carceplex Wider [«]cavitand subunits (n = 5, 6, 7) have also been recently synthesized by members of our group to be used as building blocks for larger carceplexes.47 This paved the way for the creation of a new family of carceplexes from [5]cavitands. Ai r oxidation of benzyl-thiol [5]cavitand 57 in D M F or D M A in the presence of C S 2 C O 3 afforded carceplexes 58»(DMF)2 (25 %) and 58«(DMA) 2 (16 % ) respectively (Scheme 1.12).48 It is somewhat surprising that 58»guests forms from cavitand 57, since the analogous tetra disulfide-bridged [4]carceplexes do not form under the same conditions. ' H N M R chemical shifts and NOEs between the host and guest protons suggest that the AO entrapped guests adopt an unexpected orientation within carcerand 58. The D M F guests are believed to lie stacked in parallel planes perpendicular to the long Cs-axis of the host rather than parallel as observed in other carceplexes/hemicarceplexes.48 The 7V-CH3s for the two DMFs experience similar environments and lie near the equator of the host.48 These methyls normally are pushed deep into the cavitand subunits, as is apparently the case for 10«DMF,1 6 24«DMF, 2 8 and 56»(DMF) 3 . 4 2 29 The host and guest in 58«(DMF)2 are involved in independent dynamic processes, which were both probed by N M R spectroscopy (EXSY) . 4 8 For the host, a 14.4 kcal/mole energy barrier was measured for the slow interconversion between the pairs of diastereotopic benzyl protons near the disulfide linkages, which suggests that the five disulfide linkages of carcerand 58 experience considerable constraint. Dibenzyl disulfides typically have AG*s that are too small to analyze by N M R 4 8 The energy barrier for the amide bond rotations of the entrapped DMFs in 58»(DMF)2 was also measured and compared to values reported for other carceplexes. From the AG*s listed in Table 1.3,10«DMF appears to exhibit a more non-polar or gas-phase environment,16 while the cavities of the other carceplexes (Table 1.3) are probably polarized by the additional D M F guest(s) and/or the bridging sulfur atoms.48 , Table 1.3 Energy barriers for amide methyl interconversion for free and incarcerated D M F in various sized carceplexes in nitrobenzene-afs at 77 °C 4 8 The remainder of this section (1.3) will deal with the entrapment of multiple-molecule guests in molecular containers that are held together with hydrogen-bonds or metal-ligand interactions. AG* (kcal/mol) DMF (9 %) acetal-bridged carceplex 10«DMF benzylthia-bridged carceplex 24«DMF penta-disulfide carceplex 58»(DMF)2 trimer carceplex56»(DMF)3 21.1 19.1 20.5 20.8 20.5 30 1.3.7 Rebek's "Sports-Balls" Capsules Rebek and coworkers have successfully prepared many types of fascinating self-assembling capsules, which are all related as they involve the reversible oligomerization of two or more complementary concave subunits held together by numerous hydrogen-bonds.15 Many of the subunits used incorporate glycouril functionalities at each terminus of the molecule, which are separated by a rigid spacer (Figure 1.6). The "tennis" balls (Scheme 1.13) were the first of these capsules to be reported, through the dimerization of monomers 59 or 60 (Figure 1.6), which possess small cavities that can entrap 1 2 9 X e , methane, ethane, ethylene, and CH 2 C1 2 as guests in C D C I 3 . 4 9 Capsule formation is identified by the characteristic concentration independent downfield shifted resonances assigned to the urea N - H protons and the upfield shifted signals for the protons of the bound guests in ' H N M R spectra.49 O R*)—(«R H ' N Y N ^ O Spacer Spacer a o R») («R O #9 o o b 59 spacer = a, R = C 6 H 5 60 spacer = b, R = C 6 H 5 , X = H 61 spacer = b, R = C0 2-/-pentyl, X = H 62 spacer = b, R = 4-n-heptyl phenyl, X = OH Figure 1.6 Monomers for Rebek's "Sports balls' 31 Scheme 1.13 Formation of Rebek's "Tennis" ball. 592»guest Capsules with larger cavities ("softballs") are formed from monomelic subunits that incorporate larger bis-spacers between the two glycouril termini. 5 0 The four additional hydroxyl groups of the aryl spacer of monomer 62 impart more stability to the corresponding "softbalP'capsule 622 due to the formation of a total of 16 hydrogen-bonds, as opposed to only eight for 6I2 . 5 1 Guests that are able to form a maximum number of favorable van der Waals interactions with the interior of "softball" dimers form the most stable complexes.5 0'5 1 Stabilities of these dimeric complexes also improve when the encapsulated guest possesses functionalities capable of participating in hydrogen-bonding with the seam of the host (i.e., with guests such as 1-ferroenecarboxylic acid and 1-adamantanol).50'51 Enthalpies and entropies were determined in CDCl3,/?-xylene-c?/o, and benzene-^ solvents, which revealed that the formation of "softball" capsules is entropically driven. 5 1 ' 5 2 Two solvent molecules are displaced with a single guest from the "softball" capsule interiors (Scheme 1.14). This type of behavior in organic media is rarely seen and is normally unpredictable. Multiple-guest entrapment was observed in the discovery of three unique capsules that form when glycouril functionalized monomer 62 is equilibrated in a solution of benzene-cfofluorobenzene-ds (1:1). ' 5 Dimers containing two benzenes, two fluorobenzenes, and one of each guest were identified.5 1'5 2 Similar mixtures were identified for benzene-^ :toluene-a?g mixtures as wel l . 5 3 3 ' 5 4 32 Scheme 1.14 Displacement of solvent from the interior of capsule 62 2 by a single-molecule guest. The ability to bind multiple guests in a single cavity has been exploited by Rebek and coworkers in the acceleration of Diels-Alder reactions within "softball" dimer 622. 5 3 b Capsules 622»(63)2, 622»(64)2 and 622»(63»64) were identified to form by ' H N M R spectroscopy when /?ara-quinone (63) and cyclohexadiene (64) were added to solutions containing monomer 62. 5 3 a The Diels-Alder reaction between encapsulated 63 and 64 (Scheme 1.15) at millimolar concentrations in the presence of the dimer 622 was accelerated 200-fold in comparison to the analogous reaction between the free reactants in dilute solution.53 Scheme 1.15 Diels-Alder reaction within dimer capsule 622. 62 2 . (63«64) 622.65 The accelerating effect on the reaction is believed to arise from the increase in concentration of each reactant within the capsule. From a cavity volume estimate of -300 A 3 , the reactant concentration was calculated to be 5 M , which is 1000 times greater than in free solution.53 A n effective molarity (kcJkuc) of 2.4 M was also calculated and a 174-fold 33 acceleration in rate was observed for the intermolecular inner phase Diels-Alder reaction.33 Unfortunately, capsule 62 2 fails as a catalyst as turnover is prevented due to strong product inhibition. Capsule 622*65 is extremely stable, which is indicated by its apparent stability constant, Kapp of ~10 4 M " 1 . 5 3 Displacement of a single molecule guest 65 from the capsule by two smaller molecules dienophile 63 and diene 64 does not seem very likely. True catalysis of an inner phase Diels-Alder reaction inside the "softball" 62 2 was possible upon reaction of dienophile 60 with other dienes such as dimethylthiophene dioxide (66) under the encapsulation conditions (Scheme 1.16).53,55 When dimethylthiophene dioxide (66) was used under the appropriate reaction conditions to observe catalysis, the Diels-Alder adduct (67) did form and pseudo-first order kinetics were observed, signifying that product inhibition was not evident.55 Catalytic turnover that was observed was described as only modest, and only a 10-fold acceleration in rate was observed in the absence of the capsule 62 2 . 5 5 The catalytic cycle for the Diels-Alder reaction within the inner phase of capsule 62 2, between quinone (63) and dimethylthiophene dioxide (63), is shown in Scheme 1.16.5 5 , 5 6 Scheme 1.16 Turnover cycle for the Diels-Alder reaction catalyzed by Rebek's "softball". 67 622«67 63 34 1.3.8 Cylindrical-Shaped Capsules Sometimes a great deal of information can be encoded within the simplest of subunits. A n interesting example is the dimerization of deep cavity [4] cavitands utilizing similar hydrogen-bonding interactions as in the previously described "Tennis" and "Softballs"; however, different donor and acceptor groups are utilized. Cylindrical-shaped capsules such as 682 (Figure 1. 7) form from monomers (68) that are similar in structure to Cram's velcrands.11 Like other velcrands, deep-cavity resorcin[4]arene monomer 6 8 5 7 is conformationally mobile and interconverts between pseudo-"vase" and "kite" conformations in solution. The "vase" conformer is characterized as having a deep cavity as shown for 68 in Figure 1.7, while in the "kite" conformer (not shown) the cavity is flattened as the phthalamide bridges between adjacent co phenol of the resorcin[4]arene bowl are "flapped" outwards. In capsule 682, both monomelic components prefer the "vase" conformer, which is stabilized and cinched together by a seam of eight bifurcated intermolecular hydrogen-bonds between the imide and carbonyl functionalities.59 Dimerization was observed to be templated in mesitylene-J/i, which features downfield shifted ' H N M R N - H signals of the host upon encapsulation of single large guests of appropriate dimensions or pairs of small molecule guests. In turn, the aromatic lining of the cylindrical cavity causes upfield shifts of the protons of the bound guests relative to those free in solution. Size and shape play an important role in the selective encapsulation of preferred conformations of large single-molecule guests or pairs of competing small-molecule guests.59 Single-molecule guests inhabiting capsule (68)2 can spin freely about the long axis of the host, but cannot tumble end-to-end.59 ' H N M R spectroscopic data indicate that the capsule halves are non-equivalent, upon encapsulation of a single large asymmetric guest. Size/shape 35 selectivity for single-molecule guests was also observed in the formation of (68) 2 . For example, capsule (68)2 selectively binds the trans conformer of 69 (Figure 1.7), which is unusual, as the cis conformers is thermodynamically favored over the trans for the free guests in organic solutions.590 Encapsulation provides supramolecular inner phase stabilization of the trans-isomer as no encapsulated cw-isomer is bound. 5 9 0 682»69 682«(70)2 682«(71«71') Figure 1.7 Cylindrical capsules from tetraimide-bridged [4]cavitands. Restricted guest mobility is observed for pairs of smaller guests that fill capsule 682 . 5 9 0 Space restrictions in the equatorial region of the capsule cavity does not permit guests occupying each cavitand half to exchange positions without at least partial dissociation of the assembly. Furthermore, when guests containing polar functionalities are paired, preferred orientations are suggested by the ] H N M R chemical shifts.590 The imide-functionalized equatorial region of the 36 host appears to attract the more polar functionalities of the guests, placing the more hydrophobic moieties deep within the ends of the capsule (Figure 1.7).59 This feature has led to the direct observation of encapsulated hydrogen-bonded guest dimers (i.e., of 70 in 682»(70)2, Figure 1.7) and a preference for diastereomeric complexes upon encapsulation of pairs of chiral guests (i.e., encapsulation of the pair 71/71' over 71/71 or 71771', Figure 1.7).59c'60 1.3.9 Molecular Paneling So far the molecular containers that have been presented consist mainly of the dimerization through the covalent linkage or hydrogen-bonding of two or more appropriate subunits. Metal-ligand interactions have also been used to self-assemble multiple components to form discrete novel hosts.4 4'6 1 The synthesis of many of these architectures is highly efficient, as multiple different components of appropriate geometries can self-assemble into a single structure in single steps with yields greater than 90 %. Here we will only focus upon a few interesting systems, which relate to work in this thesis (Chapters 3 and 5) involving the encapsulation of multiple guest molecules. Fujita and coworkers have constructed several nanoscale hosts via a methodology termed molecular paneling. For example, the 7d-symmetric containers 75 and 76, are self-assembled and isolated with high efficiency (90 % yield) in single steps upon mixing tridentate ligand 72 with an appropriate ligated-metals (either 73a,b or 74a,b) in a 3:2 ratio in water (Scheme 1.17).62-63 Four molecules of 72 are held together by coordinative bonds to six palladium or platinum centers to form thermodynamically and kinetically stable and isolable structures in 37 water that can bind single-molecule guests such as carboranes, adamantanes, adamantanols, 1,4-62 63 dibromobenzene, 1,3,5-tribromobenzene, perfluorobenzene, and 1,3,5-tri-tert-butylbenzene. ' Scheme 1.17 Synthesis o f molecular paneled hosts. A fascinating form of supramolecular stabilization is in the formation o f dimers within 7 5 a upon complexation of cw-azobenzene, stilbenes and related derivatives. 6 4 Inclusion o f pairs o f "C-shaped guests within 7 5 a results in the formation of spherical-shaped dimers within the hydrophobic host cavity in D 2 0 solution that has a similar topology to Rebek's "sports-balls". 6 4 In contrast to cylindrical-shaped capsule ( 6 8 ) 2 (section 1.3.8), 7 5 a selectively binds only the cis-stilbene isomers, which are complementary to the interior of the host as a hydrophobic dimer (Scheme 1.18).64 frans-Stilbenes are not encapsulated and a strong positive cooperative effect on guest binding is suspected as no 1:1 hostguest complexes were formed. 6 4 The encapsulated dimers are remarkably stable and remain intact even when the host-guest assembly is extracted into C D C I 3 . 6 4 When the host 7 5 a is in excess in CDCI3, guest exchange is observed through N O E S Y correlations between the encapsulated guests 7 7 b and the free host 7 5 a , indicating that 38 the complex 75a»(77b) 2 does exchange guests. While direct encapsulation of the trans-isomers of 77a-d were not observed, the trans-isomer of 77a could be easily generated photolytically within 75a from the entrapped ds-isomer in 75a»(77a) 2 . 6 4 a Scheme 1.18 Formation of hydrophobic dimers as guests in hosts 75a/76a. 77a A = N=N, Y = Z = Me 77b A = CH=CH, Y = Z = Me 77c A = CH=CH, Y = Z = OMe 77d A = (CO) 2 > Y = OEt, Z = OMe 75a / = \ or + 2 \ _ J 76a \ Y A Z 75a 2«(77) 2 76a 2«(77) 2 While the Pd species 76a forms efficiently from the components 72 and 73a/74a in water, the formation of the Pt species 76b does not occur so readily.6 5 Low yields of 76b were obtained along with an intractable oligomeric/polymeric mixture of byproducts after allowing the reaction between 72 and 70b to proceed for in D 2 0 several days at 100 °C. 6 5 However, in the presence of sodium adamantane carboxylate (78), high product yields of 76b»(78)4 were achieved within 24 hours.65 Multiple-molecules of 78 act as a thermodynamic template in the multi-component assembly of the Pt-bridged cage 76b.65 The empty cage was obtained by removing the guests upon acidification followed by extraction.65 Cage 76b is notably more stable than its counterpart 75a, as it is tolerant to highly acidic (pH <1) and basic (pH >11) conditions that destroy 75a.65 This property augers well for a potential molecular switch via a pH-responsive guest complexation/decomplexation process.65 Complexation of guests such as A^V-dimethyl aniline to 76b occurs under neutral or basic 39 conditions while decomplexation occurs upon protonation.65 The pH-switchable guest inclusion observed for 76b is different than for complexes like 12»guest and the cage 40»guest, as reversible guest complexation occurs without disassembly of the host.65 The utility of capsule 76b has also been demonstrated in the ability to selectively encapsulate and stabilize the normally elusive and labile cyclic alkoxysilanol trimers (79) within its cavity (below) in aqueous solution. 6 3 ' 6 6 Cyclic silanol trimers like 79 have been believed to be important intermediates in the formation of polymers from the sol-gel condensation of alkyl silanols and water.66 The trapping and spectroscopic characterization of such an intermediate (79) verifies these theories.66'67 76b»79 Dimeric capsules from "Nanobowl" 80 have also been prepared using a similar 68 methodology and similar components as in the formation of cages 75 and 76.00 X -ray crystallographic data from crystals grown from a solution of 80 and excess a's-stilbene (81) showed that a nanobowl dimer of 80 forms that encapsulates six molecules of 81 in the solid state (802«(81)6.6 8 The dimer is also believed to form in concentrated solutions containing nanobowl 80 and an excess of the guest 81. A 1:3 host:guest binding stoichiometry was determined by ' H N M R spectroscopy, which is consistent with that observed for the solid state dimer.6 8 At lower concentrations of 81, the hostrguest ratio increases. Capsule 80 2»(81)6 40 apparently dissociates to the corresponding monomers that can only bind one or two guests (i.e., 80*81 and 80«(81) 2, respectively).68 80 Figure 1.8 Fujita's coordination nanobowl 80. 1.3.10 Hexameric Resorcin[4]arene Capsules Hexameric resorcin[4]arene capsules first reported by Atwood and coworkers are perhaps one of the most spectacular self-assemblies reported to date (Figure 1.9).69 Tetramethyl-rescorcin[4]arene 82 a spontaneously self-assembles upon crystallization to form hexameric capsule (82a)6(H20)g, which is seamed together by a complicated network of 36 hydrogen-bonds (Figure 1.9).69 Both intra- and intermolecular hydrogen-bonding is evident in the X-ray crystal structure of (82a)6(H20)8 between the adjacent phenol groups of 82a and between the phenols of 82a with eight participating water molecules.69 The three dimensional shape of hexamer capsule (82a)6(H20)g conforms to that of a snub-cube that has an internal cavity volume estimated at • 1375 A 3 . 6 9 Capsule (82a)6(H20)8 was found remain intact upon dissolution of the crystalline solid in benzene-c/6-69 ! H N M R spectra of (82a)6(H20)8 in benzene-^ were in accord with the 41 hexameric structure, which featured a resonance for the eight water molecules incorporated within the hexamer framework. 6 9 (82a)6(H20)8 (82b)6 [(82c)2.(83)]+BF4 © =H 2 0 Figure 1.9 Hexameric resorcin[4]arene capsules. The dotted lines in structure 79 represent intermolecular hydrogen bonds. Hexamer capsules from pyrogallol[4]arenes (i.e., (82b)6, Figure 1.9) have also been reported in the solid state upon crystallization from acetonitrile. 7 0 The solid-state structure of (82b)6 is remarkably stable and does not require H 2 O in its framework. Hexamer (82b)6 is held together by 72 hydrogen-bonds (48 intermolecular) and has a cavity with a calculated volume o f -1515 A 3 . 7 0 ' 7 1 Extensive disorder within the supramolecular cavities o f (82a)6(H20)g and (82b)6 prevented identification and characterization of the contents o f these entities in the solid In the initial report, crystals of (82b)6 were only produced once, while other attempts 71 72 gave "wave-like" hydrogen-bonded polymeric patterns of 82b in the solid state. ' A more 42 reproducible procedure was later reported, which resulted in the entrapment of multiple polar guests such as MeOH, EtOH, and isopropyl alcohol within hexamer (82b)6, as observed by lH N M R spectroscopy.71 Eighteen MeOH guest molecules were estimated to fill the enclosed space of capsule (82b)6, which do not exchange with the free MeOH in solution or with solvent molecules, even in highly polar solvents.71 The integrity of the hexameric capsule (82b)6 is maintained even in hydrogen-bond disrupting media, such as DMSO. While previous studies reported the formation of 1:173 and 2:1 7 2 hostrguest complexes from resorcin[4]arenes in solution, more recent efforts have provided evidence that hexameric capsules can self-assemble with a suitable guest template in the appropriate resorcin[4]arene hostguest ratio.7 4 Hexameric capsules form from resorcin[4]arenes (82a,d-f) in water-saturated CDCI3 in the presence various tetra-alkyl ammonium and phosphonium salts, as well as tetra-alkyl antimony bromides (R4SbBr) as guests.74 The stability of these capsules is dependent on not only the size of the included cation, but also on the character of the associated counter-anion. 7 4 b Capsules that form in the presence of tetrabutylammonium bromide, Bi^SbBr and PrLtSbBr contain additional empty space within host cavity that is easily accommodated by additional solvent molecules or appropriate aromatic guests.74 Selectivity for particular additional neutral guests is dependent on the size, shape, polarity and polarizability of the accompanying aromatic guest upon co-inclusion with Bii4SbBr.7 4 b Capsules were not formed in the absence of the primary guests BmNBr, PhjSbBr or Bu4SbBr.7 4 b Hexameric capsules of resorcin[4]arene 82c (Figure 1.9) do not form in the solid state or in solution. The absence or presence of an additional methylene group in the pendant moieties (R, Figure 1.9) of 82b and 82c, respectively, is important as only dimeric capsules 82c2 form both in the solid state and in polar, protic solvents such as methanol-^. 7 5 X-ray structures generated from crystals of 82c2 grown from aqueous acetonitrile solution indicated that four acetonitrile molecules were encapsulated within the dimer linked through hydrogen-bonding 43 interactions with sixteen water molecules (i.e., 82c2»[(H20)i6*(MeCN)4]). / : > A stable dimer entrapping a tropylium ion was observed in solution as the tetrafluoroborate salt, [82c2»83]+BF4" (Figure 1.9).75 Several applications of some of the self-assembling capsules were summarized above in addition to their synthesis and host-guest properties. These included Diels-Alder additions in Rebek's "Softball" and the stabilization of alkoxysilanol trimers in Fujita's Pt-coordinated capsule.66 Other reported examples not discussed here include bimolecular cycloadditions in Rebek's cylinder capsules,60 Wacker oxidation in Fujita's capsule,67 and the generation and stabilization of [Me 2C(OH)PEt 3] + , 7 6 while entrapped within a tetrahedral-shaped capsule, Na ] 2[GaL 6], in aqueous media.77 The inner phase chemistry of hemicarceplexes has also been studied. Hemicarcerands have been used in the generation and stabilization of highly reactive species as alternatives to more classical approaches that involve trapping highly reactive species in inert solid matrices at low temperatures and/or inert gas phases. The next section of this chapter wil l summarize work reported studies involving the inner phase reactivity of hemicarceplexes and their entrapped guests. 1.4 The Inner Phase Reactivity of Hemicarceplexes 1.4.1 Acid/Base reactions The first reported intermolecular inner phase reactions in hemicarceplexes were conducted on guests entrapped in the tris-methylene bridged hemicarceplex l l a « g u e s t (section 44 1.2.1.1, Scheme 1.2). To demonstrate the effectiveness of the protective shell provided by the host to its entrapped guest, lla«CH2Br2 was treated with 300 equivalents of «-butyllithium for a short period of time in dry T H F at 25 ° C . 2 1 CH2Br2 was left completely untouched by the highly reactive reagent, thus demonstrating that such hosts are indeed capable of functioning as protecting groups.21 Amines complexed to hemicarcerand 11a have been subjected to acids such as CF3CO2D and CD3CO2D in CDCI3. 2 1 Treatment of lla«pyridine with C F 3 C 0 2 D failed to form any pyridinium ions. However, applying the same conditions to l la«Et 2 NH resulted in immediate decomplexation.21 Only 11a (empty) and Et 2ND2+ were observed in the latter. l la«Et2NHD + and lla«Et2ND2 + are presumed to form in low concentrations well below the ! H N M R instrument detection limits. The host shell is explained to be ineffective at solvating the bulky pyridinium ions, Et2NHD+, or Et2ND2+ 2 1 In contrast, treatment of lla*CH3(CH2)3NH2 with CF3CO2D did yield lla»CH3(CH2)3ND3+.2 1 Subsequent decomplexation did occur, and was significantly faster for the deuterated amine in l la«CH 3 (CH2)3ND 3 + (tin = 10 min) under ambient conditions, than the neutral amine in the immeasurably stable l la»CH3(CH 2 )3NH 2 . 2 1 Upon treatment of l la»Et 2 NH and l la»CH 3 (CH 2 )3NH 2 with CD3C0 2 D, guest decomplexation did not occur. H/D exchange did occur to form l la«Et 2 ND or l la«CH3(CH 2 )3ND 2 as products, respectively, clearly indicating that deuterium is incorporated faster than decomplexation. The relative orientations of the two amines in hemicarceplexes l l a » E t 2 N H and l la«CH 3 (CH 2 )3NH 2 inferred from lH N M R chemical shift data and molecular ( C P K ) model examinations, both suggest that protonation occurs through different pores in the host shell. In lla«Et2NH, the alkyl chain methyl groups of the Et2NH guest protrude deep into each of the north and southern hemispheres of the of the host, while the N H moiety is at the equator. For lla«CH3(CH 2)3NH2, the alkyl chain methyl is pushed deep into one hemisphere while the amine 45 moiety at the other end is push into the opposing hemisphere of 11a. Thus, it is believed that protonation of the former likely occurs through the large-ring portals at the equator, while in the 21 latter, protons are transferred through the small openings of the polar hemispheres. 1.4.2 Oxidation/Reduction Reactions Dihydroxybenzenes (84-88) entrapped as guests in hemicarcerand 25 were quantitatively oxidized to the corresponding incarcerated quinones (89-93). This was accomplished by refluxing in CC1 4 containing Ce(NH 4)2(N0 2)6-silica gel or T i ( 0 2 C C F 3 ) 3 (Scheme 1.19).78 Hemicarceplexes with quinone guests (89-93) could not be formed by complexation with the empty hemicarcerand 25 with heating, because 89-93 are too unstable and decompose readily in solution under these conditions.78 or^o-Quinone oxidation products in hemicarceplexes 25*91 7ft and 24*93 were more stable than the free species in solution, which decompose with time. Regeneration of the hydroxy guest derivatives 84-88 was also possible using reducing agents such as Sml 2 or Na 2 S 2 04. 7 8 Water formed in the reaction is believed to egress through the 7ft equatorial portals defined by the tetramethylene bridges. In both oxidations and reductions, the bulk phase reagent (oxidant/reductant) reacts without completely entering the host, since the cavity is too small to accommodate both reactants.78 These reactions demonstrated that electrons could be transferred chemically between a hemicarceplex guest and a reactant from the bulk 46 Scheme 1.19 Oxidation/reduction o f guests entrapped in 25»guest. 25*89 25*91 25*92 25*93 25*94 1.4.3 Alkylation Reactions Ortho, meta, and para disubstituted phenol derivatives entrapped within hemicarcerand 25 have been subjected to H / D exchange and SN2 alkylation reactions. The outcomes o f these reactions were dependent upon the portal size between tetramethylene linkages and the orientation o f the incarcerated guest. 8 0 para-Disubstituted benzene guests are oriented within 25 with the para-aiyl substituents protruding into the northern and southern hemispheres (bowls), while o f ortho and meta derivatives, one aryl substituent is in a bowl and the other is in the equatorial region (Figure 1.10).34 Thus, metal ortho substituents near the equatorial region are exposed to the bulk phase. Note that nonpolar substituents (i.e., methyl groups) prefer to sit in the bowls, while polar substituents (i.e., OH) prefer the equatorial region. 47 In 25»l ,4-CH 3 C 6 H 4 OH, the guest is more orientationally fixed than in 25*1,4-H O ( C 6 H 4 ) O H . 8 0 A s a result, in D 20-saturated CDCI3 in the presence o f D B U , H / D exchange of the hydroxyl protons occurs for 25*l ,4-HO(C 6 H 4 )OH, but for 2 5 * l , 4 - C H 3 C 6 H 4 O H . 8 0 Repulsive interactions between the polar para substituents of 1 ,4 -HO(C 6 H 4 )OH with the nonpolar aryl-lined hemicarcerand interior are believed to allow the phenolic O H protons to spend more time near the large equatorial portals where they are more accessible to bulk phase reagents.8 0 25*guest 25*guest Figure 1.10 Orientations o f aryl phenol guests and transition state o f alkylation through the equatorial portals of 25*guest For ortho and meta disubstituted benzene guests, exposure o f the aryl substituents near the equator to the bulk phase was demonstrated with through-shell S N 2 reactions. Replacement of hydrogens with deuterium in 2-HOC6H 4 OH occurs upon treatment of 2 5 * 2 - H O C 6 H 4 O H with T H F - N a H at 2 5 °C, followed by D 2 0 quenching. Also , only the phenols o f ortho or meta substituents underwent methylation with N a H / M e l in T H F (Figure 1.10).3 1 The alkoxy moieties in the equatorial region o f 25 were explained to partially solvate the linear transition state necessary for S N 2 reactions. 3 1 Alkylation using larger electrophiles was not possible. 3 1 48 1.4.4 Generation and Stabilization of Cyclobutadiene Cyclobutadiene (98) has interested chemists in regards to HiickePs theory of aromaticity. Conjugated four membered ring systems like 98 are known to be highly strained and are extremely unstable compounds with short lifetimes under normal conditions. However, while entrapped within the interior of hemicarcerand 11a, cyclobutadiene (98) has been observed to exist as a stable ground state singlet in solution even at 60 °C. 8 1 This was accomplished by photolysis of the known precursor, oc-pyrone (95) as a guest in hemicarceplex l la«95, which was prepared by complexing 95 to empty hemicarcerand 11a. Hemicarceplex l la»95 was photolyzed in both CDCI3 and THF-dg solutions. Short irradiation times gave l la»97, which extruded C O 2 to give l la»98 after longer irradiation times.81 Further irradiation of l la«98 resulted in cycloreversion, splitting cyclbutadiene 98 into two acetylene molecules, which were released from the cavity of 11a.81 When l la»98 was heated in THF-c?8 (99), guest exchange occurred to give l la»99 and free cyclooctatetraene (101). As a free guest, cyclobutadiene 98 was suggested to have dimerized to form the transient o 1 intermediate 100, which then rearranges to cyclooctatetraene 101. Hemicarceplexes l la«98 and l la«97 were also subjected to other reaction conditions.81 Exposure of hemicarceplex l la«98 to O2 resulted in the formation of l la«102. 8 1 Thermally induced rearrangements of entrapped photopyrone (97) were also demonstrated in the conversion of l la«97 to l la»95 via the intermediate lla*96.81 Hemicarceplex l la»98 represents the first stabilization of a highly reactive intermediate within the inner phase of a molecular container. 49 Scheme 1.20 Inner phase reactions conducted within hemicarceplex lla«guest. The thermal transformations in the dotted rectangle occur in free solution. 11a«95 11a«97 11a«98 11a»99 1.4.5 Generation and Stabilization of ort/io-Benzyne Another versatile yet short-lived intermediate in organic synthesis is ort/io-benzyne (104). Benzyne is involved in several nucleophillic aromatic substitution reactions and is a hybrid of the two canonical forms (A and B): A B The first report on the generation and stabilization of benzyne (104) in solution was as hemicarceplex 25*104, via a series of photolysis experiments from 25*102. Hemicarceplex 25*102, prepared by heating hemicarcerand 25 in molten 102, was photolyzed in solution at room temperature to give 25*103. Benzocyclopropanone (103) is known to be highly strained and is normally only observed free in solution at low temperatures (-78 °C). 8 3 Photolysis of 25*103 in THF-«? 8 solution (A. = 280 nm) at low temperature gave 25»104 and 25»benzene, which were detected by ' H and 1 3 C N M R spectroscopies. Once formed, 104 in 25*104 is short-lived {t\/2 - 205 s), as a regioselective Diels-Alder innermolecular addition with the interior o f 25 occurs to give adduct 105. 8 4 Scheme 1.21 Generation o f ortho-benzyne (104) in hemicarcerand 25 105 1.4.6 Generation and Stabilization of Cycloheptatriene Carbenes have also been studied within the inner phase o f hemicarceplexes. Warmuth and coworkers have conducted an ongoing investigation into the photolytic generation o f phenyl carbenes from aziridine precursors entrapped within the hemicarcerands 25 and 106b-d (Figure O f 1.11). Phenylaziridine guests (i.e., 107) extrude N 2 upon photoexcitation at low temperatures in solution to generate singlet phenylcarbenes (109 , Scheme 1.22). The highly reactive 51 phenylcarbene 109 can undergo regioselective innermolecular insertion with the C - H s o f the hemispheric acetal O C H 2 0 moieties (X) o f 25.85 The rate o f these insertions was cleverly retarded by introducing a kinetic isotope effect upon replacement the O C H 2 O hydrogens with deuterons (i.e., as in 106b).85 Further irradiation of 106b»109 at higher wavelengths (>416 nm) at low temperatures successfully afforded the incarcerated ring expansion product, cycloheptatriene 110 (31 %), which is stable in hemicarceplex 106b»110 even at 100 ° C . 8 5 Jtf o N ^ o ' ^ / ON 25 X = CH 2 , A = (CH 2 ) 4 48 X = CH 2 , A = H, H 106a X = CD 2 , A = H, H 106bX = CD 2 , A = (CH 2 ) 4 106cX = CH 2 , 106dX = CD, > A = - - X X . . (S.S) ' } A = H 2 C i : 0 ) . J H 2 C -y* 0 >CH3 ' C H 3 Figure 1.11 Hemicarcerands used in cycloheptatriene studies. Scheme 1.22 Generation and stabilization 106a«cycloheptatrienes. 106b.108 ISC = inter-system crossing. 52 In order to investigate the inner phase reactivity of entrapped cycloheptatrienes, Warmuth and coworkers measured the energy barriers for the enantiomerization between the pairs, llOa'/llOa" and llObVllOb", which had not been previously done (Scheme 1.23). 8 5 c The allene forms o f cycloheptatriene (HOaVllOa") are speculated to undergo entantiomerization via 85b a process involving a transition state resembling the intermediate carbene 111 (Scheme 1.23). Diastereomeric hemicarceplexes 106d«110a' and 106d«110a" were generated by photolyzing hemicarceplex 106d«107a, and their rates of entantiomerization were measured by variable temperature N M R spectroscopy. 8 5 0 The average energy barriers for enantiomerization measured were 20.0 kcal/mole and 21.2 kcal/mole for the hemicarceplex pairs 106d«110b7106d»110bM and 106a»110a'/106a»110a", respectively. Both energy barriers were in good agreement with theoretical calculated values. 8 5 0 The lower value for the former is consistent with the stabilizing effect invoked by the para-methy\ substituent of phenyl carbenes. 8 5 0 Scheme 1.23 Enantiomerization of cycloheptatriene. R a R = H R 110' b R = C H 3 110" Hemicarceplex 106b«110 was also subjected to several intermolecular inner phase addition reactions. 8 5 Addit ion of HC1 to 106b»110a gave the mono-chlorinated product 106b«112. 8 5 a While 106b»110a did not react directly with M e O H , 106b«113 was formed from 102b»112 under the same conditions. 8 5 3 It is unusual that M e O H does not add to 53 cycloheptatriene (110a), as free cycloheptatrienes are known to readily react with alcohols or H 2 0 in solution. 8 5 The lack of reactivity for 110 in 106b*110a was rationalized as fo l lows. 8 5 0 Cycloheptatriene 110a is believed to react via the transition state 111, which is predicted to be more stable than the allene forms (enantiomers HOaVllOa") in polar solvents due to its larger predicted dipole moment. 8 5 0 However, while entrapped in 106b, 111 is destabilized via unfavorable interactions with host, despite that *H N M R chemical shift data suggests that the guest 110a adopts an orientation in the host with the allenic carbon positioned near the equator where it is l ikely accessible to an invasive M e O H from the bulk phase. 8 5 0 20 kcal/mole o f energy is estimated to have to be provided for a reaction between 106b»110a and M e O H to Scheme 1.24 Intermolecular inner phase reactions with 106a»110. Novel reactivity between the entrapped cycloheptatriene 110 and bulk phase reactants has also been discovered. 8 5 3 Cycloheptatrienes such as 110 do not normally react with 0 2 in solution, however, exposure of 106b«110a to 0 2 resulted in the isolation o f 106b«117 (Scheme 1.25). 8 5 b This reaction was proposed to first proceed via the formation o f dioxirane 114, which Oft undergoes ring contraction to form the intermediate, norcaradiene (115). Homolytic cleavage o f the 0 - 0 bond o f the three member peroxide ring of 115 then ensues, and is followed by a 54 chelotropic retrocyclization to form 106a»117 upon release of C O 2 to the bulk phase (Scheme 1.25). 8 5 b Unfortunately, no hard experimental evidence supporting this mechanism was presented. Scheme 1.25 Proposed mechanism of intermolecular inner phase reaction between O2 with incarcerated cycloheptatriene. 106b«110a 106b*114 106b«115 106b*116 106b*117 1.5 Conclusions This chapter demonstrates recent and important developments towards the synthesis and development of large supramolecular host-guest assemblies. Rig id spherical-shaped hosts that can completely surround a complexed guest have been known for nearly two decades and a great deal o f information has been gathered that has greatly increased our understanding o f the driving forces behind self-assembly processes. Improved knowledge o f noncovalent interactions such as hydrogen-bonding, electrostatics, pi-stacking, C X - p i ( X = H , F , C l , Br , I), and van der Waals interactions have allowed chemists to create and develop larger and more sophisticated systems o f nanoscopic proportions. Container molecules such as carceplexes, hemicarceplexes, self-assembling capsules, and several other molecular cavities have provided a vehicle for the study o f template effects in chemistry. It has been shown in this chapter that the sophistication o f these systems is 55 continuously increasing to a much grander scale, enabling the entrapment o f larger and more esoteric chemical entities as single or multiple-molecule guests. The latter holds considerable interest as it may provide some indication as to whether or not solvent effects can be differentiated from multiple-molecule template effects. Although a concrete answer may not be reached, the work in this thesis presents an attempt to shed some light into this "grey" area. Closing-off a microenvironment o f solvent molecules within a discrete boundary also augers wel l for the investigation of how clusters of molecules behave. When does a cluster of molecules constitute a phase? Can phase transitions be observed for an encapsulated microenvironment within a single molecule host? Are they altered in comparison to that observed for the bulk phase? Another interesting question is about the microscopic structure of an entrapped microenvironment. Is it possible to probe how the molecules constituting a medium are ordered with respect to each other within a capsular structure? Do interactions with the host interior have an effect on this? Can solvent spheres of different orders an magnitudes about various analytes be studied within such an enclosed environment? A n improved understanding o f these phenomena could allow improved control o f specific processes that occur within them. The work represented in this thesis attempts to address some of these questions. 56 Reinhoudt, D . N . ; Stoddart, F . J.; Ungaro, R. Chem. Eur. J. 1998, 4, 1349-1351. (a) Lehn, J. - M . Science 1993, 260, 1762-1763. (b) Weber, E . Chapter 1, Introduction and Historical Perspective, Vogtle, F . V o l . Ed. ; in the series Comprehensive Supramolecular Chemistry; Lehn, J. - M . ; Atwood, J. L . ; Davies, J. E . D . ; MacNicol, D. D . ; Vogtle, F . , Ed.; Permagon: New York, 1996; V o l . 1, 1-28. Kyba, E. P., Helgeson, R. C ; Madan, K. ; Gokel, G. W.; Tarnowski, T. L. ; Moore, S. S.; Cram, D. J. Am. Chem. Soc. 1977, 99, 2564-2571. Breitenbach, J.; Boosfeld, J.; Vogtle, F. Chapter 2; Some General Strategies Towards Macrocyclic Systems; in the series Comprehensive Supramolecular Chemistry; Lehn, J. - M . ; Atwood, J. L . , MacNicol, D. D . , Vogtle, F . , Series Eds.; Permagon: New York, 1996; Vo l . 2, 29-67. Diederich, F . ; Stang, P. J. Templated Organic Synthesis; Wi ley-VCH; Weinheim, 2000. Busch, D . ; Vance, A . L . ; Kolchinski, A . G . Chapter 1; Molecular Template Effect: Historical View, Principles and Perspectives; in the series Comprehensive Supramolecular Chemistry; Lehn, J. - M . ; Atwood, J. L . , MacNicol, D . D . , Vogtle, F . , Series Eds.; Permagon: New York, 1996; Vo l . 9, 1-42. Pederson, C. J. J. Am. Chem. Soc. 1967, 89, 7017-7036. Chapter 2; Crown Ethers Bradshaw, J. S.; Izatt, R. M . ; Bordunov, A . V . ; Zhu, C. Y . ; Hathaway, J. K . Vo l . Ed. ; in the series Comprehensive Supramolecular Chemistry; Lehn, J. - M . ; Atwood, J. L . , MacNicol, D. D . , Vogtle, F . , Series Eds.; Permagon: New York, 1996; Vol . 1, 35-96. (a) Diederich,, F. Cyclophanes; The Royal Society of Chemistry: Cambridge, 1991. (b) Seel, C ; Vogtle, F . Angew. Chem. Int. Ed. Engl. 1992, 31, 528-549. Gutsche, D . C. Calixarenes; The Royal Society of Chemistry: Cambridge, 1994. Vicens, J.; Bohmer, V . Calixarenes: A Versatile Class of Macrocyclic Compounds, Kluwer Academic Publishers: Boston, 1991. Ikeda, A . ; Shinkai, S. Chem. Rev. 1997, 97, 1713-1734. 11. Cram, D . J . ; Cram, J. M . Container Compounds and Their Guests; The Royal Society of Chemistry: Cambridge, 1994. 12. Collet, A . Tetrahedron 1987, 43, 24, 5725-5759. Collet, A . . Chapter 11, Cryptophanes; Vogtle, F. Vo l . Ed. ; in the series Comprehensive Supramolecular Chemistry; Lehn, J. - M . ; Atwood, J. L . ; Davies, J. E . D. ; MacNicol, D . D . ; Vogtle, F . , Ed.; Permagon: New York, 1996; Vol . 2, 325. 13. Timmerman, P.; Verboom, W. ; Reinhoudt, D. N . Tetrahedron 1996, 52, 2663. 14. (a) Maverick, E . ; Cram, D. J. Chapter 12; Carcerands and Hemicarcerands: Hosts that Imprison Molecular Guests; Vogtle, F . Vo l . Ed. ; in the series Comprehensive Supramolecular Chemistry; Lehn, J. - M . ; Atwood, J. L . ; Davies, J. E . D . ; MacNicol, D . D . ; Vogtle, F . , Ed. ; Permagon: New York, 1996; Vo l . 2, 367-418. (c) Sherman, J. C . In Large Ring Molecules; Semlyen, J. A . , Ed. ; J. Wiley & Sons: West Sussex, England, 1996; pp 507-524. (d) Sherman, J. C. Tetrahedron 1995, 51, 3395-3422. (e) Jasat, A . ; Sherman, J. C. Chem. Rev. 1999, 931-967. 15. (a) Hof, F . ; Craig, S.; Nuckolls, C ; Rebek, J. Jr. Angew. Chem. Int. Ed. Engl. 2002, 41, 1488-1508. (b) Rebek, J. Jr. Acc. Chem. Res. 1999, 32, 278-286. (c) de Mendoza, J. Chem. Eur. J. 1998, 4, 1373-1377. (d) Conn, M . M . ; Rebek, J. Jr. Chem. Rev. 1997, 97, 1647-1668. (e) Rebek, J. Jr. Chem. Soc. Rev. 1996, 255-264. 16. Sherman, J. C ; Cram, D . J. / . Am. Chem. Soc. 1989, 111, 4527-4528. 17. Hof, F . ; Rebek, J. Proc. Nat. Acad. Sciences 2002, 99, 4775-4777. 18. (a) Warmuth, R.; Yoon, J. Acc. Chem. Res. 2001, 34, 95-105. (b) Warmuth, R. Eur. J. Org. Chem. 2001, 423-437. (c) Warmuth, R. / . Inch Phenom. and Macrocyclic Chem. 2000, 37, 1-38. (d) Goto, K . ; Okazaki, R. Liebigs Ann./Recueil 1997, 2393-2407. 19. Sherman, J. C ; Knobler, C. B . ; Cram, D. J. / . Am. Chem. Soc. 1991, 113, 2194-2204. 58 20. (a) Chapman, R. G . ; Chopra, N . ; Cochien, E. D. ; Sherman, J. C. / . Am. Chem. Soc. 1994, 116, 369-370. (b) Chapman, R. G . ; Sherman, J. C. J. Org. Chem. 1998, 63, 4103-4110. 21. Cram, D. J.; Tanner, M . E . ; Knobler, C. B. J. Am. Chem. Soc. 1991, 113, 7717-7727. 22. Chopra, N . ; Sherman, J. C. Supramol. Chem. 1995, 5, 31-37. 23. Chapman, R. G . ; Sherman, J. C. J. Am. Chem. Soc. 1995, 117, 9081-9082. 24. Fraser, J. R.; Borecka, B . ; Trotter, J.; Sherman, J. C. / . Org. Chem. 1995, 60, 1207-1213. 25. Chapman, R. G. ; Olovsson, G . ; Trotter, J . ; Sherman, J. C. J. Am. Chem. Soc. 1998, 120, 6252-6260. 26. Nakamura, K . ; Sheu, C ; Keating, A . E . ; Houk, K . N . J. Am. Chem. Soc. 1997, 119, 4321-4322. 27. Chapman, R. G. ; Sherman, J. C. J. Am. Chem. Soc. 1998, 120, 9818-9826. 28. (a) Bryant, J. A . ; Blanda, M . T.; Vincenti, M . ; Cram, D . J. J. Chem. Soc, Chem. Commun. 1990, 1403-1405. (b) Bryant, J. A . ; Blanda, M . T.; Vincenti, M . ; Cram, D. J. J. Am. Chem. Soc. 1991,113, 2167-2172. 29. Jasat, A . ; Sherman, J. C. Unpublished results. 30. Makeiff, D. A . ; Pope, D. J.; Sherman, J. C. / . Am. Chem. Soc. 2000, 122, 1337-1342. 59 31. (a) Robbins, T. A . ; Knobler, C. B . ; Bellow, D. R.; Cram, D. J. / . Am. Chem. Soc. 1994, 116, 111-122. (b) Yoon, J.; Sheu, C ; Houk, K . N . ; Knobler, C. B . ; Cram, D . J. J. Org. Chem. 1996, 61, 9323-9339. 32. (a) Ro, S.; Rowan, S. J . ; Pease, A . R.; Cram, D. J . ; Stoddart, J. F . Org. Lett. 2000 (b) Piatnitshki, E . L . ; Flowers II, R. A . ; Deshayes, K . Chem. Eur. J. 2000, 6, 999-1017. (c) Cram, D . J.; Helgeson, R. C ; Knobler, C. B . ; Maverick, E . F. Tetrahedron Lett. 2000, 9465-9470. (d) Paek, K . ; Dim, H . ; Yun, S.; Lee, H . C ; No, K . T. / . Org. Chem. 2001, 66, 5736-5743. (e) Singh, H . ; Warmuth, R. Tetrahedron 2002, 1257-1264. 33. Byun, Y . -S . ; Robbins, T. A . ; Knobler, C. B . ; Cram, D . J. / . Chem. Soc, Chem. Commun. 1995, 1947-1948. 34. Helgeson, R. C ; Paek, K . ; Knobler, C. B . ; Maverick, E . F . ; Cram, D . J. / . Am. Chem. Soc. 1996, 118, 5590-5604. 35. van Wageningen, A . M . A . ; van Duynhoven, J. P. M . ; Verboom, W.; Reinhoudt, D. N . J. Chem. Soc, Chem. Commun. 1995, 1941-1942. 36. (a) van Wageningen, A . M . A . ; Timmerman, P.; van Duynhoven, J. P. M . ; Verboom, W. ; van Veggel, F. C. J. M . ; Reinhoudt, D . N . Angew. Chem. Int. Ed. Engl. 1997, 3, 639-654. A similar carceplex to 38«guest has also been synthesized by Paek and coworkers, which features ethylene bridges (-0(CH2)20-) bridges rather than amides. See (b) Ihm, H . ; Paek, K . Bull. Kor. Chem. Soc. 1999, 20, 757-760. 37. 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. 38. (a) Jacopozzi, P.; Dalcanale, E . Angew. Chem. Int. Ed. Engl. 1997, 36, 613-615. (b) Fochi, F . ; Jacopozzi, P.; Wegelius, E . ; Rissanen, K . ; Cozzini, P.; Marastoni, E . ; Fisicaro, E . ; Manini, P.; Fokkens, R.; Dalcanale, E . / . Am. Chem. Soc. 2001,123, 7539-7552. 39. Cuminetti, N . ; Ebbing, M . H . K . ; Prados, P.; de Mendoza, J . ; Dalcanale, E . Tett. Lett. 2001, 42, 527-530. 60 40. (a) Fox, O. D . ; Dalley, K . ; Harrison, R. G. / . Am. Chem. Soc. 1998, 120, 7111-7112. (b) Fox, O. D . ; Leung, J. F. - Y . ; Hunter, J. M . ; Dalley, K . ; Harrison, R. G . ; Inorg. Chem. 2000, 39, 783-790. 41. Fox, D . O.; Dalley, N . K . ; Harrison, R. G. Inorg. Chem. 1999, 38, 5860-5863. 42. (a) Chopra, N . ; Sherman, J. C. Angew. Chem. Int. Ed. Engl. 1997. 36, 1727-1729. (b) Chopra, N . ; Naumann, C ; Sherman, J. C. Angew. Chem. Int. Ed. Engl. 2000, 39, 194-196. (c) Our group has also recently reported the formation of a tris-carceplex and tris-capsule from a cyclic hexamer of [4]cavitands. Naumann, C ; unpublished results. Mungaroo, R.; Sherman, J. C . / . Chem. Soc, Chem. Commun. 2002, 1672-1673. 43. Chopra, N . ; Sherman, J. C. Angew. Chem. Int. Ed. Engl. 1999, 38, 1955-1957. 44. For other multi-[4]cavitand assemblies with similar topologies to trimer 43 and tetramer 45 reported by other research groups, see (a) Fox, D . O.; Drew, M . G . B . ; Beer, P. D . Angew. Chem. Int. Ed. Engl. 2000, 39, 136-140. (b) Fox, D . O.; Drew, M . G . B . ; Wilkinson, E . J. S.; Beer, P. D . J. Chem. Soc, Chem. Commun. 2000, 391-392. (c) Timmerman, P.; Verboom, W. ; Van Veggel, F. C. J. M . ; van Hoorn, W. P.; Reinhoudt, D . N . Angew. Chem. Int. Ed. Engl. 1994, 33, 1292-1295. 45. Yoon, J . ; Cram, D.J.J. Chem. Soc, Chem. Commun. 1997, 2065-2066. 46. For other related multi-capsular host-guest structures, see (a) Castellano, R. K . ; Rudkevich, D . M . ; Rebek, J. Jr. Proc. Natl. Acad. Sci. 1997, 94, 7132-7137. (b) Castellano, R. K . ; Rebek, J. Jr. J Am. Chem. Soc. 1998, 120, 3657-3663. (c) Brody, M . S.; Schalley, C . A . ; Rudkevich, D. M . ; Rebek, J. Jr. Angew. Chem. Int. Ed. Engl. 1999, 38, 1640-1644. (d) Schalley, C. A . Adv. Mater. 1999, 11, 1535-1537. (e) Rebek, J . Jr. J. Chem. Soc, Chem. Commun. 2000, 637-643. (f) Castellano, R. K . ; Clark, R.; Craig, S. L . ; Nuckolls, C ; Rebek, J. Jr. Proc. Natl. Aced. Sci. 2000, 97, 12418-12421. 47. Naumann, C ; Roman, E . ; Peinador C ; Ren, T.; Patrick, B . O.; Kaifer, A . E . ; Sherman, J. C. Chem. Eur. J. 2001, 7, 1637-1645. 48. Naumann, C ; Place, S.; Sherman, J. C. J. Am. Chem. Soc. 2002, 124, 16-17. 61 49. (a) Wyler, R.; Mendoza, J.; Rebek, J. Jr. Angew. Chem. Int. Ed. Engl. 1993, 32, 1699-1701. (b) Branda, N . ; Wyler, J. Rebek, J. Jr. Science 1994, 263, 1267-1268. (c) Valdes, C ; Spitz, U . P.; Toledo, L. M . ; Kubik, S. W.; Rebek, J. Jr. J. Am. Chem. Soc. 1995,117, 12733-12745. (d) Branda, N . ; Grotzfeld, R. M . ; Valdes, C ; Rebek, J. Jr. J. Am. Chem. Soc. 1995,117, 85-88. (e) Garcias, X . ; Rebek, J. Jr. Angew. Chem. Int. Ed. Engl. 1996, 35, 1225-1228. (f) Szabo, T.; Hilmersson, G.; Rebek, J. Jr. 1998,120, 6193-9194. (g) Evans, D. A . ; Evans, J. S. / . Org. Chem. 1998, 63, 8027-8030. 50. (a) Meissner, R. S.; Rebek, J. Jr.; de Mendoza, J. Science 1995, 270, 1485-1488. (b) Tokunaga, Y . ; Rudkevich, D . M . ; Rebek, J. Jr. 1997, 36, 2656-2659. (c) Rivera, J. M . ; Martin, T.; Rebek, J. Jr. J. Am. Chem. Soc. 1998,120, 819-820. (d) Tokunaga, Y . ; Rebek, J. Jr. J. Am. Chem. Soc. 1998, 120, 66-69. (e) Tokunaga, Y . ; Rudkevich, D . M . ; Santamaria, J . ; Hilmersson, G . ; Rebek, J. Jr. Chem. Eur. J. 1998, 4, 1449-1457. (f) Hilmersson, G . ; Rebek, J. Jr. Magn. Res. In Chem. 1998, 36, 663-669. (g) Santamaria, J . ; Martin, T.; Hilmersson, G . ; Craig, S.; Rebek, J . Jr. 1999 Proc. Natl. Aced. Sci. USA, 96, 8344-8347. (h) Rivera, J. M . ; Rebek, J. Jr. 2000,122, 7811-7812. (i) Rivera, J. M . ; Craig, S.; Martin, T.; Rebek, J. Jr. 2000 Angew. Chem. Int. Ed. Engl, 39, 2130-2132. (j) Rivera, J. M . ; Martin, T.; Rebek, J. Jr. 2001 J. Am. Chem. Soc, 123, 5213-5220. 51. Kang, J.; Rebek, J. Jr. Nature 1996, 382, 239-241. 52. (a) Meissner, R.; Garcias, X . ; Mecozzi, S.; Rebek, J. Jr. / . Am. Chem. Soc. 1997, 119, 77-85. (b) Kang, J.; Rebek, J. Jr. Nature 1996, 382, 239-241. 53. (a) Kang, J . ; Hilmersson, G . ; Santamaria, J . ; Rebek, J. Jr. J. Am. Chem. Soc. 1998, 120, 3650-3656. (b) Kang, J.; Rebek, J. Jr. Nature 1997, 385, 50-52. 54. For other reversibly forming capsules monomers that incorporate glycouril functionalities, see: (a) Grotzfeld, R. M . ; Branda, N . ; Rebek, J. Jr. Science 1996, 271, 487-489. (b) O'Leary, B . M . ; Szabo, T.; Svenstrup, N . ; Schalley, C. A . ; Lutzen, A . ; Schafer, M . ; Rebek, J. Jr. J. Am. Chem. Soc. 2001,123, 11519-11533. (c) Lutzen, A . ; Renslo, A . R.; Schalley, C. A . ; O'Leary, B. M . ; Rebek, J. Jr. J. Am. Chem. Soc. 1999, 121, 7455-7456. (d) Lutzen, A . ; Starnes, S. D . ; Rudkevich, D . M . ; Rebek, J . Jr. Tetrahedron Lett. 2000, 41, 3771-3178. (e) Martin, T.; Obst, U . ; Rebek, J. Jr. Science 1998, 281, 1842-1845. (f) Nuckolls, C ; Hof, F . ; Martin, T.; Rebek, J. Jr. / . Am. Chem. Soc. 1999, 121, 10281-10285. (g) Hof, F . ; Nuckolls, C ; Rebek, J. Jr. J. Am. Chem. Soc. 2000, 122, 4251-4252. (h) Hof, F . ; Nuckolls, C ; Craig, S. L . ; Martin, T.; Rebek, J. Jr. / . Am. Chem. Soc. 2000, 122, 10991-10996. 62 55. Kang, J.; Santamaria, J . ; Hilmersson, G . ; Rebek, J. Jr. / . Am. Chem. Soc. 1998,120, 7389-7390. 56. For Diels-Alder additions within a different capsule, see Ooi, T.; Kondo, Y . ; Maruoka, K . Angew. Chem. Int. Ed. Engl. 1998, 37, 3039-3041. 57. Saito, S.; Nuckolls, C ; Rebek, J. Jr. J. Am. Chem. Soc. 2000,122, 9628-9630. 58. For other related noncapsular hosts, see: (a) Rudkevich, D. M . ; Hilmersson, G . ; Rebek, J. Jr. / . Am. Chem. Soc. 1997, 119, 9911-9912. (b) Rudkevich, D . M . ; Rebek, J. Jr. Eur. J. Org. Chem. 1999, 1991-2005. (c) Haino, T.; Rudkevich, D . M . ; Rebek, J. Jr. / . Am. Chem. Soc. 1999,121, 11253-11254. (d) Ma , S.; Rudkevich, D . M . ; Rebek, J . Jr. Angew. Chem. Int. Ed. Engl. 1999, 38, 2600-2602. (e) Tucci, F . ; Rudkevich, D . M . ; Rebek, J. Jr. J. Org. Chem. 1999, 64, 4555-4559. (f) Renslo, A . R.; Rudkevich, D . M . ; Rebek, J. Jr. / . Am. Chem. Soc. 1999,121, 7459-7460. (g) Saito, S.; Nuckolls, C ; Rebek, J. Jr. J. Am. Chem. Soc. 2000, 122, 9628-9630. (h) Renslo, A . R.; Tucci, F . ; Rudkevich, D. M . ; Rebek, J. Jr. J. Am. Chem. Soc. 2000, 122, 4573-4582. (i) Renslo, A . R.; Rebek, J. Jr. Angew. Chem. Int. Ed. Engl. 2000, 39, 3281-3283. 0") Shivanyuk, A . ; Rissanen, K . ; Korner, S. K . ; Rudkevich, D . M . ; Rebek, J. Jr. Helv. Chim. Acta 2000, 83, 1178-1790. (k) Starnes, S. D.; Rudkevich, D . M . ; Rebek, J. Jr. Org. Lett. 2000, 123, 4659-4669. (1) Lucking, U . ; Chen, J . ; Rudkevich, D . M . ; Rebek, J. Jr. / . Am. Chem. Soc. 2001,123, 9929-9934. (m) Saito, S.; Rebek, J. Jr. Bioorganic & Medicinal Chemistry Letters 11 2001, 1497-1499. (n) Gibson, C ; Rebek, J. Jr. Org. Lett. 2002, 4, 1887-1890. (o) Far, A . R. ; Shivanyuk, A . ; Rebek, J. Jr. / . Am. Chem. Soc. 2002, 124, 2854-2855. (p) Amrhein, P.; Wash, P. L . ; Shivanyuk, A . ; Rebek, J. Jr. Org. Lett. 2002, 4, 319-321. (q) Far, A . R.; Cho, Y . L . ; Rang, A . ; Rudkevich, D . M . ; Rebek, J. Jr. Tetrahedron 2002, 58, 741-755. (r) Tucci, F . ; Renslo, A . R.; Rudkevich, D . M . ; Rebek, J. Jr. Angew. Chem. Int. Ed. Engl. 2000, 29, 1076-1079. (s) Lucking, U . ; Tucci, F . ; Rudkevich, D . M . ; Rebek, J. Jr. / . Am. Chem. Soc. 2000, 122,- 8880-8889. (t) Starnes, S. D.; Rudkevich, D . M . ; Rebek, J. Jr. / . Am. Chem. Soc. 2001,123, 4659-4669. (u) Gibb, C . L . D . ; Stevens, E . D . ; Gibb, B. C. / . Am. Chem. Soc. 2001,123, 5849-5850. (v) Gibb, C . L . D . ; X i , H . ; Politzer, P. A . ; Concha, M . ; Gibb, B . C. Tetrahedron 2002, 58, 673-681. (w) Gibb, C . L . D . ; L i , X . ; Gibb, B. C. Proc. Nat. Acad. Sci. 2002, 99, 4857-4862. 59. (a) Heinz, T.; Rudkevich, D. M . ; Rebek, J. Jr. Nature 1998, 394, 764-766. (b) Tucci, F . ; Rudkevich, D . M . ; Rebek, J. Jr. / . Am. Chem. Soc. 1999,121, 4928-4929. (c) Heinz, T.; Rudkevich, D. M . ; Rebek, J. Jr. Angew. Chem. Int. Ed. Engl. 1999, 38, 1136-1139. (d) Korner, S. K . ; Tucci, F. C ; Rudkevich, D. M . ; Heinz, T.; Rebek, J. Jr. Chem. Eur. J. 2000, 6, 187-195. 63 60. For chemical applications of Rebek's cylindrical capsules, see: (a) Chen, J . ; Rebek, J. Jr. Org. Lett. 2002, 4, 327-329. (b) Chen, J . ; Korner, S.; Craig, S. L . ; Rudkevich, D. M . ; Rebek, J. Jr. Nature 2002, 415, 385-386. (c) Chen, J . ; Korner, S.; Craig, S. L . ; L in , S.; Rudkevich, D. M . ; Rebek, J. Jr. Proc. Nat. Acad. Sci. 2002, 99, 2593-2596. Also see reference 59d. 61. (a) MacGillivray, L . R.; Atwood, J. L . Angew. Chem. Int. Ed. Engl. 1999, 38, 1018-1033. (b) Olenyuk, B . ; Whiteford, J . A . ; Fechtenkotter, A . ; Stang, P. J . Nature 1999, 398, 796-799. (c) Olenyuk, B . ; Levin, M . D . ; Whiteford, J. A . ; Shield, J. A . ; Stang, P. J. /. Am. Chem. Soc. 1999, 121, 10434-10435. (d) Leininger, S.; Olenyuk, B . ; Stang, P. J . Chem. Rev. 2000, 100, 853-908. (e) Schweiger, M . ; Seidel, S. R. ; Schmitz, M . ; Stang, P. J. Org. Lett. 2000, 2, 1255-1257. (f) Caulder, D . L . ; Raymond, K . N . Acc. Chem. Res. 1999, 32, 975-962. 62. Fujita, M . ; Oguro, D . ; Miyazawa, M . ; Oka, H . ; Yamaguchi, K . ; Ogura, K . Nature 1995, 378, 469-471. 63. Kusukawa, T.; Fujita, M . Angew. Chem. Int. Ed. Engl. 1998, 37, 3142-3144. 64 (a) Kusukawa, T.; Fujita, M . / . Am. Chem. Soc. 1999, 121, 1397-1398. (b) Kusukawa, T.; Yoshizawa, M . ; Fujita, M . Angew. Chem. Int. Ed. Engl. 2001, 40, 1879-1884. 65 (a) Ibukuro, F . ; Kusukawa, T.; Fujita, M . / . Am. Chem. Soc. 1998, 120, 8561-8562. (b) Sakamoto, S.; Yoshizawa, M . ; Kusukawa, T.; Fujita, M . ; Yamaguchi, K . Org. Lett. 2001, 3, 1601-1604. 66. Yoshizawa, M . ; Kusukawa, T.; Fujita, M . ; Yamaguchi, K . / . Am. Chem. Soc. 2000, 122, 6311-6312. 67. Pt-coordinated cage 76b was also shown to behave as a reverse-phase transfer (Wacker) oxidation catalyst in the conversion of entrapped styrene to acetophenone in the presence o f 0 2 . Ito, H . ; Kusukawa, T.; Fujita, M . Chem. Lett. 2000, 598-599. 68. Y u , S. - Y . ; Kusukawa, T.; Birada, K . ; Fujita, M . / . Am. Chem. Soc. 2000, 122, 2665-2666. 64 69. MacGillvray, L . R.; Atwood, J. L . Nature 1997, 389, 469-472. 70. Gerkensmeier, T.; Iwanek, W. ; Agena, C. ; Frohlich, R.; Kotila, S.; Nather, C . ; Mattay, J. Eur. J. Org. Chem. 1999, 2257-2262. 71. Atwood, J. L . ; Barbour, L . J . ; Jerga, A . J. Chem. Soc, Chem. Commun. 2001, 2376-2377. 72. For other examples of different resorcin[4]arene arrays observed in the crystalline solid state, see: (a) Murayama, K . ; Katsuyuki, A . J. Chem. Soc, Chem. Commun. 1998, 607-608. (b) Rose, K . N . ; Barbour; L . J.; Orr, W. G . ; Atwood, J. L . J. Chem. Soc, Chem. Commun. 1998, 407-408. (c) MacGillvray, L . R.; Atwood, J. L . J. Chem. Soc, Chem. Commun. 1999, 181-182. (d) Shivanyuk, A . ; Rissanen, K . ; Kolehmainen, E . J. Chem. Soc, Chem. Commun. 2000, 1107-1108. (e) Atwood, J . L . ; Barbour, L . J . ; Ness, T. J . ; Raston, C. L . ; Raston, P. L . J. Am. Chem. Soc. 2001, 123, 7192-7193. (f) MacGillvray, L . R.; Diamente, P. R.; Reid, J. L . ; Ripmeester, J. A . J. Chem. Soc, Chem. Commun. 2001, 359-360. 73. Aoyama, Y . ; Tanaka, Y . ; Toi, H . ; Ogoshi, H . J. Am. Chem. Soc 1988, 110, 634-635. Aoyama, Y . ; Tanaka, Y . ; Sugahara, S. J. Am. Chem. Soc. 1989, 111, 5397-5404. 74. (a) Shivanyuk, A . ; Rebek, J. Jr. Proc. Nat. Acad. Sci. 2001, 98, 7662-7665. (b) Shivanyuk, A . ; Rebek, J. Jr. J. Chem. Soc, Chem. Commun. 2001, 2424-2425. 75. Shivanyuk, A . ; Rebek, J. Jr. J. Chem. Soc, Chem. Commun. 2001, 2374-2375. 76. Ziegler, M . ; Brumaghim, J. L . ; Raymond, K . N . Angew. Chem. Int. Ed. Engl. 2000, 39, 4119-4121. 77. Parac, T. N . ; Caulder, D . L . ; Raymond, K . N . J. Am. Chem. Soc. 1998,120, 8003-8004. 78. Robbins, T.; Cram, D . J. J. Am. Chem. Soc. 1993,115, 12199. 65 79. For other references regarding through-shell electron transfer from the bulk phase to a hemicarceplex guest, see (a) Parola, A . J.; Pina, F . ; Maestri, M . ; Armaroli, N . ; Balzani, V . New J. of Chem. 1994, 18, 659-661. (b) Pina, F . ; Parola, A . J . ; Ferreira, E.; Maestri, M . ; Armaroli, N . ; Balardini, R.; Balzani, V . J. Phys. Chem. 1995, 99, 12701-12703. (c) Farran, A . ; Deshayes, K . ; Matthews, C ; Balanescu, I. / . Am. Chem. Soc. 1995,117, 9614-9615. (d) Farran, A . ; Deshayes, K . J. Phys. Chem. 1996, 100, 3305-3306. (e) Parola, A . J . ; Pina, F . ; Ferreira, E. ; Maestri, M . ; Balzani, V . J. Am. Chem. Soc. 1996, 118, 11610-11616. (f) Piatnitshki, E. L . ; Deshayes, K . Angew. Chem. Int. Ed. Engl. 1998, 37, 970-972. (g) Place, I.; Farran, A . ; Deshayes, K . ; Piotrowiak, P. J. Am. Chem. Soc. 1998, 120, 12626-12633. (h) Mendoza, S.; Davidov, P .D. ; Kaifer, A . E . Chem. Eur. J. 1998, 4, 864-870. (i) Romanova, Z. S.; Deshayes, K . ; Piotrowiak, P. J. Am. Chem. Soc. 2001, 123, 2444-2445. (j) Romanova, Z. S.; Deshayes, K . ; Piotrowiak, P. / . Am. Chem. Soc. 2001,123, 11029-11036. 80. Kurdistani, S. V . ; Helgeson, R. C ; Cram, D. J. J. Am. Chem. Soc. 1995, 117, 1659-1660. 81. Cram, D.J . ; Tanner, M . E.; Thomas, R. Angew. Chem. Int. Ed. Engl. 1991, 30, 1024-1027. 82. Warmuth, R. Angew. Chem. Int. Ed. Engl. 1997, 36, 1347-1349. 83. Simon, J. G . G . ; Munzel, N . ; Schweig, A . Chem. Phys. Lett. 1993, 201, 311. 84. Beno, B . R.; Chimin, S.; Houk, K . N . ; Warmuth, R.; Cram, D . J. Chem. Commun. 1998, 301-302. 85. (a) Warmuth, R.; Marvel, M . Angew. Chem. Int. Ed. Engl. 2000, 39, 6, 1117-1119. (b) Warmuth, R.; Marvel, M . Chem. Eur. J. 2001, 7, 1209-1220. (c) Warmuth, R. / . Am. Chem. Soc. 2001, 123, 6955-6956. (d) Warmuth, R.; Kerdelhue, J. - L . ; Carrera, S. S.; Langenwalter, K . J . ; Brown, N . Angew. Chem. Int. Ed. Engl. 2002, 41, 96-99. 66 2. Template Effects In the Formation of a Tetramethylene-bridged Hemicarceplex 2.1 Introduction Recall that in the beginning of Chapter 1, template effects in the formation o f carceplex 10a»guest' and hemicarceplex l la»guest 2 (Scheme 2.1) were discussed to proceed via the formation of preorganized hydrogen-bonded complexes, 12a«guest and 13a»guest, respectively. Clearly, capsule 12a»guest can form during the reaction to give 25»guest. The question we pose is: is capsule 12«guest a good transition state model for the formation of any carceplex/hemicarceplex derived from tetrol 8a? If the GDS occurs early (first or second bridges), this should be the case. If late (third or fourth bridge), the cavity size and shape o f the transition state may deviate significantly from that of 12a«guest; this would also demonstrate that 12a»guest is not necessarily a good transition state model for all compounds for which tetrol 8a is a precursor. Hemicarceplexes 25»guest have been prepared using several different methods. Cram reported the synthesis of hemicarceplexes 25»DMA, 25»DMSO, and 25«NMP (Scheme 2.1) by templation involving bridging reactions in the respective solvents between tetrol (8a, two equiv.), base, and 1,4-butanediolditosylate or 1,4-butanedioldimesylate (four equiv.). 4 These • 1 conditions are similar to the conditions used for carceplex 10a»guest from the same subunit, tetrol 8a.5 Empty hemicarcerand 25 was then obtained from 25»DMA by expelling the D M A guest from the interior by heating in a solvent (i.e., Ph20) that is too large to enter the interior o f 25.4 A wide variety of new hemicarceplexes were prepared from the empty hemicarcerand 25 by heating in a neat guest as solvent or PI12O doped with guest.4 Complexed guests included 67 aliphatic ethers and alkyl halides, cyclic ethers, lactones, lactams, as well as mono and disubstituted benzene derivatives. 4 In a third procedure, hemicarceplex 25«guest is formed from diol 48 by reaction with 1,4-butanedioldimesylate in the presence o f a base and guest.6 This procedure was useful for preparing hemicarceplexes 25»napthalene, 2 5 » C 6 H 5 O C H 2 C H = C H 2 , 25»quinoline, and 25«indole. 4 b * 7 Scheme 2.1 Capsule 12a«guest as a transition state model in the formation o f carceplexes/hemi carceplexes. Mm 8a + HO J2 O, Guest Base H. W 8a 12a«guest 4 B r ^ ^ c i 10a«guest 25«guest 11a«guest 48»guest Cram called 25«guest the most versatile of the many hemicarceplexes made due to its ability to form strong complexes with a large family of different guests.4 This property, and the fact that good yields can be obtained, has made hemicarceplex 25«guest an attractive system for numerous studies involving intermolecular inner phase reactions, 6 , 8 supramolecular photochemistry/photophysics, 9 and the generation and stabilization o f reactive intermediates. 1 0 ' 1 1 68 Since much was known about the chemistry of 25»guest, it was chosen as the best candidate to continue our investigations into template effects in carceplex/hemicarceplex formation. In this chapter, the template effects in the formation o f tetramethylene-bridged hemicarceplex 25»guest from shell-closure reactions involving tetrol 8a, base, and linker, in the presence of suitable guest molecules are reported. In total, 83 different templates were investigated, and 30 template ratios were determined. A recent account of the work in this chapter has already been published. 1 2 2.2 Results and Discussion 2.2.1 Reaction Conditions/Synthesis of Hemicarceplexes 25«guest The ideal solvent for the determination of template ratios in the formation o f carceplexes/hemicarceplexes is one that is a noncompetitive template and possesses the necessary properties to facilitate product formation. Many o f the solvents used by Cram for the formation o f 25«guest gave reasonable hemicarceplex yields (30 % for 25«DMA, 55 % for 25«NMP and 18 % for 25»DMSO). 4 However these solvents are too competitive as templates to be used for screening other potential guests (templates) at low concentrations. Several different solvents were tested by doping reactions with low concentrations (1 mole % of the solvent) o f a known template (e.g., D M A ) . 1 2 ' 1 3 N F P was determined to be the best out o f fifteen different solvents screened ( D M A , D M F , N M P , N F P , l-acetyl-3-methylpiperidine, tetramethylene sulfone, ethyl acetate, D M P U , nitrobenzene, cyclohexane, acetonitrile, acetone, 2-butanone, T H F and D M S O ) as it was the least competitive template that gave the best hemicarceplex yields 69 (-13 % for 25»DMA in N F P ) . 1 2 ' 1 3 Yields o f hemicarceplex 25»guest in N F P were generally 5-10%. Conditions used for screening potential guest molecules were: addition o f tetrol 8a, C s 2 C 0 3 (19 equiv. per tetrol 8a), K I (0.8 equiv. per linker), 1,4-dibromobutane (10 equiv. per tetrol 8a) and guest (1 to 50 mol % of the solvent) in N F P at 80 °C for 48 hours (Scheme 12 13 • 2.2). ' A l l hemicarceplexes obtained from shell-closure reactions in N F P were isolated as mixtures containing 25»guest and 25»NFP. Scheme 2.2 Templated synthesis of hemicarceplex 25«guest. R R R R C s 2 C 0 3 P H ;o KI, NFP o- . Br(CH 2) 4Br <S ^ > 3 80 °C Guest R R R R 25»guest 2.2.2 Unsuccessful templates A total of eighty-three guests were screened in separate reactions in N F P . The guests chosen for screening were molecules that were commercially available, inexpensive, and requiring little purification prior to use. Potential guests screened also had to be inert under the standard set o f reaction conditions (Scheme 2.1). Guests were chosen that appeared complementary to the interior o f 25 based on Corey-Pauling-Koltun ( C P K ) models, or that have 70 been reported by Cram as 25»guest.4 These include hemicarceplexes 25«guest produced by either templating or non-templating procedures.4"8 Of the total 83 guests screened, 36 were found to behave as unsuitable guests (Chart 2.1). These molecules are qualitatively categorized as being potentially too large in some dimension, too reactive, too small, too basic or too apolar to act as suitable templates in the formation of hemicarceplex 25»guest. Chart 2.1 Unsuitable guests for the templated formation of hemicarceplex 25»guest. Structures in the boxes represent the guests screened for this thesis. A l l other guests screened were previously reported. ' Too large in some dimension ; N H V ^ C H O H 3 C ^ C H 3 H 3 C A O H 3 H 3 C l 0 - C H 3 °U (^O CH3CN CH3OH CH3CH2OH CH 3CH 2CH 2OH Too reactive Too small H0HC^°H H 3C . 0 A 0.CH 3 Too basic Too apolar 71 2.2.3 Successful Templates Forty-seven o f the total 8 3 guests screened were found to act as suitable templates in the formation of 25»guest. Only 3 0 1 4 o f these 4 7 were used for the determination o f template ratios (Table 2 . 1 ) . Hemicarceplex products obtained from shell closure reactions in N F P were isolated as mixtures containing 25»guest and 25»NFP. In most cases, the hemicarceplex 25»guest products could not be separated by chromatography on normal phase silica gel due to similar retention factors. A l l hemicarceplex products present in the reaction mixtures were identified by ' H N M R spectroscopy and M A L D I mass spectrometry. Pure samples o f each new host-guest complex were prepared via guest exchange from 25»DMA while heating at high temperatures above 1 4 0 °C, either in Ph20 (a solvent too large to enter hemicarcerand 25) doped with a high concentration o f guest, or in neat guest (similar to Cram's method, 4 3 see experimental section). *H N M R assignments for the host signals of new complexes (see experimental section, Table 2 . 3 ) formed were based on analogy to similar reported hemicarceplexes (25»guest), which differ only in the imprisoned guest 4 - 8 The chemical shift difference ( A 8 ) between the free ( A S f t e e , hemicarcerand 25) and bound host ( A S b o u n d , 25«guest) proton signals are relatively small, ranging between - 0 . 0 9 and 0 . 2 5 ppm (see experimental section). In contrast, large upfield shifts ( 0 . 0 7 - 4 . 4 3 ppm) are observed for the protons of the entrapped guests due to the highly shielding magnetic environment provided by the arene-lined pi electron-rich walls of 25 . 4 a Better dispersion o f the guest proton signals is also characteristic for encapsulated species. For the remaining 1 7 o f the 4 7 suitable templates, ! H N M R spectroscopic and/or M A L D I mass spectrometric data on the product mixtures isolated from 1 7 separate reactions in N F P with the addition of these guests each provided evidence suggesting the templated formation of 7 2 25«guest. These 17 guests were not included in the templation study because homogeneous samples could not be obtained for any of the corresponding hemicarceplexes 25»guest. 1 5 2.2.4 Guest Orientation and Mobility Information regarding each guest's preferred orientation within 25 can be gleaned from the ' H N M R A8 values in Table 2.3 (experimental section). It has been previously demonstrated, that for carceplexes and hemicarceplexes, protons buried deep within the aryl lined hemispheres show large, positive A8 values, while protons near the equatorial region show smaller, less positive A5 values. 4 In this report, we found that A8 values for methyl and/or methoxy protons ofpara-xylene, 1,4-bromotoluene, 1,4-chlorotoluene, 1,4-methylanisole, 1,4-chloroanisole are large (4.14-4.43), while those for the ortho and meta aryl protons are much smaller (0.94-1.22). Other /?ara-disubstituted benzene guests with non-proton bearing substituents (para-dibromobenzene, 1 -bromo-4-chlorobenzene, jpara-dichlorobenzene, 1 -iodo-4-bromobenzene, and l-iodo-4-chlorobenzene), have relatively smaller AS values for the ortho and meta aryl protons as well (0.77-1.05). Therefore/?ara-disubstituted benzene guests are orientated within the host with the para substituents deep within the northern and southern hemispheres, while the aryl hydrogens are located near the equator (Figure 2.1). Two sets of host signals are also observed for hemicarceplexes 25«guest with /?ara-disubstituted benzene guests with unlike substituents, arising from the lack of rotation of the guest about the short (horizontal) axis o f the host on the 'FI N M R timescale. This interpretation is consistent with ' H N M R and x-ray crystallographic data gathered on 25»guest with other /;ara-disubstituted guests, provided by the Cram group 4 a 73 mono and para disubstituted benzenes large AS aliphatic alcohols 25«1,3-dimethyl-2- 25«isopropyl imidazolininone acetate 25.NFP Figure 2.1 Predicted guest orientations o f substituted benzene and aliphatic alcohol in 25»guest. For mono-substituted benzenes in this study, A5 values range from 3.32-3.76 forpara-hydrogens, 1.50-1.90 for meta-hydrogens, and 1.00-2.10 for ort/io-hydrogens. ' H N M R signals o f the proton bearing substituents of mono-substituted benzenes (anisole, thioanisole, and toluene) show large A5 values (3.09-3.99), indicating that they are situated deep within the hemispheres. This suggests that mono-substituted benzenes are orientated along the long axis o f the host such that the substituents and para hydrogens are located deep within opposing hemispheres, while the meta and o/t/io-hydrogens are near the equator (Figure 2.1). Spectra of 25»iodobenzene 4 a and 25«anisole also show significant broadening o f several host signals, which sharpen up upon heating. These guests experience restricted rotation about the horizontal axis o f the host below ambient temperatures. 74 Similar analysis of host-guest complexes formed with the remaining guests in Table 2.3 led to the following conclusions regarding their orientations inside host 25. Aliphatic alcohols (3-pentanol, 2-butanol, 2-pentanol, 2,4-pentanediol, and 3-hexanol) are positioned such that each terminal methyl group o f the guest lie deep within the two opposing hemispheres, and the methylenes, methines, and hydroxyls attached to carbon chain are closer to the equatorial region o f the host (Figure 2.1). ' H N M R guest signal assignments were made for each o f these complexes with the help o f 2D C O S Y experiments. ' H N M R spectra for the larger, bulkier alcohols, 3-hexanol and 2,4-pentanediol also showed doubling of the host resonances, which suggests that these guests do not enjoy any rotational freedom about the equator o f 25. In 25«l,3-dimethyl-2-imidazolidinone, the guest must be orientated with its methyl groups in the polar regions and the methylenes of the ring near the equatorial region. The acetyl methyl group o f 25«isopropyl acetate prefers to sit in the polar regions o f 25, while the isopropyl hydrogens are closer to the equator. In 25»NFP, the long axis of the guest is aligned with that of the host placing the aldehyde and C4 methylene protons deeper within the polar regions, while the C2 , and C3 methylene protons of the ring are in the equatorial region. The preferred orientations of N M P , D M S O , and D M A are described elsewhere.4 2.2.5 Determination of Template Ratios 2.2.5.1 Competition experiments The template ratios in Table 2.1 were determined from the product ratios measured from each o f a series of head-to-head competition reactions between different pairs o f guests. In each experiment, guests possessing similar templating ability (i.e., those directly above and below 75 each guest in Table 2.1) were competed to obtain the most precise ratios possible. Product ratios were determined by integration o f each set o f guest signals in the ' H N M R spectra. Guests (templates) were added at concentrations that yield a relative integration o f approximately 1:1, and the template ratios were adjusted accordingly. Cross-check competition experiments between the non-adjacent guests in Table 2.1 confirmed the accuracy o f the method. For example, competition o f the best guest para-xylene directly against N F P gave a template ratio o f 3600, which is identical to the value listed in Table 2.1. 76 Table 2.1 Template ratios in the formation of 25»guest. Template Ratio G u e s t 25«guest 10a«guesta para-xylene 3600 -4-bromotoluene 2800 -/>ara-dibromobenzene 2100 -4-chlorotoluene 2000 -anisole 1900 -1 -bromo-4-chlorobenzene 1700 -/?ara-dichlorobenzene 840 -1 -bromo-4-iodobenzene 730 -1 -chloro-4-iodobenzene 620 -4-methylanisole 580 -3-pentanol 490 -iodobenzene 460 -4-chloroanisole 450 -thioanisole 440 -bromobenzene 240 -2-butanol b 200 2800 toluene 140 -2-pentanol b 140 -benzene 110 2400 chlorobenzene 110 -2,4-pentanediolb 100 -3-hexanol b 99 -cyclohexane 58 -fluorobenzene 39 -N M P 28 1 D M A 26 20 D M S O 17 180000 1,3-dimethyl-2-imidazolidinone 16 -isopropyl acetate 10 -N F P 1 -'From reference 1. b Racemic mixtures were used. 2.2.5.2 Control Experiments Hemicarcerand 25 posseses four large portals through which smaller guest molecules can pass. Although both complexation and decomplexation rates are quite slow at high temperatures,4 3 we wanted to confirm that the template ratios in Table 2.1 are due to a kinetic template effect; i.e., that the reaction (guest encapsulation) was irreversible in all cases. Guest exchange subsequent to host formation was checked under the set of conditions used for the competition reactions. Thus, control experiments were performed for each hemicarceplex 25»guest, which involved separately subjecting each to the standard reaction conditions (Scheme 2.2) in the presence of two or more potentially competitive guests at concentrations o f 1 mole % of the solvent. Competing guests were chosen such that one guest was a slightly better template and the other a slightly worse template. This was because these guests, or guests with similar templating abilities were used in the determination of the template ratios. Guest exchange was investigated by ' H N M R spectroscopy. In the thirty control experiments performed, guest exchange was only observed for hemicarceplex 25«NFP in the presence o f para-xylene and 2-butanol, where <14 % of 25»2-butanol formed. Thus, N F P could potentially be a better template than the template ratios suggest. However, the results of various cross-check experiments indicated that this was not the case. 2.2.5.3 General Trends in Templating Abilities From the results shown in Table 2.1, various trends are apparent that correlate to the structures of each guest. Para-disubstituted benzene derivatives are the best guests/templates and there is a trend in their selectivity. In general, the ability of^ara-disubstituted benzene 78 derivatives to facilitate the formation of 25«guest is enhanced with increasing substituent size, until a limit is approached, past which selectivity drops. Derivatives with C H 3 substituents are better templates, which suggests that interactions between CH3S and the forming host are important in the formation o f 25»guest. A similar trend to /?ara-disubstituted benzenes is evident for monosubstituted benzenes. Templating abilities increase with substituent size (as F < CI < B r < I and CH3 < O C H 3 and SCH3), however, there are some exceptions. The templating abilities o f mono substituted benzene derivatives increase in the order of fluorobenzene < benzene = chlorobenzene < toluene < bromobenzene < thioanisole < iodobenzene < anisole. Templating ability drops five-fold upon exchange o f the O atom in anisole with the larger S atom. Dipolar effects may be at play for these templates as possible C H - p i interactions between the CH3 hydrogens o f thioanisole and the pi electron clouds of the interior of the arene-rich host are expected to be less favorable than for anisole. A greater negative charge density (repulsive) may reside on the CH3 protons o f thioanisole caused by the less electron withdrawing S atom (compared to O). Another exception is the three-fold drop in selectivity from benzene to fluorobenzene as a H atom is replaced with an F atom. Size is not likely to be a major contributing factor for these o 1 c two guests since hydrogen and fluorine have similar van der Waals radii (1.20 and 1.35 A , respectively). Unfavorable interactions may exist between the highly electronegative fluorine substituent and pi-electron rich interior o f the forming host. If this were the case, then further substitution with other F atoms would lead to a greater reduction in templating ability. Hexafluorobenzene failing as a suitable template under the set of conditions 1 3 is consistent with this. It is also noteworthy that preliminary template ratios (in parentheses) were determined for /rara-difluorobenzene (5), l-chloro-4-fluorobenzene (42), 1 -bromo-4-fluorobenzene (77), and 4-fluorotoluene (97). These template ratios are smaller than those for their respective protio 79 analogues (flourobenzene, chlorobenzene, bromobenzene, and toluene, respectively). These guests were omitted from this study because pure samples could not be obtained. Selectivity for aliphatic alcohols increases in the order o f 3-hexanol < 2,4-pentanediol < 2-pentanol < 2-butanol < 3-pentanol. Size and shape appear to be the most important factors governing their templating abilities. For example, introduction o f an extra methylene unit between C1 and C2 , and C2 and C3 in 2-butanol results in increases and decreases in selectivity, respectively. Insertion between C I and C2 improves templating ability by a factor of 2.5 for 3-pentanol, while addition between C2 and C3 results in a slight decrease in selectivity as for 2-pentanol. The more symmetric 3-pentanol is able to form more optimal van der Waals contacts with the host where the two terminal methyls are placed deeply within the polar hemispheres o f the host. Stabilization may be possible through CH-7C interactions between the methyl protons and the arenes pi-faces inside the host. Inserting an oxygen atom on the C4 o f 2-pentanol, as in 2,4-pentandiol, results in only a slight decrease in templating ability. Extending the carbon chain length from five to six carbons reduces templating power only slightly as well . Additional methyl groups introduced at C2 and C4 o f 3-pentanol, as in 2,4-dimethyl-3-pentanol (Chart 2.1), result in no observable templating ability. In summary, larger and more highly branched aliphatic alcohols form less favorable stabilizing interactions with the forming host shell. The low selectivities observed for guests near the bottom of Table 2.1 is attributed to the irregular shapes, small sizes, or higher polarities of these templates resulting in a less complementary fit to the forming host. Overall, size and shape o f the guests appear to be most important at governing the relative abilities of each guest at templating the formation of 25«guest. Stabilization of the forming host by various guests is l ikely to mainly arise from the maximization of van der Waals contacts, CH -7t interactions, and hydrogen-bonding. 80 2.3 Conclusions The apparent lack o f correlation between template ratios for hemicarceplex 25»guest and carceplex 10a«guest demonstrates that capsule 12a«guest is not a good transition state model for the GDS in the formation of hemicarceplex 25«guest. Therefore, although capsule 12a«guest can form in the presence o f suitable guests during the reaction to produce carceplex 10a«guest, hemicarceplex l la»guest, and hemicarceplex 25«guest, it is only relevant in the formation o f the former two (see Scheme 2.3). From the results presented in this chapter and from the examination of molecular ( C P K ) models, it appears that the cavity size in the GDS is much larger than that of capsule 12a»guest. It is speculated that the GDS could involve the formation o f the fourth bridge, in a tris(tetramethylene)-bridged intermediate (Scheme 2.3a), similar in structure to that previously isolated by Cram (i.e., triol 48, Scheme 2.3).1 7 Prior to the GDS, guest exchange is l ikely to be fast (i.e., during formation of the first and second bridges). Although no attempts were made to elucidate the mechanism o f formation of 25»guest for this thesis, preliminary evidence supporting this theory has been recently obtained by another member in the Sherman group. 81 Byproducts (decomposition, poly/oligomers or overalkylated intermediates) (a) 25*guest GDS 48«guest 12a«guest 10a«guest Scheme 2.3 Proposed mechanisms in the formation o f 25»guest and 10a«guest.' (a) Template effect in the formation o f hemicarceplex 25»guest only occurs during the formation o f either the third or fourth bridge. Precursors (e.g., mono-bridged intermediate) could readily polymerize since no preorganization is imparted by the template/guest, (b) Template effect is at play throughout the reaction in the formation of carceplex 10a»guest. Polymerization is minimized due to preorganization imparted by the template/guest in all stages. The observed template effect in the formation hemicarceplex 25»guest is vastly different from the template effect previously observed in the formation of carceplex 10a»guest. Replacement o f the methylene with tetramethylene bridges results in a severe drop in the range o f selectivity, decreasing from a million-fold to 3600-fold. However, the drop in the range o f selectivity could be attributed to the N F P solvent used in the reaction, which is always a competitor in the formation of 25«guest. The low selectivity observed for the guests in Table 2.1 8 2 in the formation of 25«guest compared to that found for 10a«guest prompted us to investigate the effect o f temperature on selectivity. Reported competition experiments in the formation o f 10a«guest that gave a one million-fold range in selectivity were conducted at 60 °C . 1 Para-xylene was competed against N F P in separate reactions at various temperatures ranging from ambient to 80 °C. N o significant changes in selectivity were observed. The low yields obtained for each o f the 30 hemicarceplexes are attributed to the lack o f preorganization between the reactive functionalities (phenoxides) o f the reaction intermediates formed leading up to the GDS (Scheme 2.3a). Installation o f a bulky tetramethylene linkage would most l ikely disrupt the hydrogen-bonds in a capsule such as 12a«guest. Guests would not be expected to bind strongly at this point. 3 c Subsequent intermediate species formed would be poorly preorganized for hemicarceplex formation (Scheme 2.3a). For example, in a mono-bridged intermediate, the bowls can rotate freely and thus would be poised for poly/oligomerization upon further reaction. For the larger guests (than D M S O , D M A , N M P , benzene, 2-butanol) in Table 2.1, encapsulation to form a preorganized (hydrogen-bonded) intermediate complex is impossible, as such a species would be highly strained since maximum intermolecular distance between hydrogen-bonding groups would be exceeded. It has been demonstrated here that larger guest molecules can template the formation o f 25»guest, than those observed in the formation o f carceplex 10a»guest. In general, para-disubstituted benzene derivatives are preferred, followed by mono-substituted benzenes, and aliphatic alcohols. The shapes and sizes o f these guests are complimentary to the interior o f the forming host, maximizing non-covalent host-guest interactions such as van der Waals contacts, hydrogen-bonding, C H - p i , and possibly pi-stacking. Note also that several guests that bind strongly to 254 did not show any observable template effect. Although these guests are 83 complementary to the interior o f the formed host 25, they evidently are non-complementary to the transitions state occurring during the GDS. This study has provided further insight into the nature o f the noncovalent interaction involved in the templated formation o f host-guest systems, carceplexes and hemicarceplexes in particular. This information has given us further knowledge leading to the design and creation o f much larger assemblies, which may eventually reach the complexity o f those found in nature. Examples o f larger and more sophisticated assemblies w i l l be presented in chapters 3 to 5. 84 2.4 Experimental Section 2.4.1 General Experimental A l l reagents were purchased from the Aldr ich Chemical Co. , Inc., and were used without further purification unless stated otherwise. N F P , D M A , and D M S O were distilled and stored over 4 A molecular sieves under an N2 atmosphere. A l l other reagents were commercially available at >98 % purity and were used without further purification. A l l reactions were carried out under a positive pressure o f N 2 , unless stated otherwise. Sil ica gel ( B D H , 230-400 mesh) was used for column chromatography. Silica gel thin layer chromatography was performed on glass-backed plates (Aldrich, silica gel 60, F254, 0.25 mm). ! H N M R spectra were acquired using a Bruker WH-400 spectrometer in CDCI3 at ambient temperature using the residual ' H signal as the reference. Mass spectra were recorded on a Kratos Concept II H Q (DCI) and a V G Tofspec in reflectron mode ( M A L D I ) . M A L D I samples were prepared using 2,5-dihydroxybenzoic acid ( D H B ) as the matrix. Bovine insulin protein standard (molecular weight = 5733.55 amu.) was used with the matrix, sinapinic acid (3,5-dimethoxy-4-hydroxy cinnamic acid, molecular weight = 224.21 amu.), to calibrate the spectrometer prior to use. The molecular weights reported (ranging 2250-2533 amu.) are averaged and are not exact due to the limited resolution o f the M A L D I M S instrument used. Refer to structure 25»guest (Scheme 2.1) for host proton labels, and Table 2.3 for guest proton labels. 85 2.4.2 Synthesis of Hemicarceplexes 25«guest 2.4.2.1 General Templating Procedure for the Synthesis of 25«guest The synthesis and characterization of the hemicarceplexes 25»NFP, 25«NMP, 25»DMA, 25»DMSO, using 1,4-dibromobutane in N F P is described elsewhere. 1 2 ' 1 3 The synthesis o f the rest were done using a similar procedure and their characterization by *H N M R spectroscopy and M A L D I mass spectrometry are described below. Refer to Scheme 2.2 and Table 2.3 (vide supra) for host and guest proton labels, respectively. 25»l,3-dimethyl-2-imidazolidinone. Procedure A . Tetrol 8a (47.7 mg, 0.047 mmol), CS2CO3 (0.536 g, 1.65 mmol, 35 equiv.) and K I (67.5 mg, 0.407 mmol, 0.8 equiv. per molecule o f linker) were mixed with N F P (10 mL) and l,3-dimethyl-2-imidazolidinone (10 mL) . The reaction was stirred at 80 °C for 10 min, and then 1,4-dibromobutane (60 \iL, 0.50 mmol, 11 equiv.) was added. The reaction was stirred further at 80 °C for 48 h, the solvent was removed in vacuo, and the yellow-brown residue was resuspended in 2 M HC1 (20 m L ) and extracted with CHCI3 (3 x 10 mL) . The C H C 1 3 extracts were combined and dried over MgSC>4, filtered, and the solvent was removed in vacuo. The resulting crude yellow-brown oil was then passed through silica gel (230-400 mesh) eluting with either C H C 1 3 , C H 2 C 1 2 , or CH 2 Cl 2 :hexanes (6:1). Precipitation o f the product from CHCb/hexanes gave a white solid (1.8 mg, 3.2 %). ' H N M R (400 M H z , C D C I 3 ) 5 7.23-7.18 (m, 24H, Qj/fc), 7.18-7.09 (br m, 16H, C ^ ) , 6.81 (s, 8H, Hp), 5.73 (d, 8H, J = 7.3 Hz , H 0 ) , 4.84 (t, 8H, J = 8.2 Hz , H m ) , 4.44 (d, 8H , J = 7.3 H z , Hi), 86 3.94 (br s, 16H, O G t f 2 C H 2 ) , 2.67 (m, 16H, C H 2 C 7 f 2 C 6 H 5 ) , 2.47 (m, 16H, C / / 2 C H 2 C 6 H 5 ) , 1.95 (br s, 16H, OCH 2 C77 2 ) , 1.77 (s, 4 H , H b ) , -1.06 (s, 6H , H a ) . M S ( M A L D I ) m/z (rel intensity) 2387 ( ( M « C 5 H 1 0 N 2 O + N a + ) + ; 100), calcd for C i 4 9 H i 4 6 N 2 0 2 4 » N a + = 2388. 25«cyclohexane. Similar to procedure A , except that a solvent mixture o f N F P (20 m L ) and cyclohexane (20 mL) was used instead of N F P (10 mL) and l,3-dimethyl-2-imidazolidinone (10 mL) . Tetrol 8a (107.2 mg, 0.105 mmol), C s 2 C 0 3 (0.628 g, 1.93 mmol, 18 equiv.) and K I (127.1 mg, 0.766 mmol, 0.8 equiv. per molecule of linker), 1,4-dibromobutane (120 | i L , 1.00 mmol, 10 equiv.), cyclohexane (10 mL) , N F P (10 mL) . 25«cyclohexane (3.3 mg, 2.7 %) was obtained as a white solid. *H N M R (400 M H z , CDC1 3 ) 5 7.20 (m, 24H, C ^ s ) , 7.15 (m, 16H, Q Z / s ) , 6.84 (s, 8H , H p ) , 5.79 (d, 8H , J = 6.9 H z , H 0 ) , 4.81 (t, 8H, J = 7.9 Hz , H m ) , 4.17 (d, 8H , J = 6.9 H z , HO, 3.90 (br s, 16H, O C 7 / 2 C H 2 ) , 2.67 (m, 16H, C H 2 C / f 2 C 6 H 5 ) , 2.48 (m, 16H, C / / 2 C H 2 C 6 H 5 ) , 1.92 (br s, 16H, O C H 2 C 7 / 2 ) , -0.57 (s, 12H, H a ) . M S ( M A L D I ) m/z (rel intensity) 2358 ( ( M « C 6 H i 2 + N a + ) + ; 100), calcd for Ci5oH , 4 8 0 2 4»Na + = 2358. 25«fluorobenzene. Similar to procedure A , except that fluorobenzene was used instead of 1,3-dimethyl-2-imidazolidinone. Tetrol 8a (54.8 mg, 0.054 mmol), C s 2 C 0 3 (0.303 g, 0.930 mmol, 17 equiv.) and K I (61.0 mg, 0.367 mmol, 0.7 equiv. per molecule o f linker), 1,4-dibromobutane 87 (60 U.L, 0.50 mmol, 10 equiv.), fluorobenzene (10 mL) , N F P (10 mL) . 25«fluorobenzene (5.7 mg, 9.0 %) was obtained as a white solid. ! H N M R (400 M H z , CDC1 3 ) 5 7.23-7.18 (m, 24H, CeH5), 7.18-7.13 (br m, 16H, C ^ s ) , 6.93 (s, 8H, Hp), 5.68 (d, 8H , J = 7.1 Hz , H 0 ) , 5.24 (m, 2 H , H c ) , 5.18 (dd, 2 H , J - 7.6 Hz , H b ) , 4.83 (t, 8H, J = 7.6 H z , H m ) , 4.05 (d, 8H , J = 7.1 Hz , H ; ) , 3.84 (br s, 16H, O G t f 2 C H 2 ) , 3.45 (t, 1H, J = 7.2 Hz , H a ) , 2.67 (m, 16H, C H 2 C / / 2 C 6 H 5 ) , 2.50 (m, 16H, C / / 2 C H 2 C 6 H S ) , 1.93 (br s, 16H, O C H 2 G t f 2 ) . M S ( M A L D I ) m/z (rel intensity) 2370 ( ( M » C 6 H 5 F + N a + ) + ; 100), calcd for Ci 5oHi4iF0 24»Na + = 2370. 2.4.2.2 General Guest Exchange Procedure for the Synthesis of 25«guest A l l other 25»guest were prepare using Procedure B (Similar to Cram's procedure). 4 3 Hemicarceplex 25«DMA (10-20 mg) was dissolved in either neat guest, or a mixture o f P h 2 0 and guest (see Table 2.2) in a 10 m L round bottom flask equipped with a condenser. The mixture was heated in a sand bath to temperatures ranging between 140-200 °C for 2 to 5 d. (Table 2.2), after which the solvent was removed in vacuo. The crude residues were purified by precipitation from CHCI3 solutions with hexanes or methanol. If the crude mixture appeared dark in color, the sample was eluted through a pad o f silica gel with C H 2 C 1 2 prior to precipitation. Spectroscopic characterization of 25*benzene, 25«para-xylene, 25»para-dibromobenzene, 25«iodobenzene were all in accord to that reported elsewhere. 4 3 *H N M R and 88 M A L D I mass spectroscopic data for all other hemicarceplexes are reported below. A8 values for each guest are also summarized below in Table 2.3. Table 2.2 Conditions for the synthesis of hemicarceplexes 25»guest using procedure B . Guest Solvent Temperature C Q Time (days) % Yield para-xylene Guest/Ph 2 0 150 2 86 4-bromotoluene Neat guest 180 4 76 /?ara-dibromobenzene Guest/Ph 2 0 146 3 67 4-chlorotoluene Neat guest 160 4 60 anisole Neat guest 160 4 83 1 -bromo-4-chlorobenzene Neat guest 200 5 60 />ara-dichlorobenzene Guest/Ph 2 0 155 3 59 4-bromo-1 -iodobenzene Guest/Ph 2 0 145 3 70 /xara-methylanisole Guest/Ph 2 0 155 3 43 4-chloro-1 -iodobenzene Guest/Ph 2 0 145 3 68 iodobenzene Guest/Ph 2 0 150 2 87 3-pentanol Guest/Ph 2 0 140 4 67 4-chloroanisole Neat guest 160 5 59 thioanisole Neat guest 200 5 57 bromobenzene Neat guest 155 3 62 2-butanol Guest/Ph 2 0 160 6 49 toluene Guest/Ph 2 0 170 5 57 chlorobenzene Guest/Ph 2 0 150 5 91 2-pentanol Guest/Ph 2 0 155 3 77 benzene Guest/Ph 2 0 150 5 94 2,4-pentanediol Guest/Ph 2 0 200 5 81 3-hexanol Guest/Ph 2 0 140 4 62 isopropyl acetate Guest/Ph 2 0 160 3 4 25«4-bromotoluene ' H N M R (400 M H z , CDC1 3 ) 8 7.32-7.18 (m, 24H, C ^ s ) , 7.18-7.10 (m, 16H, C ^ s ) , 6.87 (s, 4 H , Hp), 6.81 (s, 4 H , Hp), 6.37 (d, 2H , J = 8.1 Hz , H a ) , 5.93 (d, 2 H , J = 8.1 Hz , H b ) , 5.67 (d, 4 H , J = 5.4 H z , H o ) , 5.66 (d, 4 H , J = 6.7 Hz , H 0 ) , 4.86 (t, 4 H , J = 7.8 Hz , H m ) , 4.84 (t, 4 H , J = 7.8 H z , 89 H m ) , 4.16 (d, 4 H , J = 7.1 Hz , Hi), 4.08 (d, 4 H , J = 7.1 Hz , HO, 3.89 (br s, 16H, O G r Y 2 C H 2 ) , 2.70 (m, 16H, C H 2 C / f 2 C 6 H 5 ) , 2.50 (m, 16H, C / / 2 C H 2 C 6 H 5 ) , 1.86 (br s, 16H, O C H 2 C / / 2 ) , -2.15 (s, 3H , H c ) . M S (LSIMS) 2418.91487 (2418.91527). M S ( M A L D I ) m/z (rel intensity) 2445 ((M»C 7 H 7 Br + N a + ) + ; 100), calcd for Ci5iHi430 2 4Br«Na + = 2445. 25«4-chlorotoluene ' H N M R (400 M H z , CDC1 3 ) 5 7.22 (m, 24H, Ctfs), 7.19 - 7.09 (m, 16H, Qtfs), 6.88 (s, 4 H , Hp), 6.85 (s, 4 H , Hp), 6.13 (d, 2 H , J = 8.1 Hz , H a ) , 5.98 (d, 2 H , J = 8.0 H z , H b ) , 5.68 (d, 4 H , J = 7.1 H z , H 0 ) , 5.66 (d, 4 H , J = 7.1 Hz , H 0 ) , 4.87 (t, 4 H , J = 8.1 Hz , H m ) , 4.84 (t, 4 H , J = 8.1 Hz , H m ) , 4.12 (d, 4 H , J = 7.1 Hz , HO, 4.08 (d, 4 H , J = 7.1 Hz , HO, 3.89 (br s, 16H, O C 7 / 2 C H 2 ) , 2.70 (m, 16H, C H 2 C / / 2 C 6 H 5 ) , 2.51 (m, 16H, C / / 2 C H 2 C 6 H 5 ) , 1.88 (br s, 16H, OCH 2 C77 2 ) , -2.08 (s, 3H , H c ) . M S ( M A L D I ) m/z (rel intensity) 2400 ( (M»C 7 H 7 C1 + N a + ) + ; 100), calcd for C i 5 i H i 4 3 C 1 0 2 4 » N a + -2400. 25*anisole *H N M R (400 M H z , CDC1 3 ) 8 7.23-7.19 (m, 24H, Cffls), 7.19-7.13 (m, 16H, CaHs), 6.95 (s, 4 H , Hp), 6.83 (s, 4 H , Hp), 5.70 (br s, 4 H , H 0 ) , 5.61 (br s, 4 H , H 0 ) , 5.54 (d, 2 H , J = 8.2 H z , H c ) , 5.12 (dd, 2 H , J = 8.0 H z , H b ) , 4.85 (br s, 8H, H m ) , 4.23 (br s, 4 H , HO, 4.19 (br s, 4 H , H f ) , 3.85 (br s, 90 16H, OCH2CH2), 2.84 (t, 1H, J = 8.0 Hz , H a ) , 2.69 (m, 16H, CH 2Gr7 2C 6H5), 2.51 (br s, 16H, O C H 2 C / / 2 ) , 1.90 (br s, 16H, OCH 2 C77 2 ) , -0.54 (s, 3H, H d ) . ' H N M R (400 M H z , C 6 D 6 , 348 K ) 5 7.15 (br s, 24H, CeH5), 7.11 (m, 16H, CeH5 and H p ) , 5.80 (d, 2 H , J = 8.1 H z , H c ) , 5.74 (d, 8H , J = 6.7 H z , H 0 ) , 5.39 (dd, 2 H , J = 7.9 H z , H b ) , 5.22 (t, 4 H , J= 7.8 H z , H m ) , 4.46 (d, 4 H , J = 6.7 Hz , HO, 3.93 (br s, 16H, O C / / 2 C H 2 ) , 3.19 (t, 1H, J = 6.8 H z , H a ) , 2.72 (m, 16H, C H 2 C / f 2 C 6 H 5 ) , 2.57 (m, 16H, C / / 2 C H 2 C 6 H 5 ) , 1.83 (br s, 16H, O C H 2 C # 2 ) , -0.19 (s, 3 H , H d ) . M S (LSIMS) 2357.00457 (2356.99967). M S ( M A L D I ) m/z (rel intensity) 2382 ( ( M « C 7 H 8 0 + N a + ) + ; 100), calcd for C ,5 iH 1 440 2 5 «Na + = 2382. 25* l-bromo-4-chlorobenzene ' H N M R (400 M H z , CDC1 3 ) 8 7.21 (m, 24H, Q 5 / / 5 ) , 7.18-7.10 (m, 16H, Q / / 5 ) , 6.84 (s, 4 H , H p ) , 6.81 (s, 4 H , Hp), 6.45 (d, 2 H , J = 8.5 Hz , H a ) , 6.15 (d, 2 H , J = 8.5 Hz , H b ) , 5.67 (d, 4 H , J - 6.8 H z , H 0 ) , 5.66 (d, 4 H , J = 6.8 H z , H 0 ) , 4.84 (t, 4 H , J = 7.9 H z , H m ) , 4.83 (t, 4 H , J = 7.9 H z , H m ) , 4.11 (d, 4 H , J - 6.8 Hz , Hi), 4.06 (d, 4 H , J = 6.8 Hz , H O , 3.91 (br s, 16H, O C 7 / 2 C H 2 ) , 2.69 (m, 16H, CH 2 Cr7 2 C 6 H 5 ) , 2.49 (m, 16H, C 7 7 2 C H 2 C 6 H 5 ) , 1.86 (br s, 16H, OCH2Cr72). M S ( M A L D I ) m/z (rel intensity) 2465 ( (M»C 6 H 4 BrCl + N a + ) + ; 100), calcd for Ci5oHi4o0 2 4 BrCl«Na + = 2465. 91 25»/;ai*a-dichlorobenzene *H N M R (400 M H z , CDC1 3 ) 8 7.20 (m, 24H, C ^ s ) , 7.16 (s, 16H, CeHs), 6.84 (s, 8H , H p ) , 6.20 (s, 4 H , H a ) , 5.67 (d, 8H , J = 6.8 H z , H 0 ) , 4.84 (t, 8 H , J = 7.8 H z , H m ) , 4.06 (d, 8 H , J = 6.8 H z , HO, 3.90 (br s, 16H, O C # 2 C H 2 ) , 2.70 (m, 16H, C H 2 C / 6 G 6 H 5 ) , 2.50 (m, 16H, C / / 2 C H 2 C 6 H 5 ) , 1.87 (brs, 16H, O C H 2 C 7 / 2 ) . M S ( M A L D I ) m/z (relative intensity) 2419 ( ( M » C 6 H 4 C 1 2 + N a + ) + ; 100), calcd for C 1 5 oH,4o0 2 4Cl 2 »Na + = 2421. 25»l -bromo-4- iodobenzene ' H N M R (400 M H z , CDC1 3 ) 8 7.32-7.18 (m, 24H, Csfls), 7.18-7.11 (m, 16H, C ^ s ) , 6.80 (s, 4 H , H p ) , 6.76 (d, 2 H , J = 8.4 Hz , H a ) , 6.74 (s, 4 H , H p ) , 6.31 (d, 2 H , J - 8.4 H z , H b ) , 5.66 (d, 4 H , J = 6.7 Hz , H 0 ) , 5.65 (d, 4 H , J = 6.7 Hz , H 0 ) , 4.86 (t, 4 H , J = 7.8 Hz , H m ) , 4.85 (t, 4 H , J = 7.8 H z , H m ) , 4.26 (d, 4 H , J = 6.7 Hz , Hi), 4.10 (d, 4 H , J= 6.7 H z , HO, 3.94 (br s, 16H, O C # 2 C H 2 ) , 2.70 (m, 16H, C H 2 C 7 7 2 C 6 H 5 ) , 2.48 (m, 16H, C 7 / 2 C H 2 C 6 H 5 ) , 1.82 (br s, 16H, O C H 2 C / / 2 ) . M S ( M A L D I ) m/z (rel intensity) 2556 ( (M«C 6 H 4 BrI + N a + ) + ; 100), calcd for C I 5 oH 1 4 o0 2 4BrI«Na + = 2556. 92 25» 1 -chloro-4-iodobenzene 'H N M R (400 M H z , CDC1 3 ) 8 7.20 (m, 24H, Crfs), 7.16 (m, 16H, CM), 6.84 (s, 4 H , Hp), 6.78 (d, 2 H , J = 8.5 Hz, H a ) , 6.75 (s, 4 H , Hp), 6.02 (d, 2 H , J = 8.5 Hz, Hb), 5.66 (d, 4 H , J = 6.7 Hz, H0), 5.65 (d, 4 H , J - 6.7 Hz, H0), 4.86 (t, 4 H , J = 8.1 Hz, Hm), 4.84 (t, 4 H , J = 8.2 Hz, Hra), 4.28 (d, 4 H , J = 6.7 Hz, H;), 4.06 (d, 4 H , J = 6.7 Hz, HO, 3.94 (br s, 8H, O C / / 2 C H 2 ) , 3.91 (br s, 8H , O C 7 / 2 C H 2 ) , 2.70 (m, 16H, CH 2 C7 / 2 C 6 H 5 ) , 2.48 (m, 16H, C / / 2 C H 2 C 6 H 5 ) , 1.84 (br s, 16H, O C H 2 C r Y 2 ) . M S ( M A L D I ) m/z (rel intensity) 2512 ((M«C6H4C1I + Na+)+; 100), calcd for Ci5oHi4o0 2 4 ClI»Na + = 2512. 25*3-pentanol ! H NMR (400 MHz, CDC1 3 ) 8 7.32-7.19 (m, 24H, C^s), 7.19-7.10 (m, 16H, br), 6.81 (s, 8H , Hp), 5.77 (d, 8H, J = 7.1 Hz, H0), 4.82 (t, 8H, J - 7.9 Hz, Hm), 4.25 (d, 8H , J = 7.1 Hz, H O , 3.93 (br s, 16H, O C # 2 C H 2 ) , 2.68 (m, 16H, CH 2Gtf 2C6H 5), 2.48 (m, 16H, C / / 2 C H 2 C 6 H 5 ) , 1.96 (br s, 16H, O C H 2 C # 2 ) , 1.56 (br m, 1H, Hc), 1.21 (d, 1H, J = 3.6 Hz, Hd), -0.56 (m, 2 H , H D ) , -0.70 (m, 2 H , Hb), -2.77 (t, 6 H , J = 7.4 Hz, Ha). MS ( M A L D I ) m/z (rel intensity) 2362 ( (M«C 5 Hi 2 0 + Na+)+; 100), calcd for Ci49Hi48025»Na+ = 2362. 93 25»4-chloroanisole ' H N M R (400 M H z , CDC1 3 ) 6 7.32-7.14 (m, 24H, Ctfs), 7.14-7.10 (m, 16H, Q ^ ) , 6.84 (s, 4 H , Hp), 6.82 (s, 4 H , Hp), 6.02 (d, 2 H , J = 8.7 Hz , H a ) , 5.87 (d, 2 H , J = 8.7 H z , H b ) , 5.69 (d, 4 H , J = 7.0 H z , H 0 ) , 5.64 (d, 4 H , J = 7.0 Hz , H 0 ) , 4.88 (t, 4 H , J = 7.9 Hz , H m ) , 4.83 (t, 4 H , J = 7.9 H z , H m ) , 4.19 (d, 4 H , J = 7.0 Hz , Hi), 4.14 (d, 4 H , J = 7.0 Hz , H O , 3.91 (br s, 8H, OCH2CH2), 3.89 (br s, 8H, OC//2CH2), 2.70 (m, 16H, Cn2CH2C6n5), 2.50 (m, 16H, CH2CH2C6H5), 1.85 (br s, 16H, OCH2C//2), -0.37 (s, 3H, H c ) . M S ( M A L D I ) m/z (rel intensity) 2415 ( ( M « C 7 H 7 C 1 0 + N a + ) + ; 100), calcd for C 1 5 1 H , 4 3 0 2 5 C l « N a + = 2416. 25*thioanisole ' H N M R (400 M H z , CDCI3): 7.32-7.19 (m, 24H, C(JHS), 7.19-7.08 (m, 16H, CeH5), 6.95 (s, 4 H , H p ) , 6.84 (s, 4 H , Hp), 5.89 (d, 2 H , J - 7.5 Hz , H c ) , 5.68 (d, 4 H , J = 7.0 H z , H 0 ) , 5.59 (d, 4 H , J = 7.0 H z , H o ) , 4.98 (dd, 2 H , J = 7.7 Hz , H b ) , 4.87 (t, 4 H , J - 7.7 Hz , H m ) , 4.83 (t, 4 H , J = 7.7 H z , H m ) , 4.60 (d, 4 H , J = 7.0 Hz , H i ) , 4.17 (d, 4 H , J = 7.0 Hz , H O , 3.96 (br s, 8H, OCH2CR2), 3.84 (br s, 8H , OC//2CH2). 2.99 (t, 1H, J = 7.4 Hz , H a ) , 2.67 (m, 16H, CH2C//2C6H5), 2.50 (m, 16H, C//2CH2C6H5), 1.88 (br s, 16H, O C H 2 C / f 2 ) , -1.65 (s, 3H , H d ) . M S ( M A L D I ) m/z (rel intensity) 2398 ( ( M » C 7 H 8 S + N a + ) + ; 100), calcd for Ci5 iHi440 2 4 S«Na + = 2398. 94 25»bromobenzene ' H N M R (400 M H z , CDC1 3 ) 5 7.23 (m, 24H, CM), 7.19-7.09 (br s, 16H, CeH5), 6.89 (s, 8H, Hp), 6.30 (d, 2 H , J = 9.7 Hz , H c ) , 5.65 (d, 8H , J = 7.0 H z , H 0 ) , 5.61 (dd, 2 H , J = 7.8 H z , H b ) , 4.84 (t, 8H , J = 8.1 Hz , H r a ) , 4.14 (d, 8H , J = 7.0 H z , HO, 3.88 (br s, 16H, O C 7 / 2 C H 2 ) , 3.37 (t, 1H, J = 7.4 Hz , Ha), 2.69 (m, 16H, C H 2 C i / 2 C 6 H 5 ) , 2.52 (m, 16H, C r 7 2 C H 2 C 6 H 5 ) , 1.89 (br s, 16H, OCH2CH2). M S ( M A L D I ) m/z (rel intensity) 2431 ( ( M « C 6 H 5 B r + N a + ) + ; 100), calcd for C i 5 0 H i 4 i O 2 4 B r » N a + = 2431. 25»2-butanol ' H N M R (400 M H z , CDC1 3 ) 8 7.20 (m, 24H, Q/Ys), 7.14 (m, 16H, C ^ s ) . 6.81 (s, 8H , H p ) , 5.81 (d, 8H , J = 7.1 H z , Ho), 4.81 (t, 8H, J = 7.4 Hz , H m ) , 4.19 (d, 8H , J = 7.1 Hz , HO, 3.91 (br s, 16H, O C / / 2 C H 2 ) , 2.67 (m, 16H, C H 2 C # 2 C 6 H 5 ) , 2.47 (m, 16H, C / ^ C H ^ e H s ) , 1.99 (br s, 16H, OCH 2 C77 2 ) , 1.71 (br m, 1H, H c ) , 1.00 (d, 1H, H d ) , -0.52 (br m, 1H, H b ) , -0.61 (brm, 1H, H b ) , -2.35 (d, 3 H , J = 5.8 H z , H e ) , -2.76 (t, 3H, J = 6.9 Hz , H a ) . M S ( M A L D I ) m/z (rel intensity) 2348 ( ( M » C 4 H i 0 O + N a + ) + ; 100), calcd for C i 4 8 H 1 4 6 0 2 5 « N a + = 2348. 95 25»toluene ' H N M R (400 M H z , CDC1 3 ) 8 7.21 (m, 24H, Ctfs), 7.16 (m, 16H, Crfs), 6.91 (s, 8H , H p ) , 5.83 (d, J = 7.2 Hz , H c ) , 5.65 (m, 10H, H 0 and H b ) , 4.84 (t, 8H , J - 7.4 Hz , H m ) , 4.12 (d, 8H , J = 7.2 H z , H O , 3.86 (br s, 16H, O C i / 2 C H 2 ) , 3.33 (t, 1H, J - 6.8 Hz , H a ) , 2.69 (m, 16H, C H 2 C # 2 C 6 H 5 ) , 2.51 (m, 16H, C/ /2CH 2 C 6 H 5 ) , 1.90 (br s, 16H, O C H 2 C t f 2 ) , -1.79 (s, 3H , H d ) . M S ( M A L D I ) m/z (rel intensity) 2365 ( ( M « C 7 H 8 + N a + ) + ; 100), calcd for Ci5iHi44024«Na+ = 2366. 25«chlorobenzene ' H N M R (400 M H z , CDCI3) 5 7.21 (m, 24H, Csfls), 7.16 (m, 16H, CeH5), 6.90 (s, 8H, H p ) , 6.01 (d, 2 H , J = 7.6 H c ) , 5.67-5.63 (m, 10H, H 0 and H b ) , 4.83 (t, 8H , J = 7.4 H z , H r a ) , 4.11 (d, 8H , J = 7.0 Hz, H O , 3.87 (br s, 16H, O C # 2 C H 2 ) , 3.36 (t, 1H, J - 7.4 Hz , H a ) , 2.68 (m, 16H, C ^ C / ^ Q H s ) , 2.51 (m, 16H, C i 7 2 C H 2 C 6 H 5 ) , 1.90 (br s, 16H, OCH 2 C77 2 ) . M S ( M A L D I ) m/z (rel intensity) 2386 ((M«C 6 H 5 C1 + N a + ) + ; 100), calcd for Ci 5 0 Hi4iClO 24»Na + = 2386. 25«2-pentanoI ' H N M R (400 M H z , CDCI3) 8 7.23-7.17 (br s, 24H, C^s), 7.16 - 7.11 (br s, 16H, C ^ s ) , 6.81 (s, 8H , H p ) 5.76 (d, 2 H , J - 7.1 H z , H 0 ) , 4.82 (t, 8H, J = 7.9 Hz , H m ) , 4.26 (d, 8H , J = 7.1 H z , Hi), 96 3.96 (br s, 16H, O C 7 / 2 C H 2 ) , 2.67 (m, 16H, C H 2 C 7 / 2 C 6 H 5 ) , 2.47 (m, 16H, C / / 2 C H 2 C 6 H 5 ) , 1.95 (br s, 16H, O C H 2 C # 2 ) , 1.85 (m, 1H, H d ) , 0.76 (d, 1H, J = 4.2 H z , H e ) , -0.21 (m, 1H, H c ) , -0.36 (m, 1H, H c ) , -0.56 (m, 1H, H b ) , -0.68 (m, 1H, H b ) , -2.62 (d, 3H , J = 6.1, H f ) , -2.68 (t, 3H , J = 7.5 Hz , H a ) . M S ( M A L D I ) m/z (rel intensity) 2363 ( ( M » C 5 H , 2 0 + N a + ) + ; 100), calcd for Ci49Hi480 25«Na + = 2362. 25«2,4-pentanediol ! H N M R (400 M H z , CDC1 3 ) 8 7.23-7.19 (br m, 24H, C ^ s ) . 7.19-7.08 (m, 16H, CeHs), 6.82 (s, 8H , Hp), 5.74 (d, 8H, J = 7.1 Hz , H 0 ) , 4.82 (t, 4 H , J = 7.7 Hz , H m ) , 4.81 (t, 4 H , J = 7.7 H z , H m ) , 4.44 (d, 4 H , J = 7.1 Hz , Hj), 4.40 (d, 4 H , J = 7.1 Hz , H;), 4.20 - 3.78 (m, 16H, O G r 7 2 C H 2 ) , 2.67 (m, 16H, C H 2 C / / 2 C 6 H 5 ) , 2.48 (m, 16H, C 7 7 2 C H 2 C 6 H 5 ) , 2.22 (m, 1H, H e ) , 1.94 (br s, 17H, O C H 2 G r 7 2 and H c ) , 1.15 (d, 1H, J = 4.0 Hz , H b ) , 0.85 (d, 1H, J = 3.6 Hz , H f ) , 0.07 - 0.22 (br m, 2 H , H d ) , -2.46 (d, 3 H , J = 6.3 H z , H g or H a ) , -2.51 (d, 3 H , J = 6.3 H z , H a or Hg). M S ( M A L D I ) m/z (rel intensity) 2379 ( ( M « C 5 H 1 2 0 2 + N a + ) + ; 100), calcd for C i4 9 Hi480 2 6 »Na + = 2378. 25«3-hexanol ' H N M R (400 M H z , CDC1 3 ) 8 7.32-7.18 (m, 24H, CeHs), 7.18-7.09 (br s, 16H, Crfs), 6.79 (s, 8H , H p ) , 5.74 (d, 2 H , J = 7.0 Hz , H 0 ) , 4.83 (t, 8H , J = 7.9 Hz , H m ) , 4.30 (d, 8H , J = 7.0 H z , HO, 3.96 (br s, 16H, O C # 2 C H 2 ) , 2.67 (m, 16H, C H 2 C # 2 C 6 H 5 ) , 2.47 (m, 16H, C # 2 C H 2 C 6 H 5 ) , 2.05 (m, 1H, Hd) 1.92 (br s, 16H, O C H 2 C / / 2 ) , 0.92 (d, 1H, J = 4.5 H z , H e ) , 0.00 (m, 1H, H f ) , -0.13 (m, 1H, H f ) , -0.34 (m, 1H, H c ) , -0.50 (m, 3H, H b and H c ) , -3.17 (t, 3H , J = 7.4 Hz , H a ) , -3.31 (t, 3H , J - 7.5 Hz , Hg). M S ( M A L D I ) m/z (rel intensity): 2375 ( ( M « C 6 H , 4 0 + N a + ) + ; 100), calcd for Ci5oH 1 5 o0 2 5»Na + - 2376. 25»isopropyl acetate ] H N M R (400 M H z , CDC1 3 ) 6 7.20 (m, 24H, C ^ s ) , 7.15 (m, 16H, C ^ s ) , 6.81 (s, 8H , H p ) , 5.68 (d, 8 H , J = 6.5 H z , H 0 ) , 4.89 (t, 8H , J = 8.1 H z , H m ) , 4.17 (d, 8H , J = 6.5 H z , HO, 3.99 (br s, 16H, O C / / 2 C H 2 ) , 3.31 (m, 1H, H c ) , 2.68 (m, 16H, C H 2 C / / 2 C 6 H 5 ) , 2.48 (m, 16H, C f f 2 C H 2 C 6 H 5 ) , 1.92 (br s, 16H, O C H 2 C / / 2 ) , -0.42 (d, 6H, J = 6.0 Hz , H b ) , -2.38 (s, 3H, H a ) . M S ( M A L D I ) m/z (rel intensity) 2374 ( ( M » C 5 H 1 0 O 2 + N a + ) + ; 100). Calcd for C 1 4 9 H 1 4 6 0 2 6 » N a + = 2376. 98 Table 2.3 ' H N M R chemical shifts for free and bound guests o f hemicarceplex 25»guest in CDCI3 at ambient temperature. Guest structure Proton 8free (multiplicity) 8 b 0 u n d (multiplicity) A S /mra-xylene H h H a 2.32 (s)1 -2.08 (s)1 4.40 1 W a ' .  H 3 C - f V c H 3 H b 7.07 (s)1 5.90 (s)1 Br—(^3~CH3 M a H b 1 -bromo-4-iodobenzene H a H para-dibromobenzene a. 1.171 4-bromotoluene H a v H a 7.34(d) 6.37(d) 0.97 H b 7.02(d) 5.93(d) 1.09 H c 2.28 (s) -2.15 (s) 4.43 4-chlorotoluene H a v , H  H a 7.20(d) 6.13(d) 1.07 H b 7.08(d) 5.98(d) 1.10 H c 2.30 (s) -2.08 (s) 4.38 H a 7.21(d) 6.31(d) 0.90 B r — ( 7 V-1 H b 7.53(d) 6.76(d) 0.77 H a 7.29 (s)1 6.40 (s)1 0.89 1 B r ~ ^ ~ ~ y ~ B r 4-chloro-l-iodobenzene H H 3 W b H a 7.07(d) 6.02(d) 1.05 C I — f 7 — I H b 7.59(d) 6.78(d) 0.81 l-bromo-4-chlorobenzene H H 3 W b H a 7.19(d) 6.15(d) 1.04 C I — f V B r H b 7.40(d) 6.45(d) 0.95 /;ara-dichlorobenzene Ha. H a 7.25 (s) 6.20 (s) 1.05 99 Guest structure Proton 5free (multiplicity) 8 b 0 u n d (multiplicity) A5 4-methylanisole Hh H r \= / O C H , H a H b H c H d 2.29(s)' 7.09 (d)1 6.81 (d)1 3.78 (s)1 -2.1 l(s) 1 5.87(s)1 5.87(8)' -0.35 (s)1 4.40 1 1.221 0.941 4.13 1 4-chloroanisole H a . H b C l — ^ V- O C H 3 H a H b H c 7.22 (d) 6.81 (d) 3.77 (s) 6.02 (d) 5.87 (d) -0.37 (s) 1.20 0.94 4.14 thioanisole H b He S C H , H a H b H c H d 7.10 (t) 7.24 (t) 7.24 (d) 2.44 (s) 2.99 (t) 4.98 (t) 5.89 (d) -1.65 (s) 4.11 2.26 1.35 4.09 anisole HL \ / c O C H , H a H b H c H d 6.95 (t) 7.29 (t) 6.91 (d) 3.80 (s) 3.19 (t) 5.39 (t) 5.80 (d) -0.19 (s) 3.76 1.90 1.11 3.99 iodobenzene H b H c « - 0 -H a H b He 6.85 (t)1 7.40 (t)1 7.65 (d)1 3.32 (t)1 5.38 (t)1 6.65 (d)1 3.53 1 2.02 1 LOO 1 bromobenzene H b H c H a ^ ^ B r H a H b He 7.28 (t) 7.22 (t) 7.49 (d) 3.37 (t) 5.61 (t) 6.30 (d) 3.91 1.61 1.19 chlorobenzene H a H Q - c , H a H b H c 7.30 (t) 7.25 (t) 7.18(d) 3.36 (t) 5.66 (t) 6.01 (d) 3.94 1.59 1.17 toluene H b H c CHo H a H b H c H d 7.30 (t) 7.18 (t) 7.25 (d) 2.30 (s) 3.32 (t) 5.65 (t) 5.83(d) -1.79(s) 3.73 1.50 1.24 4.09 100 Guest structure Proton 8free (multiplicity) S b o u n d (multiplicity) A8 fluorobenzene H a H b H c 7.13(t) 7.06 (t) 7.34 (m) 3.45 (t) 5.18 (t) 5.24 (m) 3.68 1.88 2.10 benzene 0*H- H 7.37 (s)1 4.70 (s)1 2.67 1 3-hexanol H BH a OH g H: H H a H b H c H d H e H f Hp 0.90 (t)1 1 1.41 (m) 1 1 1.41 (m) 1 1 3.51 (m) 1 1 2.07 (br s)" 1.36 (m)" 0.91 (t)1 1 -3.17 (t) -0.50 (m) -0.50 (m), -0.34 (m) 2.05 (m) 0.92 (d) -0.13 (m), 0.00 (m) -3.31 (t) 4.07 1.91 1.91, 1.75 1.46 1.16 1.49, 1.36 4.22 2,4-pentanediol b f HC- H c H e r OH H3C CH3 H d H a 1.20 or 1.17(d)" -2.51 (d) 4.42-3.68 H b 3.01 (brs)1 1 1.15 (d) 1.86 H c or H e 4.12 or 4.00 (m)" hidden -H d 1.58-1.43 (m)" 0.07-0.22 (m) 1.51-1.21 H e or H c 4.12 or 4.00 (m) 1 1 2.22 (m) 1.90-1.78 H f 3.01 (br s)" 0.85 (d) 2.16 H a or H g 1.20 or 1.17(d)" -2.46 (d) 4.42-3.68 3-pentanol d a H c OH H 3 C v ^ ^ \ ^ C H 3 H H a H b H c H d 0.91 (t)11 1.49 (m), 1.40 (m) 1 3.43 (m)" 1.41 (brs)" -2.77 (t) -0.70 (m), -0.56(m) 1.56 (m) 1.21 (d) 3.68 2.19, 1.96 1.87 0.20 2-pentanol HbMd UM H a C ^ ^ ' X H ; H c 2-butanol a H P H H ; H a " 0.90 (t)" -2.68 (t) 3.58 H b 1.45-1.21 (m) 1 1 -0.56 (m), -0.68 (m) -He 1.45-1.21 (m) 1 1 -0.21 (m), -0.36 (m) -H d 3.78 (m)" 1.85 (m) 1.93 H e 1.42 (brs)1 1 0.76 (d) 0.66 H f 1.15(d)" -2.62 (d) 3.77 H a 0.90 (t)" -2.76(t) 3.66 H b 1.47 (m)" -0.52 (m), -0.61 (m) 1.99, 2.08 H c 3.69 (m)" 1.71 (m) 1.94 H d 1.47 (d)1 1 1.00 (d) 2.94 H e 1.15(d)" -2.35 (d) 3.46 101 Guest structure Proton 8 f r e e (multiplicity) 8 b 0 u n d (multiplicity) A S isopropyl acetate b O C H 3 H 3 C ^ O ^ C H , H a H b He 1.98 (s) 1.20 (d)* 4.95 (m)* -2.38 (s) -0.42 (d)* 3.31 (m)* 4.36 1.62* * 1.64 dimethylsulfoxide O II a H a 2.46 (s)1 -0.49 (s)1 2.95 1 cyclohexane H a 1.44 -0.57 2.01 A^,iV-dimethylacetaniide ° * C H 3 a / - N c H 3 C C H 3 H a H b H c 2.08 (s)1 2.94 (s)1 3.02 (s)1 -1.64 (s)1 1.61 (s)1 -0.42 (s)1 3.72 1 1.331 3.441 7V-methyl-2-pyrollidinone Ha , "N d H c C H 3 H a H b He H d 2.23 (t) 1.90 (m) 3.26 (t)' 2.70 (s) m m in -0.59 (t) -0.78 (m) hidden -0.89 (s) HI III III III 2.82 2.68 3.59 m 1,3-dimethyl-2-imidazolidinone O a jl H h1 H a H b 3.22 (s) 2.73 (s) -1.06 (s) 1.77 (s) 4.28 0.96 7V-formylpiperidine H b H e H a H b He H d He H f 1-49 (m)* -1.22 (m)* 2.71 ' 1.33 (m)* -0.03 (m) or 0.18 (m)* 1.20-1.36* 1.38 (m)* -0.03 (m) or 0.18 (m)* 1.20-1.36* 3.12 (m)* 2.04 (m) or 2.09 (m)* 1.08-1.18* 3.27 (m)* 2.04 (m) or 2.09 (m)* 1.08-1.18* 7.80 (s)* 4.47 (s)* 3.33* 'From reference 4a. "Chemical shifts are concentration dependent. Values were recorded at -260 m M concentrations. H I F r o m reference 4b. From reference 13. 102 2.4.3 Competition experiments Tetrol 8a (50.0 mg, 0.049 mmol), K I (60.0 mg, 0.361 mmol, 7.4 equiv.), C s 2 C 0 3 (300 mg, 0.921 mmol, 19 equiv.) were mixed in 20 m L of JV-formylpiperidine, followed by the addition o f guest 1 (Gi) and guest 2 (G 2 ) at 80 °C with stirring. The relative ratios o f G i : G 2 added to the reaction mixture were chosen so as to obtain close to a 1:1 ratio o f hemicarceplexes to optimize integration in the ' H N M R spectra. The reactions were allowed to stir for at least 10 min before adding 1,4-dibromobutane (60.0 (iL, 0.493 mmol). After further stirring at 80 °C for 48 h, the solvent was removed in vacuo. The product mixture was then resuspended in CHCI3 and triturated before filtering through a pad o f Celite. The filtrate was evaporated, and the residue was "dry-loaded" onto a pad of silica gel and eluted with C H 2 C 1 2 . Solvent was removed in vacuo, and a solid was precipitated from CHCI3 by addition o f hexanes. The hemicarceplex product mixture was then dried at 0.01 mm H g at 70 °C for 24 h Product ratios were calculated from the ' H N M R spectra by integration of each set of guest signals. The error in the integration is estimated to be ± 10 %. ' 103 2.4.4 Template Ratios The results from each competition experiment from which the template ratios in Table 2.1 were calculated are listed in Table 2.4. Results o f the crosscheck experiments conducted to check the validity o f the template ratios in Table 2.1 are shown in Table 2.5. Crosscheck experiments were conducted using the same procedure as for the competition experiments. A sample calculation for the determination o f template ratios is as follows: Directly measured template ratios (adjusted for starting ratios o f guests): 25»isopropyl acetate/25»NFP = 9.69/1; 25«l,3-dimethyl-2-imidazolidinone/25»isopropyl acetate = 1.56/1. Therefore the template ratio for 25»l,3-dimethyl-2-imidazolidinone/25»NFP - (9.69/1)(1.56/1) = 16/1. The same procedure was used to calculate template ratios for the rest. Some competition experiments between directly adjacent pairs o f guests in Table 2.1 could not be performed because o f overlapping signals in the ' H N M R spectra. For these cases, the closest above or below neighbours were competed instead. For example: para-xylene guest signals overlapped with those o f 4-bromotoluene, 4-chlorotoluene, and /?ara-dibromobenzene. Therefore, para-xylene was competed against anisole. 104 Table 2.4 Competition experiment results. Guest 1 Guest 2 Guest 1 Guest para-xylene anisole 1.85 1.00 4-bromotoluene anisole 1.46 1.00 />-dibromobenzene anisole 1.10 1.00 4-chlorotoluene anisole 1.02 1.00 anisole 1 -bromo-4-chlorobenzene 1.13 1.00 1 -bromo-4-chlorobenzene /rara-dichlorobenzene 2.02 1.00 />ara-dichlorobenzene 1 -bromo-4-iodobenzene 1.16 1.00 1 -bromo-4-iodobenzene 1 -chloro-4-iodobenzene 1.17 1.00 1 -chloro-4-iodobenzene 4-methylanisole 1.08 1.00 4-methylanisole iodobenzene 1.26 1.00 3-pentanol iodobenzene 1.02 1.00 iodobenzene thioanisole 1.05 1.00 4-chloroanisole thioanisole 1.03 1.00 thioanisole bromobenzene 1.79 1.00 bromobenzene 2-butanol 1.23 1.00 2-butanol toluene 1.41 1.00 toluene 2-pentanol 1.04 1.00 2-pentanol benzene 1.20 1.00 benzene chlorobenzene 1.01 1.00 chlorobenzene 2,4-pentanediol 1.07 1.00 2,4-pentanediol 3-hexanol 1.05 1.00 3-hexanol fluorobenzene 2.55 1.00 cyclohexane fluorobenzene 1.49 1.00 fluorobenzene N M P 1.38 1.00 N M P D M A 1.10 1.00 D M A D M S O 1.53 1.00 D M S O D M I 1.08 1.00 D M I isopropyl acetate 1.56 1.00 isopropyl acetate N F P 9.69 1.00 Table 2.5 Crosscheck experiments results. Guest 1 Guest 2 Guest 1 Guest 2 para-xylene /?ara-dibromobenzene 1.23 1.00 para-xylene 4-bromotoluene 1.43 1.00 para-xylene 4-chlorotoluene 1.67 1.00 para-xylene N F P 3550.00 1.00 4-bromotoluene 4-chlorotoluene 1.46 1.00 4-bromotoluene jpara-dichlorobenzene 3.81 1.00 />ara-dibromobenzene iodobenzene 5.18 1.00 4-chlorotoluene bromobenzene 8.53 1.00 anisole 1 -chloro-4-iodobenzene 3.02 1.00 anisole />ara-dichlorobenzene 2.07 1.00 anisole bromobenzene 8.49 1.00 anisole thioanisole 5.48 1.00 1 -bromo-4-chlorobenzene /jara-dichlorobenzene 2.02 1.00 /rara-dichlorobenzene thioanisole 2.17 1.00 /rara-dichlorobenzene 4-methylanisole 1.37 1.00 4-methylanisole thioanisole 1.64 1.00 1 -chloro-4-iodobenzene benzene 6.44 1.00 iodobenzene N F P 445.00 1.00 iodobenzene thioanisole 1.20 1.00 3-pentanol bromobenzene 1.90 1.00 3-pentanol N F P 443.00 1.00 thioanisole toluene 2.64 1.00 thioanisole N F P 370.00 1.00 bromobenzene chlorobenzene 2.00 1.00 2-butanol chlorobenzene 1.57 1.00 2-butanol N F P 150.00 1.00 2-butanol 2-pentanol 1.33 1.00 toluene N F P 134.00 1.00 toluene chlorobenzene 1.26 1.00 toluene benzene 1.13 1.00 2-pentanol N F P 115.00 1.00 benzene N F P 95.60 1.00 benzene 2,4-pentanediol 1.08 1.00 benzene 3-hexanol 1.27 1.00 2,4-pentanediol N F P 72.00 1.00 3-hexanol N F P 57.00 1.00 fluorobenzene D M A 1.78 1.00 D M A D M I 1.65 1.00 D M S O Isopropyl acetate 1.37 1.00 D M S O N F P 14.00 1.00 2.4.5 Control experiments Hemicarceplex 25«guest 1 (25«G1, 5.0 mg), K I (12 mg, 0.072 mmol), C s 2 C 0 3 (90 mg, 0.275 mmol) were dissolved in o f ./V-formylpiperidine (4 ml). Guest 2 and guest 3 (G2 and G3 respectively, 1 mol % each) were added under N 2 , stirring at 80 °C before adding o f 1,4-dibromobutane (12 |j.L, 0.099 mmol). Competing guests were chosen so that the template ratios for G3 < G l < G2. The mixture was allowed to stir for two days at 80 °C before removing the solvent in vacuo. Purification of the crude product mixture was identical to the procedure used for the competition reactions. The results of these experiments are shown in Table 2.6. 107 Table 2.6 Control experiment results. ^ % guest Hemicarceplex Competing guests exchange 25«para-xylene 4-bromotoluene,^ara-dibromobenzene, N F P 0 25»4-bromotoluene para-xylene, /?ara-dibromobenzene, N F P 0 25«/?ara-dibromobenzene 4-bromotoluene, 4-chlorotoluene, N F P 0 25«4-chlorotoluene /?ara-dibromobenzene, anisole, N F P 0 25»anisole 4-chlorotoluene, l-bromo-4-chlorobenzene, N F P 0 25»l-bromo-4-chlorobenzene anisole, 1-bromo-4-iodobenzene, N F P 0 25»para-dichlorobenzene 1-bromo-4-chlorobenzene, l-bromo-4- 0 iodobenzene, 4-methylanisole, 3-pentanol, N F P 25»l-bromo-4-iodobenzene l-bromo-4-chlorobenzene, l-chloro-4- 0 iodobenzene, N F P 25«l-chloro-4-iodobenzene l-bromo-4-iodobenzene, iodobenzene, N F P 0 25«4-methylanisole />ara-dichlorobenzene, 4-chloroanisole, N F P 0 25«3-pentanol 4-chloroanisole, iodobenzene, N F P 0 25«iodobenzene 1-chloro-4-iodobenzene, 3-pentanol, N F P 0 25«4-chloroanisole 4-chloroanisole, bromobenzene, 4-methylanisole, 0 thioanisole, N F P 25»thioanisole 4-chloroanisole, bromobenzene, N F P 0 25«bromobenzene thioanisole, 2-butanol, N F P 0 25»2-butanol bromobenzene, toluene, N F P 0 25«toluene 2-butanol, chlorobenzene, N F P 0 25«2-pentanol toluene, benzene, chlorobenzene, N F P 0 25«benzene 2-pentanol, 2,4-pentanediol 0 25*chlorobenzene 2-pentanol, 3-hexanol, toluene, N F P 0 25»2,4-pentanediol benzene, 3-hexanol, N F P 0 25«3-hexanol benzene, 2,4-pentanediol, cyclohexane, N F P 0 25«cyclohexane 3-hexanol, fluorobenzene, N F P 0 25»fluorobenzene cyclohexane, N M P 0 25«NMP fluorobenzene, D M A , N F P 0 25*DMA N M P , D M S O , N F P 0 25»DMSO D M A , l,3-dimethyl-2-imidazolidinone, N F P 0 25»l,3-dimethyl-2- isopropyl acetate, D M S O , N F P 0 imidazolidinone 25«isopropyl acetate 2-butanol, benzene, N F P 1 3 0 1 3 25«NFP />ara-xylene, 2-butanol, N F P 14 108 2.5 References 1. (a) Chapman, R. G . ; Chopra, N . ; Cochien, E . D . ; Sherman, J. C. J. Am. Chem. Soc. 1994, 116, 369-370. '(b) Chapman, R. G . ; Sherman, J. C. J. Org. Chem. 1998, 63, 4103-4110. 2. Chopra, N . ; Sherman, J. C . Supramol. Chem. 1995, 5, 31-37. 3. (a) Chapman, R. G . ; Sherman, J. C. J. Am. Chem. Soc. 1995,117, 9081-9082. (b) Chapman, R. G . ; Olovsson, G . ; Trotter, J,; Sherman, J. C. J. Am. Chem. Soc. 1998,120, 6252-6260. (c) Chapman, R. G . ; Sherman, J. C. J. Am. Chem. Soc. 1998,120, 9818-9826. 4. (a) Robbins, T. A . ; Knobler, C. B . ; Bel low, D . R.; Cram, D . J. J. Am. Chem. Soc. 1994, 116, 111-122. (b) Yoon , J.; Sheu, C ; Houk, K . N . ; Knobler, C . B . ; Cram, D . J. J. Org. Chem. 1996, 61, 9323-9339. 5. Sherman, J. C ; Knobler, C. B . ; Cram, D . J. J. Am. Chem. Soc. 1991,113, 2194-2204. 6. Kurdistani, S. K . ; Helgeson, R. C ; Cram, D . J. Am. Chem. Soc. 1995,117, 1659-1660. 7. Place, D . ; Brown, J.; Deshayes, K . Tetrahedron Lett. 1998, 39, 5915-5918. 8. Robbins, T.; Cram, D . J. J. Am. Chem. Soc. 1993,115, 12199. 9. Romanova, Z . S.; Deshayes, K . ; Piotrowiak, P. J. Am. Chem. Soc. 2001,123, 11029-11036. 10. (a) Warmuth, R. Angew. Chem. Int. Ed. Engl. 1997, 36, 12, 1347-1350. (b) Warmuth, R. J. Chem. Soc, Chem. Commun. 1998, 59-60. (c) Beno, B . R.; Chimin, S.; Houk, K . N . ; Warmuth, R.; Cram, D . J. Chem. Commun. 1998, 301-302. 11. (a) Warmuth, R.; Marvel , M . Angew. Chem. Int. Ed. Engl. 2000, 39, 1117-1119. (b) Warmuth, R.; Marvel , M . Chem. Eur. J. 2001, 7, 1209-1220. (c) Warmuth, R. J. Am. Chem. Soc. 2001,123, 6955-6956. (d) Warmuth, R.; Kerdelhue, J. - L . ; Carrera, S. S.; Langenwalter, K . J.; Brown, N . Angew. Chem. Int. Ed. Engl. 2002, 41, 96-99. 12. Makeiff, D . A . ; Pope, D . J.; Sherman, J. C. / . Am. Chem. Soc. 2000,122, 1337-1342. 109 13. Pope, D . J. Masters thesis, University of British Columbia. 14. Thirteen o f the thirty guests used in the templation study, (para-xylene, para-dibromobenzene, j^ara-dichlorobenzene, 2-butanol, toluene, benzene, chlorobenzene, cyclohexane, N M P , D M A , D M S O , isopropyl acetate, N F P ) were screened by Pope. The rest were screened for this thesis (see reference 13). 15. Only nine o f these seventeen templates (tetramethylene sulfoxide, 1,4-dioxane, 1,4-dithiane, 4-fluorotoluene,7?ara-difluorobenzene, l-fluoro-4-iodobenzene, l-bromo-4-fluorobenzene, l-chloro-4-fluorobenzene, and 4-bromoanisole) were discovered in work for this thesis. The remaining eight (2-propanol, «-propyl acetate, ethyl acetate, cyclohexanone, T H F , 1,4-thioxane, thiophene, pyrazine) were discovered by Pope (reference 13). 16. Eisenberg, D . ; Crothers, D . Physical Chemistry with Applications to the Life Sciences. 1979, Forkner, M . Ed. ; Benjamin/Cummins Publishing Company: California, 508. 17. Yoon , J. ; Sheu, C ; Houk, K . N . ; Knobler, C. B . ; Cram, D . J. J. Org. Chem. 1996, 61, 9323-9339. 18. Mungaroo, R.; Sherman, J. C. unpublished results. 110 3. Investigation into the Formation of a Carceplex Derived From Three Cavitand Subunits 3.1 Introduction The first two chapters of this thesis discussed studies by the Sherman group involving templation in the formation of two-bowl carceplexes 10a»guest ' and hemicarceplexes l l a « g u e s t (Chapter 1) and 25»guest (Chapter 2). 3 Each of these systems are formed via Cram's methodology involving the bridging o f two tetrol bowls (8a) with four linker groups. 4 Larger systems have also been synthesized using Cram's method, but they are not suitable for templation studies because the longer interbowl linkages often result in large holes in the host's shell, which preclude guest containment. 4 , 5 Therefore, as a movement to extend our template studies to the encapsulation of larger and higher order guests, we had to synthesize larger hosts made from rigid and non-porous subunits. 10a«guest 11a«guest 25»guest Our group has recently succeeded in creating larger carceplexes by expanding the size (i.e., widening) of the bowl subunit 6 and by linking three bowls together7 to form hosts whose molecular recognition properties are just beginning to be explored. 8 , 9 The first three-bowl carceplex, 56*(DMF)3 (36 %), was synthesized by Chopra and Sherman by sealing off the cavity of cyclic trimer 43 with mesityl caps. 9 From the same reaction, an intermediate byproduct, 111 trimer cavitand 114 ( 1 0 %) was also isolated. 1 0 The large chamber within carceplex 56»guests makes it an excellent candidate for further explorations into the chemical and physical properties o f confined media and mechanisms of carceplex formation/guest entrapment. Therefore, for this work we pursued increasing the repertoire of encapsulated guests, and studying their relative templating abilities in the formation of 56»guests. We knew that solvent could be entrapped, but wondered i f larger single-molecule guests could be as well . Scheme 3.1 Synthesis of trimer carceplex 56«guest from tetrol 8a. 56*guests 114 (10-35%) (10-25%) Carceplex 56«guests is distinct from other reported systems as it offers the potential to study (kinetic) template effects between suitable guests (templates) that differ in molecularity. One o f the first questions that comes to mind is: can multiple-molecules template the formation o f a capsular host such as 56«guests? Are entrapped clusters o f small-molecules brought 1 1 2 together in a random (statistical) fashion, or does the binding o f one component to a host facilitate the nucleation of the others (cooperatively) in the templation o f a larger host? Furthermore, can templation by multiple-molecules be distinguished from a (micro)solvent effect? Template effects and solvent effects represent a "grey" area that few researchers have attempted to delineate, largely because of the lack of means to do so. Carceplex 56»guests may provide such a means in its ability to sequester multiple solvent/template molecules from the bulk reaction medium. This chapter is broken down into four main sections, each o f which presents studies pertaining to 56»guests and related hosts to further our understanding o f the driving forces behind carceplex formation. In the next three sections, the synthesis and characterization o f trimer carceplex 56»guests and its precursors w i l l be discussed (3.2 and 3.3), followed by templation studies (section 3.4). Section 3.5 deals with investigations o f several A,C-trimer derivatives as potential transition state models for the GDS in the formation of 56»guest. Finally the chapter ends with a short section (3.6) describing the complexation o f multiple-molecule guests to hexa-benzyl trimer 42 (Scheme 3.1). 3.2 Improved Synthesis of A,C-Trimer 42 The synthesis of trimer carceplex 56«guests' 1 was first reported by Chopra and Sherman in four steps from tetrol 8a (Scheme 3.1). 7 , 1 2 In the initial report, tetrol 8a was benzylated ( D M F , K2CO3, benzyl bromide) to give A , C - d i o l 41 in 6 % yield, which was cyclized in D M F (K2CO3, B r C H 2 C l ) to give hexa-benzyl trimer 42 in low yield (5 %) . 7 Hydrogenolysis o f 42 ( H 2 , Pd/C, 4 arm., 90:5:5 benzene:MeOH:acetone, 2 d) quantitatively gave hexa-hydroxyl derivative 113 (43).7 Since trimer 43 was only available in 0.1 gram quantities from grams o f starting material (-0.03 % in seven steps from commercially available resorcinol) and multi-gram quantities o f 8a have been required in the past for templation studies of carceplexes 10a»guest and 25«guest (chapters 1 and 2), hexa-hydroxyl trimer 43 had to be synthesized more efficiently. Since the initial report, we have recently improved the efficiency o f several o f the steps in the synthesis of hexa-hydroxyl trimer 43. The yield o f A , C - d i o l 41 was improved (by Christoph Naumann o f our group) from 6 % to 10-15 % using different conditions ( D B U , acetone, benzyl bromide). 1 3 No selectivity was observed in this reaction as the A , B - d i o l (see Chapter 5), mono-, tris-, and tetra-benzyl by-products were also formed. A n eight-fold increase (40 %) in yield o f the benzylated trimer 42 was attained by the 24-hour addition o f separate D M S O solutions o f diol 41 and B r C E ^ C l to a stirring mixture o f D M S O and K2CO3 at 60 °C. Deprotection o f hexa-benzyl trimer 42 was also carried out more efficiently under milder (H2, Pd/C, 1:1 benzene:methanol, 1 arm., ~3 h). 3.3 Synthesis and Characterization of Trimer Carceplexes 56*Guests 3.3.1 Carceplexes From Neat Solvent To begin templation studies, a reaction solvent suitable for guest screening and conducting competition experiments was sought. Just as for previous templation studies (e.g., in Chapter 2), the reaction solvent must be of sufficient polarity, and a poor template either as a single-molecule or a multiple-molecule guest. Many single-molecule guests that were considered and tested as potential templates (molecules too large for the cavity) were high-114 melting solids, which are unsuitable solvents and are not commercially available in liter quantities. Carceplexes 56«(guest)x were isolated from reactions in neat D M F , D M A , N M P , N F P , D M S O , butyrophenone, acetophenone and hexanophenone. Vigorous conditions (80-90 °C, 4-5 days) were required to form 56»butyrophenone in neat butyrophenone in. comparable yields (~25 %) to other 56«(guest)j;. Reactions in D M S O , acetophenone, and hexanophenone gave poor yields of 56»(DMSO) 3, 56«(acetophenone) 2, 56*hexanophenone,14 respectively, which were isolated from messy reaction mixtures. Thus, butyrophenone, D M S O , acetophenone, and hexanophenone were not good reaction solvents. 1 5 Good yields (-35 %) o f carceplexes 56«(DMF) 3, 56«(DMA) 2, 56«(NMP) 2 and 56«(NFP)^ (x = 1 and 2 ) 1 6 were obtained in neat D M F , D M A , N M P , and N F P , respectively, at room temperature after two days. Therefore, further reactions were conducted with binary solvent mixtures of D M F , D M A , N M P , and N F P to determine which were the poorest templates. N F P was determined to be the best reaction solvent because o f its relatively poor templating ability on its own. For example, a reaction in N F P : D M F (1:1) gave only 56«(DMF) 3, while reactions in N F P spiked with D M F , D M A , N M P , or D M S O (>1 % v/v) furnished only mixtures o f 56»(NFP«guest) and 56«(guesf)JC (guest = D M F , D M A , N M P , D M S O ) ; no 56«NFP or 56«(NFP) 2 were obtained under these conditions. A l l pure carceplexes 56»(guest) x and carceplex 56»guests mixtures were characterized by both ! H N M R spectroscopy and M A L D I mass spectrometry. 115 3.3.2 Trimer Carceplex Water Complexes 3.3.2.1 Characterization of Complexes 56»[Guests»(H 20) ),] Initially, seemingly pure samples of 56«guests (according to T L C and M A L D I mass spectrometry) isolated gave unexpectedly complicated ' H N M R spectra. A s described in this section, the complex spectra were attributed to the formation o f water complexes with 56»guests, 56«[guests«(H20) >,] (y = 0, 1, 2, 3,...), which are in slow exchange on the lH N M R timescale (Table 3.1). In wet organic solvents, multiple sets o f resonances for each type o f host, guest, and water protons are observed, which correspond to various carceplex hydrates and the anhydrous carceplex. However, in sieve-dried solvents, only a single species, the anhydrous carceplex is observed. Water complexation has been previously reported with hemicarcerands in solution (i.e., hemicarceplex l l a « ( H 2 0 ) x ) 1 7 and the solid state (i.e., hemicarceplexes 25«H20 and 25«(H 2 0) 2 ) , but never for carceplexes or hemicarceplexes (i.e., the guest plus a water molecule). Table 3.1 lists the lH N M R chemical shifts o f bound water resonances (singlets) in various trimer carceplex complexes, 56»[guests«(H20)y]. The chemical shift differences between free and bound water (A8) values range between 0.51 and 3.26 ppm and appear to be dependent on the permanently entrapped carceplex guest. These values l ikely reflect the amount o f unoccupied space and the orientation of the permanently entrapped guest(s). In most cases, between one and three bound waters for each hydrate are estimated, but this could not be determined precisely because o f poor signal dispersion o f the many overlapping host and guest resonances. Ironically, the first trimer carceplex ever synthesized, 56«(DMF)3, does not show 116 any evidence for the formation of complexes 56«[(DMF) 3 »(H 2 0) y ] , even in water-saturated solvents ( H 2 0 and D 2 0 ) . Table 3.1 ' H N M R chemical shift data for bound H 2 0 in carceplexes 56«[guests»(H 20)^] in CDCI3 at 3 0 0 K . Guest(s) Sbound (H 20) 8 A 5 ( 8 f ree -$bound) Hexanophenone 0 . 6 1 , - 1 . 2 4 0 . 9 3 , 2 . 7 8 Butyrophenone 0 . 7 3 , - 0 . 4 3 , - 1 . 5 7 0 . 8 1 , 1 . 9 7 , 3 . 1 1 1,3,5-tris(ethynyl)benzene - 0 . 2 0 1 . 7 4 , 2 . 8 4 Trimethyl-1,3,5-benzenetricarboxylate - 0 . 1 4 , - 0 . 8 0 1 . 6 8 , 2 . 3 4 1,3,5-trimethoxybenzene 0 . 6 1 , 0 . 1 1 , - 0 . 1 1 1 . 4 3 , 1 . 6 5 , 2 . 1 5 1,3,5-triethylbenzene 1 . 0 3 , 0 . 6 5 , - 1 . 0 0 0 . 5 1 , 0 . 8 9 , 2 . 5 4 N F P 0 . 6 0 , 0 . 1 0 , - 0 . 7 8 , - 1 . 7 2 0 . 9 4 , 1 . 4 4 , 2 . 3 2 , 3 . 2 6 N F P - D M S O 0 . 6 4 , 0 . 6 1 , - 0 . 2 4 , - 1 . 4 8 0 . 9 0 , 0 . 9 3 , 1 . 7 8 , 3 . 0 2 ( N M P ) 2 0 . 4 7 , - 0 . 9 8 1 . 0 7 , 2 . 5 2 ( D M A ) 2 0 . 4 3 , - 0 . 5 9 1 . 1 1 , 2 . 1 3 5free ( H 2 0 ) in CDCI3 is taken to be 1 . 5 4 ppm (relative to the residual CHCh signal at 7 . 2 4 ppm). a In H 2 0 saturated CDCI3. Semi-quantitative estimates of guest exchange rates between hydrated and non-hydrated 56«guests are minutes or less at room temperature: when D 2 0 ( 1 - 2 U.L) is added to CDCI3 solutions containing complexes 56«[guests«(H 20) >,], the bound water signals disappear within the time required to record a single *H N M R spectrum, while the host signals remain virtually unchanged. Weak exchange correlations between the free and bound water were detected by I D E X S Y ( N O E S Y ) at 3 0 0 K , which implies that exchange rates are near the lower limits o f the time regime applicable to E X S Y (rate constants o f ~ 1 0 " 2 s"1).1 9 Increasing the temperature did not improve this because of poor thermodynamic stability; heating results in a decrease in the intensities the hydrate signals. Anomalous to the hydrated carceplexes listed in Table 3 . 1 is 56»(DMSO)3. ' H N M R spectra of 56«(DMSO)3 in water-saturated CDCI3 do show a major (dry species) and a minor (a 1 1 7 hydrate) set o f host and guest resonances. However, no bound water signals are evident. Bound and free water may have identical chemical shifts, or 56»(DMSO)3 may only bind many waters (i.e., three or more, cooperatively): the loss or gain of a single water molecule from 56»[(DMA)3»(H20),,] may be fast on the N M R timescale, whereas the loss or gain o f many waters is kinetically slow. Molecular ( C P K ) models show that the structure of 56 is extremely rigid and possesses only very small holes, through which small molecules such as water can pass. The largest holes appear to be near the methyl groups of the mesitylene caps (six in total) and between the four arenes (near H p s ) at the lower r im of each of the three [4] cavitand subunits. C P K models suggest that water can only fit through the latter o f these holes. The next two sections w i l l present a case study describing carceplex 56»(DMA)2 and its hydrates (56»[(DMA) 2 »(H 2 0) r ] , y = 1 and 2). 3.3.2.2 Trimer Carceplex Complexes 56»[(DMA) 2»(H 20),] Trimer carceplex 56*(DMA)2 was first synthesized by Chopra in neat D M A solvent ( K 2 C 0 3 , K I , 2,4,6-tris(bromomethyl)mesitylene).1 0 Although M A L D I M S data suggested that the isolated carceplex 56«(DMA)2 was pure, lU N M R spectra at room temperature (Figure 3.1) were complicated in comparison to spectra for 56«(DMF)3.10 This was initially attributed to asymmetry in carceplex 56»(DMA)2 and/or the entrapment of unknown guests (i.e., trace solvent impurities that could not be removed by distillation). 1 0 118 (a) (b) (c) i" . 4.60 4.50 4.40 4.30 4.20 b' h JJ c' c" 1 1 1 JL J i S b" \ b l 0L c" C | Wl b c 1.60 1.50 1.40 0.70 0.60 0.50 0.40 (ppm) _i a a a i JJ\ U -0.60 -0.80 -1.0 -1.20 H 3 C O b N - ( a H3C CH3 Figure 3.1 Expanded regions of N M R spectra (400 M H z ) of 5 6 » ( D M A ) 2 . (a) H 20-saturated CDCI3, (b) D 20-saturated CDCI3, and (c) sieve-dried C D C 1 3 . a, b, c, and i (Hi) are for 56»(DMA) 2 . a', b', c', i ' (Hi'), and * (bound H 2 0 ) are for 5 6 « [ ( D M A ) 2 » H 2 0 ] . a", b", c", i " (Hj'), and *' (bound H 2 0 ) are for 5 6 » [ ( D M A ) 2 « ( H 2 0 ) 2 ] . h = free H 2 0 . h' = free H D O . bowl 56, FT = CH 3 , R = C H 2 C H 2 C 6 H 5 115, R' = H, R = C H 2 C H 2 C 6 H 5 Figure 3.2 Proton peak labels for trimer carceplex 56«guest. 119 To pick up where Chopra left off, several ' H N M R studies were conducted to fully characterize carceplex 56»(DMA)2, which eventually led to the discovery o f hydrated carceplexes. In sieve dried CDCI3, only a single set o f resonances (labeled i , a, b, and c in Figure 3.1c) appear for each type o f host and guest proton. For solutions o f 56*(DMA)2 containing H 2 O , the spectra are much more complicated, displaying two other major sets o f signals for the host and guest protons (labeled i ' , a', b', c', * and i " , a", b", c", *', Figure 3.1a). We conclude that the original set o f signals in dry CDCI3 are from dry 56«(DMA)2, while the new sets are from 56»[(DMA) 2 «(H 2 0) > ,] . The bound H 2 0 signals at 0.43 and -0.59 ppm (Figure 3.1a), were identified from their absence in spectra of 56«(DMA)2 recorded in D 2 O saturated CDCI3 (Figure 3.1b). Integration o f the ' H spectrum and 2D N O E S Y correlations 2 0 suggest that the carceplex hydrates contain one (-0.59 ppm) and two water molecules (0.43 ppm). Considering the equilibria: 56«(DMA) 2 + H 2 0 ~ 56«(DMA) 2 «H 2 0 56»(DMA) 2 »H 2 0 + H 2 0 ^ 56«(DMA) 2 «(H 2 0) 2 equilibrium constants for the stepwise binding o f one and two water molecules (Aj and K2, respectively, in M" 1 ) are: K _ [ 5 6 . ( D M A ) 2 * H 2 Q ] ' [ 56»(DMA) 2 ] [H 2 0] K [ 5 6 « ( D M A ) 2 * ( H 2 Q ) 2 ] 2 [56«(DMA) 2 • H 2 0 ] [ H 2 0 ] For 56«(DMA) 2 (1.85 m M , in H 20-saturated CDCI3, 2 1 300 K ) , Kx and K2 were calculated to be 25 M " 1 and 11 M " 1 , respectively. Binding of the first water molecule makes it less favorable for a second water to b ind . 2 2 This is l ikely because there is less space available in the carceplex cavity for the second water molecule. 120 To test how in/out exchange rates of water are affected when larger holes are present in the trimer carceplex shell, benzene-capped carceplex 115«(DMA)2 was synthesized by capping trimer 43 with l,3,5-tris(bromomethyl)benzene instead o f 2,4,6-tris(bromomethyl)-mesitylene). In water saturated solution, no bound water signals are observed for 115»(DMA)2. Instead, significant broadening o f the H ; (host) and D M A guest proton resonances occurs. These broad signals sharpen with heating, or in dry solutions containing 115»(DMA)2 (at 300 K ) . Thus, carceplex 115«(DMA)2 also forms hydrates (i.e., 115»[(DMA)2»(H20) > , ] ) , in which the absence o f the cap methyls allows water to enter/exit the inner phase faster than for 5 6 « [ ( D M A ) 2 » ( H 2 0 ) , ] . 115*guests 3.3.2.3 Effects of Acid and Base in the Complexation Water The discovery o f hydrated carceplexes, 56«[guests»(H20) > , ] , led to further investigations into their stability in the presence of acid and base using *H N M R spectroscopy. The effect of acid was tested by titrating trifluoroacetic acid (TFA) to water-saturated C 6 D 6 ( [ H 2 0 ] - 3 9 m M ) solutions containing 5 6 » ( D M A ) 2 . 2 1 For testing the effect o f base, hydroxide was generated in situ by adding D B U to 5 6 » ( D M A ) 2 into water-saturated C 6 D 6 . Carceplex 5 6 « ( D M A ) 2 was used for these experiments because it forms more stable hydrates than other 56«[guests»(H20) > , ] , and 121 because 56»[(DMA) 2 «(H20) > , ] ( y = 1,2) were the best characterized o f all 56»[guests»(H20) y ] . For both acid and base experiments, equilibrium was reached within minutes, as ' H N M R spectra recorded did not change significantly within the 24 hour time period after the initial mixings. Increasing the concentration o f acid (H + /H30 + ) and base ( O H ) (in separate experiments) both decreases the concentration of the hydrates 56«[(DMA)2*(H20) y] ( y = 1,2) and increases the concentration of the dry carceplex. This was monitored by observing the change in the relative integration of the signals for each species 56«[(DMA)2»(H 20)y] ( y = 0, 1,2) in the l H N M R spectra. The intensities o f both host and guest signals corresponding to the hydrated species decrease, while the signals for the anhydrous species increase with increasing concentrations of acid or base (e.g., the acetyl methyl proton signals in Figure 3.3a). For example, when T F A (5 equiv. per water) was added to a water-saturated C6D6 solution o f 56»(DMA) 2, the ratio of 56«(DMA)2:56«[(DMA)2*H20]:56»[(DMA)2»(H20)2] went from 1:2.5:4.2 (Figure 3.3a) to 20:5.2:1 (Figure 3.3b). Similarly, the ratio changed from 1:2.4:4.7 to 23:3.7:1 (Figure 3.3c) after adding D B U (20 equiv. per water). Four times less acid ( T F A ) was required to achieve the same effect with base ( O H ) . We conclude that acid and base both inhibit the formation of the trimer carceplex hydrates through the formation of H s O + or "OH, respectively. CF3CO2" H 3 0 + has a stronger need for hydration in C 6 D6 than does D B U H + "OH. These results further support the theory that 1 7 charged species are poorly stabilized within the inner phase o f carceplexes/hemicarceplexes. Protonation/deprotonation o f entrapped guests has also been observed to affect the thermodynamic stabilities o f other related host-guest complexes. 1 7 , 2 4 The inhibited binding o f water to 56»guests is important to studies presented in Chapter 4. 122 -0.2 -0.3 -0.4 -0.S -0.6 -0.7 -0.8 -0.9 (ppm) Figure 3.3 *H N M R spectra (500 M H z , H 20-saturated C 6 D 6 ) of the acetyl methyl protons o f 56»[(DMA)2»(H 20)j,]. (a) 56«[(DMA.) 2«(H 20) J,]. (b) 56«[(DMA) 2»(H 20),,] and 180 m M T F A (5 equiv. per H 2 0 ) . (c) 56«[(DMA) 2 »(H 2 0) y ] and 670 m M D B U (20 equiv. per H 2 0 ) . * = bound H 2 0 . a = 56«(DMA) 2. a'= 56»[(DMA) 2 »H 2 0]. a" = 56«[(DMA) 2 «(H 2 0) 2 ] . [ H 2 0 ] is estimated to be 39 m M . 2 1 The remainder o f section 3.3 w i l l discuss the characterization o f several single- and two-molecule carceplexes. Conformational and orientational mobility for a few entrapped guests w i l l also be examined. 3.3.3 Single-Molecule Guest Carceplexes 3.3.3.1 Synthesis and Characterization Single-molecule templates screened in the formation o f 56»guest were chosen based on their commercial availability or synthetic accessibility. C3-symmetric guests such as 1,3,5-trisubstituted benzenes were considered because of their similar C3-symmetry to the host cavity. 123 A r y l ketones were screened with subsequent photolysis (Norrish type II cleavage) experiments in mind (see Chapter 4 ) . Molecular (CPK) models were used to get a feel for the complementarity between potential guests with the size and shape of the interior of 56«guest. Models o f all suitable single-molecule guests (Chart 3 . 1 ) fit easily in the interior o f 56«guest, with the exception o f trimethyl-l,3,5-benzene tricarboxylate (117), which fits more tightly in the cavity of 56 than other single-molecule guests (in Chart 3 . 1 ) . A total o f nine single-molecule guest (Chart 3 . 1 ) trimer carceplexes (56»guest) were synthesized. A l l 56»guest were formed in N F P , doped with an appropriate guest (see experimental), except for 56«butyrophenone (vide supra). Seven o f the nine suitable single-molecule guests were isolated in high purity ( > 9 0 %); 56»valerophenone and 56«heptanophenone were isolated as carceplex mixtures with 56«(NFP) X (x - 1 , 2 ) . Table 3 . 2 ' H N M R chemical shifts (sieve-dried CDCI3) for various 1,3,5-trisubstituted benzene derivative guests in 56»guest. Guest Proton Sfree (ppm) Abound (ppm) A 5 (ppm) Trimethyl ester 117 H a 3 . 9 6 - 0 . 5 8 4 . 5 4 A * a H b 8 . 8 5 ~ 7 . 2 4 a - 1 . 6 1 Tris-acetylene 116 H a 3 . 0 8 - 1 . 2 4 4 . 3 2 * a H b 7 . 5 5 5 . 9 1 1 . 6 4 H a 1 . 2 3 - 2 . 0 3 3 . 2 6 Triethyl 119 H b 2 . 6 0 0 . 7 8 1 . 8 2 H c 6 . 8 6 5 . 1 2 1 . 7 4 Trimethoxy 118 H a 3 . 7 5 0 . 5 3 3 . 2 2 H b 6 . 0 8 4 . 3 4 1 . 7 4 Signal is hidden under A r H protons of the carceplex feet. See Chart 3 . 1 for proton labels. 1 2 4 Chart 3.1 Single-molecule guests for the templated formation of carceplex 56*guest. Molecular ( C P K ) models and the A8 values in Table 3.2 both suggest that the trisubstituted benzene guests are all oriented in 56*guest with the host and guest C3-axes aligned. The protons at the termini (i.e., H a for each C3-symmetric guest in Chart 3.1) of the aryl substituents are directed deep into the shielding bowl subunits, while the aryl protons are situated in the less shielded central region. This is consistent with the larger upfield shifts of the terminal protons (A5s range 3.22-4.54 ppm), which increase with substituent length, and the smaller upfield shifts for the corresponding aryl protons (A8s range 1.61-1.74). The predicted orientations from A8 values are in agreement with M M 2 minimized structures (Figure 3.4), which illustrate the complementarity between these guests and the host cavity. C P K models also suggest that guest rotation about the C3-axis of the host l ikely occurs freely for 56*118 and 56*119, but is severely hindered for 56*116 and 56*117. A discussion of the predicted orientations of the guests in 56*121 and 56*123 (Figure 3.4) is deferred until Chapter 4. 125 56.120 56 Figure 3.4 M M 2 minimized space filling models of carceplexes 56»guest. To simplify calculations, the pendant groups (R = C H 2 C H 2 C 6 H 5 , see Figure 3.2) of the [4]cavitand subunits were replaced with Hs. For clarity, some atoms have been removed to show the geometries of each entrapped guest inside the host. 3.3.3.2 Trimer Carceplex 56«Trimethyl-l,3,5-benzenetricarboxylate Several investigations into the molecular and conformational mobilities o f guests 25 entrapped in carceplexes and related capsular hosts by N M R spectroscopy have been reported. M a n y o f these systems were mentioned in Chapter 1, which included the interconversion 26 between carceroisomers in Reinhoudt's calix[4]arene-resorcin[4]arene hybrid carceplex and rotational barriers for the amide bonds of entrapped amide guests (i.e., D M F ) in several carceplexes (Chapter 1). ' The slowed interconversion between the axial and equatorial protons o f six-membered rings such as in thioxane 2 9 and cyclohexane 3 0 within the confining environments of reversibly forming capsules have also been reported. In this section, the slowed rotation about the aryl-carbonyl carbon bonds of trimethyl-l,3,5-benzene tricarboxylate (117) in 56»117 is discussed. Carceplex 56«117 is the only single-molecule guest carceplex that has a broad ' H N M R spectrum at 300 K in CDCI3 (Figure 3.5c), which reflects the restricted conformational mobility imposed on the largest single-molecule guest, trimethyl ester 117. A t 250 K , there are two sets o f host resonances for H 0 , H m , H x , H; , and H P 2 while only single sharp singlets are observed for H p i , H a c , and the ester methyl protons of bound 117 (Figure 3.5e). The doubling of the host resonances at 250 K is unique to 56»117, as other carceplexes 56»guests only show a single set o f broad resonances. A t higher temperatures (Figures 3.5a and b), the pairs o f host resonances broaden and coalesce into single signals (Figure 3.5b-e). A t 400 K in nitrobenzene-^ (Figure 3.5a), the spectrum is sharp and only a single set of signals appear for each type o f host and guest proton. 127 H x T 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I 1 I I I I I I 7.0 6.6 6.2 5.8 5.4 5.0 4.6 4.2 -0.10 -0.70 (ppm) Figure 3.5 *H N M R spectra (500 M H z ) of 56-117. (a) 56»117 in nitrobenzene-J 5 at 400 K and in C D C 1 3 at (b) 330 K , (c) 300 K , (d) 290 K , (e) 250 K . Note that for the proton labels in (e): H p 2 = H p 2 ' and H p 2 " , H 0 = H 0 * and H 0 " , H m = H m ' and H m " , H x = H x * and H x " , H ; = H i ' and Hj". * = C 0 2 C / / 3 . See Figures 3.1 and 3.6 for proton labels. In theory, tris-ester 117 can adopt several conformations via rotations about the aryl-carbonyl carbon ( A r - C O ) and the C - 0 bonds of the ester substituents. 180° rotations about the A r - C O bond gives rise to two rotamers that differ in the direction of one of the three ester carbonyls (i.e., CCC and CCC, Scheme 3.2). The energy barrier for the A r - C O bond rotation in methyl benzoate has been previously measured to be 4.92 kcal/mole. 3 1 For each CCC and CCC, the three ester substituents can adopt cis (E) or trans (Z) conformations, which interconvert by rotations about the C - 0 bonds of the ester substituents.3 2 Usually the E form is negligible at 128 room temperatures,3 2 and the energy barrier for EIZ interconversion for simple esters is generally 1-2 kcal /mole . 3 2 ' 3 3 Scheme 3.2 Conformational interconversion processes for 117. 117 (CCQ 117 (CCC) o o X — A . C H 3 R O R O I CH 3 cis (£) trans (2) The ! H N M R spectrum at 250 K is most consistent with 56»117 containing the rotamer 117 ( C C C ) . Each individual bowl in the structure of 56»117 is equivalent, while the pairs o f interconverting protons ( H x ' / H x " , H P 7 H P " , H o / H 0 ' , H m / H m ' , H i / H , ' ) on each bowl are nonequivalent due to the direction of the protruding carbonyl (Figure 3.6). This assumes that each carbonyl is planar to the arene o f the guest (and the arene caps o f the host), which is reasonable because aryl carbonyl groups typically prefer to be planar (conjugated) with directly bonded aromatic rings. The spectrum of 56*117 in Figure 3.5e is not consistent with the less symmetric rotamer 56»117 (CCC), which is expected to show a more complicated spectrum. However, coincidental chemical shifts for 56»117 (CCC) and 56*117 (CCC), as well as fast exchange between the two cannot be ruled out. Furthermore, the numerous E and Z isomers (four for CCC and six for CCC) cannot be distinguished by ' H N M R spectroscopy; either each gives coincidental chemical shifts or all are in fast exchange. 3 4 129 Figure 3.6 Symmetry o f the host imposed by the guest in 56»117. The red arrow indicates the direction o f the carbonyl group, (a) Looking into one of the [4]cavitand subunits with peak labels for the various host protons. The arrow is arbitrarily shown from right to left, (b) Top view (schematic) of 56»117 (CCC) with proton labels for H x and H a c . (c) Top view (schematic) of56«117 (CCC). CCC and CCC are shown in Scheme 3.2. Note that for (b) and (c) the C 3 -axis o f 56 (and 117) is perpendicular to the plane of the page. The energy barrier for the interconversion of H, ' and H / ' was measured to be 14.2 kcal/mole at 267 K in C D 2 C 1 2 by I D E X S Y (rate constant is 13.0 s"1).3 5 This is approximately three times greater than the barrier for the A r - C O bond rotation for both methyl benzoate (4.92 kcal/mole) 3 1 and/?ara-methoxybenzoate (4.6 kcal/mole). 3 6 This energy barrier is interpreted to correspond to the process whereby CCC interconverts to an equivalent CCC via concerted 180° rotations of all three A r - C O bonds. A s was observed with a penta-disulfide bridged carceplex 58«guests (Chapter 1, section 1.3.6), although the intrinsic barrier for each rotation in 56*117 is 130 ~5 kcal/mole, the requirement for multiple rotations in concert in the constrained carceplex environment results in a substantive energy barrier (14.2 kcal/mol). To our knowledge, the rotational barrier for 117 free in solution has not been reported. 3.3.4 Carceplexes 56«(NFP»Guest) 3.3.4.1 Carceplex 56«(NFP»DMSO) Carceplex 56«(NFP«DMSO) was isolated from the trimer carceplex reaction in N F P doped with D M S O (<5 mol % of the solvent), which was characterized by ' H N M R spectroscopy and M A L D I mass spectrometry. Preliminary reactions also suggested that carceplexes 56»(NFP*guest) were also formed from similar reactions with D M F , D M A , and N M P . Guest mobility in 56»(NFP*DMSO) was probed by variable (low) temperature ' H N M R spectroscopy. A t 200 K in CD2CI2, the spectrum (Figure 3.7e) is broad and complicated, especially in the host region between 4.5-7.0 ppm. The symmetry o f the host appears to be considerably reduced due to restricted guest mobility. This is also indicated by the presence o f two broad resonances of equal intensity at 0.69 and -1.89 ppm that correspond to two nonequivalent bound D M S O methyl proton environments that are in slow exchange (relative to the N M R timescale). The downfield signal is probably from D M S O methyl protons located in the central region of the host while the signal further upfield is from D M S O methyl protons sticking into the bowls (see Scheme 3.3). Both signals broaden and coalesce into a single broad resonance at -0.38 ppm (at 300 K , Figure 3.7a). Coalescence is estimated to occur at -250 K , 131 which corresponds to an activation energy of -10 .5 kcal/mole (Av -section for details on the calculation). 1032 Hz , see experimental | I I I I | | 1 I I I I I I I I | 1 I I I | I I I t I I I 1 1 | I I I I I t I I I | I I I I M I 1 I | 1 1 I 1 | 1 1 I I I I I 1 I | I I 7 6 5 4 3 2 1 0 -1 -2 (ppm) Figure 3.7 ' H N M R spectra (400 M H z , C D 2 C 1 2 ) o f 56«(NFP»DMSO) in C D 2 C 1 2 at various temperatures, (a) 300 K , (b) 275 K , (c) 250 K , (d) 225 K , (e) 200 K . * = D M S O methyl protons, n = 56«NFP impurity. The dynamic process observed in the spectra shown in Figure 3.7 is interpreted to be due to either (I) the migration of D M S O from one bowl to the next, or (II) a 180° rotation about the C 2 -axis of the D M S O guest bound to a bowl subunit (Scheme 3.3). The analogous dynamic process (Scheme 3.3) for 56«(DMSO)3 is faster than in 56«(NFP»DMSO). In C D 2 C 1 2 , the single D M S O methyl resonance o f 5 6 « ( D M S O ) 3 only broadens into the baseline upon cooling to 185 K . Temperatures well below 185 K are required 132 to freeze out the different D M S O methyl environments for each guest. Thus, an upper limit of~7.7 kcal/mole on the activation energy is estimated for 56«(DMSO)3. 3 7 The observed interconversion process for D M S O is more sterically hindered with an accompanying N F P guest, compared to with two accompanying D M S O s by -2.8 kcal/mole. Energy barriers for the rotation o f D M S O about the C2-axes of carceplex 10a«DMSO and complex 12b»DMSO have been reported to be 12.7 and 12.6 kcal/mole, respectively. 3 8 Guest mobility in 56«(NFP»DMSO) and 5 6 » ( D M S O ) 3 is greater than in the smaller, more confining containers, 10a»DMSO and 12b»DMSO. D M S O mobility in carceplexes with be revisited again in Chapter 5. Scheme 3.3 D M S O mobility in 56«(NFP»DMSO). 133 3.3.4.2 Carceplexes 56»(NFP«Guests) (Guest = Aryl Ketones) Carceplexes 56»(NFP«guest) were prepared at 70-80 °C in N F P doped with aryl ketones (acetophenone, propiophenone, 4'-ethylacetophenone), while little or no carceplex products formed in reactions at room temperatures (not even 56«NFP or 56«(NFP)2). While homogeneous samples o f 56»(NFP»4'-ethylacetophenone) could be obtained, trimer carceplexes 56»(NFP»acetophenone), 56»(NFP»propiophenone) were isolated and characterized as mixtures with 56»NFP. Carceplex 56»(NFP»acetophenone) was prepared free o f significant amounts o f 56»NFP from a reaction in NFP:acetophenone (8:2), however, 56«(NFP»acetophenone) under these conditions forms along with 56«(acetophenone)2 as an inseparable mixture (~ 1.3:1, respectively). Attempts to prepare 56»(NFP«propiophenone) in N F P at higher guest concentrations were unsuccessful. 3.4 Templation Studies 3.4.1 Single-Molecule Guest Competitions Equipped with a suitable array of guests for 56«guests that differ in structure and molecularity, competition experiments were conducted to quantify the relative templating abilities. To illustrate how TRs are calculated for competing templates of different molecularities, we w i l l first examine the simplest case: TRs for single-molecule versus single-molecule competitions (77?ns). The formation of carceplexes (and hemicarceplexes) involves several stepwise reactions proceeding through many different intermediates that resemble the 134 final structure o f the host. Since template effects are only important during the guest determining step (GDS), only the GDS itself and the formation of the intermediate involved need to be considered. To keep things simple, a single intermediate host (H*) is used that undergoes an intramolecular cyclization reaction during the GDS: K k H* + G ^ H*« G — 56« G where, K = K^1 (3.1) [ H ] [ G ] and k is the rate constant for the GDS. The rate o f formation of 5 6 » G from complex H * » G is: Rate = AfH*«G] (3.2) Substituting equation 3.1 into 3.2 gives: Rate = A3fe[H*][G] (3.3) For two competing guests, G A and G B , the equilibrium and rate constants are K\, k\ and K2, k2, respectively. Integration gives the product ratio, which can be expressed as: [ S 6 « G A ] _ * A f c A [ G A ] (3.4) [ 5 6 . G B ] KBkE[GB] Hence, 77? n (single-molecule versus single-molecule templates) are: T R [ 5 ^ G J [ G J = £ A ( 3 5 ) [ S 6 . G . I G J KBkB TR\\s are simply calculated by first measuring the carceplex product ratios (56»GA:56»GB) , and adjusting according to the initial concentrations o f each G A and G B . 77? I is should be constant at any concentration of either of the two competing guests. TRns for the formation of 56»guest are listed in Table 3.3 for seven single-molecule templates at ambient temperature and at 70 °C. Competitions (see experimental section 3.8.5) for 56«guest were done similarly to competitions for hemicarceplex 25»guest (Chapter 2): trimer 135 43 was reacted with 2,4,6-tris(bromomethyl)mesitylene in the presence of two or more competing guests (at concentrations ranging from -0.05 (15 equiv. per host) to 5 mole % o f the solvent), K2CO3, and K I in N F P solvent. To simplify the measurements, a large excess o f each guest (G) relative to the starting material was always added (to ensure that the steady state approximation applies). Size and shape appear to be the governing factors in determining the templating abilities. The more effective single-molecule templates in Table 3.3 have like symmetry (C3) to the host. Tris-acetylene 116 is the best template, which is not surprising as C P K models suggest the most optimal van der Waals contacts can form between 116 and the trimer interior o f 56. Smaller guests (i.e., 118,120,121, and 123) have poorer templating abilities, and l ikely have less van der Waals contacts with the forming host. Trimethyl ester 117 is only slightly better than the guests below it in Table 3.3, despite its large size. The potential for 117 to form a greater van der Waals contact area with the host is l ikely offset by unfavorable repulsive interactions due to the tight fit in the host cavity. Table 3.3 Template ratios (77?n, unitless) in the formation of 56«guest from single-molecule templates. Template Ratio Guest T R n r.t. 70 °C 1,3,5-Triethynylbenzene (116) 860 11200(440) Trimethyl-1,3,5-benzene tricarboxylate (117) 55 280 (11) 1,3,5-Trimethoxybenzene (118) 25 260 (10) 1,3,5-Triethylbenzene (119) 23 -Hexanophenone(123) 5 50 (2) Butyrophenone (121) 1 25 1-Formylpiperidine (120) - 1 'Bracketed values are relative to butyrophenone. r.t. = room temperature. 136 Temperature does not have a huge effect on the relative templating abilities o f the single-molecule guests (Table 3.3). Only a two-fold drop in the range o f selectivity is observed for guests 116,118, and 123 (relative to 121) upon increasing the reaction temperature -45 °C, while a five-fold drop is observed for 117. 3.4.2 Template Ratios for 56»(NFP»Guest) 3.4.2.1 Competitions Between Two-Molecule Pairs in the Formation of 56»(NFP«Guest) Competition experiments were also conducted between acetophenone, propiophenone, and 4'-ethylacetophenone in N F P to determine the relative templating abilities o f the two-molecule templates, NFP«guests . Template ratios for two-molecule versus two-molecule templates, TR22,39 calculated similarly to 77?ns, are shown in Table 3.4 for the formation of carceplexes 56»(NFP»aryl ketone). Size and shape complementarity between N F P paired with aryl ketones appears to be the most important factor in determining the relative templating abilities. The best pair, N F P and 4'-ethylacetophenone combine to provide the best template, seven-fold over the next best guest acetophenone. The additional para-ethyl substituent appears to enable 4'-ethylacetophenone to interact strongly with the host interior, as it can easily span between two bowl subunits (Figure 3.8). C P K models of this guest in the cavity suggest that it is held rigidly in place and can form favorable van der Waals contacts between two different bowls. Further stabilization may occur between overlapping p-orbitals between the C = 0 orbitals o f N F P and the arene o f 4'-ethylacetophenone (Figure 3.8). Addit ion o f a methylene unit to the alkyl group o f acetophenone, as in propiophenone, decreases templating ability two-fold. Whi le the 137 slight increase in guest size does not appear to increase favorable van der Waals interactions between the aryl ketone and the forming host, it may increase steric repulsions with N F P (Figure 3.8). Overall, for two-molecule guests, templating ability is better for guests that maximize favorable host-guest and guest-guest interactions, while minimizing unfavorable host-guest and guest-guest interactions. Table 3.4 Template ratios (7R 2 2 ) for 56«(NFP«guest). Guest (TR22) 4'-ethylacetophenone 15 acetophenone 2 propiophenone 1 56«(NFP«acetophenone) 56»(NFP«propiophenone) 56«(NFP»4'-ethyl-acetophenone) Figure 3.8 M M 2 minimized structures of trimer carceplexes 56«(NFP»aryl ketone). To simplify calculations, the pendant groups (R = C H 2 C H 2 C 6 H 5 , see Figure 3.2) o f the [4]cavitand subunits were replaced with Hs. For clarity, the mesityl caps have been removed to show the geometries of each entrapped guest inside the host. 138 3.4.2.2 Single-Molecule Templates Versus NFP»Guest A problem with comparing the templating abilities for guests o f different molecularity is that the TR values w i l l have units (e.g., M for 77?i2). Assuming the same host intermediate is involved in the GDS for both single- and two-molecule templates, KA H* + G A = H*«G A H % G A + G B ^ H* . (G A .G B ) H%(G A .G B ) ^ 56 .(G A .G B) Analogous to equation 3.3, the rate of formation of 5 6 » ( G A * G B ) is: Rate = /?&AB[H*][GA][GB] (3-6) where ft is the overall binding constant, J3= KAKAB- Dividing equation 3.3 by 3.6, followed by algebraic manipulation and integration gives the template ratio for the competition between a single-molecule versus a two-molecule template, TR\2 (in M ) . Note that G = G A , G B , or G C ; K = KA, KQ, or KC; and k = kA, kB, or kc. T R ^ [ S 6 » G ] [ G A ] [ G B ] = Kk 1 2 [ 5 6 . ( G A . G B ) ] [ G ] BkAB To compare the relative templating abilities o f the two-molecule guests with single-molecule guests, separate competitions were conducted between the guests in Table 3.4 and selected guests from Table 3.3 (NFP, 118 and 123) in N F P (Table 3.5). The relative TRns for 118 and 123 in Table 3.5 (bracketed values) agree with the corresponding TRus (70 °C) in Table 3.3. Similarly, TR2xs for NFP»4'-ethylacetophenone, NFP»acetophenone, NFP»propiophenone (Bracketed values) agree with the TR22s in Table 3.4. The better single-molecule templates (in Table 3.3) compete more effectively against the same two-molecule template, and vice versa. 139 Table 3.5 Template ratios (TRn (G/NFP»G B ) ) for single-molecules versus NFP«aryl ketone ( N F P « G B ) at 70 °C. G G B TRn (G/NFP«GB) (M) a 7*i?2ib (M" 1 ) 118 4'-ethylacetophenone 123 4'-ethylacetophenone N F P 4'-ethylacetophenone N F P acetophenone 2.4 (280) 0.32 (40) 0.0087 (1) 0.059 0.42 3.1 N F P propiophenone 0.15 110 (16) 17 (2) 7 (1) a Normalized values relative to TRn (NFP/NFP«4'-ethylacetophenone) are bracketed. b77?2i is two-molecule versus single-molecule templates (77?2i = l/77?i2)- Normalized values relative to 77?2i (NFP»propiophenone/NFP) are bracketed. Note that in equation 3.7, G A = N F P . 77?i2s are essentially equilibrium effective molarities. 4 0 We interpret the chemical meaning o f 77? 12s as the concentration of G B (or GA) needed for the multiple-molecule template, G A * G B , to equal the templating ability o f a single-molecule template, G , at a specific concentration of G A (or G B ) . For example, the two-molecule template NFP»4'-ethylacetophenone competes very well with the single-molecule template N F P ; only 8.7 m M o f 4'-ethylacetophenone is required (with a constant concentration o f N F P ) to equal the templating ability o f N F P . Larger TRn values correspond to greater templating power for the lower molecularity template ( G ) , while large 77?2i values correspond to greater templating power for the higher molecularity templates. 3.4.3 Other TR22s, TR23S, and TR33s Several multiple-guest versus multiple-guest 77?s (i.e., TR22, TR23, and ^ 33) were measured from separate mixed solvent reactions (Table 3.6). These values were calculated in the 140 same manner as other TRs: the product ratios were measured by N M R spectroscopy and multiplied by a correction factor that accounts for the relative initial concentrations of each guest. Qualitatively, the relative templating abilities of the homogeneous clusters o f entrapped guests at room temperature increases in the order of: ( D M A ) 2 = ( N M P ) 2 > ( D M S O ) 3 > ( D M F ) 3 . Note that in Table 3.6, the concentration o f D M F required for ( D M F ) 3 to compete effectively with ( D M A ) 2 , at a specific concentration o f D M A , is physically unattainable (neat D M F ~13 M ) . On the other hand, D M S O competes well at much lower concentrations. The relative templating abilities between multiple D M S O molecules relative to multiple D M F molecules w i l l be discussed more in the next section. Table 3.6 TRs (TR22, TR32, TR-^) for multiple-molecule templates in the formation of 56»guests. Solvent Mixture Guest A Guest B TR21 TR23/M TR33 (v/v) ( G A ) ( G B ) ( G A / G B ) ( G A / G B ) ( G A / G B ) D M S O : D M F ( l : 9 ) ( D M S O ) 3 ( D M F ) 3 _ _ 480 D M A : D M S O (9:1) ( D M A ) 2 ( D M S O ) 3 - 0.17 -D M A : D M S O (8:2) ( D M A ) 2 ( D M S O ) 3 - 0.17 -D M A : D M F (3:7) ( D M A ) 2 ( D M F ) 3 - 11 -D M A : D M F (1:1) ( D M A ) 2 ( D M F ) 3 - 15 -D M A : N M P ( 1 : 1 ) ( D M A ) 2 ( N M P ) 2 1.1 - -3.4.4 TRus (as well as TRns and TR23S) 3.4.4.1 Competitions Between Tris-Acetylene 116, (DMSO)3, and (DMF)3 Experiments were performed to see how well three-molecule guests fair against single-molecule guests in their relative templating abilities. 77?i3S measured between tris-acetylene 116, 141 ( D M S O ) 3 , and ( D M F ) 3 in different solvent mixtures are shown in Table 3.7. Tris-acetylene 116 was the only viable single-molecule guest (amongst those in Chart 3.1) because it does not pair with D M S O or D M F to form 56»(116»guest), and the guest signals (i.e., acetylene protons o f 56*116) do not overlap with any of the guest signals for 56»[(DMSO) J C»(DMF) ) ,] (x, y = 0-3). Table 3.7 Template ratios (77?13) for ( D M S O ) 3 and ( D M F ) 3 against tris-acetylene 116. Guest A ( G A ) Guest B ( G B ) Solvent TRn ( G A / G B ) TRnxl03 ( G B / G A ) 116 a ( D M F ) 3 D M S O : D M F ( l : 9 ) 490000 M2 0.02 M " 2 116 a ( D M S O ) 3 D M S O : D M F ( l : 9 ) 1100 M2 0.9 M " 2 116 b ( D M S O ) 3 D M S O : N F P ( 1 : 1 0 ) 43 M2 140 M " 2 a[116] = 28.2 m M . b[116] = 6.54 m M 77?i3s are interpreted as the concentration of a three-molecule template, squared, required to equal the templating ability of a single-molecule template. Since neat D M S O and D M F are only ~13 M , only ( D M S O ) 3 can ever physically equal the templating ability o f tris-acetylene 116 (6.54 m M ) under the conditions examined. The concentrations of D M S O and D M F ( » 1 3 M ) needed for ( D M S O ) 3 and ( D M F ) 3 to equal the templating ability o f 116 in D M S O : D M F (1:9) are not physically attainable. 77? i 3 for ( D M S O ) 3 is 27 times greater in D M F than in N F P at similar concentrations of D M S O , which demonstrates a significant solvent effect. Free tris-acetylene may be solvated better in N F P (less polar) than in D M F (more polar), and thus may be less l ikely to be bound to the intermediate host prior to the GDS in the formation o f 56«116. A s expected from the 77?3 3 shown in Table 3.6 (480), ( D M S O ) 3 competes more effectively with tris-acetylene 116 than ( D M F ) 3 . A8 values for the guests in 5 6 » ( D M S O ) 3 and 56«(DMF) 3 (see Table 3.16, experimental section), suggest that the methyl groups o f each guest 142 are oriented into the bowls o f the host. The size and shape o f D M S O may allow for a more complementary fit between its methyls and the bowls than the methyls o f D M F . C H - p i interactions between the guest methyls and pi-electron rich host arenes l ikely stabilize these orientations, which are likely stronger with more polarized C - H s o f D M S O . Also , steric repulsions between the other A^-methyl and the formyl group of D M F with the host may significantly offset any stabilization gained from C H - p i interactions. The greater templating ability o f D M S O (than D M F ) in the formation of 56»guests is consistent with other two 1" 3 and six-bowl (Chapter 5) carceplexes. 56.(DMF)3 56.(DMSO) 3 3.4.4.2 Temperature Effects on TRX3, TRU, and TR23 To investigate the effect of temperature on TRs between guests o f different molecularities, 77?s were measured for the competing guests ( D M S O ) 3 , D M S O » N F P and 116 at 30,40, and 50 °C in N F P (Table 3.8). For these experiments, the concentrations o f D M S O (607 m M ) and 116 (8.91 m M ) were chosen so that each of the three carceplexes, 56»(DMSO)3, 56«(DMSO«NFP) and 56*116 formed in comparable amounts. A s expected, 77?i3, 77?23 and 77? 12 each increase with temperature (at least doubling over a 20 °C range), and the greatest increase is observed for 77?o, which has the greatest differential in molecularity. Simplistically, 143 the overall entropic cost of organizing templates of higher molecularities should be larger than for templates o f lower molecularities, which is consistent with the values in Table 3.8. Table 3.8 Dependence of the template ratios (77?i3, TRi2 and TR23) on temperature (7). T (°C) a TRl3 (M2) TRn (M) 7*23 (M) 30 51 2300 0.022 40 97 3100 0.032 50 213 3900 0.058 "Temperature fluctuation: ± 1.0 °C over the course of the reaction. [NFP] = 8.62 M , [ D M S O ] = 607 m M , [116] = 8.91 m M . Plots of ln(77?) against 1/T (see experimental section) for each 7V?i3, TRX2, and TR23 gave enthalpic (AAPP+AAH*) and entropic (AA,S°+AA,y*) values from the slope and intercepts, respectively (Table 3.9). For each set of competitions, both enthalpic and entropic components are favorable. Note that the individual thermodynamic/kinetic components cannot be separated from the values listed in Table 3.9. However, the kinetic components (AA//* , AAS*, AAG*) are believed to provide little contribution to the overall values (remember that the overall values here are differences). This comes in light of the belief that template ratios are due to ground state effects: the rate o f the GDS is determined by the relative energies o f the complexes formed between the host intermediate and the competing guests (AAH°, AAS°, AAG°). This is supported by the good correlations between the binding abilities with appropriate GDS transition state models and the template ratios in previously reported carceplex studies, 2 9 , 4 1 which w i l l be addressed for 56»guests in the next section (3.5) of this chapter. 144 Table 3.9 Thermodynamic/kinetic values for template ratios for one vs. three, one vs. two, and two vs. three-molecule templates. AMP+AA/y* AAS°+AASt T(AAS°+AAS*) AAG°+AAGt (kcal/mole) (cal/moleK) (kcal/mol)d (kcal/mole) TRn* TRnb TR23C -14.1 -5.3 -9.5 54.3 32.7 23.7 16.2 9.7 7.1 -30.3 -15.0 -16.6 a 56»116/56«(DMSO) 3 . b 56»116/56«(NFP»DMSO). c 56»(NFP«DMSO)/56«(DMSO) 3 . dTemperafure = 298 K . This section (3.4) demonstrated the following with regard to the templated formation of carceplex 56«guests. Overall, the best single-molecule templates in the formation of 56»guests maximize noncovalent interactions with the interior of the forming host, such as van der Waals, pi-stacking, and C H - p i interactions. Tris-acetylene 116 was the best template and the most complementary to the trimer cavity of C P K models o f 56»guests. For multiple-molecules, the best templates maximize favorable and minimize repulsive noncovalent interactions between host-guest and guest-guest. Finally, lower molecularity guests are better templates than higher molecularity guests, which is most l ikely due to the lower entropic cost of binding one component as opposed to several during the GDS. This also is evident in the thermodynamically more stable complexes that form between structurally related hosts to 56»guest and single-molecule guests, than with multiple-molecule guests, which w i l l be discussed in sections 3.5 and 3.6. 145 3.5 Complexation of Single-Molecule Guests 3.5.1 Trimer Complexes 42«Guest and 43«Guest 3.5.1.1 Potential Transition State Models for the GDS In previous templation studies (Chapter 1), suitable transition state models were prepared whose relative thermodynamic stabilities of complexes with various guests correlated nicely with the template ratios for the corresponding carceplexes (derived from similar subunits). 2 ' 4 1 For this work, a suitable transition state model could be potentially used to screen guests and determine their relative templating abilities in the formation of 56«guests, while reducing the total consumption of starting material 43. Unfortunately, this idea did not work out as wel l as initially planned (vide infra). Nonetheless, several new large organic hosts were discovered in the process that form thermodynamically stable complexes with various neutral organic guests. These compounds also shed some light on the transition state of the GDS in the formation o f 56»guests. C P K models suggest that trimers 42 and 43 have fairly rigid "barrel-shaped" cavities that have two large openings at the top and bottom through which potential guests can readily pass. The presence o f the benzyl groups lining the upper and lower rims of 43 may offer some steric resistance to guest entrance/departure. Although models of 42 and 43 are significantly more rigid than cyclic tetramers,7 pentamers,4 2 and hexamers 1 3 ' 4 3 o f [4]cavitand arrays, there is still a significant amount of flexibility with respect to 56»guest. Thus, the greater flexibility o f 42 and 43 may offer a more complementary fit between their interiors and a more diverse range o f guests. 146 The ideal host for screening potential guests for the formation of 56«guest is one that is easy to synthesize and forms strong complexes with relevant guests to yield binding affinities that correlate with the 77?s. In this study, hexa-benzyl trimer 42 was found to fit this description the best. Binding studies between 42 (and 43) with various single-molecule guests w i l l be discussed in the next few sections, followed by the examination of other related hosts (trimer cavitands), which better resemble the proposed transition state 129 (Scheme 3.4) than 42/43. Scheme 3.4 Speculated mechanism and intermediates in the formation o f 56«guest. 42 | GDS? 56*guest 3.5.1.2 Guests Tha t D i d Not F o r m Complexes wi th 42 Suitable templates from Table 3.3, and other molecules with similar structural attributes (i.e., functionalities, C3-symmetry) to those in Table 3.3 were screened for complex formation with 42. Seven o f these guests (not in Table 3.3) showed no sign o f complex formation (Chart 3.2). Compounds 130 and 135 may not bind because they are too polar for the host cavity. C P K models o f benzophenone and tris-acetyl 131 suggest poor complementarity with the cavity o f 42. Repulsive interactions between the electronegative carbonyl oxygens o f 131 and the arenes of 147 the bowls also probably contribute to the lack of binding of this guest. Compounds 132,133 and 134 may be too small to interact strongly with the host. Therefore, no attempts were made to form 56*guests in the presence of these templates. Chart 3.2 Guests that did not form complexes with 42 in nitrobenzene-Js at 300 K . 130 131 132 133 HOOC O 134 135 136 3.5.1.3 The Stabilities of Trimer Complexes 42»Guest and 43*Guest When the effective single-molecule templates: tris-acetylene 116, tris-ester 117, tris-ether 118, tris-ethyl 119, and bis-ester 137 (Table 3.3) were added to solutions containing hexa-benzyl trimer 42 at 300 K , thermodynamically stable complexes 42»guest formed in which guest exchange is slow on the ' H N M R timescale. For example, addition o f the best template, tris-acetylene 116 (0.5 equiv.), to 42 in nitrobenzene-^ (Figure 3.9a) gave a ~1:1 mixture o f free and bound host (Figure 3.9b). When equal concentrations of host 42 and guest 116 are present, only complex 42*116 is observed (Figure 3.9c). The formation of complex 42*116 was also observed in several other solvents (Table 3.10). 148 0 Hi O a Q . C H 3 H 3 137 To compare the effect o f benzyls versus hydroxyl groups on the upper and lower rims of the A,C-tr imer cavity, binding o f 116 to hexa-hydroxyl trimer 43 was also studied. These studies were limited to only two solvents due to the low solubility o f 43 (Table 3.10). In contrast to 42, 43 appears to aggregate in nitrobenzene-ds and gives rise to a broad ' H N M R spectrum that is difficult to interpret (Figure 3.9d). 4 4 Upon addition of tris-acetylene 116, the host signals (i.e., H 0 , H a c , H m and Hj) sharpen considerably, and a broad signal for the bound acetylene protons is observed upfield at -0.13 ppm for complex 43*116 (Figure 3.9e). This signal is broad compared to the sharp signal at -0.32 ppm for 42*116 (Figures 3.9c and b, respectively). Although guest exchange is slow on the N M R timescale for both 43*116 and 42*116, it is clearly faster for complex 43*116 than for 42*116. The bulky benzyl moieties in 42*116 may either sterically hinder guest exchange or form favorable interactions with the guest. The larger A8 for the acetylene protons in 43*116 may also suggest that the benzyls cause a slight contraction o f the trimer cavity, and/or provide additional shielding to the guest, acting as caps. 149 1 1 1 1 1 1 1 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 l l I 7 6 5 4 3 2 1 0 (ppm) Figure 3.9 lH N M R (nitrobenzene-^, 300 K ) spectra o f A,C-trimers 42 and 43 and complexes with tris-acetylene 116. (a) 42, (b) 42 plus 116 (-0.5 equiv.), (c) 42 plus 116 (-1 equiv.), (d) 43, (e) 43 plus 116 (-11 equiv.). * = acetylene protons o f bound 116. R — CH^CH^CgHs Figure 3.10 A,C-Trimers 42 and 43 Stability constants (Kss) for both complexes 42*116 and 43*116 were measured in various deuterated N M R solvents (Table 3.10). Kss were calculated from the equilibrium: 150 Host + 116 Host*116 where, ^ H o s t ^ l l 6 ] _ [Hostomu [Host*116], [Hostjfree, and [116]^ were calculated from the relative integration in *H N M R spectra, knowing [Host] totai and [116]totai-Clearly, the most stable complex is 42*116 in nitrobenzene-ds (Table 3.10). The exceptional stability measured for 42*116 in nitrobenzene-rfs is l ikely due to the poor ability of nitrobenzene-c?5 at solvating the trimer cavity and/or the free guest 116. In contrast, 42*116 is the least stable in CDCI3, a solvent known to interact well with the cavity o f [4]cavitands. 1 0 ' 4 5 Affinity o f CDCI3 (CHCI3) was also evident in the formation of carceplexes 56*[(CHC1 3) 2*DMF] and 56*[CHC1 3*(DMF) 2] from the solvent mixture C H C 1 3 : D M F (1:1). For hexa-hydroxyl complex 43*116, the opposite trend in Kss is observed (Table 3.10). Complex 43*116 is nearly five times less stable in nitrobenzene-^ than in CDCl3:CD30D (19: l ) . 4 6 In the latter solvent, CD3OD l ikely helps stabilize complex 43*116 by solvating the polar hydroxyl groups. For complex 42*116 in nitrobenzene-^, the benzyl groups l ikely provide additional stability in the form of van der Waals interactions (i.e., C H - p i and pi-stacking) with the resident guest 116, which are obviously not present in 43*116. Complex 43*116 may be less stable than 42*116 in nitrobenzene-ck because nitrobenzene-^ solvates the cavity o f hexa-hydroxyl 43 better than hexa-benzyl 42. Possible hydrogen-bonding between the nitro group oxygens of nitrobenzene-ck and the hydroxyls of 43 may make desolvation more difficult than for 42 in the same solvent. 151 Table 3.10 Stability constants (Ks) for complexes 42»116 and 43»116 in various deuterated solvents. 3 Solvent 42»116 43*116 (R = OBn) (R = OH) nitrobenzene-^ toluene-^8 35000* 41 C D 2 C 1 2 C 6 D 6 C D C 1 3 : C D 3 0 D ( 1 9 : 1 ) C D C 1 3 1260 180 120 15 13 190 "Error is estimated to be ± 10 %. *Could not be measured directly. Estimated based on KKi for 42-117. Since hexa-benzyl trimer 42 formed the most stable complex with 116 in nitrobenzene-^ at 300 K , Kss were also measured for 42«guest under these conditions for other guests (Table 3.11). Only a modest correlation in Figure 3.11 (r 2 = 0.835) was observed between the Kss for 42«guest (in nitrobenzene-J5) and 77?ns (in N F P ) in the formation of 56»guest for the guests 116,117,118, and 119, in comparison to correlations for other reported systems (r 2 > o.92). 2 9 ' 3 8 ' 4 1 ' 4 5 The main deviation in the plot of l o g ( i Q versus log(77?i 1) appears for tris-ethyl 119, whose 77? n (23) is lower than predicted (-70) from the Ks o f complex 42*119 (Figure 3.11). 4 7 Therefore, in order to find a better correlation, binding studies between guests 116-119 and 137 and other trimer derivative hosts were examined that more closely resemble intermediates further along in the reaction forming 56»guest. 152 Table 3.11 Stability constants (Kss) for 42«guest (nitrobenzene-^, 300 K ) . Guest TRu l,3,5-tris(ethynyl)benzene (116) 35000 860 1,3-dimethylpthalate (137) 1080 -1,3,5-triethylbenzene (119) 460 23 trimethyl-1,3,5-benzenetricarboxylate (117) 100 55 1,3,5-trimethoxybenzene (118) 45 25 *TR\ i determined at room temperature in N F P solvent. 3.00 H 2.00 -t 1 — i < 1 1 . 1 1 1 1 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 ln(TR) Figure 3.11 Plot o f \n(Ks) against ln(77?n). ^ s are stability constants determined for complexes 42»guest (nitrobenzene-^) and TR\\ are template ratios for carceplex 56»guest (formed in N F P ) . r 2 = 0.835. 153 3.5.2 T r i m e r Cav i t and Complexes 114»Guest , 125«Guest , and 126*Guest 3.5.2.1 Synthesis of T r i m e r Cavitands 114,125, and 126 Trimer cavitand 114 was initially isolated by Chopra as a byproduct in the formation o f 56«(DMF )3 . 1 0 For this thesis, attempts were made to optimize the formation o f 114, since trimer cavitand 114 is not easily separated from trimer 43. Therefore, conditions were employed that would completely consume all o f trimer 43. Slow (24 h) addition o f 2,4,6-tris(bromomethyl)-mesitylene (2 equiv.) to a mixture of hexa-hydroxyl trimer 43, K I , and K2CO3 in D M F gave 114 (20 % ) , 56«(DMF)3, and other byproducts. Trimer cavitand 114 was also synthesized (25 % yield) free o f 43 from 56«(DMF)3 via the acid-catalyzed hydrolysis o f the mesityl ether caps with T F A in CH2CI2 (see experimental). 4 8 Attempts were also made to purify cavitand 114 from trimer 43 by derivatization to the corresponding benzyl and cyanobenzyl ethers, 125 and 126, respectively. Mixtures containing 114 and 43 were benzylated and cyanobenzylated using similar procedures in the presence of K2CO3 and benzyl bromide and 4-bromomethyltoluonitrile i n D M F at 60 °C, respectively. Tris-cyanobenzylation was the better method as tris-cyanobenzyl 126 was isolated in high purity (by column chromatography), while tris-benzyl 125 could not be separated from hexa-benzyl 42. Pure tris-benzyl 125 was later prepared from pure tris-hydroxyl 114. ,OR' R'O, 154 C P K models o f all three trimer cavitands have similar "cookie-jar" shapes, which are structurally more rigid than trimers 42 and 43, as the interconnected bowl subunits cannot twist with respect to one another as easily when a mesityl cap is in place. In addition, models also suggest that trimer cavitand cavities must be completely empty (desolvated) before a large single-molecule guest can enter, as the single portal is not wide enough to allow the simultaneous passage of both guest and solvent. Solution state conformational features and complexation properties of cavitands 114,125, and 126 w i l l be examined in some detail in the next two sections. 3.5.2.2 Structure, Binding, and Dynamic Features of Trimer and Trimer Cavitands and Their Complexes Comparison of the ' H N M R spectra of trimer cavitands 114,125, and 126 in various organic solvents revealed some interesting structural behavior. This behavior was evident in the observed chemical shift changes for the diastereotopic H a c i/Hac2 protons (Figure 3.12) o f the interbowl acetal moieties, which are the most flexible portions o f these molecules, according to C P K models. Four main conformations for the trimer cavitands are proposed (Scheme 3.5): H a c i / H a C 2 may both be oriented inside (C) or outside (A) the trimer cavity, or H a c i is in and H a C 2 is outside the cavity (B), or vice versa (D). Conformations A -D, and intermediates between any pair all interconvert rapidly on the N M R timescale via rotations of the interbowl acetal A r - O bonds, and thus only the equilibrium average is observed. 2 D N O E S Y data o f 114,125, and 126 suggest that conformer D is not present in significant amounts, based on the lack o f N O E s observed between H a c i / H a c 2 and protons of the mesityl cap. Trimers 42 and 43 can also adopt conformations A -D, but the conformations are less distinguishable by ' H N M R because H a c s are not longer diastereotopic, so there are no A5s. 155 Scheme 3.5 Conformational equilibria for trimer cavitand derivatives. For trimer cavitands 125 and 126, the view is from the top o f the cavity. Conformations like A - C also apply for trimers 42 and 43. Conformers A - C were identified based on ' H N M R chemical shift data on H a c i / H a C 2 obtained from solutions o f 114,125, and 126 and their complexes with 116 in various solvents and at various temperatures (e.g., Table 3.12). In principle, conformations similar to A should be recognized by a relatively small A8 between H a c i and H a C 2 , which are both shifted downfield, as they are both outside the shielding trimer cavity. Conformations similar to B should be 156 recognized by a relatively large A8 ( H a c i / H a C 2 ) , where H a c i should be significantly more shielded than H a c 2, since H a c i is oriented further into the trimer cavity than H a C 2. Finally, for conformations like C, A8 ( H a c i / H a c 2 ) is expected to be smaller than for B, and both H a c i and H a C 2 should be shifted further upfield relative to H a c i and H a c 2 in A and B. Although H a C 2 is oriented further into the shielding trimer cavity in C than B, it is difficult to distinguish the two based on chemical shift and A5 values alone. Therefore, B and C are taken as a combination (i.e., B/C). Table 3.12 ! H N M R chemical shift data for H a c protons of A,C-trimer derivatives and their complexes with 116 at 300 K . Host Solvent 5 H a c (free) AS H a c l / H a c 2 (free) 5 H a c ^ (bound) AS H a c i / H a c 2 (bound)* AS (free-bound) 42 C D C 1 3 5.60 0.00 5.75 0.00 -0.15 C D 2 C 1 2 5.58 0.00 5.73 0.00 -0.15 C 6 D 6 5.33 0.00 5.90 0.00 -0.57 toluene-i/g 5.27 0.00 5.87 0.00 -0.67 nitrobenzene-^ 5.58 0.00 6.15 0.00 -0.57 43 C D C 1 3 : C D 3 0 D ( 1 9 : 1 ) 5.72 0.00 5.84 0.00 -0.12 114 C D C 1 3 5.85/5.71 0.14 - - -nitrobenzene-^ 6.02/5.94 0.08 6.22/6.15 0.07 -0.28/-0.21 3 125 C D C 1 3 5.81/5.68 0.13 - - -nitrobenzene-^ 5.81/5.02 0.79 6.17/6.05 0.12 -1.15/-1.03 3 toluene-^ 5.55/4.54 1.01 - - -126 C D C 1 3 5.82/5.72 0.10 - - -nitrobenzene-*^ 5.85/5.09 0.76 6.18/6.10 0.08 -1.09/-1.01 3 toluene-^ 5.60/4.74 0.86 *Bound refers to complexes between the respective hosts and tris-acetylene 116. a Since H a c i / H a C 2 could not be unambiguously assigned, the two AS values are upper and lower limits. A l l o f the hosts in Table 3.12 appear to favor conformation A in C D C 1 3 (or C D C l 3 : C D 3 O D ) based on the small AS values and/or downfield shifted 8 values observed for H a c i / H a C 2 (Table 3.12). Tris-hydroxyl 114 also favors A in nitrobenzene-^5 (Figure 3.13a). In contrast, tris-benzyl 125 and tris-cyanobenzyl 126 both appear to favor B/C in less competitive 157 binding solvents such as toluene-fife or nitrobenzene-fife (e.g., Figures 3.13b and c) based on the larger A8 values observed between H a c i / H a C 2 in these solvents than in CDCI3. Hexa-benzyl trimer 42 also l ikely favors conformations B /C in toluene-fife, C6D 6 , and nitrobenzene-fife. Figure 3.12 Trimer cavitands 114,125 and 126. For 125 and 126 in nitrobenzene-fife or toluene-fife, complexation o f tris-acetylene guest 116 shifts the equilibrium towards conformation A (Scheme 3.5 and Figure 3.13d). In contrast, upon complexation o f tris-acetylene 116 to tris-hydroxyl 114, both H a c i / H a C 2 only shift slightly downfield and A5 (H a c i /H a C 2) does not change significantly relative to unbound 114. Hexa-benzyl trimer 42 and hexa-hydroxyl 43 both also appear to shift more towards conformation A with binding of tris-acetylene 116 (Table 3.12). The more negative A8 (free-bound) values for H a c in complex 42*116 compared to 43*116 also suggests that unbound 42 favors conformations B /C more than unbound 43 does. CDCI3 also l ikely binds like 116 to hosts 125 and 126, causing conformation A to be favored. This suggestion is based on the low stability of complex 42*116 observed in CDCI3 relative to nitrobenzene-fife and toluene-fife (Table 3.10); C D C 1 3 appears to be more effective at solvating (competing) the trimer cavity of 42. Further support o f CDCI3 binding was also 158 obtained in the isolation of carceplexes 56«[(CHC1 3) 2»DMF] and 56»[CHC1 3«(DMF) 2] that were characterized as a mixture. 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 6.6 6.4 6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8 4.6 4.4 -0.40 -0.60 (ppm) Figure 3.13 ! H N M R (400 M H z , nitrobenzene-d5, 300 K ) spectra o f trimer cavitand derivatives, (a) tris-hydroxyl 114, (b) Tris-cyanobenzyl 126. (c) Tris-benzyl 125. (d) Complex 125»116. * = H a c = H a c i or HaC2- Note that in spectrum (b), the upfield H a c overlaps with one o f the H m signals. The conformational changes observed for benzylated trimer 42 and trimer cavitands 125 and 126 (cyano-benzyl) are explained as follows. In solvents like toluene-dg and nitrobenzene-Js the trimer cavity may be either poorly solvated or empty. Thus, the equilibrium in Scheme 3.5 is shifted towards B/C in order to reduce the void in the cavity: C P K models o f the benzylated species suggest that the trimer cavities contract slightly in conformations B/C, relative to A. Conformations B/C reduce the exposed van der Waals surface area with respect to A when the 159 cavities are empty or occupied with solvent (nitrobenzene-fife or toluene-fife). C H - p i and pi-stacking interactions between adjacent benzyl moieties may also be maximized as they are brought into closer proximity in B/C. The benzyl moieties may also be able to solvate the cavity (i.e., v ia self-complexation) more effectively in B/C than in A. Upon binding a highly complementary guest (i.e., 116) to the trimer cavity, the stabilizing interactions present in conformers B/C are sacrificed by more favorable interactions formed between the host and 116, in A. The trimer cavity expands to accommodate the larger guest 116, which causes H a C[/Hac2 to " f l ip" outside o f the cavity. In conformation A, steric interactions between H a c i / H a c 2 and guest 116 are also less than for B/C. Also , the entropy of binding may be favorable as several (more than one) less mobile benzyls that may be complexed to the interior o f cavity in B/C are forced out of the cavity to be replaced by a single molecule o f 116 bound in A. Exposed van der Waals surface area between the host and guest is less in A than between the host and its self-complexed benzyls in B/C. The equilibrium in Scheme 3.5 for 125 and 126 can also be shifted from B/C further towards A by increasing temperature. Variable temperature ! H N M R spectroscopy was also used to probe the dynamic nature o f the trimer cavitands 125/126 in nitrobenzene-fife (Figure 3.14). A t lower temperatures, the A8 between H a c i / H a c 2 is large, but becomes smaller with increasing temperatures. Both doublets for H a c i / H a C 2 also continuously shift downfield with increasing temperature, with the upfield o f the two shifting much further. A t 420 K (Figure 3.14b), the spectrum o f 125 in nitrobenzene-fife looks very similar to 114 at 300 K (Figure 3.13a). Conformation A appears to be favored over B/C at higher temperatures, which may be due to decomplexation of the self-solvating benzyls from the interior of the cavity. 160 (a) (b) (c) (d) (e) * * L A . -i 1 1 1 1 r ~i ' 1 ' r 6.2 6.0 5.8 5.6 5.4 5.2 (ppm) 5.0 4.8 4.6 4.4 Figure 3.14 * H N M R spectra of trimer cavitand 125. (a) 125 in C D C 1 3 at 300 K ; and in nitrobenzene-Js at (b) 420 K , (c) 385 K , (d) 350 K and (e) 300 K . * = interbowl acetal protons ( H a c s , H a d and H a c 2 ) . 3.5.2.3 The Stabilities of Trimer Cavitand Complexes 114*Guest, 125»Guest, and 126«Guest Slow guest exchange rates for trimer cavitand complexes 114«116,126»116,125»116, 125«117, 4 9 and 125«119 in nitrobenzene-c/5 relative to the * H N M R timescale allowed stability constants to be easily determined from integration of the corresponding ! H N M R spectra (Table 3.13). Spectra o f solutions containing 125 mixed and the above guests are shown in Figure 3.15. The characteristic upfield-shifted bound guest signals are clearly visible for each complex, except for 125*118, which is in fast exchange on the N M R timescale. Therefore, Ks for 125*118 was not measured. 161 I I I I I I I I I 1 I I I I I I I I I I I 1 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 1 I I ' I 1 I ' I 1 I' 'I I I ' I ' ' I ' I ' ' 1 ' ' ' ' I ' ' 1 ' I 1 1 1 > I 1 1 1 1 1 ' 1 1 1 I 1 1 1 1 7.5 6.5 5.5 4.5 3.5 2.5 1.5 0.5 -0.5 -1.5 (ppm) Figure 3.15 ' H N M R (500 M H z , nitrobenzene-^, 300 K ) spectra o f trimer cavitand 125 and complexes, (a) 125, (b) 125 and 118 (49 equiv.), (c) 125 and 119 (135 equiv.), (d) 125 and 117 (129 equiv.), (e) 125 and 137 (79 equiv.). * = guest proton signals. Comparing the three trimer cavitand hosts, complex stability increases in the order of 114«116 < 126«116 < 125»116. While there is little difference in stability between these three complexes, they are at least 30 times less stable than complex 42»116. Also , comparing relative 162 stabilities o f complexes 125*116,125*117, and 125*119, tris-benzyl 125 does not appear to be a better or worse transition state model than hexa-benzyl 42, only a poorer binding host. Table 3.13 Stability constants for trimer cavitand complexes and various guests (nitrobenzene-fife, 300 K ) . Host Guest Ks ( M ) tris-cyanobenzyl 126 116 tris-hydroxyl 114 116 tris-benzyl 125 116 tris-benzyl 125 119 tris-benzyl 125 117 250 200 1200 3 * 2-9 *Tris-ester was observed to form two different complexed species (Figure 3.15). 4 9 Therefore a range is reported based on Kss estimated for each of the two different species. To summarize single-guest binding studies, the following has been demonstrated. First, the stability o f complexes formed between neutral organic guests and derivatives of trimer 43 can be altered with the addition of different protecting groups to the hydroxyls on the portal rim(s). Benzyl groups provide slightly better complex stability than cyanobenzyl or hydroxyl groups under the conditions investigated. Second, binding affinity o f the trimer cavity is significantly decreased at least 30-fold when a mesityl cap is introduced to one side (i.e., as in 114,125, and 126). The large difference in the Kss between the complexes 42*116 and 125*116 (35000 M " 1 and 1200 M " 1 , respectively) indicates that capping one of the openings o f the trimer cavity has a significant effect on complex stability. This may arise from electrostatic repulsions between the electron-rich arene of the mesityl cap and the arene o f the guest in 125*116. Favorable noncovalent interactions (host-host and host-guest) between twice as many benzyl groups with the bound guest in 42*116 (six) compared to 125*116 (three) may also contribute. 163 3.5.2.4 Complexation/Decomplexation Rates for Complexes 42*116 and 125*116 Because o f the large discrepancy in the Kss between 42*116 and 125*116, complexation/ decomplexation rates were measured for 42*116 and 125*116 in order to gain further insight into the driving forces behind complexation. Decomplexation rate constants (k<i, in s"1) were first measured by I D E X S Y (nitrobenzene-c/5, 330 K ) for the observed exchange process (H = 42 or 125): K H + 116 = ^ FM16 The rate constant for complexation (kc, in M'V 1 ) , was then calculated from the stability constant Ks using the expression Ks = kjk^. Table 3.14 Stability (Ks) and rate (kc, kd) constants for complexes 42*116 and 125*116 (nitrobenzene-^, 330 K ) . Host Ks (M 1) kc (M'V1) AGC* (kcal/mole) Ms'1) AGd* (kcal/mole) 42 125 21100 1200 57000 2640 12.2 14.2 2.7 2.2 18.7 18.9 For complexes 42*116 and 125*116, decomplexation rate constants are similar, but complexation rates differ greatly. The 20-fold faster rate of complexation for 42*116 may result from the ability o f the incoming guest 116 to push solvates occupying the interior of 42 through the opposite opening o f the trimer cavity. This is not possible for 125*116, as guest 116 cannot enter the host cavity until it is completely desolvated. For decomplexation, the superior intrinsic binding in 42*116 appears to be offset by the single exit in 125*116. The enthalpy o f activation for decomplexation is l ikely high for 42*116, as favorable noncovalent host-guest interactions 164 are disrupted upon departure of the guest. The entropy of activation for decomplexation is l ikely less favorable for 125*116, since a single large guest evacuates 125, and is possibly replaced by many smaller solvent molecules. 3.5.2.5 meta-Xylyl Capped Trimer Cavitand 127 Bis-capped trimer cavitand intermediates (Figure 3.16) were also considered as potential transition state models, since they show greater structural resemblance to the speculated transition state, 129«guests, in the formation of carceplex 56«guests. meta-Xylyl 127 was synthesized from 114 via a reaction with a,a'-dibromo-/neta-xylene in the presence o f K2CO3, K I , and (the template) l,3,5-tris(ethynyl)benzene (116) in D M F at 60 °C. Host 127 is the least symmetric structure of all trimer/trimer cavitand derivatives, as it features only a single mirror plane. A s a result, there are two types o f bowls (X: Y = 1:2), each o f whose upper and lower halves are non-equivalent due to the different mesityl and bis-meto-xylyl caps. Hence, J H N M R spectra are more complicated than the other trimer cavitand hosts. For example, in CDCI3, there are: six signals (1:1:1:1:1:1) for each H 0 , H m , H i , and pendant group CH2S; seven H p singlets (1:1:2:2:2:2:2); two doublets (1H each) and a singlet ( 4 H ) for the interbowl acetals; two singlets for the cap methyls (2:1, 9 H total); and two coupled doublets ( 4 H ) , and a singlet (2H) for the C H 2 protons ( 6 H total) on the mesityl cap. A detailed structure with all proton labels (Figure 3.21) and ' H N M R spectra (Figure 3.22) for 127 can be found in experimental section 3.8.3.1. 165 127 1 2 9 « g u e s t F igure 3.16 Bis-capped trimer cavitands. Complex 127*116 is observed to form between 127 and 116 nitrobenzene-afe. N M R spectra o f solutions containing 127 and tris-acetylene 116 (nitrobenzene-^, 300 K ) show two broad upfield resonances at -0.53 and -0.61 ppm in a 1:2 ratio, which correspond to the acetylene protons protruding into bowls X and bowls Y in 127*116, respectively (Figure 3.16 and 3.17). Exchange between the acetylene protons in bowls X and Y o f complex 127*116 is evident by variable temperature ' H N M R spectroscopy. The signals broaden with increasing temperature and coalesce at 330 K (Figure 3.17). From the coalescence temperature, a free energy o f activation was calculated to be A G * = 16.4 kcal/mole (Av= 39.0 H z at 283 K , see experimental section 3.8.4.2 for calculation details)). 5 0 This energy barrier was initially interpreted as the energy required to rotate the guest about its C3-axis, while arene of 116 is lying parallel to the mesityl cap in the trimer cavity (Figure 3.18, left). C P K models suggest that while the guest is in this orientation such a rotation is extremely sterically hindered, i f not impossible. The interconversion seems more l ikely to occur via tilting at an angle 9 with respect to the mesityl cap and rotated about the guests pseudo C3-axis while still in the cavity o f host 127 (Figure 3.18, right). 166 m 4s w m u IP) Figure 3.17 Expanded regions of ! H N M R spectra o f complex 127*116 in nitrobenzene-^5 at various temperatures, (a) 340 K , (b) 330 K , (c) 325 K , (d) 320 K , (e) 315 K , (f) 320 K and (g) 283 K . "parallel" "elevated" Figure 3.18 Guest mobility in complex 127*116. Ks for complex 127*116 was measured to be -40 M , which is significantly lower than hexa-benzyl 42*116 (35000 M" 1 ) and tris-benzyl 125*116 (1200 M" 1 ) . The more rigid bis-benzyl cap does not provide the same stabilizing interactions that the more flexible benzyl groups do. Electronic repulsions between the pi-electron clouds of the guest and both mesityl and meta-xy ly l caps may contribute to the decrease in stability relative to 42*116 and 125*116. The 167 relatively poor stability of complex 127*116 deterred further complexation studies between 127 and other guests. meta-Xylyl trimer cavitand 127, may or may not be a good transition state model for GDS in the formation of 56*guests. The potential of meta-xylyl 127 was not fully explored because o f the low stability of complex 127*guest and the complicated ' H N M R spectra. Bis-capping o f the trimer cavitand cavity opening also has a profound effect on guest exchange rates (kjki), which are estimated to be considerably slower (rate constants on the order of minutes-hours) compared those measured for 42*116 and 125*116 (milliseconds), 5 1 which is attributed to the smaller portal in meta-xylyl 127. Finally, the low symmetry o f 127 enabled guest mobility in complex 127*116 to be probed, whereas it was not possible in the other trimer and trimer cavitand complexes. 3.6 Complexation of Multiple-Molecule Guests To Trimer 42 In addition to single-molecule guests, the complexation of multiple small-molecules were also investigated. D M S O was used as the guest since ( D M S O ) 3 was found to be the best multiple-molecule template in the formation of 56*guests. Binding studies were conducted in toluene-ofe because o f its high and low temperature capabilities. The initial plan was to: (a) see i f multiple small-molecules can form stable complexes with 42 (i.e., 42*(DMSO) x), and i f so to (b) compare the stability o f complex 42*(DMSO) x to that of 42*116, and (c) investigate the driving forces into the formation of complexes like 42*(DMSO) x. Unfortunately, (b) and (c) could not be done (vide infra). 168 Complex formation between trimer 42 and D M S O is evident at 300 K in toluene-dg by ' H N M R spectroscopy. A t 300 K , guest exchange is fast on the N M R timescale for D M S O complexes with 42 as only a single averaged set of host and (free) guest signals is observed whose chemical shifts change as a function of D M S O concentration (Figure 3.19). The most prominent feature in ' H N M R spectra o f 42 and D M S O is the signal for the H a c protons, which broadens, shifts downfield, and resharpens as more D M S O is added (Figure 3.19). Relatively weak complex(es) between 42 and D M S O appear to form in toluene-dg, as hundreds o f equivalents of D M S O are required to see significant changes in the chemical shifts o f the protons o f the host 42. h 7 6 5 4 3 2 1 0 (ppm) Figure 3.19 *H N M R spectra (400 M H z , toluene-Jg, 300 K ) o f trimer 42 with D M S O . (a) 42 and D M S O (470 equiv.), (b) 42 and D M S O (47 equiv.), and (c) 42 (3.03 m M ) . h = H 2 0 . s = residual protio solvent signal (-CHD2). Free D M S O protons are at -2.65 ppm (overlapping with methylene protons o f the [4] cavitand pendant groups) 169 A t 250 K , exchange rates between free and bound host and guest signals are slow on the N M R timescale. A broad signal is observed at 0.43 ppm in addition to the signal for free D M S O at 1.67 ppm (Figure 3.20c), which was confirmed to be bound D M S O by I D E X S Y experiments. In addition, a new set o f broad host resonances is also observed for each type o f host proton at different chemical shifts than those of the free host 42 (Figure 3.20b). Host:guest integration gives a 3.5:1 ratio for 42:bound D M S O . Either every 41 binds an average o f 3-4 D M S O molecules bind all o f the time, or there is mixture of complexes in which 3-4 D M S O molecules are bound on average (the two possibilities are indistinguishable). Complexation o f solvent (toluene-fife) in addition to D M S O also could not be ruled out. \ i i i j 1 1 1 1 1 1 1 1 1 1 1 1 1 1 i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 II i ' I * ' i 7 6 5 4 3 2 1 0 (ppm) Figure 3.20 ' H N M R spectra (400 M H z , toluene-fife) of trimer 42 with D M S O at various temperatures, (a) 42 at 300 K , (b) 42 at 250 K , (c) 42 and D M S O (350 equiv.) at 250 K . h = H 2 O . * = bound D M S O methyl protons. See Figure 3.10 for proton labels. 170 Multiple-molecule guests (i.e., D M S O ) do form thermodynamically stable, but weak complexes with trimer 42. Since these complexes could not be fully characterized in solution, further studies were not pursued. 3.7 Summary The following was demonstrated in this chapter. (1) Carceplexes 56»guests were successfully synthesized that entrap: (a) single-molecule guests, and (b) two different guests, along with (c) new carceplexes entrapping two or three copies of the same guest in addition to those previously reported, 56«(DMF )3 7 and 56»(DMA )2 . 1 0 In addition, the complexation o f water to the interior o f carceplexes 56»guests was also observed for the first time. The formation o f these hydrates, as well as the restricted mobility of guests greatly hindered the characterization o f many 56«guests by the most reliable tool, ' H N M R spectroscopy. (2) Template effects were observed by both single and multiple-molecule templates in the formation o f 56»guests. For single-molecule templates, 1,3,5-trisubstituted benzene derivatives were the best guests. Size/shape complementarity between guest and host interior, pi-stacking interactions between guest and host cap arenes, and C H - p i interactions between benzene substituents and the arenes o f the host appear to be important attributes for the best templates. For multiple-molecules, templates that maximize host-guest and guest-guest interactions within the forming host cavity are better. A s in previous studies, quantification o f the relative templating ability o f single-molecule guests is relatively easy, while for multiple-molecule competitions, quantification is more complex. Complex product mixtures usually result from the latter, and comparison between many different multiple-guest templates is hindered in the 171 difficulty in finding a common reaction solvent. Overall, guests of lower molecularity were better templates in the formation of 56«guests. In many cases the concentrations required for higher molecularity guests to compete equally with low molecularity guests is physically unattainable. Table 3.15 Stability constants for trimer derivative complexes and template ratios (TR\i) for carceplex 56«guests with various guests. 56»(guest)x 42 114 125 127 tris-acetylene 116 860 35000 200 1200 40 trimethyl ester 117 55 100 - 2-9 trimethoxy 118 25 45 - 3 t r i e t h y l l l 9 23 460 - - -a From reactions in N F P between 43 2,4,6-tris(bromomethyl)mesitylene capping reagent in the presence of K I , K 2 C O 3 at ambient temperatures (~298 K ) . bDetermined in nitrobenzene-fife at 300 K . (3) The greater templating power o f single-molecule over multiple-molecule guests was evident in the formation o f a new family o f stable host-guest complexes involving the hosts 42, 43,114, and 125-127, which form and dissociate slowly on the N M R timescale. Binding ability was found to decrease in the order of 42 > 125 > 114 = 126 > 127 = 43. Structures that show greater resemblance to the speculated transition state model (i.e., 129, Scheme 3.3) in the formation of 56»guests formed the weakest complexes. Binding studies between the best hosts, 42 and 125, and the best templates in the formation of 56»guests revealed that a satisfactory correlation between the template ratios (TRu) and the stability constants (Ks) exists (Table 3.15). Differences between TR\\S and Kss may be due to solvent effects, or TRs were measured from reaction in N F P , while complexation was studied in nitrobenzene-fife. Multiple-molecules also 172 form complexes with trimer and trimer cavitand hosts, which are thermodynamically stable, but relatively weak and exchange rapidly on the N M R timescale at room temperature. Chapter 4 w i l l deal with applications of 56»guests in the stabilization of reactive intermediates and Chapter 5 reports efforts to create larger [4]cavitand-derived carceplexes that can entrap larger guests and greater numbers o f small-molecule guests. 173 3.8 Experimental Section 3.8.1 General Experimental A l l reactions were conducted under an N 2 atmosphere unless stated otherwise. Triethyl-1,3,5-benzenetricarboxylate and 2,4,6-tris(bromomethyl)mesitylene were used as purchased from Aldr ich . 10 % Pd/C was purchased from Aldr ich. D M S O , D M F , and N F P were distilled at reduced pressure and stored over activated 4 A molecular sieves prior to use. N M P was stored over activated 4 A molecular sieves prior to use. T H F used was B D H glass distilled grade and was distilled and dried over sodium/benzophenone ketyl prior to use. A l l other reagents and chromatography solvents were used as purchased without further treatment. Computer modeling on carceplexes 56«guests was done using the software CS Chem 3D Pro (ver. 3.5). 3.8.2 MALDI Mass Spectrometry M A L D I mass spectrometry is a relatively new and soft method for determining the molecular masses of large macromolecular compounds. Relatively little fragmentation occurs using this method, which greatly simplifies the interpretation o f spectra o f large molecules. The method generally involves laser ablation of an area of a sample target on which the analyte is deposited, while embedded in a solid crystalline matrix. Most of the energy is absorbed by the 52 matrix and passed on to the analyte by a process that is not fully understood. For our compounds, positive molecular ions are usually detected as sodium or potassium adducts, unless other additives are present in the matrix (e.g., Ff+ or A g + cations). This method has been 174 extremely useful to the Sherman group and is the only successful method for obtaining masses for some o f the large nonpolar molecules under investigation. 2,5-dihydroxybenzoic acid ( D H B ) was used as the matrix for all o f the compounds reported below. Masses reported as silver adducts were generated from samples ionized using D H B doped with silver trifluoroacetate (-100 equiv. per mole o f analyte). Refer to the General Experimental section o f Chapter 2 for more details about M A L D I M S . 3.8.3 Synthesis and Characterization 3.8.3.1 A,C-Trimer Derivatives A,C-diol 41. Tetrol 8a (7.25g, 7.13 mmol) was dissolved in 725 m L o f acetone and stirred in a 1L, single neck round bottom flask. D B U ( 3 . 2 0 m l , mmol, equiv.) was added and the solution was stirred. After 2 h, a thick white precipitate formed, and excess benzyl bromide was added (4.20 m L , 29.4 mmol, 4.1 equiv.). After -30 min, the reaction mixture turned to a clear solution and was allowed to stir overnight, before the solvent and excess benzyl bromide were removed by rotary evaporation in vacuo. The yellow-brown crude syrup was resuspended in 2 M HC1 (100 mL) and extracted with CHCI3 (3 x 100 mL) . CHCI3 extracts were combined, dried over MgS04 , filtered, and the CHCI3 was removed by rotary evaporation. The crude syrup was purified by flash chromatography on silica gel (230-400 mesh) eluting with ethyl acetate:hexanes (1:8 then 1:2), to afford A , C - d i o l 41 as a glassy white solid that was recrystallized from acetone/hexanes to give white crystals (1.09 g, 13 %). ' H N M R spectroscopic and M A L D I M S spectrometric data were identical to previously data reported. 9 ' 1 0 175 Hexa-benzyl t r imer 42. Solution A : 1.09 g of A , C - d i o l 41 was dissolved in 40 m L of D M S O (degassed only). Solution B : 0.600 m L of bromochloromethane was dissolved in 20 m L o f D M S O . Solutions A and B were slowly added to a stirring suspension of K2CO3 (4.23g, mmol) i n 540 m L of D M S O at 60 °C over a period o f 48 h, after which the reaction was stirred for an additional 24 h, before removing the solvent in vacuo. The crude yellow syrup obtained was resuspended in 2 M HC1 (100 mL) and extracted with C H C 1 3 (3 x 100 mL) . The combined extracts were dried over MgS04 , filtered, and the C H C 1 3 removed by rotary evaporation. CHCI3 extracts were combined, dried over M g S 0 4 , filtered, and then the CHCI3 was removed by rotary evaporation. The crude yellow oi l was recrystallized from C ^ C b / e t h y l acetate to give a white crystalline solid, which was purified by flash chromatography on sil ica gel (10-40 |Ltm, " T L C grade"), eluting with CHCbihexanes (2:1). A,C-trimer 42 was isolated as a glassy white solid that was recrystallized from C ^ C V e t h y l acetate to give a white crystalline solid (452 mg, 42 % after drying in vacuo overnight in a toluene reflux pistol). *H N M R spectroscopic and M A L D I M S spectrometric data were identical to previously reported data. 9 , 1 0 Hexa-hydroxyl A , C - t r i m e r 43. Hexa-benzyl A,C-trimer 42 (410 mg, 0.113 mmol) was dissolved in benzene (25 mL) in a 250 m L round bottom flask. Methanol (20 m L ) and 10 % Pd/C (117.3 mg) were added and the flask was sealed with a rubber septum. H2 gas was then bubbled through the solution for ten minutes and the vessel was sealed under H2 (gas) at 1 arm. The reaction was allow to stir for three h, after which a white precipitate (trimer 43) was observed. The flask was opened to air and T H F was added to completely dissolve the white precipitate before filtering the mixture through Celite. Rotary evaporation gave hexa-hydroxyl trimer 43 as a pale yellow solid, which was precipitated from THF/methanol to give a white powder (315 mg, 90 %). *H N M R spectroscopic and M A L D I M S spectrometric data were identical to previously reported data. 9 ' 1 0 176 Tr i s -hydroxy l t r imer cavitand, 114. Procedure i . Hexa-hydroxyl trimer 43 (106.8 mg, 34.6 ^mol) , K 2 C 0 3 (269.5 mg, 1.95 mmol), and K I (336.0 mg, 2.02 mmol) were mixed in D M F (50 m L ) in a round bottom flask. 2,4,6-tris(bromomethyl)mesitylene (27.9 mg, 70.0 | imol) in 2.8 m L o f D M F was then added over 12 h by syringe pump, followed by stirring for an additional 24 h before removing the solvent in vacuo. The crude yellow solid was then suspended in 2 M HC1 and extracted with CHCI3. The extracts were combined, dried over MgSCU, filtered, and the solvent was removed by rotary evaporation. Tris-hydroxyl trimer cavitand was purified by flash chromatography on silica gel (230-400 mesh) eluting with CHCl3:hexanes (4:1) and then C H C l 3 : M e O H (98:2) and then precipitated from C H C l 3 / M e O H to give a white solid (28.3 mg, 25 %). ' H N M R spectroscopic and M A L D I M S spectrometric data were identical to previously reported data. 1 0 *H N M R (500 M H z , C D C 1 3 , 300 K ) 5 7.27-7.10 (m, 60H, A r H (feet)), 6.93 (s, 3 H , H p (cap)), 6.83 (s, 6 H , H p (interbowl acetal), 6.65 (s, 3H , H p (OH)), 6.04 (d, 6H , H 0 or H 0 ' ) , 5.90 (d, 6 H , H D or H 0 ' ) , 5.88 (d, 3H , H a c i or H a c 2 ) , 5.88 (d, 3H , H a c , or H a c 2 ) , 5.44 (brs, 3 H , OH) , 5.20 (s, 6 H , H x ' ) , 4.93 (t, 6 H , H m or H m ' ) , 4.81 (t, 6 H , H m or H m ' ) , 4.43 (d, 6 H , H ; or Hi*), 4.41 (d, 6 H , H ; or HO, 2.89 (s, 9 H , CH3 (cap)), 2.75-2.38 (m, 24H, C ^ C T ^ C e H s and C / /2CH 2 C 6 H 5 ) . Refer to Figure 3.12 for proton labels. Procedure i i . A mixture containing 56«(DMF )3 and other trimer carceplex reaction byproducts (140 mg) was dissolved in T F A : C H 2 C 1 2 (1:1) and allowed to stir for 30 min, before removing the solvent by rotary evaporation. 4 8 Tris-hydroxyl trimer cavitand 114 was then purified by column chromatography on silica gel (230-400 mesh) eluting with C H C 1 3 and then C H C l 3 : M e O H (98:2), and then precipitated from C H C ^ / M e O H to give a white solid (26.1 mg, 21 %). 177 Tris-benzyl trimer cavitand 125. Tris-hydroxyl trimer cavitand 114 (29.1 mg, 8.97 umol) and C s 2 C 0 3 (103.3 mg, 317 umol) were mixed in D M F (2 m L) for at least 2 h at 60 °C before adding benzyl bromide (20 u L , 168 umol). After stirring for another 12 h at the same temperature, the solvent was removed in vacuo. The crude yellow solid was then suspended in 2 M HC1 and extracted with CHCI3. The extracts were combined, dried over MgSCU, filtered, and the solvent was removed by rotary evaporation. Tris-benzyl trimer cavitand 125 was purified by flash chromatography on silica gel (230-400 mesh) eluting with CHCl3:hexanes (4:1) and then recrystallized from C H 2 C l 2 / E t O A c to give a white solid (18.0 mg, 56 %). *H N M R (400 M H z , C D C 1 3 , 300 K ) 5 7.29-7.04 (m, 75H, A r H (feet) and A r H (benzyl)), 6.90 (s, 3 H , H p 2 ) , 6.82 (s, 9H , H p l and H p 3 ) , 6.01 (d, 6H , H D or H 0 ' ) , 5.81 (d, 3H , H a c l or H a c 2 ) , 5.80 (d, 6 H , H 0 or H 0 ' ) , 5.68 (d, 3H , H a c l or H a c 2 ) , 5.18 (s, 6H, H x ' ) , 4.92 (t, 6 H , H m or H m ' ) , 4.89 (s, 6 H , H x ) , 4.80 (t, 6 H , H m or H m ' ) , 4.40 (d, 6 H , H i or Hi'), 4.37 (d, 6H , H ( or H;'), 2.88 (s, 9 H , CH2 (cap)), 2.75-2.61 (m, 24H, C H 2 C / / 2 C 6 H 5 ) , 2.56-2.40 (m, 24H, C 7 / 2 C H 2 C 6 H 5 ) . Refer to Figure 3.12 for proton labels. 2D N O E S Y and C O S Y spectra were also recorded. M S ( M A L D I ) m/z (rel intensity) 3537 ( (M»C 2 2 gHi 9 g03 6 + N a + ) + ; 100), calcd for C 2 2 g H 1 9 8 0 3 6 « N a + = 3537. Tris-cyanobenzyl trimer cavitand 126. Tris-hydroxyl trimer cavitand 114 (38.9 mg, 12.0 umol) and C s 2 C 0 3 (125 mg, 385 umol) were mixed in 2 m L D M F at 60 °C for 2 h beforepara-bromotoluonitrile (22.8 mg, 116.3 umol) was added. After stirring for another 12 h at the same temperature, the solvent was removed in vacuo. The crude yellow solid was then suspended in 2 178 M HC1 and extracted with CHCI3. The C H C 1 3 extracts were combined, dried over MgSC>4, filtered, and the solvent was removed by rotary evaporation. Tris-cyanobenzyl trimer cavitand 126 was purified by flash chromatography on silica gel (230-400 mesh) eluting with CHCI3 and then recrystallized from CH 2 Cl2 /EtOAc to give a white solid (10.3 mg, 24 %). ' H N M R (400 M H z , CDCI3, 300 K ) 5 7.37 (d, 6H , H 2 ) ) , 7.30 (d, 6H , H,)), 7.27-7.10 (m, 60H, A r H (feet)), 6.91 (s, 3 H , H p i or H p 3 ) , 6.86 (s, 3H , H p l or H p 3 ) , 6.83 (s, 6 H , H p 2 ) , 6.03 (d, 6 H , H 0 or Ho'), 5.90 (d, 6 H , H 0 or H 0 ' ) , 5.82 (d, 3H , H a c i or H a c 2 ) , 5.72 (d, 3H , H a c i or H a c 2 ) , 5.19 (s, 6H , H x or H x ' ) , 4.96 (s, 6 H , H x or H x ' ) , 4.93 (t, 6H , H m or H m ' ) , 4.80 (t, 6H , H m or H m ' ) , 4.39 (d, 6 H , H i or H O , 4.38 (d, 6 H , H i or Hi*), 2.89 (s, 9H , CH3 (cap)), 2.71 (brm, 12H, C H 2 C # 2 C 6 H 5 ) , 2.65 (brm, 12H, C H 2 C # 2 C 6 H 5 ) , 2.59-2.41 (m, 24H, C t f 2 C H 2 C 6 H 5 ) . Refer to Figure 3.12 for proton labels. 2D N O E S Y and C O S Y spectra were also recorded. M S ( M A L D I ) m/z (rel intensity) 3612 ( (M«C 2 3 iH i950 3 6N 3 + N a + ) + ; 100), calcd for C 2 3 iH 1 9 5 036N3»Na + = 3612. (Bis meta-xylyl) T r i m e r cavitand, 127. Trimer cavitand 114 (26.1mg, 8.05 umol), K 2 C 0 3 (99.0 mg, 717 umol), K I (90.4 mg, 545 umol) and l,3,5-tris(ethynyl)benzene (20.0 mg, 133 umol) were mixed in N F P (20 mL) for 2 h a,a'-Dibromo-meta-xylene (2.1 mg, 7.96 umol) was then added in N F P (1 mL) and the reaction was stirred for 12 h The crude yellow-orange solid was suspended in 2 M HC1 and extracted with CHCI3. The C H C 1 3 extracts were combined, dried over MgS04 , filtered, and the solvent was removed by rotary evaporation. Bis meta-xylyl trimer cavitand 127 was then purified by flash chromatography on silica gel (10-40 um, " T L C grade") 179 eluting with CHCI3 and then precipitated from CHCb/hexanes to give a white solid (5.2 mg, 19 %). ' H N M R (400 M H z , CDCI3, 300 K ) 5 7.46 (s, 1H, H™,), 7.36 (t, 1H, H ^ ) , 7.32-7.08 (m, 62H, A r H (feet) and H ^ ) , 6.94 (s, 2 H , H p 2 ) , 6.93 (s, 1H, H p 7 ) , 6.88 (s, 2 H , H p 3 ) , 6.85 (s, 2 H , H p 6 ) , 6.80 (s, 2 H , Hp4), 6.77 (s, 2H , H p l ) , 6.69 (s, 1H, H p 5 ) , 6.28 (d, 1H, H a c t ) , 6.11 (d, 2 H , H o b l ) , 6.01 (d, 2 H , H o t 3 ) , 5.99 (d, 2H , H o b 2 ) , 5.96 (d, 2 H , H o t 2 ) , 5.92 (d, 2H , H o b 3 ) , 5.73 (s, 4 H , H a c ) , 5.68 (d, 2 H , H o t 2 ) , 5.56 (d, 1H, H a c b ) , 5.50 (s, 1H, OH) , 5.35 (d, 2H , H x l ) , 5.21 (d, 2 H , H x 2 ) , 5.01 (s, 2 H , H x 3 ) , 4.97 (d, 2 H , H ^ ) , 4.97 (t, 2H , H m b l ) , 4.92 (t, 2H , H m b 2 ) , 4.89 (t, 2 H , H m b 3 ) , 4.87 (t, 4 H , H m t 3 and H m t 2 ) , 4.83 (t, 2 H , H m t l ) , 4.78 (d, 2H , H ^ ) , 4.59 (d, 2 H , H i b l ) ,4 .58 (d, 2 H , H i t 3 ) , 4.46 (d, 2 H , H i t 2 ) , 4.39 (d, 2 H , H i b 2 ) , 4.29 (d, 2H , H i t l ) , 4.14 (d, 2 H , H i b 3 ) , 2.89 (s, 6H , CH3 (b) cap), 2.81 (s, 3 H , CH3 (a) cap), 2.77-2.60 (brm, 12H, C H 2 C 7 / 2 C 6 H 5 ) , 2.60-2.39 (brm, 12H, CH 2 C/f 2 C6Hs). See Figure 3.21 for proton labels. Proton assignments were made with the help o f 2D N O E S Y , C O S Y , and long range C O S Y spectra recorded. H p i . p 7 were assigned from long range coupling correlations to Hm ti-3 and H m b i . 3 . H i b i - 3 were assigned based on N O E S Y correlations with the methyl protons (a and b) o f the mesityl cap. M S ( M A L D I ) m/z (rel intensity) 3366 ((M*C 2 , 5Hi8 6036 + N a + ) + ; 100), calcd for C 2 1 5 Hi860 3 6«Na + = 3365. 180 Hmx3 Figure 3.21 Proton labels and structure of 127. t = top. b = bottom. ' H N M R spectra of 127 in C D C 1 3 and nitrobenzene-J 5 and 127 and tris-acetylene 116 are shown in Figure 3.22. 181 j i i i i i i i i i ] * H T r r i i i t | i 1 i i i i i i i " | i l i i i i i i i | i i i r*r i T i i | i i i i i i 1 1 i i | i i i i r ' l i i I i i i i ) i"T i i i i i i i i " i ~ r i i i | 7 6 5 4 3 2 1 0 (ppm) Figure 3.22 ' H N M R spectra of 127 and its complexes, (a) 127 and 116 (3 equiv.) in nitrobenzene-c/5. (b) 127 in nitrobenzene-J5. (c) 127 in CDCI3. 3.8.3.2 Single-Guest Trimer Carceplexes (56«guest and 115»guest) Refer to Figure 3.2 for all host proton labels. Mesitylene-capped carceplex 56»NFP. Procedure A . Tris-hydroxyl trimer 43 (10.2 mg, 3.30 umol), K2CO3 (106.0 mg, 767.0 umol), K I (39.0 mg, 235.0 umol) and N F P (9 m L ) were added to a single neck round bottom flask and stirred at 70 °C for at least 1 h 2,4,6-tris(bromomethyl)-mesitylene (5.2 mg, 13.0 umol) was then added in N F P (1 mL) and the reaction was allowed to stir overnight. The solvent was then rotary evaporated in vacuo and the yellow-orange residue 182 was resuspended in 2 M HC1 (10 mL) and extracted with C H C 1 3 ( 3 x 1 0 m L ) . The CHCI3 extracts were combined, dried over MgS04 , filtered, and the solvent was removed by rotary evaporation. Carceplex 56»NFP was then purified by flash chromatography on silica gel (10-40 [ i m , " T L C grade") eluting with CHCl 3 :hexanes (2:1) and then precipitated from CHCb/hexanes to give a white solid. Carceplex 56»NFP was isolated as an inseparable mixture that also contained -15 % 56«(NFP) 2 . ' H N M R (400 M H z , sieve-dried C D C 1 3 , 300K) 6 7.32-7.14 (m, 60H, A r H (feet)), 6.97 (s, 6 H , H p l ) , 6.78 (s, 6H , H p 2 ) , 6.23 (s, 1H, H f ) , 6.00 (d, 12H, H 0 ) , 5.79 (s, 6H , H a c ) , 5.13 (s, 12H, H x ) , 4.92 (t, 12H, H m ) , 4.10 (d, 12H, HO, 2.80 (s, 18H, CH3 (cap)), 2.71 (m, 24H, C H z G f / ^ H s ) , 2.63-2.39 (m, 24H, C H 2 C 7 / 2 C 6 H 5 ) , 1.68 (m, 2 H , H d or H e ) , 1.46 (m, 2H, H d or H e ) , -0.11 (m, 2 H , H b or H c ) , -0.23 (m, 2 H , H b or H c ) , -0.30 (m, 2H, H a ) . 2D C O S Y spectra were also recorded. M S ( M A L D I ) m/z (rel intensity) 3537 ( ( M « C 2 2 5 H 2 0 3 O 3 7 N + N a + ) + ; 100), calcd for C 2 2 5 H 2 o 3 0 3 7 N . N a + = 3536. Mesitylene-capped carceplex 5 6 » b u t y r o p h e n o n e . Procedure B . Hexa-hydroxyl trimer 43 (50.4mg, 16.3 umol), K I (251.6 mg, 1.52 mmol), and K 2 C 0 3 (289.5 mg, 2.09 mmol) were allowed to stir in 40 m L or butyrophenone at 85-95 °C under a N 2 atmosphere for l h , after which 2,4,6-tris(bromomethyl)mesitylene was added (21.4 mg, 52.6 jumol) in 1 m L butyrophenone. After 4 d, the solvent was then rotary evaporated in vacuo and the yellow-orange residue was 183 resuspended in 2 M HC1 (10 mL) and extracted with C H C 1 3 ( 3 x 1 0 m L ) . The C H C 1 3 extracts were combined, dried over MgS04, filtered, and the solvent was removed by rotary evaporation. Carceplex 56»NFP was then purified by flash chromatography on silica gel (10-40 um, " T L C grade") eluting with CHCbihexanes (2:1) and then precipitated from CHCb/methanol and then CHCl3/hexanes to give a white solid (13.2 mg, 23 %). *H N M R (400 M H z , sieve-dried C D C 1 3 , 300 K). 8 7.27-7.15 (m, 24H, A r H (pendant groups) and residual CHCI3), 7.00 (s, 3H , H p l ) , 6.78 (s, 3H , H p 2 ) , 6.32 (d, 2 H , H , ) , 5.90 (d, 12H, H 0 ) , 5.71 (s, 12H, H a c ) , 5.20 (s, 12H, H x ) , 5.15 (t, 2 H , H 2 ) , 4.91 (brt, 12H, H m ) , 4.77 (t, 1H, H 3 ) , 4.03 (d, 12H, Hi), 2.83 (s, 18H, CH3 (cap)), 2.68 (t, 24H, C H 2 C / 7 2 C 6 H 5 ) , 2.49 (t, 24H, C / f 2 C H 2 C 6 H 5 ) , 0.94 (m, 2 H , H a ) , -0.02 (m, 2 H , H b ) , -2.28 (t, 3H , H c ) . 2D C O S Y spectra were also recorded. MS (MALDI) m/z (rel intensity) 3655 ( (M«C 2 i 9 H ,9 2 0 3 6 + Ag + ) + ; 100), calcd for C 2 i 9 Hi9 2 036«Ag + = 3656. Benzene-capped carceplex 115*butyrophenone. Procedure B, except 1,3,5-tris(bromomethyl)benzene53 (5.4 mg, 15.1 umol, ~5 equiv.) was used instead of 2,4,6-tris(bromomethyl)mesitylene, with hexa-hydroxyl trimer 43 (9.5 mg, 3.08 umol), K 2 C03 (24.6 mg, 178 umol, 57 equiv.), KI (11.7 mg, 70.5 umol, 2.4 equiv.) and butyrophenone (10 mL). Carceplex 115«butyrophenone was isolated as a white solid (2.4 mg, 22 %). 184 lH N M R (500 M H z , C D C 1 3 , 300 K ) . 5 7.44 (s, 6H , A r H (cap)), (m, 24H, A r H (pendant groups) and residual CHC1 3 ) , 6.97 (s, 3H , H p l ) , 6.78 (s, 3H , H p 2 ) , 6.28 (d, 2 H , H i ) , 5.93 (d, 12H, H 0 ) , 5.82 (s, 12H, H a c ) , 5.22 (t, 2 H , H 2 ) , 4.98 (s, 12H, H x ) , 4.89 (brt, 12H, H m ) , 4.07 (d, 12H, H;), 2.69 (brm, 24H, C H 2 C # 2 C 6 H 5 ) , 2.51 (brm, 24H, C / f 2 C H 2 C 6 H 5 ) , 0.97 (m, 2 H , H a ) , -0.14 (m, 2 H , H b ) , -2.33 (t, 3 H , H c ) . H 3 is hidden. M S ( M A L D I ) m/z (rel intensity) 3503 ( ( M . C 2 2 3 H i 9 2 0 3 7 + K + ) + ; 100), calcd for C 2 2 3 Hi 9 2 037«K + = 3503. Mesitylene-capped carceplex 56*valerophenone. Procedure A . Hexa-hydroxyl trimer 43 (9.7 mg, 3.14 umol), K 2 C 0 3 (42.2 mg, 306.0 | imol), K I (30.8 mg, 186.0 umol), N F P (10 m L ) , valerophenone (140 uL) , 2,4,6-tris(bromo- methyl)mesitylene (6.6 mg, 16.4 umol). Precipitation from CHCb/hexanes gave 56«valerophenone as a white solid (1.6 mg, 22 %). 56»Valerophenone was isolated as a mixture with 56«NFP and 56«(NFP) 2. *H N M R (500 M H z , sieve-dried CDCI3, 300 K ) 8 7.30-7.15 (m, 60H, A r H (feet)), 6.97 (s, 3H , H p ) , 6.78 (s, 3H, H p ) , 6.45 (d, 2 H , H i ) , 6.00 (d, 12H, H 0 ) , 5.79 (s, 6 H , H a c ) , 5.16 (t, 2 H , H 2 ) , 5.13 (s, 12H, H x ) , 4.91 (t, 12H, H m ) , 4.82 (t, 1H, H 3 ) , 4.10 (d, 12H, Hi), 2.80 (s, 18H, CH3 (cap)), 2.70 (t, 12H, C H 2 C # 2 C 6 H 5 (feet)), 2.53 (m, 12H, C 7 7 2 C H 2 C 6 H 5 (feet)), 1.46 (t, 2 H , H a ) , 0.96 (m, 2 H , H b ) , 0.98 (m, 2 H , H c ) , -3.06 (t, 2H , H d ) . 2D C O S Y spectra were also recorded. Hi O HK HH • 11 1 u 1 185 M S ( M A L D I ) m/z (rel intensity) 3587 ( ( M » C 2 3 o H 2 o 6 0 3 7 + N a + ) + ; 100), calcd for C 2 3 oH 2 o60 3 7«Na + = 3585. Mesitylene-capped carceplex 56*hexanophenone. Procedure A . Hexa-hydroxyl trimer 43 (10.2 mg, 3.30 umol), K 2 C 0 3 (106.0 mg, 767.0 umol), K I (39.0 mg, 235.0 umol), N F P (6.5 m L ) , hexanophenone (3.5 mL) , 2,4,6-tris(bromomethyl)mesitylene (5.0 mg, 12.5 umol). Precipitation from CHCl 3 /hexanes gave 56«hexanophenone as a white solid (2.6 mg, 22 %). 56«hexanophenone was isolated as a mixture with 13 % 56«NFP. ' H N M R (500 M H z , sieve-dried CDC1 3 , 300 K ) 5 7.30-7.14 (m, 60H, A r H (feet)), 7.20 (s, 3 H , Hp), 6.80 (s, 3H , Hp), 6.40 (d, 2H , H, ) , 5.87 (d, 12H, H 0 ) , 5.74 6.01 (s, 6 H , H a c ) , 5.19 (s, 12H, H x ) , 5.00 (t, 2 H , H 2 ) , 4.91 (t, 12H, H m ) , 4.36 (t, 1H, H 3 ) , 4.12 (d, 12H, HO, 2.81 (s, 18H, CH2 (cap)), 2.71 (m, 12H, C H 2 C / / 2 C 6 H 5 (feet)), 2.52 (m, 12H, C # 2 C H 2 C 0 H 5 (feet)), 1.06 (m, 2 H , H a ) , 0.05 (m, 2 H , H b ) , -0.60 (m, 2 H , H c ) , -0.97 (m, 2H, H d ) , -2.58 (t, 2 H , H e ) . 2D C O S Y spectra were also recorded. ' H N M R (500 M H z , nitrobenzene-^, 400 K ) 5 7.68 (s, 3 H , H p ) , 7.52 (s, 3 H , H p ) , 7.32 (d, 24H, H o r t h o (feet)), 7.21 (t, 24H, H m e t a (feet)), 7.14 (t, 12H, H p a r a (feet)), 6.60 (d, 2 H , H i ) , 6.04 (d, 12H, H 0 ) , 6.01 (s, 6 H , H a c ) , 5.40 (t, 2 H , H 2 ) , 5.34 (s, 12H, H x ) , 5.27 (brt, 12H, H m ) , 4.68 (t, 1H, H 3 ) , 4.46 (d, 12H, HO, 2.91 (brm, 66H, CH3 (cap) and C H 2 C # 2 C 6 H 5 (feet) and C t f 2 C H 2 C 6 H 5 (feet)), 1.32 (m, 2 H , H a ) , -0.38 (m, 2H, H b ) , -0.26 (m, 2 H , H c ) , -0.60 (m, 2 H , H d ) , -2.08 (t, 2 H , H e ) . 186 M S ( M A L D I ) m/z (rel intensity) 3601 ( (M»C 2 2 iH 2 08O 3 7 + N a + ) + ; 100), calcd for C 2 2 i H 2 o 8 0 3 7 » N a + = 3599. Mesitylene-capped carceplex 56* 1,3,5-trimethoxybenzene. Similar to procedure A . Hexa-hydroxyl trimer 43 (11.5 mg, 3.73 umol), K 2 C 0 3 (51.3 mg, 372 umol), K I (38.7 mg, 233 umol), N F P (10 m L ) , 1,3,5-trimethoxybenzene (884.6 mg, 5.26 mmol), 2,4,6-tris(bromomethyl)-mesitylene (9.8 mg, 24.6 umol). Precipitation with CHCl 3 /hexanes gave 56» 1,3,5-trimethoxybenzene as a white solid (2.8 mg, 19 %). *H N M R (400 M H z , C D C 1 3 , 300 K ) 5 7.30-7.14 (m, 60H, A r H (feet)), 7.00 (s, 6 H , H p ] ) , 6.81 (s, 6 H , H p 2 ) , 6.00 (d, 12H, H 0 ) , 5.87 (s, 6H , H a c ) , 5.05 (s, 12H, CH2 (cap)), 4.95 (t, 12H, H m ) , 4.34 (s, 3H , A r H (guest)), 4.31 (d, 12H, HO, 2.72 (m, 24H, C H 2 C 7 / 2 C 6 H 5 ) , 2.67 (s, 18H, CH2 (cap)), 2.53 (m, 24H, C H 2 C / / 2 C 6 H 5 ) , 0.53 (s, 9H , OCH3 (guest)). M S ( M A L D I ) m/z (rel intensity) 3592 ( ( M » C 2 3 i H 2 0 4 O 3 9 + N a + ) + ; 100), calcd for C 2 2 i H 2 0 4 O 3 9 . N a + = 3591. Mesitylene-capped carceplex 56»l,3,5-triethylbenzene. Similar to procedure A . Hexa-hydroxyl trimer 43 (11.2 mg, 3.63 umol), K 2 C 0 3 (63.0 mg, 457 umol), K I (58.9 mg, 355 umol), N F P (8 mL) , 1,3,5-triethylbenzene (2 m L , umol), 2,4,6-tris(bromomethyl)mesitylene (5.0 mg, 187 12.5 umol). Precipitation with CHCl 3 /hexanes gave 56*l,3,5-triethylbenzene as a white solid (7.3 mg, 56 %). ' H N M R (400 M H z , C D C 1 3 , 300 K ) 6 7.29-7.15 (m, 60H, AriZ(feet)), 7.00 (s, 6 H , H p l ) , 6.80 (s, 6 H , H p 2 ) , 5.98 (d, 12H, H 0 ) , 5.86 (s, 6H , H a c ) , 5.12 (s, 3H , AxH(guest)), 5.11 (s, 12H, H x ) , 4.94 (t, 12H, H m ) , 4.28 (d, 12H, Hi), 2.78-2.63 (m, 24H, C H 2 C / / 2 C 6 H 5 ) , 2.70 (s, 18H, CH3 (cap)), 2.63-2.41 (m, 24H, C H 2 C / / 2 C 6 H 5 ) , 0.77 (q, 6H , CH2CH3 (guest)), -2.03 (s, 9 H , CU2CH3 (guest)). M S ( M A L D I ) m/z (rel intensity) 3585 ( ( C 2 3 i H 2 i 0 O 3 6 + N a + ) + ; 100), calcd for C 2 3 i H 2 , 0 O 3 6 « N a + = 3585. Mesitylene-capped carceplex 56«trimethyl-l,3,5-tribenzenecarboxylate. Similar to procedure A . Hexa-hydroxyl trimer 43 (10.1 mg, 3.27 umol), K 2 C 0 3 (50.7 mg, 367 umol, 112 equiv.), K I (38.4 mg, 231 umol, 71 equiv.), N F P (10 mL) , 1,3,5-trimethylbenzenecarboxylate (962.9 mg, 3.82 mmol), 2,4,6-tris(bromomethyl)mesitylene (6.0 mg, 15.0 umol, -4.5 equiv.). Precipitation with CHCl 3 /hexanes gave 56»trimethyl-l,3,5-tribenzenecarboxylate as a white solid (1.9 mg, 16%). ' H N M R (nitrobenzene-Js, 500 M H z , 400 K ) 5 7.62 (s, 6H , H p l or H p 2 ) , 7.57 (s, 3 H , A r H (guest)), 7.34 (d, 24H, H o r t h o (feet)), 7.22 (t, 24H, H m e t a (feet)), 7.15 (t, 12H, H p a r a (feet)), 6.25 (s, 6 H , H a c ) , 6.00 (d, 12H, H 0 ) , 5.36 (t, 12H, H m ) , 5.08 (s, 12H, H x ) , 4.79 (d, 12H, HO, 3.03-2.86 (m, 66H, C H 2 C 7 / 2 C 6 H 5 and CH3 (cap) and C H 2 C 7 7 2 C 6 H 5 ) , -0.10 (s, 9H , C O O C 7 / 3 (guest)). Note, signal(s) for the other six H p protons are hidden under nitrobenzene-^ residual proton signals. 188 *H N M R (CDCI3, 500 M H z , 250 K ) 5 7.29-7.14 (brm, 63H, ArH (feet) and ArH (guest) and H i and H 2 and H 3 ) , 6.90 (s, 12H, H p i ) , 6.85 (s, 6 H , H p 2 * or H p 2 " ) , 6.66 (s, 6 H , H p 2 ' or H p 2 " ) , 6.06 (brd, 6 H , H 0 or H 0 ' ) , 6.98 (s, 6H , H a c ) , 5.85 (brd, 6 H , H 0 or H 0 '),4.98 (t, 6H , H m or H m ' ) , 4.94 (t, 6 H , H m or H m *), 4.92 (d, 6 H , H x ' or H x " ) , 4.83 (d, 6H , H x ' or H x " ) , 4.58 (d, 6 H , H ; or Hf'), 4.30 (d, 6 H , H i or Hi'), 2.72 (brm, 24H, C H 2 C / / 2 C 6 H 5 ) , 2.53 (s, 18H, CH3 (cap)), 2.53 (brm, 24H, C r 7 2 C H 2 C 6 H 5 ) , -0.66 (s, 9H , H a ) . See Figure 3.6 for host proton labels. See Chart 4.1 for guest proton labels. M S ( M A L D I ) m/z (rel intensity) 3677 ( ( C 2 2 7 H 2 0 4 O 4 2 + N a + ) + ; 85), calcd for C 2 2 7 H 2 0 4 O 4 2 « N a + = 3675. Mesitylene-capped carceplex 56»l,3,5-tris(ethynyl)benzene. Procedure C . Procedure A , except at room temperature. Hexa-hydroxyl trimer 43 (28.6 mg, 9.26 umol), K 2 C 0 3 (167.9 mg, 1.22 mmol), K I (167.5 mg, 1.01 mmol), N F P (10 mL) , 1,3,5- tris(ethynyl)benzene 5 4 (51.4 mg, 325 umol), 2,4,6-tris(bromomethyl)mesitylene (15.6 mg, 39.1 umol). Precipitation with CHCl 3 /hexanes gave 56»l,3,5-tris(ethynyl)benzene as a white solid (12.4 mg, 40 %). *H N M R (500 M H z , C D C 1 3 , 300 K ) 8 7.29-7.16 (m, 60H, A r / f (feet)), 6.98 (s, 3 H , H p l ) , 6.80 (s, 3 H , H p 2 ) , 5.99 (d, 12H, H 0 ) , 5.91 (s, 3H, H b (guest)), 5.87 (s, 6H , H a c ) , 5.06 (s, 12H, H x ) , 4.94 (t, 12H, H m ) , 4.37 (d, 12H, HO, 2.72 (m, 24H, C H 2 C / f 2 C 6 H 5 (feet)), 2.67 (s, 18H, CH3 (cap)), 2.53 (m, 24H, C H 2 C H 2 C 6 / Y 5 (feet)), -1.24 (s, 3 H , H a ) . See Chart 4.1 for guest proton labels. M S ( M A L D I ) m/z (rel intensity) 3574 ( ( C 2 3 1 H i 9 8 0 3 6 + N a + ) + ; 100), calcd for C 2 3 i H , 9 8 0 3 6 . N a + = 3573. 189 Benzene-capped carceplex 115»l,3,5-tris(ethynyl)benzene. Procedure C , except that 1,3,5-tris(bromomethyl)benzene 5 3 (14.0 mg, 39.3 umol, 4.0 equiv.) was used instead o f 2,4,6-tris(bromomethyl)mesitylene. Hexa-hydroxyl trimer 43 (31.0 mg, 0.010 mmol), K2CO3 (152.01 mg, 1.10 mmol), K I (124.7 mg, 0.75 mmol, 75 equiv.), N F P (30 m L ) , 1,3,5-tris(ethynyl)benzene 5 4 (58.0 mg, 0.384 mmol). Carceplex 115*1,3,5-tris(ethynyl)benzene was isolated as a white solid (21.0 mg, 60 %). ' H N M R (500 M H z , CDCI3, 300 K ) 8 7.27-7.08 (m, 66H, A r H (cap) and A r H (feet)), 6.94 (s, 3 H , Hp,), 6.80 (s, 3 H , H p 2 ) , 6.00 (d, 12H, H 0 ) , 5.99 (s, 6 H , H a c ) , 5.87 (s, 3 H , H b ) , 4.91 (t, 12H, H m ) , 4.79 (s, 12H, H x ) , 4.33 (d, 12H, HO, 2.71 (brm, 24H, CH2GH2C6H5 (feet)), 2.52 (brm, 24H, C H 2 C H 2 C 6 # 5 (feet)), -1.18 (s, 3H , H a ) . See Chart 4.1 for guest proton labels. M S ( M A L D I ) m/z (rel intensity) 3491 ((C225H186O36 + N a + ) + ; 100), calcd for C 2 2 5 H i 8 6 0 3 6 * N a + = 3489. 3.8.3.3 Two-Guest Carceplexes (56«(guest)A; and 115»(guest)w x = 2) Mesitylene-capped carceplex 56«(NFP) 2. Procedure C was used with hexa-hydroxyl trimer 43 (15.6 mg, 5.05 umol), K 2 C 0 3 (41.2 mg, 299 umol), K I (52.0 mg, 313 umol), 2,4,6-tris(bromomethyl)mesitylene (12.9 mg, 32.3 umol), and N F P (10 mL) . Carceplex 56»(NFP)2 was isolated as a white solid (6.1 mg, 26 %). Note that 56»(NFP) 2 was isolated as a mixture with 56»NFP (-2:1 56«(NFP)2:56«NFP). 190 ' H N M R (400 M H z , sieve-dried C D C 1 3 , 300 K ) 5 7.28-7.16 (m, 60H, A r H (feet)), 6.99 (s, 3H , H p l ) , 6.77 (s, 3 H , H p 2 ) , 5.99 (d, 12H, H 0 ) , 5.84 (s, 6H , H a c ) , 5.19 (s, 12H, H x ) , 4.93 (t, 12H, H m ) , 4.30 (d, 12H, Hi), 2.84 (s, 18H, CH3 (cap)), 2.72 (m, 24H, C H 2 C / 7 2 C 6 H 5 (feet)), 2.52 (m, 24H, C / / 2 C H 2 C 6 H 5 (feet)), 1.70 (m, 4 H , H d or H e ) , 1.51 (m, 4 H , H d or H e ) , 0.13 (m, 4 H , H b or H c ) , -0.01 (m, 4 H , H b or H c ) , -0.32 (m, 4 H , H a ) . 2D C O S Y spectra were also recorded. CHO is hidden under H 0 for 56»(NFP) 2 at 6.00 ppm. This is based on the observation that the CHO shifts downfield from the Hi protons in spectra in nitrobenzene-ds at temperatures above 350 K . M S ( M A L D I ) m/z (rel intensity) 3650 ((C 2 3 iH 2 i 4 038N 2 + N a + ) + ; 100), calcd for C 2 2 7 H 2 i o 0 3 8 N 2 « N a + - 3649. Mesitylene-capped carceplex 56« (DMA) 2 . Procedure C was used with hexa-hydroxyl trimer 43 (22.9 mg, 7.40 umol), K 2 C 0 3 (53.3 mg, 384 umol), K I (57.4 mg, 346 umol), D M A (10 mL) , and 2,4,6-tris(bromomethyl)mesitylene (9.9 mg, 39.1 umol). Carceplex 5 6 « ( D M A ) 2 was obtained as a white solid (8.8 mg, 33 %). ' H N M R (400 M H z , sieve-dried CDCI3, 300 K ) 5 7.27-7.15 (m, 60H, AriZ(feet)), 6.99 (s, 3 H , H p i ) , 6.78 (s, 3H , H p 2 ) , 5.96 (d, 12H, H 0 ) , 5.83 (s, 6H , H a c ) , 5.13 (s, 12H, H x ) , 4.93 (t, 12H, H m ) , 4.31 (d, 12H, HO, 2.78 (s, 18H, CH3 (cap)), 2.71 (m, 24H, C H 2 C i / 2 C 6 H 5 (feet)), 2.54 (m, 24H, C # 2 C H 2 C 6 H 5 (feet)), 1.44 (s, 6H , NGrY 3 ) , 0.56 (s, 6H , NC7/ 3 ) , -0.97 (s, 6 H , COCH3). 2D C O S Y spectra were also recorded. 191 r M S ( M A L D I ) m/z (rel intensity) 3596 ((C227H210O38N2 + N a + ) + ; 100), calcd for C2 27H2io0 3 8N2«Na + = 3597. Benzene-capped carceplex 115»(DMA) 2. Procedure C, except l,3,5-tris(bromomethyl)-benzene 5 3 (4.9 mg, 13.7 umol) was used instead of 2,4,6-tris(bromomethyl)mesitylene with hexa-hydroxyl trimer 43 (10.8 mg, 3.50 umol), K 2 C 0 3 (51.0 mg, 370 umol), K I (37.4 mg, 225 umol), D M A (10 mL) . Carceplex 115«(DMA)2 was obtained as a white solid (9.9 mg, 81 %). lH N M R (400 M H z , C D C 1 3 300 K ) 8 7.44 (s, 6 H , Ar i / (cap) ) , 7.27-7.15 (m, 60H, A r / f (feet)), 6.97 (s, 3H , Hpi), 6.75 (s, 3H , H p 2 ) , 5.97 (d, 12H, H 0 ) , 5.91 (s, 6H , H a c ) ) , 4.95 (s, 12H, H x ) , 4.91 (t, 12H, H m ) , 4.30 (d, 12H, HO, 2.71 (m, 24H, C H 2 C # 2 C 6 H 5 (feet)), 2.51 (m, 24H, C / 7 2 C H 2 C 6 H 5 (feet)), 1.41 (s, 6H , N G t f 3 ) , 0.44 (s, 6 H , N C i / 3 ) , -0.94 (s, 6H , COC7/ 3 ) . M S ( M A L D I ) m/z (rel intensity) 3514 ((C227H2io038N2 + N a + ) + ; 100), calcd for C 227H2io038N 2«Na+ = 3513. Mesitylene-capped carceplex 56»(NMP)2. Procedure C was used with hexa-hydroxyl trimer 43 (14.9 mg, 4.83 umol), K 2 C 0 3 (28.6 mg, 172 umol), K I (34.4 mg, 249 umol), N M P (13 m L ) , 2,4,6-tris(bromomethyl)-mesitylene (7.3 mg, 183 umol). Carceplex 56»(NMP)2 was obtained as a white solid (8.8 mg, 33 %). ' H N M R (400 M H z , sieve-dried C D C 1 3 , 300 K ) 8 7.30-7.10 (m, 60H, Ar/7(feet)), 6.99 (s, 3 H , Hpi), 6.77 (s, 3H , H p 2 ) , 5.96 (d, 12H, H 0 ) , 5.82 (s, 6H , H a c ) , 5.14 (s, 12H, H x ) , 4.93 (t, 12H, H m ) , 192 4.31 (d, 12H, HO, 2.79 (s, 18H, CH3 (cap)), 2.70 (m, 24H, CU2CH2C6U5 (feet)), 2.49 (m, 24H, C f Y 2 C H 2 C 6 H 5 (feet)), 1.87 (m, 4 H , H a ) , 0.16 (m, 4 H , H b ) , 0.06 (m, 4 H , H c ) , -0.53 (s, 6 H , H d ) . 2D C O S Y spectra were also recorded. M S ( M A L D I ) m/z (rel intensity) 3620 ((C229H210O38N2 + Na +) +; 100), calcd for C229H2io038N2«Na+ = 3621. Mesitylene-capped carceplex 56«(ace tophenone) 2 . Procedure C was used with hexa-hydroxyl trimer 43 (9.1 mg, 2.95 umol), K2CO3 (38.1 mg, 276 umol), K I (13.8 mg, 831 umol), and 2,4,6-tris(bromomethyl)mesitylene (5.9 mg, 14.8 umol). Carceplex 56»(acetophenone) 2 was obtained as a white solid (0.5 mg, 4 %). ' H N M R (400 M H z , CDCI3, 300 K ) 8 7.30-7.10 (m, 60H, A r H (feet)), 7.05 (s, 3 H , Hp>), 6.83 (s, 3H , H p 2 ) , 5.94 (d, 12H, H 0 ) , 5.93 (s, 6H, H a c ) , 5.61 (d, 2H , Hi), 5.31 (t, 1H, H 3 ) , 5.05 (s, 12H, H x ) , 4.96 (t, 12H, H m ) , 4.43 (d, 12H, H ; ) , 2.64 (s, 18H, CH3 (cap)), 2.74 (m, 24H, CU2CH2C6U5 (feet)), 2.53 (m, 24H, C / / 2 C H 2 C 6 H 5 (feet)), -1.27 (s, 3H , H a ) . 193 3.8.3.4 Two-Guest Carceplexes (56«(NFP«guest)) Mesitylene-capped carceplex 56»(NFP«DMSO). Procedure A was used with hexa-hydroxyl trimer 43 (10.8 mg, 3.50 umol), K 2 C 0 3 (42.1 mg, 305 umol), K I (32.3 mg, 195 umol), N F P (10 m L ) , D M S O (60 uL) , 2,4,6-tris(bromo- methyl)mesitylene (10.3 mg, 25.8 umol). Carceplex 56«(NFP»DMSO) was obtained as a white solid (4.7 mg, 37 %). Note that 56»(NFP»DMSO) was isolated with ~7 % 56»NFP. *H N M R (400 M H z , sieve-dried C D C 1 3 , 300 K ) 8 7.30-7.13 (m, 60H, AriZ(feet)), 6.98 (s, 3H , Hpi), 6.77 (s, 3H, H p 2 ) , 6.43 (s, 1H, CHO), 5.97 (d, 12H, H 0 ) , 5.80 (s, 6 H , H a c ) , 5.16 (s, 12H, H x ) , 4.93 (t, 12H, H m ) , 4.29 (d, 12H, Hi), 2.81 (s, 18H, CH3 (cap)), 2.71 (m, 24H, C H 2 G f Y 2 C 6 H 5 (feet)), 2.51 (m, 24H, C r Y 2 C H 2 C 6 H 5 (feet)), 1.73 (brm, 2H, H d or H e ) , 1.36 (brm, 2 H , H d or H e ) , -0.13 (brm, 2 H , H b or H c ) , -0.02 (brm, 2H, H b or H c ) , -0.40 (brs, 6H , ( C / / 3 ) 2 S O ) , -0.59 (brm, 2 H , H a ) . 2D C O S Y spectra were also recorded. M S ( M A L D I ) m/z (rel intensity) 3615 ((C 2 2 7Hi 980 3 8NS + N a + ) + ; 100), calcd for C 2 2 7Hi980 3 8NS »Na + = 3614. Mesitylene-capped carceplex 56»(NFP»4'-ethylacetophenone). Procedure A was used with hexa-hydroxyl trimer 43 (10.1 mg, 3.27 umol), K 2 C 0 3 (42.5 mg, 308 umol), K I (42.8 mg, 258 umol), N F P (10 mL) , 4'-ethylacetophenone (1 mL) , 2,4,6-tris(bromomethyl)mesitylene (9.4 mg, 194 23.6 umol). Carceplex 56«(NFP«4'-ethylacetophenone) was obtained as a white solid (1.2 mg, 10%). ' H N M R (500 M H z , nitrobenzene-^, 400 K ) 5 7.72 (s, H p ) , 7.68 (s, H p ) , 7.58 (s, H p ) , 7.55 (s, H p ) , 7.35-7.12 (brm, 60H, Ar77(feet)), 6.63 (brd, 2 H , H , ) , 6.33 (brd, 2 H , H 0 ) , 6.20 (brs, 2 H , H a c ) , 6.15 (brm, 4 H , H a o and H 0 ) , 6.11 (brd, 2 H , H 0 ) , 6.08 (brs, 2 H , H a c ) , 6.03 (brd, 2 H , H c ) , 6.00 (brd, 2 H , H 0 ) , 5.93 (brd, 2 H , H 2 ) , 5.91 (brd, 2 H , H 0 ) , 5.40-5.25 (brm, 20H, H x s and H m s ) , 5.20 (brd, 2 H , H x ) , 5.10 (brd, 2H , H x ) , 5.06 (brd, 2 H , HO, 4.76 (brd, 2 H , H ( ) , 4.67 (brd, 2 H , HO, 4.63 (brd, 2 H , HO, 4.57 (brd, 2 H , HO, 4.54 (brd, 2H , HO, 4.42 (brd, 2 H , HO, 2.97 (brm, 66H, C H z C / ^ C g H s and C / / 2 C H 2 C 6 H 5 and C H 3 (cap)), 1.66 (brm, 2H, H d or H e ) , 1.47 (brm, 2 H , H d or H e ) , 1.09 (q, 2 H , C//2CH3), -0.02 (brm, 2 H , H b or H c ) , -0.41 (brm, 2H, H b or H c ) , (s, 3 H , C(CO)Gt f 3 ) , -1-15 (brm, 2 H , H a ) , -1.89 (t, 3H , C H 2 C / / 3 ) . 2 D C O S Y spectra were also recorded. M S ( M A L D I ) m/z (rel intensity) 3684 ( ( C 2 3 5 H 2 i 5 0 3 8 N + N a + ) + ; 100), calcd for C 2 3 5 H 2 i 5 0 3 8 N » N a + = 3684. Mesitylene-capped carceplex 56»(NFP»acetophenone). Procedure A was used with hexa-hydroxyl trimer 43 (12.5 mg, 4.05 umol), K 2 C 0 3 (41.2 mg, 299 umol), K I (31.9 mg, 192 umol), N F P (10 m L ) , acetophenone (100 uL) , 2,4,6-tris(bromomethyl)mesitylene (9.8 mg, 24.6 umol). Carceplex 56»(NFP*acetophenone) was obtained as a white solid (2.2 mg, 10 %). Note that 56«(NFP«acetophenone) was isolated as a mixture containing 52 % 56«NFP. 195 ] H N M R (500 M H z , nitrobenzene-^, 400 K ) 8 7.69 (s, 6H , H p l or H p 2 ) , 7.53 (s, 6 H , H p , or H p 2 ) , 7.32 (brd, 24H, Kortho (feet)), 7.21 (brt, 24H, Hmeta (feet)), 7.14 (brt, 12H, Hpara (feet)), 6.73 (d, 2 H , H i ) , 6.12 (s, 6H , H a c ) , 6.11 (d, 12H, H 0 ) , 5.89 (t, 2 H , H 2 ) , 5.69 (t, 2 H , H 3 ) , 5.53 (s, 1H, CHO (NFP guest)), 5.29 (s, 12H, H x ) , 5.27 (t, 12H, H m ) , 4.66 (d, 12H, H 0 ) , 3.00-2.85 (brm, 66H, C H 2 G / 7 2 C 6 H 5 and C J f Y 2 C H 2 C 6 H 5 and C H 3 (cap)), 1.67 (brm, 2 H , H d or H e ) , -0.33 (brm, 2 H , H b or H c ) , -0.46 (brm, 2 H , H b or H c ) , -0.60 (s, 3 H , COCr7 3 ) , -0.80 (brm, 2 H , H a ) . Note that either H d or H e is hidden under the H 2 0 signal at ~1.85ppm, as indicated in the C O S Y spectrum. M S ( M A L D I ) m/z (rel intensity) 3664 ( ( C 2 3 3 H 2 1 I 0 3 8 N + N a + ) + ; 100), calcd for C 2 3 3 H 2 i , 0 3 8 N . N a + = 3665. Mesitylene-capped carceplex 56«(NFP«propiophenone). Procedure A was used with hexa-hydroxyl trimer 43 (10.9 mg, 3.53 umol), K 2 C 0 3 (46.0 mg, 333 umol), K I (32.0 mg, 193 umol), N F P (10 m L ) , propiophenone (60 uL) , 2,4,6-tris(bromomethyl)mesitylene (13.1 mg, 32.9 umol). Carceplex 56«(NFP»propiophenone) was isolated as a white solid (1.9 mg, 15 %). Note that carceplex 56«(NFP»propiophenone) was isolated as a mixture containing 65 % 56»NFP and 56»(NFP) 2). ' H N M R (500 M H z , nitrobenzene-^, 400K) 8 7.69 (s, 6H , H p , or H p 2 ) , 7.67 (s, 6 H , H p i or H p 2 ) , 7.33 (brd, 24H, Rortho (feet)), 7.21(brt, 24H, Hmeta (feet)), 7.14 (brt, 12H, Hpara (feet)), 6.90 (d, 196 2 H , H i ) , 6.14 (d, 12H, H 0 ) , 6.07 (s, 6H , H a ) , 5.50 (t, 2H , H 2 ) , 5.34 (s, 12H, H x ) , 5.27 (t, 12H, H f f l ) , 5.14 (s, 1H, H f ) , 5.04 (t, 1H, H 3 ) , 4.64 (d, 12H, H 0 ) , 2.93 (brm, 66H, CH2C//2C6H5 and C / f 2 C H 2 C 6 H 5 and C H 3 (cap)), 1.71 (m, 2 H , H d or H e ) , 1.33 (m, 2 H , H d or H e ) , 0.19 (q, 2 H , H a ' ) , -0.24 (brm, 2 H , H b or H c ) , -0.36 (brm, 2H, H b or H c ) , -1.01 (brm, 2 H , H a ) , -2.24 (t, 3H , H b ' ) . 2D M S ( M A L D I ) m/z (rel intensity) 3669 ( ( C 2 3 4 H 2 i 3 0 3 8 N + N a + ) + ; 100), calcd for C 2 3 4 H 2 i 3 0 3 8 N « N a + = 3670. Mesitylene-capped carceplex 56« (NFP»DMF) . Procedure C was used with hexa-hydroxyl trimer 43 (10.7 mg, 3.47 umol), K 2 C 0 3 (39.9 mg, 289.0 umol), K I (27.1 mg, 163 umol), N F P (10 mL) , D M F (70 uL) , 2,4,6-tris(bromo- methyl)mesitylene (8.5 mg, 21.3 umol). Carceplex 56»(DMF»NFP) was isolated as a white solid (2.2 mg, 18 %). Note that carceplex 56»(DMF«NFP) contained 56«NFP and 56«(NFP) 2 (<10 %). *H N M R (400 M H z , C D C 1 3 , 335 K ) 8 7.33-7.11 (brm, 60H, A r H (feet)), 7.01 (s, 6 H , H pi), 6.80 (s, 6 H , H p 2 ) , 6.10 (s, 1H, H a or H f ) , 6.07 (s, 1H, H a or H f ) , 6.00 (d, 12H, H G ) , 5.83 (s, 6 H , H a ) , 5.15 (s, 12H, H x ) , 4.94 (t, 12H, H m ) , 4.30 (d, 12H, H ; ) , 2.81 (s, 18H, CH3 (cap)), 2.72 (t, 24H, CH 2 C / / 2 C6H 5 ) , 2.53 (m, 24H, C//2CH2C6H5), 1.71 (m, 2 H , H d or H e ) , 1.50 (m, 2 H , H d or H e ) , 1.21 (s, 3 H , H b ) , 0.09 (s, 3H , H c ) , -0.26 (m, 4 H , H b and H c ) , -0.45 (m, 2 H , H a ) . 2D C O S Y spectra were also recorded. C O S Y spectra were also recorded. 197 ' H N M R (500 M H z , nitrobenzene-ds, 370 K ) 8 7.51 (s, 6H , H p ) or H p 2 ) , 7.32 (brd, 24H, Uorlho), 7.20 (bit, 24H, Hmeta), 7.13 (brt, 12H, Upara), 6.36 (brs, 1H, CHO (NFP)), 6.36 (brs, 1H, CHO (DMF)) , 6.14 (brd, 12H, H 0 ) , 6.08 (brs, 6H , H a c ) , 5.27 (brm, 24H, H x and H m ) , 4.60 (brd, 12H, Hi), 2.96 (brm, 48H, C H 2 C / 7 2 C 6 H 5 and G t f 2 C H 2 C 6 H 5 ) , 2.92 (s, 18H, CH3 (cap)), 2.11 (brm, 2 H , H d or H e ) , 1.74 (brm, 2 H , H d or H e ) , 1.46 (s, 3H, N C / / 3 ) , (s, 3 H , N C t f 3 ) , 0.01 (brm, 4 H , H b and H c ) , -0.14 (brm, 2 H , H a ) . M S ( M A L D I ) m/z (rel intensity) 3609 ( ( M » C 2 2 8 H 2 1 0 O 3 8 N 2 + N a + ) + ; 100), calcd for C 2 2 8 H 2 i o 0 3 8 N 2 « N a + = 3609. 3.8.3.5 Three-Guest Carceplexes (56»(guest)^ and 115»(guest)^, x = 3) Mesitylene-capped carceplex 56»(DMSO)3. Procedure C was used with hexa-hydroxyl trimer 43 (11.0 mg, 3.56 umol), K 2 C 0 3 (61.4 mg, 445 umol), K I (52.3 mg, 315 umol), N F P (8 m L ) , D M S O (2 m L ) , 2,4,6-tris(bromomethyl)mesitylene (4.9 mg, 12.3 umol). Carceplex 56»(DMSO) 3 was obtained as a white solid (3.1 mg, 13 %). ! H N M R (400 M H z , sieve-dried C D C 1 3 , 300 K ) 8 7.30-7.15 (m, 60H, A r H (feet)), 7.00 (s, 3H , H p l ) , 6.74 (s, 3H, H p 2 ) , 5.91 (d, 12H, H 0 ) , 5.78 (s, 6H, H a c ) , 5.19 (s, 12H, H x ) , 4.94 (t, 12H, H m ) , 198 4.50 (d, 12H, HO, 2.73 (s, 18H, CH3 (cap)), 2.72 (m, 24H, G t f 2 C H 2 C 6 H 5 (feet)), 2.51 (m, 24H, Gf7 2CH 2C6H5 (feet)) -0.06 (brs, 18H, (C / / 3 ) 2 SO) . M S ( M A L D I ) m/z (rel intensity) 3659 ((C225H2io037S3 + N a + ) + ; 100), calcd for C 2 2 5H2io0 3 7 S 3 «Na + = 3657. Benzene-capped carceplex 115»(DMF )3. Procedure C was used, except 1,3,5-tris(bromomethyl)benzene 5 3 (2.2 mg, 6.18 umol) was used instead o f 2,4,6-tris(bromomethyl)-mesitylene with A,C-trimer 43 (5.0 mg, 1.62 umol), K 2 C 0 3 (20.3 mg, 0.147 mmol , 24 equiv.), K I (4.7 mg, 0.028 mmol, 4.5 equiv.), D M F (5 mL) . Carceplex 115»(DMF) 3 was obtained as a white solid (2.3 mg, 40 %). ' H N M R (500 M H z , C D C 1 3 , 300 K ) 8 7.42 (s, 6H , A r H (cap)), 7.30-7.10 (m, 60H, A r H (feet)), 6.96 (s, 3 H , H p i ) , 6.75 (s, 3 H , H p 2 ) , 5.94 (d, 12H, H 0 ) , 5.90 (s, 3 H , H a ) , 5.88 (s, 6 H , H a c ) , 4.96 (s, 12H, H x ) , 4.92 (t, 12H, H m ) , 4.47 (d, 12H, H ; ) , 2.70 (m, 24H, C H 2 C i / 2 C 6 H 5 (feet)), 2.50 (m, 24H, C / / 2 C H 2 C 6 H 5 (feet)), 0.97 (s, 9H , H c ) , -0.17 (s, 9H , H b ) . c H 3 C O b N - ^ a H 3 C H M S ( M A L D I ) m/z (rel intensity) 3560 ( ( C 2 2 2 H 2 o i 0 3 9 N 3 + N a + ) + ; 100), calcd for C 2 2 5 H 2 i o 0 3 7 S 3 « N a + = 3558. 199 Table 3.16 Chemical shift differences (A8 = 8bOUnd-8free) for protons o f various guests in 56»(guest) x in C D C 1 3 . Carcep lex p ro ton 8 f r e e ( p p n i ) (ppm) AS (ppm) A 8 (ppm) 25»gues t A 8 (ppm) 10»guest 56»(DMSO) 3 O I I a H3C CH3 56«(DMA) 2 C H 3 C O b N - ^ a H 3 C C H 3 56»(DMF) 3 H 3 C O b N ^ a H 3 C H 56»(NMP) 2 O d C H 3 5 6 - N F P H b H e I I ' I 5 6 » ( N F P ) 2 2.46 -0.06 2.52 2.95 3.70 H a 2.08 -0.97 3.05 3.72 3.54 H b 3.02 1.44 1.58 1.33 1.98 H c 2.94 0.56 2.38 3.44 4.40 H a 7.99 5.97 2.02 3.71 H b 2.86 -0.11 2.97 - 3.88 H c 2.94 1.04 1.90 2.98 H a 2.23 1.87 2.76 2.82 4.01 H b 1.90 0.16 1.84 2.68 4.02 H c 3.26 0.06 3.10 - 1.79 H d 2.70 -0.53 0.83 3.59 4.49 H a 1.49 -0.30 1.79 2.71 H b or H c 1.38/1.33 -0 . i l / -0 .23 1.44-1.61 1.20-1.36 -H d o r H e 3.27/3.12 1.68/1.46 1.59-1.79 1.08-1.18 -H f 7.80 6.23 1.57 3.33 H a 1.49 -0.32 1.81 2.71 H b or H c 1.38/1.33 -0.01/0.13 1.20-1.39 1.20-1.36 -H d or H e 3.27/3.12 1.51/1.70 1.42-1.76 1.08-1.18 -H f 7.80 h - 3.33 h = hidden. 200 Table 3.17 lH N M R chemical shifts (CDC1 3 ) for aryl ketone guests in 56»(NFP«guest) . Guest Proton 8 free (ppm) 8 bound (ppm) A8 (ppm) H a 2.59 -0.92 3.51 H , 7.94 6.51 1.43 acetophenone H 2 7.55 5.59 1.96 H 3 7.45 5.36 2.09 H a 2.59 -1.27 3.86 (acetophenone)2 H i 7.94 5.61 2.33 H 2 7.55 -H 3 7.45 5.31 2.14 H a ' 2.98 -2.80 5.78 H b ' 1.22 -1.41 2.63 propiophenone H , 7.95 6.71 1.24 H 2 7.68-7.28 -H 3 7.68-7.28 4.77 H a 2.56 -1.22 3.78 H b 2.70 0.86 3.56 4'-ethylacetophenone H c 1.25 -2.39 3.64 H i 7.87 6.40 1.47 H 2 7.27 5.60 1.67 H 2 Hi Hi 0 b' H a ' H 2 cH y > H 3 C \ = ~ J C H 3 3.8.4 N M R spectroscopy 3.8.4.1 Genera l N M R spectra were recorded on Bruker Avance 400 or A M X 500 spectrometers equipped with inverse-gradient probes. Prior to beginning experiments at different temperatures and at different solvents, the probe was tuned. Before starting all 2D ( N O E S Y , R O E S Y , C O S Y , H M Q C ) experiments and I D E X S Y experiments, the 90° pulse width was optimized. Chemical shifts are reported relative to the residual proton signals in the deuterated solvent used (i.e., CDCI3, C D 2 C I 2 , (CDC1 2 ) 2 , CeD6, toluene-fife, pyr idine-^ and nitrobenzene-fife). A l l deuterated solvents used were purchased from Cambridge Isotopes Inc. Temperature calibrations 5 5 were performed for all kinetic runs ( I D E X S Y experiments and coalescence temperature measurements). Calibration standards (4 % methanol in methanol-fi?4 (T< 300 K ) or 80 % ethylene glycol in DMSO-afe (T> 300)) provided by Bruker were equilibrated in the probe o f the spectrometer for at least 10 min before recording a ' H N M R spectrum. The temperature dependent AS values between the hydroxyl protons and the methylene (ethylene glycol) or methyl protons (methanol), respectively, were measured and the actual probe temperature was interpolated from a calibration graph provided by Bruker. The temperatures reported were the temperatures that were measured and should not differ from the actual temperature by more than 1-2 °C. 202 3.8.4.2 Coalescence Temperature (TC) Measurements Activation energies ( A G C J ) in kcal/mole for the two-site exchange processes in 56»117 and 56«(NFP»DMSO) were calculated using equation 3.8, A G C J = RTc[\n(kBfnh) + ln(rc/Av)] (3.8) where R (universal gas constant) = 1.9872 cal K" 1 mole" 1; kB (Boltzman constant) = 3.2995x10" 2 4 cal K " 1 ; h (Planck's constant) = 1.5836x10"34 cal s; Tc is the coalescence temperature in K ; and A v i s the chemical shift difference, in H z . 5 6 This equation only applies to exchange between two equally populated noncoupled spin systems. The activation energy (AG*) for the exchange process observed in complex 127*116 was calculated using the method reported by Shanan-Atidi and B a r - E l i . 5 0 This method provides a quick and convenient way o f determining the activation energies from the coalescence temperature for two unequally populated noncoupled spin systems that are in exchange. X(X= 27rAvr), where r i s the average lifetime of the two spin states and A v i s the frequency difference in Hz) is first is calculated from the population difference AP, with the equation 3.9: AP = PA-PB = I X2-2^ X (3.9) For 127*116, A P = 2 -1 = 1, and therefore, X= 2.83. The free energies o f exchange between both spin systems are then calculated using equations (3.10) and (3.11), AG/ = RTC In TCY X ^ A v I 1 - A P (3.10) A G n J = RT In f hn X A v 1 1 + A P (3.11) A G A * is undefined, but A G B * can be solved by inserting A v = 39.0 H z , Tc = 330 K , a n d X = 2.83. 203 3.8.4.3 I D E X S Y Experiments A l l I D E X S Y ( N O E S Y ) experiments were performed on a Bruker Avance 400 M H z spectrometer. The pulse sequence used was selnogp.2 (Avance- version- 00/02/07), which is a I D N O E S Y that uses selective refocusing with a shaped pulse. Dipolar coupling may be due to N O E or chemical exchange. N O E was discounted for carceplex 56*117 and complexes 42*116 and 125*116 from the broadening and coalescence of exchanging signals observed by variable temperature * H N M R spectroscopy. For each specific set of experiments, a relaxation delay (d l ) o f 5 T i s o f the exchanging nuclei was used. T i s were measured by the inversion-recovery method using the Bruker pulse program called invgs (avance- version- 00/02/07). A total of ten experiments were performed at different relaxation delays of 0.001 s, 0.003 s,'0.010 s, 0.030 s, 0.100 s, 0.300 s, 1.00 s, 3.00 s, 10.0 s, 30.0 s. Peak heights were measured using the Bruker xwinnrnr software and T is were calculated using the simfit command from the equation: I, = I 0 + Pe~' / T | (3.12) where L is the peak height at time t (s), Io is the initial peak height, P is a constant and T i is the longitudinal relaxation time constant (in s). Each set o f E X S Y experiments involved the selective irradiation o f the signals for each exchanging nucleus at a series of at least five mixing times (tm). The choice o f tms depended on the observed rate o f exchange, which could be adjusted by changing the temperature. For this thesis, temperatures were set so that integration of the irradiated (Im-) and response (I r) signals in each E X S Y spectrum gave ratios (Iin-:Ir) between 10:1 and 1:1 at tms below 400 ms. Relative peak areas were obtained from the integration values of both irradiated and response signals in each I D ' H - ' H E X S Y spectrum obtained using Bruker W i n N M R I D software (ver. 6.0). The 204 equilibrium magnetization (Mo) was measured from the integration of the signals o f interest in the I D 'FX spectrum. Rate constants were calculated using matrix analysis 1 9 from the equation: M M j 1 = e~RI'm (3.13) where the variables in bold are n x n (n = 1, 2, 3...) square matrices: M is the matrix o f integration intensities at a particular mixing time (rm, in s), Mo is the matrix of equilibrium magnetization values and R is the matrix containing the site-to-site rate constants. The matrix R was calculated from the linearized form of equation 3.13: Rfm= - l n [ M M 0 _ 1 ] = -X ( lnA )X _ 1 (3.14) where X is the square matrix that diagonalizes M M o " 1 to A , so that In A is a diagonal matrix whose elements are the logarithms of the eigenvalues of M M o " 1 . A l l matrix operations were performed using Maple 5.0 software. The individual first order site-to-site rate constants (in s"1) that make up the matrix R, were obtained graphically by plotting each element o f the matrix-X ( lnA )X _ 1 versus tm (see Figures 3.23-3.25). 5 7 Note that for two-site exchange, X ( l n A ) X _ 1 = Additional E X S Y Data for Complexes 42*116 and 125*116. The kinetics o f guest exchange for complexes 42*116 and 125*116 were examined by I D N O E S Y ( E X S Y ) spectroscopy at 330 K in nitrobenzene-fife (see Tables 3.18 and 3.19, and Figures 3.23 and 3.24). A sample calculation is shown for complex 42*116 from the E X S Y data for tm = 0.250 s. an al2 a2\ a22 205 Table 3.18 Additional I D E X S Y data for complex 42»116 (nitrobenzene-^, 330 K ) . *m(s) I n (G f ree ) Il2 ( G b o u n d ) I21 ( G f r e e ) I22 ( G b o u n d ) «12 «21 0.050 7.90 1.00 1.00 6.79 0.14 0.14 0.100 4.08 1.00 1.00 3.37 0.27 0.28 0.150 2.82 1.00 1.00 2.34 0.41 0.41 0.200 2.18 1.00 1.00 1.78 0.55 0.56 0.250 1.90 1.00 1.00 1.47 0.68 0.69 0.300 1.71 1.00 1.00 1.28 0.81 0.83 0.350 1.56 1.00 1.00 1.17 0.94 0.96 0.400 1.49 1.00 1.00 1.08 1.05 1.07 ly (i = j) = Integration intensities of selectively irradiated signal. I,y (/ =j) = Integration intensities response signals, ay = X ( l n A ) X " ! matrix elements. G f ree = free guest. G b o u n d = bound guest. Integration o f ' H N M R spectrum gave the ratio 1.02:1 for the signals at G f r e e : G b o u n d -Sample calculation. From the integration in the I D ' H N M R spectrum, the ratio of free:bound (If:Ig) guest 116 was measured to be 1.02:1.00. Therefore, M 0 = A t tm = 0.250 s, 'If 0 " " 1 . 0 2 0 . 0 0 " and M ; 1 = ~\IIf 0 " 1 / 1 . 0 2 0 . 0 0 = = 1 . 0 0 0 . 0 0 1 . 0 0 0 0 . 0 0 M = In "1.90 1.00" J21 122 _ 1.00 1.47 Therefore, M M : 1 = 1.87 0.98 1.00 1.47 Using Maple 5.0 the eigenvalues of M M o " 1 were calculated to be 0.47 and 2.49 and the eigenvectors o f M M o " 1 were calculated to be "-0.63" and "0.77" -0 .78 0.64 206 Therefore, X = 0.78 0.77" "0.47 0.00" and A = 0.64 0.00 2.49 Using Maple 5.0, X " 1 , InA, and then X ( l n A ) X _ 1 were calculated, X ( l n A ) X - 1 = where aX2 = 0.68 and a2x = 0.69 (Table 3.18). Note that an an "0.42 0.68" _a2l a22_ 0.69 0.15 R = lr k ""11 "-I2 k k "-21 "-22 where the elements kX)- =1 ,2 ) are the rate constants for exchange from site i to j, which can be calculated from the average of slopes o f the two plots of an and 021 vs tm (equation 3.14, Figure 3.23). hj can also be determined from single point calculations at each tm: X ( l n A ) X - ! R=R = 0.42/0.25 0.68/0.25 0.69/0.25 0.15/0.25 1.68 2.72 2.76 0.60 where k\2 - 2.7 s"1 and k2\ = 2.8 s"1. 207 1.4 1.2 0 I 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 tm(s) Figure 3.23 Plot o f X ( l n A ) X " 1 matrix elements (al2 and 0121) versus mixing time (tm) for complex 42*116 (nitrobenzene-^5, 330 K ) . [42] = 6.37 m M , [116] = 12.4 m M and [117] = 318 m M . • = ai2 (slope (kl2) = (2.63 ± 0.8) s"1, ^-intercept = 0.0143, r 2 = 0.999). • = a2] (slope (£21) = (2.69 ± 0.8) s"1, ^-intercept = 0.0121, r 2 = 0.999). Errors in jfeys were based on 95 % confidence limits from regression analysis. Table 3.19 Additional I D E X S Y data for complex 125*116 (nitrobenzene-tfs, 330 K ) . tm (S) I n (Gfree) I12 (Gbound) I21 (Gfree) I22 (Gbound) «12 «21 0.050 10.12 1.00 9.83 1.00 0.11 0.09 0.100 4.27 1.00 5.03 1.00 0.22 0.20 0.150 3.16 1.00 3.13 1.00 0.35 0.30 0.200 2.41 1.00 2.40 1.00 0.46 0.41 0.250 2.05 1.00 1.99 1.00 0.61 0.53 0.300 1.71 1.00 1.74 1.00 0.71 0.63 0.350 1.60 1.00 1.61 1.00 0.80 0.71 0.400 1.44 1.00 1.44 1.00 0.93 0.82 h (i = j ) = Integration intensities o f selectively irradiated signal. I,y (/ = j) = Integration intensities response signals. a{j = X ( l n A ) X _ 1 matrix elements. G f r e e = free guest. Gb0und = bound guest. Integration o f J H N M R spectrum gave the ratio 1:1.14 for the signals at Gfree:Gbound. 208 1.20 1.00 H 0.80 :5 0.60 ] 0.40 ] 0.20 0.00 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 tm(s) Figure 3.24 Plot o f X ( l n A ) X _ 1 matrix elements (an and a2i) vs. mixing time for complex 125»116 (nitrobenzene-^, 330 K ) . [125] = 5.67 m M and [116] = 8.22 m M . • = ax2 (slope (kx2) = 2.35 ± 0.11) s"1, ^-intercept = 0.005, r 2 = 0.998). - = a2i (slope (k2X) = (2.09 ± 0.08) s"1, y-intercept = 0.009, r 2 = 0.999). Errors in £yS were based on 95 % confidence limits from regression analysis. Add i t iona l I D E X S Y Data for T r imer Carceplex 56»117. The individual elements a§, (ij = 1, 2) o f the matrix X ( l n A ) X " ' were calculated at various randomized mixing times at 267 K in CDCb for 56»117 are shown in Table 3.20. The rate constants were obtained from the linear plots shown in Figure 3.25. 209 Table 3.20 Additional I D E X S Y data for trimer carceplex 56«117 (CD 2 C1 2 , 267 K ) . 'm(s) In (4.87 ppm) Il2 (4.69 ppm) I21 (4.87 ppm) I22 (4.69 ppm) «12 «21 0.005 13.10 1.00 1.00 13.64 0.08 0.07 0.015 4.44 1.00 1.00 4.86 0.22 0.21 0.030 2.35 1.00 1.00 2.36 0.44 0.48 0.045 1.81 1.00 1.00 1.89 0.64 0.59 0.060 1.48 1.00 1.00 1.44 0.83 0.88 0.075 1.28 1.00 1.00 1.45 0.96 0.93 0.100 1.14 1.00 1.00 1.15 1.31 1.37 hj (i =j) = Integration intensities of selectively irradiated signal, ly (i = j) = Integration intensities response signals. a<j = X ( l n A ) X _ 1 matrix elements. Integration of ' H N M R spectrum gave the ratio 1:1 for the signals at 4.87 ppm:4.69 ppm. 1.6 0 0.02 0.04 0.06 0.08 0.1 0.12 tm(s) Figure 3.25 Plot of X ( l n A ) X " ' matrix elements (a\2 and a2\) versus mixing time for 56»117 ( C D 2 C 1 2 , 267 K ) . • = a n (slope (kl2) = (12.8 ± 1.8) s"1, ^-intercept = 0.0386, r 2 = 0.997). • = a2\ (slope (k2l) = (13.2 ± 1.8) s"1, ^-intercept = 0.0241, r 2 = 0.986). Errors in fcjS were based on 95 % confidence limits from regression analysis. 210 Free energies o f activation (AG*, in kcal/mol) were calculated from k (in s"1) using the equation: A G * = -RTln(kh/kBT) (3.15) 3.8.5 Template Ratios 3.8.5.1 Competition Experiments Competition experiments were conducted using a procedure similar to procedure A , except that the competing guests, guest 1 (Gi) and guest 2 (G2), were added at concentrations ranging from 0.1 to 5.0 mole % of the solvent, N F P . Separate reactions were conducted both at ambient conditions (298 K ) and at 70 °C. The results of all competition experiments at room temperature and 70 °C are listed in Table 3.21 in Table 3.22, respectively. Table 3.21 Single-guest competition experiments at room temperature. G , G 2 TR ( G O TR ( G 2 ) 1,3,5-tris(ethynyl)benzene 1,3,5-triethylbenzene 50.1 1.00 1,3,5-tris(ethynyl)benzene Trimethyl-1,3,5-benzene 2.25 1.00 tricarboxylate 1,3,5-tris(ethynyl)benzene 1,3,5-trimethoxybenzene 38.9 1.00 Trimethyl-1,3,5-benzene 1,3,5-trimethoxybenzene 4.46 1.00 tricarboxylate 1,3,5-trimethoxybenzene 1,3,5-triethylbenzene 1.08 1.00 1,3,5-trimethoxybenzene hexanophenone 4.70 1:00 1,3,5-trimethoxybenzene butyrophenone 24.6 1.00 211 Table 3.22 Single-guest competition experiments at 70 °C. G i G2 TR (GO TR (G 2) 1,3,5-tris(ethynyl)benzene Trimethyl-1,3,5-benzene tricarboxylate 38.8 1.00 1,3,5-tris(ethynyl)benzene 1,3,5-trimethoxybenzene 42.4 1.00 Trimethyl-1,3,5-benzene 1,3,5-trimethoxybenzene 1.09 1.00 tricarboxylate Trimethyl-1,3,5 -benzene N F P 264 1.00 tricarboxylate 1,3,5-trimethoxybenzene N F P 256 1.00 1,3,5-trimethoxybenzene hexanophenone 5.70 1.00 1,3,5-trimethoxybenzene butyrophenone 24.1 1.00 hexanophenone butyrophenone 10.9 1.00 hexanophenone N F P 47.0 1.00 Competitions between NFP»guest l versus NFP«guest2 (Table 3.23), and NFP»guest versus single-molecule guests (Table 3.24) were also conducted using procedure A . Table 3.23 Competition results for 56»(NFP»guest) at 70 °C. G i G 2 TR22 (Gi) (G 2) 4'-ethylacetophenone propiophenone 15.5 . 1.00 4'-ethylacetophenone acetophenone 6.3 1.00 acetophenone propiophenone 2.3 1.00 212 Table 3.24 Single- versus two-molecule template competition results at 70 °C. ( G / N F P » G B ) ( G / N F P « G B ) 4'-ethylacetophenone (0.0130) N F P (8.70) 0.73 0.00952 M 4'-ethylacetophenone (0.0133) N F P (8.96) 0.60 0.00781 M 4'-ethylacetophenone (0.0132) N F P (8.87) 0.62 0.00847 M 4'-ethylacetophenone (0.0132) N F P (8.85) 0.62 0.00813 M 4'-ethylacetophenone (0.00669) N F P (8.92) 1.36 0.00909 M 4'-ethylacetophenone (0.0267) N F P (8.89) 0.33 0.00893 M 4'-ethylacetophenone (0.0132) 118 (0.0618) 1.28 2.81 M 4'-ethylacetophenone (0.0132) 118 (0.0802) 0.96 1.91 M 4'-ethylacetophenone (0.00669) 118 (0.0551) 2.07 2.26 M 4'-ethylacetophenone (0.0267) 118 (0.0549) 0.63 2.76 M 4'-ethylacetophenone (0.0132) 123 (0.058) 0.23 0.32 M acetophenone (0.0254) N F P (8.92) 2.64 0.0671 M acetophenone (0.0256) N F P (8.96) 1.98 0.0513 M propiophenone (0.0521) N F P (8.92) 2.92 0.155 M propiophenone (0.233) N F P (8.70) 0.63 0.147 M "Bracketed values in the first two columns correspond to [G B ] and [G], respectively, in M . PR = product ratio. 213 3.8.5.2 Solvent Competit ions Table 3.25 M A L D I mass spectrometric data on carceplex product mixtures in different binary solvent mixtures at room temperature. Solvent Rat io Carceplex m/z m/z M i x t u r e (v/v) products expected found D M F : N M P 1:1 56« »(NMP) 2 3621a 3622 a 56* »[(DMF) 2 «NMP] 3668 a 3668a 56* >(DMF) 3 3642a 3643 a D M F : D M S O - c i 6 9:1 56* •(DMSO-6>6)3 3760b 3760 b 56* »[(DMSO-J 6 ) 2 «DMF] 3749b 3749 b 56* • [DMSO-o>(DMF) 2 ] 3738 b 3738 b 56* >(DMF) 3 3727b 3728 b D M F : D M A 7:3 56* »(DMA) 2 3682b 3679 b 56* »[(DMA) 2 »DMF] 3755 b 3756b 56* »[DMA»(DMF) 2 ] 3745 b 3742b 56* »(DMF) 3 3727 b 3727b D M A : D M S O 1:9 56* • ( D M S O ) 3 3742 b 3743b 56* • [DMSO«(DMA) 2 ] 3760 b 3760 b 56* • [ D M S O D M A ] 3673 b 3674 b 56* • ( D M A ) 2 3682 b 3686b D M A : N M P 1:1 56* »(DMA) 2 3682b 3683 b 56* »(DMA) 2 3594b 3696 b 56* »(NMP) 2 3706 b 3708b D M S O : N M P 1:9 56* »(DMSO) 3 3657a 3660 a 56* »[(DMSO) 2 «NMP] 3678 a 3678 a 56* »[DMSO«(NMP) 2 ] 3699 a 3701a 56* »(NMP) 2 3621a 3623 a D M A : N F P 1:9 56* »(DMA) 2 3597a 3598 a 56* »[DMA«NFP] 3623 a 3624 a N M P : N F P 1:1 56* »(NMP) 2 3621a 3623 a 56* »[NMP«NFP] 3635 a 3637 a D M F : N F P 1:1 56* »(DMF) 3 C - -D M F : C H C 1 3 1:1 56* »[(CHC1 3 ) 2 «DMF] 3819b 3820 b 56* »[CHC1 3«DMF] 3773b 3773 b v/v = volume/volume. a N a adducts. b A g adducts. cDetected by ! H N M R spectroscopy. 214 3.8.5.3 Temperature Dependence of the Template Ratios TRn, TRn, and 7*23 Procedure A was used for competitions between tris-acetylene 116, N F P « D M S O , and (DMSO)3 , except different temperatures were used. Reactions were heated at constant temperatures using thermostated silicon o i l baths. Template ratios determined from each experiment at each temperature were the average of two runs. Template ratios were calculated by considering the following equations: * T H* + T * D T H* + D K2D H*« D H*« D + D ^DN H*« D 2 H*« D+ N ^3D H*« (D«N) H*« D 2 + D *T H*. D 3 H*.T ^ND 56* T H*» D* N • 56* (D« N) ^3D H*. D 3 • 56* D 3 where T, D , and N are the respective concentrations (in M ) of 116, D M S O and N F P . H * is the host intermediate prior to the GDS. KT, KD, K2D, Km, and K3D are equilibrium constants and kj, £ND, and A:3D are the rate constants for the GDSs o f the corresponding carceplexes. TRn, TRi2, and 77?23 are derived similarly to equation 3.5: [56*T][Df = _ ^ A M 2 1 3 [56 .D 3 ][T] fi3Dk3D TRn = \ ^ m m = _ K I k ^ M ( 3 . 1 7 ) [ 5 6 . N . D ] [ T ] £ N D £ N D 215 T R _ [ 5 6 « N « D ] [ D ] 2 _ / ? N D A : N D M 1 2 [ 5 6 « D 3 ] [ N ] 03Dk3D (3.18) where the overall binding constants are fiiD = £ D ^ 2 D ^ 3 D and /3ND = KDK-N. Enthalpic (AAfF+AAH1) and entropic (AAS^ +AAS1*) values were obtained from the slopes and ^-intercepts, respectively, of plots of R\n(TRmn) versus l/T. Note that: Rln(TRmn) = -(AA//°+AA#*)/r-(AAS°+AAS*) (3.19) where m = 1,2, n = 2,3. The plots are shown below as Figures 3.26-3.28. Refer to Table 3.8 for the template ratio data. Figure 3.26 Plot o f \n(TRl3) versus 1/7/for 116/ (DMSO) 3 competitions. Slope = AMT+AAH1 = -(14.1 ± 1.6) kcal/mol, ^-intercept = AA^+AAS4 = (54.3 ± 5.2) cal mol" 1 K " 1 , r 2 = 0.987. Errors are the standard errors of one standard deviation. 216 3.05 3.2 (1 /T)*1000 Figure 3.27 Plot of ln(77?i2) versus 1/7/for 116/(NFP«DMSO) competitions. Slope = AAFr+AAH* = -(5.25 ± 0.07) kcal/mol, ^-intercept = AAS°+AASt = (32.7 ± 0.2) cal mol" 1 K " 1 , r 2 = 1.000. Errors are the standard errors o f one standard deviation. (1 fT)*1000 (1/K) Figure 3.28 Plot o f ln(77?23) versus 1/7/for (NFP»DMSO) / (DMSO) 3 competitions. Slope = AAH°+AAH% = -(9.49 ± 1.82) kcal/mol, ^-intercept = AAS°+AASl = (23.7 ± 5.8) cal mol" 1 K " 1 , r 2 = 0.965. Errors are the standard errors o f one standard deviation. 217 3.8.6 Host-Guest Complexes with Trimer Derivatives 3.8.6.1 General Complexation Experiments ' H N M R complexation experiments were conducted for all hosts and guests listed in sections 3.8.6.2 and 3.8.6.3 as follows. Separate stock solutions o f host and guest were prepared separately in each solvent. Aliquots of each host and guest stocks were added to an N M R tube. The tube was shaken to mix the two solutions and then a ' H N M R spectrum was recorded immediately. Samples were equilibrated within minutes, as no further changes in the relative intensities of the signals after longer periods of time (except for meto-xylyl 127 and tris-acetylene 116). Association constants, Kss, were calculated from the relative integration o f the free and bound host and guest signals measured in the lH N M R spectra, using the equation: H + G - H«G * s = ^ l (3.19) [H][G] H is the host, G is the guest and H«G is the host-guest complex. 218 3.8.6.2 A , C - T r i m e r s 42 and 43 See Figure 3.9 for host proton labels. Table 3.26 *H N M R chemical shifts (ppm) of 42 in various deuterated solvents at 300 K . Solvent A r H (benzyl) A r H (feet) Hp H„ H a c H x H m H i C H 2 s (feet) C D C l 3 a 7.27, 7.03, * * 6.85, 6.82 5.88 5.60 4.85 4.83 4.36 2.66, 2.47 C D 2 C l 2 b ** 7.03 ** ** 6.95, 6.92 5.84 5.58 4.89 4.81 4.22 2.68, 2.53 C 6 D 6 b $ $ $ 5.96 5.33 5.19 5.21 4.44 2.68, 2.52 toluene-c?8B 7.40, 7.21 # # # 6.12 5.27 5.04 5.10 4.55 2.60, 2.48 nitrobenzene-^ 3 7.57, 7.22, 6.98 7.19, 7.13, 7.11 7.76 7.70 6.08 5.58 5.30 5.16 4.65 2.83, 2.77 pyridine-tf*5A 7.63, 7.36, @ 7.66, h 6.23 5.75 5.31 5.26 4.70 2.92-2.73 a 500 M H z . b 400 M H z . *Signals appear as a multiplet partially hidden under residual proton signals of the solvent, between 7.25-7.10 ppm. **Signals appear as a multiplet partially hidden under residual proton signals of the solvent, between 7.30-7.10 ppm. $ = signals appear as a multiplet partially hidden under residual proton signals o f the solvent, between 7.54-6.84 ppm. # = signals appear as a multiplet partially hidden under residual proton signals of the solvent, 7.18-6.96 ppm. @ = signals appear as a multiplet partially hidden under residual proton signals o f the solvent, 7.32-7.13 ppm. 219 Table 3.27 *H N M R chemical shifts (ppm) for selected protons of 42»116 in various deuterated solvents at 300 K . a Solvent H„ H a c H x H m H i Guest C D C l 3 b 5.82 5.75 4.63 4.83 4.29 C D 2 C l 2 b 5.84 5.73 4.70 4.81 4.29 C 6 D 6 b 6.17 5.90 4.77 5.22 4.44 toluene-a*8b 6.12 5.87 4.74 5.16 4.67 nitrobenzene-<i5d 6.20 6.15 4.95 5.19 4.76 6.62, -0.84 6.67, -0.82 -0.47 -0.52 -0.32 a The aryl protons and the methylene protons o f the feet (R) o f the host are not listed because either they are hidden under solvent or other host signals, or do not change in chemical shift significantly from that o f the free host in the respective solvents. b 400 M H z . c H b o f guest 116 is hidden. d 500 M H z . Table 3.28 lH N M R chemical shifts (ppm) of complex 43*116 in various deuterated solvents at 300 K . Solvent A r H (feet) Hp H„ H a c H m Hj C H 2 s (feet) Hb, H a (116) C D C l 3 a 7.22- 6.86, 5.91 5.84 4.82 4.25 2.64, 6.44, 7.07 6.67 2.45 -0.86 nitrobenzene-a'5 7.24, h 6.42 6.25 5.19 4.88 2.82, h, 7.10 2.49 -0.13 5% (v/v) C D 3 O D added, h = hidden. 220 Table 3.29 *H N M R chemical shifts o f selected protons of complexes 42»guest (nitrobenzene-' s , 300 K ) . a Guest H 0 H a c H x H m Hi Guest 116 6.20 6.15 4.95 5.19 4.76 -0.32 137 a 6.16 6.20 5.22 5.19 4.73 8.02, 7.98, h, 0.55 117 b 6.27 6.27 5.26 4.90 4.83 7.99, 0.45 119 c 6.28 6.12 5.01 5.15 4.64 6.15, 1.64,-1.24 118 d 6.18 5.94 5.16 5.09 4.68 5.50, 1.48 'Addit ional signals: 5 4.56 (d, A r H (feet)), 7.64 (s, H p l or H p 2 ) , 7.52 (s, H p , or H p 2 ) , 7.41 (t, A r H (feet)), 7.24 (t, A r H (feet)), 7.21 (d, A r H s (feet)), 7.14 (t, A r H s (feet)), 7.12 (t, A r H s (feet)), 2.80 (brm, CH2s (feet)). Add i t i ona l signals: 8 7.65 (s, H p i or H p 2 ) , 7.64 (s, H p l or H p 2 ) , 7.33-7.02 (m, A r H s (feet and benzyls)), 2.83 (m, C772s (feet)). Addi t iona l signals: 8 7.64 (s, H p I or H p 2 ) , 2.77 (brm, CH2s (feet)). Note that the other H p i or H p 2 is hidden under another signal. A d d i t i o n a l signals: 8 7.70 (s, H p , or H p 2 ) , 7.53-7.06 (m, A r H s (feet and benzyls)): 2.77 (brm, C772s (feet)). Note that the other H p i or H p 2 is hidden under another signal, h = hidden. 221 Table 3.30 Additional data for Ks measurements for complexes 42»116 and 43»116. Host (H) Solvent [H] total (mM) [116] total (mM) [H»116]:[H] Vtotal (uL) Ks (M 1) 42 toluene-a^3 3.08 , 3.67 2.08 1.00 400 1300 42 toluene-ag 3.08 3.67 1.87 1.00 400 1100 42 toluene-fife 3.08 3.67 2.09 1.00 400 1130 42 toluene-fife 1.80 1.78 1.16 1.00 416 1400 42 C 6 D 6 b 1.63 7.35 1.03 1.00 400 160 42 C 6 D 6 2.29 10.1 1.33 1.00 434 150 42 C 6 D 6 1.58 7.48 0.84 1.00 400 120 42 C 6 D 6 1.58 7.48 0.80 1.00 400 120 42 C 6 D 6 1.58 7.48 0.83 1.00 400 120 42 C D 2 C 1 2 C 2.16 4.77 0.70 1.00 460 180 42 C D C l 3 d 3.25 67.7 0.86 1.00 400 13 42 C D C l 3 d 3.25 67.7 0.86 1.00 400 13 42 C D C l 3 d 3.25 67.7 0.86 1.00 400 13 42 C D C l 3 : C D 3 O D (19 Dd 3.10 64.5 0.95 1.00 420 15 42 C D C l 3 : C D 3 O D ( 1 9 Dd 3.10 64.5 0.96 1.00 420 15 42 C D C l 3 : C D 3 O D ( 1 9 Dd 3.10 64.5 0.96 1.00 420 15 43 G D C 1 3 : C D 3 0 D ( 1 9 De 2.75 13.7 1.95 1.00 400 130 43 C D C 1 3 : C D 3 0 D (19 De 2.75 13.7 1.80 1.00 400 120 43 C D C l 3 : C D 3 O D (19 l ) e 2.67 14.8 1.80 1.00 370 110 43 nitrobenzene-fifef 2.05 24.8 0.95 1.00 440 40 aRelative integration from H a c , H b , H and H a signals for 42*116, and H a c and H ; for 42. bRelative integration from H 0 , H b , H and H a (116) signals for 42»116, and H a c and Hj for 42. cRelative integration from A r H (116), H a c , H b and H a (116) signals for 42»116, and H a c for 42. d Relative integration from A r H (116), H a c , H b and H a (116) signals for 42»116, and H a c for 42. Rela t ive integration from A r H (116), H 0 and H a (116) signals for 42«116, and H a c and H 0 or 42. fRelative integration from free and bound 116 signals (H a ) . [ H ] t o t a i = total host added. [116] totai= total guest 116 added. V t otai = total volume. 222 Table 3.31 Additional data for Ks measurements for complexes 42»guest in nitrobenzene-^-T (K) Guest (G) [H] total (mM) [G] total (mM) [H.G]:[H]a Vtotal (UL) ( M 1 ) 300 b 137 1.12 1.53 1.04 1.00 500 1100 300 b 119 1.60 9.93 3.37 1.00 535 460 300 b 119 1.83 2.98 1.12 1.00 535 540 300 b 119 1.83 2.98 0.69 1.00 535 540 300 b 117 5.02 26.7 4.23 1.00 535 130 315 b 117 5.02 26.7 2.84 1.00 535 90 330 b 117 5.02 26.7 1.76 1.00 535 50 300 b 117 2.11 20.4 0.15 1.00 400 110 300 b 117 1.64 8.94 0.83 1.00 535 100 315 b 117 1.64 8.94 0.71 1.00 535 88 330 b 117 1.64 8.94 0.39 1.00 535 47 315 c 117 5.84 33.7 3.18 1.00 500 100 315° 117 1.78 6.94 0.81 1.00 500 130 300 b 118 2.01 29.9 1.02 1.00 535 34 300 b 118 1.64 42.1 1.45 1.00 548 35 300 b 118 1.96 8.44 0.45 1.00 445 .58 283 b 118 1.96 8.44 0.51 1.00 445 65 300 b 118 1.82 13.7 0.66 1.00 430 51 'From integration. b A t 500 M H z field strength. c 4 0 0 M H z . H = host. G = guest. V = Volume. 223 Table 3.32 Additional data for relative stability measurements (KKi) o f 42«guest in nitrobenzene-^. Guest 1 (Gi) Guest 2 (G2) T (K) [Host] (mM) [Gi] (mM) [G2] (mM) [H.Gi]: [H.G,] [G2]free: [Gl]free Vtotal (UL) KTe\ 116 117 300 a 2.23 2.18 161 2.00: 1.00 168:1.00 407 336 116 117 300 a 5.46 2.79 220 0.93:1.00 371:1.00 535 345 116 117 300 a 2.09 2.11 11.3 1.65: 1.00 221:1.00 427 365 116 117 315 b 6.37 12.4 318 7.07:1.00 60.2:1.00 498 426 116 117 315 b 6.37 12.4 318 6.63:1.00 60.1:1.00 498 398 116 117 315 a 5.46 2.79 220 0.97:1.00 308:1.00 535 299 116 117 330 b 6.37 12.4 318 6.99:1.00 61.0:1.00 498 426 116 117 330 b 6.37 12.4 318 6.57:1.00 60.9: 1.00 498 400 116 117 330 a 5.46 2.79 220 1.00:1.00 418:1.00 535 418 117 118 300 a 2.23 161 254 1.06:1.00 1.43:1.00 407 1.5 a 5 0 0 M H z . b 4 0 0 M H z . H = host. V = volume. 224 3.8.6.3 T r i m e r Cavitand Complexes Refer to Figure 3.12 for proton labels. Table 3.33 *H N M R chemical shifts (ppm) of trimer cavitand 125 and complex 125*116 in (nitrobenzene-^, 500 M H z , 300 K ) . Proton 125 125*116 A r H (feet and benzyl) h (d), 7.45 (t), 7.29 (d), 7.35-7.05 h (d), 7.37-7.06 (m) H p 7.81, 7.71, h 7.72, 7.65, h H 0 6.07, 6.07 6.19, 6.08 H a c 5.81,5.02 6.17, 6.05 H b 5.37 4.88 H m 5.25, 5.08 5.28,5.14 H b ' 5.37 5.13 H i 4.62, 4.37 4.69, 4.56 C H 3 (cap) 3.07 2.88 C H 2 s (phenyl ethyl feet) 2.99-2.67 3.00-2.70 (m) guest - 6.67 (s), -0.54 (s) h = hidden (either under other host signals or residual protio solvent signals). 225 Table 3.34 lU N M R chemical shifts (ppm) of trimer cavitand 126 and complex 126*116 in nitrobenzene-^ (500 M H z ) Proton 125 125*116 Hi h 7.39 (d) H 2 7.76 (d) 7.45 (d) ArH (feet and benzyl) 7.28 (d), 7.18 (m), 7.09 (m) 7.25-7.04 (m) Hp 7 .81(s) ,h ,h 7 .72(s) ,h ,h H 0 6.19(d), 6.08(d) 6.18 (d), 6.18 (d) H a c 5.85 (d), 5.09 (d) 6.18 (d), 6.10 (d) H b 5.42 (s) 5.09 (s) H m 5.38 (t), 5.25 (t) 5.13 (t), 5.13 (t) H b ' 5.09 (s) 5.29 (s) Hi 4.66 (d), 4.37 (d) 4.68 (d), 4.64 (d) CH3(cap) 3.06 (s) 2.91 (s) CH 2s (phenyl ethyl feet) 2.88 (m), 2.82 (m), 2.76 (m) 2.91 (m), 2.81 (m) guest - 6.79 (s), -0.56 (s) h = hidden (either under other host signals or residual protio solvent signals). Table 3.35 *H N M R chemical shifts (ppm) of trimer cavitand 114 and complex 114*116 in nitrobenzene-^ (400 M H z ) . Proton 126 126*116 OH 6.61 (s) 6.78 (s) Ar H (feet and benzyl) 7.39-7.01 (m) 7.39-7.01 (m) H p 7.79 (s), 7.69 (s), 7.47 (s) 7.42 (s), h, h H„ 6.21 (d), 6.08 (d) 6.22 (d), 6.22 (d) H a c 6.02 (d), 5.94 (d) 6.22 (d), 6.15 (d) H m 5.22 (t), 5.15 (t) 5.22 (t), 5.15 (t) H b ' 5.28 (s) 5.28 (s) Hi 4.66 (d), 4.43 (d) 4.74 (d), 4.43 (d) C H 3 (cap) 2.95 (s) 2.92 (s) CH 2s (phenyl ethyl feet) 3.05-2.73 (m) 2.92 (m), 2.81 (m) guest - 6.88 (s), -0.53 (s) h = hidden (either under other host signals or residual protio solvent signals). 226 Table 3.36 Additional data fovKs measurements for complexes 114»116,125*116 and 126»116 in nitrobenzene-fife at different temperatures (7). >st ) Guest (G) T(K) [H] total (mM) [G] total (mM) [H.G]:[Hf [H.G]:[G]* Vtotal QiL) (M" 1 ) 114 116 300 2.81 5.04 1.49 0.21 422 177a 114 116 300 2.35 4.60 2.57 0.24 420 233b 114 116 305 5.92 5.16 1.04 0.72 500 192a 114 116 300 1.27 1.29 - 0.21 408 210 b 125 116 300 2.17 4.60 3.42 0.58 420 1180b 125 116 315 5.67 8.22 4.42 1.13 502 1076 b 125 116 330 5.67 8.22 5.00 1.14 502 1206 b 126 116 300 2.65 2.18 0.43 0.46 399 249 b •Determined from integration of *H N M R spectrum. a400 M H z . b 500 M H z . 227 3.9 References 1. (a) Sherman, J. C ; Knobler, C. B . ; Cram, D . J. J. Am. Chem. Soc. 1991,113, 2194-2204 (b) Chapman, R. J.; Chopra, N . ; Cochien, E . D . ; Sherman, J. C . J. Am. Chem. Soc. 1994, 116, 369-370. (c) Chapman, R. J.; Sherman, J. C. Org. Chem. 1998, 63, 4103-4110. 2. Chopra, N . ; Sherman, J. C . Supramol. Chem. 1995, 5, 31-37. 3. Makeiff, D . A . ; Pope, D . J. ; Sherman, J. C . J. Am. Chem. Soc. 2000,122, 1337-1342. 4. Cram, D . J. ; Cram, J. M . Container Compounds and Their Guests; The Royal Society o f Chemistry: Cambridge, 1994. 5. Gibb, C . L . D . ; Stevens, E . D . ; Gibb, B . C. Chem. Commun. 2000, 363-364. 6. Naumann, C ; Place, S.; Sherman, J. C. J. Am. Chem. Soc. 2002,124, 16-17. 7. Chopra, N . ; Sherman, J. C. Angew. Chem. Int. Ed. Engl. 1997. 36, 1727-1729. 8. Chopra, N . ; Naumann, C ; Sherman, J. C. Angew. Chem. Int. Ed. Engl. 2000, 39, 194-196. 9. Chopra, N . ; Sherman, J. C. Angew. Chem. Int. Ed. Engl. 1999, 38, 1955-1957. 10. Chopra, N . Ph.D thesis, University of British Columbia. 11. The following terminology w i l l be used for carceplexes for the remainder of this thesis (Chapters 3-5). 56»guest refers to carceplexes with only single-molecule guests. . 56«(guest) x refers specifically to carceplexes with two or more (x = 2, 3, 4,...) copies of the same guest (i.e., 56»(DMA)2). 56»guests is general to all carceplexes, where guests may be one or more copies o f the same guest, or one or more different guests. 12. Tetrol 8a was made using a literature procedure in ~11 % overall yield starting from the commercially available resorcinol. For the original procedure, see Sherman, J. C ; Knobler, C . B . ; Cram, D . J. J. Am. Chem. Soc. 1991,113, 2194-2204. 228 13. Naumann, C. unpublished results. 14 Kodumoru, V . ; unpublished results. 15. Aside from being poor reaction solvents, butyrophenone, D M S O , acetophenone and hexanophenone are also modest templates in the formation of 56»guests. 16. Carceplex reactions in N F P at room temperature formed mixtures o f 56»(NFP)2 and 56«NFP (~2:1), which could not be separated. A t higher temperatures (i.e., 70 °C), the formation of 56«NFP was favored (56«(NFP) 2 :56«NFP = 15:85). 17. Cram, D . J.; Tanner, M . E . ; Knobler, C. B . J. Am. Chem. Soc. 1991,113, 7717-7727. 18. Robbins, T. A . ; Knobler, C . B . ; Bellew, D . R.; Cram, D . J. J. Am. Chem. Soc. 1994,166, 111-122. 19. Perrin, C . L . ; Dwyer, J. Chem. Rev. 1990, 90, 935-967. 20. 2D N O E S Y spectra of 56« (DMA) 2 show correlations between the sets o f guest ( D M A and H 2 O ) protons for each water complex with the same Hi signal o f the host, thus connecting each bound water to a single D M A guest within the same host. 21. The concentration of H 2 O (59.4 m M ) in H 2 O saturated CDCI3 was based on the mutual solubility (wt %) for C H C 1 3 : H 2 0 (99.891:0.109) measured at 31 °C, reported in : Stephen, H . ; Stephen, T. Solubilities of Inorganic and Organic Compounds V o l . 1, Part 1, Permagon Press, New York, 1963. 370. 22. Stepwise binding constants, K\ - 35 M " 1 and K2 - 18 M " 1 , were also determined for carceplex 56«(NMP) 2 (56»[(NMP) 2 «(H 2 0)j , ] , y = 1, 2) in H 20-saturated C D C 1 3 (300 K ) . 23. Benzene-capped trimer carceplexes 115»(DMF) 3 ,115«butyrophenone , and 115»1,3,5-tris(ethynyl)benzene were also synthesized in addition to 115»(DMA)2 (see experimental section). 24. For a few references in which protonation o f amine guests destabilizes the stability of the host-guest (guest = amines) complex in solution, see: Ibukuro, F. ; Kusukawa, T.; Fujita, 229 M . Am. Chem. Soc. 1998,120, 8561-8562. Marquez, C ; Werner, M . N . Angew. Chem. Int. Ed. 2001,40,3155-3160. 25. Pons, M . ; Mil le t , O. Prog. Nucl. Magn. Reson. Spectrosc. 2001, 38, 267-324. 26. 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. 27. Sherman, J. C ; Cram, D . J. J. Am. Chem. Soc. 1989, 111, 4527-4528. 28. Paek, K . ; Ihm, H . ; Y u n , S.; Lee, H . C ; No , K . T. J. Org. Chem. 2001, 66, 5736-5743. 29. Chapman, R. G . ; Sherman, J. C . J. Org. Chem. 2000, 65, 513-516. 30. O'Leary, B . M . ; Grotzfeld, R. M . ; Rebek, J. Jr. J. Am. Chem. Soc. 1997,119, 11701-11702 31. Pawar, D . M . ; Wilson, K . K . ; Noe, E . A . J. Org. Chem. 2000, 65, 1552-1553. 32. E l i e l , E . L . ; Wilen , S. H . Stereochemistry of Organic Compounds. Wiley-Interscience: New York , 1994,618-619. 33. Spellmeyer, D . C ; Grootenhuis, P. D . J.; Mil ler , M . D . ; Kuyper, L . G . ; Kol lman, P. A . J. Phys. Chem. 1990, 94, 4483-4491. 34. ! H N M R spectra (400 M H z ) o f 56*117 in C D 2 C 1 2 were recorded at 185 K . Severe line broadening of both host and guest prevented a meaningful interpretation. 35. The energy barrier, A G C * =14.1 kcal/mole, for the interconversion of H i 1 and H;" ( A v m a x = 142.8 Hz , 250 K ) was also measured from the coalescence temperature in CDCI3 (303 K , 500 M H z ) . This value is in agreement with the value determined by I D E X S Y . See experimental section for details of the calculation. 36. Drakenberg, T.; Sommer, J.; Jost, R. / . Chem. Soc, Perkin II1980, 363. 230 37. The activation energy of - 7.7 kcal/mole for 5 6 » ( D M S O ) 3 was estimated assuming a Av similar to that o f 56»(NFP»DMSO) and a coalescence temperature o f 185 K . 38. Chapman, R. G . ; Sherman, J. C. J. Am. Chem. Soc. 1999,121, 1962-1963. 39. 77?22S were calculated according to the following equation for the competition between 4'-ethylacetophenone«NFP and acetophenone»NFP were calculated using: f 4 ' -EtAcPhe • N F P ^ _ [56 • 4 ' -EtAcPhe • N F P ] [AcPhe] TR22 AcPhe • N F P [56 • AcPhe • NFP][4 ' -EtAcPhe] where 4'-EtAcPhe = 4'-ethylacetophenone and AcPhe = acetophenone. 40. Effective molarity, E M , is defined as the concentration of a catalytic group required to make an intermolecular reaction go at the observed rate of the corresponding intramolecular process. Quite often, EM-values correspond to concentrations that are physically unattainable by several powers o f 10. See Mandolini , L . Adv. Phys. Org. Chem. 1986, 22, 1-111. Kirby, A. Adv. Phys. Org. Chem. 1980,17, 183-278 41. (a) Chapman, R. G . ; Sherman, J. C . J. Am. Chem. Soc. 1995,117, 9081-9082. (c) Chapman, R. G . ; Sherman, J. C . J. Am. Chem. Soc. 1998,117, 9818-9826 42. Naumann, C ; Chopra, N . unpublished results. 43. Mungaroo, R.; Sherman, J. C. J. Chem. Soc, Chem. Commun. 2002 , 1672-1673. 44. The broad spectra for 43 occurs exclusively in nitrobenzene-ck. Sharp spectra o f 4 3 « 1 1 6 were recorded in C D C l 3 : C D 3 O D (95:5). 45. Chapman, B . Ph.D. thesis, University of British Columbia. 46. 5 % C D 3 O D in C D C 1 3 is required to get hexa-hydroxyl trimer 42 to completely dissolve. 47. To address solvent effects between TRs (determined in N F P ) and Kss, competitions guest competitions in the formation o f 56»guest were attempted in nitrobenzene between 118 and 119 (room temperature, 2 d). Unfortunately, no carceplex 56«guest products were formed under these conditions. However, another member in our group has recently 231 synthesized carceplex 56*nitrobenzene in neat nitrobenzene using different conditions: at (1) room temperature for 5 d, or (2) 80 °C for 2 d (Mungaroo, R. unpublished results). 48. I would like to thank Rajesh Mungaroo for useful discussions regarding the removal of the mesityl caps from carceplex 56*guests using T F A . 49. Tris-ester 117 was observed to form two complexes in nitrobenzene-^ by ' H N M R spectroscopy, 125*117 (partial) and 125*117 (full). In the former, 117 is only partially inside the host cavity: two esters moieties protrude into the cavity, while the third out the portal. This is based on the ester methyl proton signal for this species is at 0.38 ppm, which is very similar to the same protons for bis-ester 137, which come at 0.37 ppm. In the latter, 125*117 (full), the guest is completely inside the host cavity (i.e., all three ester moieties protrude into the three bowls). The chemical shift o f the methyl protons for this species is further upfield at 0.08 ppm. Stability constants for 125*117 (partial) and 125*117 (full) were measured to be 9 and 2 M " 1 , respectively. Similarly, partial and fully bound species were also observed to form between hexa-benzyl trimer 42 and tris-ester 117, which are in slow exchange at temperature below 300 K , in the less favored binding solvent, toluene-ag. 50. Shanan-Atidi, H . ; Bar -E l i , K . H . J. Phys. Chem. 1970, 74, 961-963. 51. Complexation for 127*116 occurs in the order of minutes (as opposed to milliseconds for 42*116 and 125*116), as equilibrium is reached after ~30 minutes upon addition of 116 to meta-xylyl 127 (in nitrobenzene-^ at 300 K ) . Decomplexation is also substantially slower for complex 127*116 than 42*116 and 125*116. M u c h higher temperatures (-420 K ) are required to detect exchange between free and bound 116 for 127*116 than for 42*116 and 125*116 (room temperature) using I D E X S Y . The activation energy for decomplexation (la) is estimated to be 19-20 kcal/mol. 52. (a) Vestal, M . L . Chem. Rev. 2001,101, 371-373. (b) M o s i , A . A . ; Eigendorf, G . K . Current Org. Chem. 1998, 2,154-155. (c) Jolliffe, K . A . ; Calama, C . M . ; Fokkens, R.; Nibbering, N . M . M . ; Timmerman, P.; Reinhoudt, D . N . Angew. Chem. Int. Ed. 1998, 37, 1247-1251. (d) Castro, J. A . ; Koster, C ; Wilkins , C. Rapid Commun. Mass Spectrom. 1992,5,239-241. 53. 1,3,5-tris(bromomethyl)benzene was synthesized in two steps from triethyl-1,3,5-benzene tricarboxylate. See Mitchel l , M . S.; Walker, D . - L . ; Whelan, J. ; Bosnich, B . Inorg. Chem. 1987, 26, 396-400. 232 54. l,3,5-tris(ethynyl)benzene was synthesized in two steps from 1,3,5-tribromobenzene using a reported procedure. See Weber, E . ; Hecker, M . ; Koepp, E . ; Orlia, W . / . Chem. Soc., Perkin Trans. 7/1988, 1251-1257. 55. Braun, S.; Kalinowski , H . -O. ; Berger, S. 100 and More Basic NMR Experiments. John Wi ley & Sons: New York, 1996, 112-117. 56. Friebolin, H . Basic One- and Two-Dimensional NMR Spectroscopy. John W i l e y & Sons: New York, 1993, 287-314. 57. For additional details on I D E X S Y experiments, see: (a) Naumann, C ; Roman, E . ; Peinador, C ; Ren, T.; Patrick, B . O.; Kaifer, A . E. ; Sherman, J. C . Chem. Eur. J. 2001, 7, 1637-1645. (b) Naumann, C ; Patrick, B . O.; Sherman, J. C. Tetrahedron 2002, 787-798. (c) Naumann, C ; Patrick, B . O.; Sherman, J. C. Chem. Eur. J. 2002, 8, 51X1-1123. (d) Naumann, C . Ph.D. thesis, University of British Columbia. 233 4. The Generation, Stabilization, and Ketonization of Acetophenone Enol as a Guest in Trimer Carceplex 56«Guests 4.1 Introduction 4.1.1 The Importance of Enols Enol intermediates are involved in numerous important organic processes involving carbonyl compounds subjected to acid or base catalysis. 1 In order to be able to predict and control the outcomes of these reactions, a detailed understanding of enol chemistry is fundamental. Formed from a variety o f organic (often photochemical) reactions, enols are the primary (kinetic) products that rapidly tautomerize to the more thermodynamically stable ketone tautomers. The energy barrier for this conversion is generally quite small; indeed, the study of many enols has been hampered by their transient existence, along with low equilibrium concentrations.1 Keto-enol tautomerism has been studied extensively (thermodynamic and kinetics) in work dating back well over one hundred years, and is important in the history o f many ideas about the nature of acid-base catalysis, for which it is considered a classical field o f physical organic chemistry. The importance o f enols also extends to many biological processes in which enols and their anions (enolates) are intermediates in racemizations, isomerizations, carboxylations/ decarboxylations, aldol and Claisen condensations, additions, eliminations, oxidations and reactions with molecular oxygen. 3 Many o f these reactions occur both in the presence and absence o f enzymes. 3 Enols have also been studied in organometallic systems.4 Metal-enol 234 complexes are believed to be involved as intermediates in both biochemical (i.e., in metallo-enzymes) 4 and chemical processes (e.g., the Wacker process).5 4.1.2 Simple Enols Simple enols have the shortest lifetimes and are defined as enols that have no special stabilizing functionalities, such as electron-withdrawing groups or hydrogen-bonding substituents.6 The double-bond substituents of "simple" enols are typically hydrogen, alkyl, or aryl groups. Therefore, 1,3-diketones, p-keto esters, nitro ketones, cyano ketones, sulfonyl ketones, phenols and other related derivatives are not simple enols. A number o f simple enols have been isolated as pure substances by kinetically "trapping" them by introducing large bulky substituents into the a and |3 v inyl carbons of the enol group. 6 Enols that do not possess bulky substituents P to the hydroxyl group are considerably less kinetically stable and cannot be isolated as pure substances under normal conditions. Structural information has been obtained on a variety of these simple enols in the gas phase or at low temperatures in solution by spectroscopy (microwave, IR, U V and N M R ) . 1 Solution state (low temperature) N M R methods are amongst the most powerful tools used and have yielded detailed structural information and lifetimes of transient simple enols. ' Under ambient conditions, the a-hydroxy radicals of v inyl enol 9 and acetophenone enol 1 0 have been detected from radical reactions i n acetone-do by ' H N M R from the effect o f chemically induced dynamic nuclear polarization ( C I D N P ) . 1 1 235 4.1.3 The Norrish Type II Photocleavage of Aryl Ketones One well-known method for generating simple enols from aryl ketones with y-hydrogens (i.e., butyrophenone 121) is the Norrish type II photocleavage reaction. 1 2 Photoexcitation o f 121 (k > 300 nm)) produces both singlet (1211) and triplet (1213) carbonyl states that interconvert v ia intersystem crossing (Scheme 4.1). 1 2 Abstraction o f a y-hydrogen then occurs exclusively from the triplet state, followed by intersystem crossing to form a singlet diradical (1381). Diradical 1381 can then: (1) revert back to the starting ketone 121, (2) cyclize to form a cyclobutanol (141), or (3) cleave to give acetophenone enol (139) and ethylene (140). In solution, the kinetic cleavage product (enol 139) normally tautomerizes rapidly to the more (thermodynamically) stable ketone (142). Disproportionation o f all 1,4-biradicals (generated by Norrish II cleavage) to starting material can occur by back abstraction of the hydroxyl hydrogen. 1 2 Efficient cleavage requires the appropriate geometry (transoid) o f the 1,4-biradical intermediate (Scheme 4.2). Stereoelectronic overlap of the rj-bond that w i l l be broken must occur between both half-filled 12 orbitals. Cyclization occurs when the biradical adopts a cisoid geometry in which both o f the half-filled orbitals overlap. 1 2 Although the different geometries rapidly interconvert for aryl ketones free in solution, this may not be the case when they are trapped in a confining environment, and may lead to different cleavage:cyclization product ratios (i.e., 142:141). 236 Scheme 4.1 Norrish II photocleavage/cyclization of butyrophenone (121). 4.1.4 Okazaki ' s Endohedra l E n o l Okazaki 's lantern-shaped endohedral molecule (144) with an "introverted" enol functionality is the only reported generation and stabilization of a simple enol that does not possess protective substituents P to the enol hydroxyl group. 1 4 In a "mother molecule-daughter molecule" inner phase reaction, enol 144 was generated by photolysis o f methylthiaacetal precursor 143. Norrish II cleavage gave "mother" molecule 139 and its "daughter", 237 thiaformaldehyde ( S C H 2 ) , which exits from the inner phase. 1 4 Tautomerization to the corresponding bis-meta disubstituted acetophenone, 145, occurred after three days i n CDCI3 at room temperature in the presence o f trifluoroacetic acid ( T F A ) . 1 4 The cavitand moiety in endohedral compound 144 provides a stabilizing effect similar to other kinetically stable enols with bulky substituents at the P position of the enol double bond. 4.1.5 Objectives of This Chapter The stability o f the enol in compound 144 inspired investigations into the generation and stabilization o f similar enols (i.e., acetophenone enol 139) within the confines o f trimer carceplex 56«guests (Scheme 4.3) as work for this thesis. It should be possible to generate kinetically stable simple enols within 56«guests (were guests = suitable enol precursors) v ia Norrish II photocleavage. Since trimer 42 was already available from the templation study (Chapter 3), carceplexes 56»guests encapsulating suitable enol precursors could be prepared with relative ease. These compounds (56«guests) may deliver the first indefinitely stable "free-standing" enol 139 under ambient conditions. The simple enol in Okazaki's compound 144 is not "free-238 standing" as it is covalently attached to the host. In addition, we were also interested in examining the inner phase reactivity of entrapped enols as guests in 56»guests in much greater detail than was done for Okazaki's enol (144). These experiments turned out to be far more interesting than initially anticipated, and wi l l be the prime focus of this chapter. Scheme 4.3 Synthesis o f trimer carceplex 56»guests. K 2 C 0 3 42 56«guests 141 142 4.2 Synthesis and Character izat ion of T r i m e r Carceplexes wi th A r y l Ketone Guests Chapter 3 reported the encapsulation of the following aryl ketones as guests in trimer carceplex 56»guests: acetophenone, propiophenone, 4'-ethylacetophenone, butyrophenone (121), valerophenone, hexanophenone (123), and heptanophenone.1 5 Since 56»121 and 56*123 were 2 3 9 the only carceplexes of this group that could be easily obtained free o f other carceplex byproducts, they were the only ones photolyzed for this study. Carceplexes 56»121 and 56*123 were both isolated from separate A,C-trimer capping reactions in neat 121 and the solvent mixture NFP:123 (8:2), respectively, at temperatures above 80 °C (see experimental section, Chapter 3). 4.3 Photolysis of Carceplexes 56»121 and 56«123 4.3.1 Photolysis of Carceplex 56*121 Carceplex 56«121 was photolyzed (k > 300 nm, 5-6 h) i n degassed C6D6 solutions at ambient temperatures (see experimental) to yield carceplex mixtures of 56«(139»140) and 56»141 (Scheme 4.1). ! H N M R spectra of a sample in C6D6 containing crushed 4 A sieves before and after photolysis are shown in Figures 4.2a and b, respectively. The signals for bound 121 disappear (Figure 4.2a) and new guest signals for bound enol 139 and ethylene (140) in 56«(139»140) and bound 141 in 56*141 emerge (Figure 4.2b). Spectra were initially acquired in CeD6 as this was the photolysis solvent (i.e:, the photolysis could be done in a N M R tube). However, better signal-to-noise was obtained when spectra o f 56»(139«140)/56«141 were acquired in CDCI3, because solubility is greater in CDCI3 than in CeD6. Expansions of the spectral regions where the guest signals appear are shown in Figure 4.2c. Integration revealed a 5.6:1 ratio of 56»(139«140):56«141. This is not significantly different than the ratio of 6.3:1 (142:141) measured by *H N M R spectroscopy for our "control" photolysis of free 121 under the same conditions (k > 300 nm, 70 min) . 1 6 240 (a) (b) J) LLJ O H JtUUJ • 5.60 5.50 J 1.5 l"T" I 1 1 1 1 1 ( 1 1 ' T T " | I I I I I I I I I | I -2 (C) F T 11 I I I I I 1 I I - T - T T ' T T I 1 1 1 ' ' ' ' T " ' m 1 ' ' ' I ' [ ' T " | " " T " T " 1" I' I I I | I I I 1 T " I 1 I ' 1 | I ' 1 ' 1 1 1 1 '"P 3 , 2 (ppm) H i " O H H , , H 2 ' u H 3 " 1 2 J'V. H t a —1 r — T : 1 1 1 i 1 1 1 1 1 1 1 r- I 6.4 6.2 ' 6.0 5.8 5.6 5.4 5.2 5.0 2.0 1.8 I ! ! i I i i i I I 0.9 0.7 0.5 0.3 0.1 -0.1 (ppm) H O H „ H , - O H h H , ' O H I J. i_i i H H u " H t a 139 H H r 140 H , " LiH, 141 Figure 4.2 1 H N M R spectra of 56»121 and 56«(139«140)/56«141. (a) 56»121 before photolysis in C 6 D 6 . (b) 56«121 after photolysis (56»(139«140)/56«141) in C 6 D 6 . H 2 ' is hidden under the signal at 5.3 ppm (indicated by C O S Y ) , (c) 56«(139«140)/56«141 in CDCI3. 241 The aryl protons of both enol and cyclobutanol guests were assigned based on the characteristic coupling pattern for the spin system (AA'BB'Q o f a mono-substituted benzene ring. Hidden aryl resonances (i.e., for 141 in 56»141) were located by 2D C O S Y experiments. The geminal v inyl protons of the enol guest in carceplex 56«(139»140), H t a and H c a , appear at 1.73 and 0.94 ppm (Figure 4.3), respectively, which show coupling by 2D C O S Y . These protons in Okazaki 's enol (144) were reported at 3.33 and 0.51 ppm. 1 4 The geminal relationship between Hta and H c a o f 139 in carceplex 56»(139«140) was also confirmed in the ' H - 1 3 C H M Q C spectrum, which showed one-bond coupling between each with the same carbon. N O E s were also observed between H t a and H c a o f the enol guest (139) by 2D N O E S Y (Figure 4.3). The enol hydroxyl proton appears as sharp singlet at 5.31 ppm, which shows N O E s to both H t a and H c a . The vinyl proton signals were assigned as cis (H c a ) and trans (Hta) based on the relative intensities of N O E cross peak correlations with the O H signal. When the lower contour levels in 2D N O E S Y spectra are plotted, N O E s between the pairs OH(enol) /H c a , OH(enol) /H t a , Hi'/Hta and Hi ' /H c a are all present. However, at higher contour levels, only N O E s between OH(enol ) /H c a and Hi'/Hta are observed, suggesting that these pairs o f nuclei are situated closer in space than OH(enol) /H t a and Hi ' /H c a . The vinyl protons o f entrapped ethylene (140) are a sharp singlet at 2.01 ppm (Figure 4.3c) in CDCI3, which integrates to a total of four protons, and shows innermolecular N O E s to the same host protons ( H 0 and Hi) l ining the interior of the host cavity as the enol 139 in 56«(139«140) (see experimental section 4.7.2, Table 4.3). Also , a weak, but distinct N O E is observed between Hi' and ethylene (Hj, Figure 4.3), which confirms that enol 139 and 140 occupy the same carceplex chamber. For carceplex 56*141, three multiplets appear at 0.47, 0.14, and -0.13 ppm in a 3:2:1 ratio for the cyclobutanol methylene protons (Figure 4.2c, inset). The signal at 0.47 ppm consists o f 242 two overlapping multiplets for Hf and H g or H n . H g and H n could not be distinguished. H e and either Hh or H g appear upfield as multiplets at 0.14 and -0.13 ppm, respectively. 2D C O S Y coupling was observed between all four cyclobutane ring protons. The hydroxyl proton o f 141 is a sharp singlet at 2.14 ppm, which shows a weak N O E to H e by 2D N O E S Y . Figure 4.3 Intra and intermolecular N O E s observed between 139 and 140 in carceplex 56«(139»140). Photocleavage o f 56»121 was also monitored by IR spectroscopy. Carceplex 56»121 was photolyzed (k > 300 nm) in a K B r pellet for 90 min The IR spectrum recorded before photolysis shows a carbonyl stretch at 1684 cm"1 for guest 121, which is not present after photolysis. The absence o f a carbonyl stretch indicates that no ketone (e.g., acetophenone (142)) remains in the product mixture. The spectroscopic data above is consistent with carceplex 56»(139»140), and exemplifies the first reactive intermediate successfully generated and stabilized within the inner phase o f a carceplex (as opposed to a hemicarceplex). In addition, the first stabilization o f a free-standing simple enol under ambient conditions is also demonstrated. In the next few sections, the photolysis o f 56»123 wi l l be discussed, followed by the ketonization of 139 in 56»(139»140). 243 4.3.2 Photolysis of Trimer Carceplex 56*123 Carceplex 56*123 was photolyzed under similar conditions used for 56*121. N o changes were observed in ' H N M R spectra of a C6D6 solution of 56*123 that was irradiated (k > 300 nm, ~7 h) for longer than the time required to convert all o f free 121 and 56*121 to the corresponding photo-products. 1 7 Many factors may contribute to this (e.g., sensitization by the host), but most l ikely the geometry of the entrapped guest 123 in 56*123 is not ideal for Norrish II cleavage. To gain further insight into the guest geometries in both 56*121 and 56*123, J H N M R chemical shift data (A5s) and M M 2 minimized structures were analyzed. According to the AS (Sbound-Sfree) values summarized in Table 4.1, guests 121 and 123 are predicted to be oriented in the trimer cavity extended between two adjacent cavitand subunits (Figure 4.4). The arenes occupy one bowl, the methyl protons of the aliphatic chain extend into another, while the third bowl is unoccupied. This interpretation is based on the large AS values for the H 3 and H c o f both bound guests (Table 4.1). Smaller ASs are observed for (aliphatic methylene) protons in less shielded (central) regions of the host. Reorientation o f the two ends of the guest between different pairs of adjacent bowls is rapid with respect to the ' H N M R timescale. 244 Table 4.1 Chemical shift data (CDCI3, 300 K ) for guest protons of butyrophenone (121) and hexanophenone (123). Guest proton Sfree (ppm) S b o u n d (ppm) A S (ppm) H 1 ° H b Hi O H b H d H , 7.84 6.32 1.52 H 2 7.42 5.15 2.27 H 3 7.52 4.77 2.75 H a 2.93 0.94 1.99 H b 1.78 -0.02 1.80 H e 0.99 -2.28 3.27 H i 7.94 6.40 1.54 H 2 7.53 5.00 2.53 H 3 7.43 4.36 3.07 H a 2.96 1.06 1.90 H b 1.73 0.05 1.68 H c 1.35 -0.60 1.90 H d 1.35 -0.97 2.32 H e 0.89 -2.58 3.47 A S - Sfree-Sbound The guest orientations in M M 2 minimized structures calculated for 56*121 and 56*123 are in agreement with those predicted from A S values. In general, aryl ketones can potentially adopt two basic geometries in the ground state (A and B, Figure 4.4, bottom). 1 8 Geometry A features the abstractable y-hydrogen in close proximity to the carbonyl oxygen, while in B, the y-hydrogens are further away. In 56*121 and 56*123, M M 2 calculations favor geometry B over A by ~5 and ~20 kcal/mole, respectively. 1 9 Thus, for 56*121, one might expect a fairly low concentration of geometry B needed for Norrish type II cleavage, and therefore a slow reaction. For 56*123, one might expect essentially no reaction. Both expectations were met. 245 56.121(B) 56.121(A) 56.123(B) B Figure 4.4 M M 2 minimized structures of carceplexes 56*121 and 56*123. To simplify calculations, the pendant groups of the cavitand subunits were replaced with Hs. For clarity, the mesityl caps have removed to show the geometries of each entrapped guest inside the host. 4.4 The Stability of Acetophenone Enol in the Inner Phase of 56*(139*140) 4.4.1 Objectives for the Ketonization of 56*(139*140) The stability o f entrapped enol 139 was examined under a variety o f conditions to investigate i f ketonization to the more stable tautomer, acetophenone (142), is possible within the inner phase of 56*(139*140). I f 142 forms, we also wanted to look at what types o f conditions (i.e., presence o f water, acid, and base) are necessary to facilitate ketonization, as well as probing possible mechanisms for comparison with the analogous process studied extensively for free enol 139 in aqueous media. The next few sections w i l l discuss experiments performed to ketonize 246 entrapped enol 139 with H2O and D2O in nitrobenzene solution. This is followed by a separate study investigating the effects of acid (H + ) and base ( O H ) on ketonization in benzene solution. 4.4.2 Ketonization of Acetophenone Enol in Carceplex 56«(139»140) To compare the stability of the enol 139 in 56«(139«140) to Okazaki's enol 144, the ketonization of carceplex 56»(139»140) under similar conditions were initially explored. Clearly the pores o f the host shell i n 56«(139»140) are smaller than i n 144 since thiaformaldehyde (CH 2 =S) can exit 144 readily at room temperature, whereas the similar sized ethylene (140, CH2=CH2) cannot leave 56»(139»140) even at high temperatures. Okazaki 's enol 144 was reported to completely tautomerize to ketone 145 within days in the presence o f T F A in CDCI3. 1 4 T F A may react the enol 144 as a complexed guest, or in a through-shell reaction without completely entering the endohedral space of 144. However, T F A is larger than 140 and l ikely cannot enter 56»(139*140). When carceplex 56*(139«140) was subjected to T F A in C D C 1 3 at room temperature, very little 56*(142«140) was observed to form over long periods o f time (several weeks). Therefore, we needed a different set of conditions to influence the tautomerization o f the entrapped enol in carceplex 56«(139»140). Recall from Chapter 3 that trimer carceplexes 56«guests were reported to bind one or more water molecules within the same chamber as entrapped guests (i.e., 56»[guests*(H20) y]). In fact, only a single carceplex, 56»(DMF)3, did not show any sign o f water complexation. Evidence has been obtained for bound water in 56«(139»140) (see Figure 4.12, experimental section). I f a water molecule is held within close spatial proximity to the entrapped enol in 247 56»[139»140«(H2O)>,], one might expect that the tautomerization could be effected. That is, a proton source is required in order to facilitate ketonization. To investigate the ketonization of 56«(139«140) by water, initial reactions were conducted in dry and water-saturated CDCI3 solutions. For 56»(139»140), no tautomerization was observed to occur under dry conditions (vide infra). In water-saturated CDCI3, ! H N M R spectra showed very little changes over a period of several weeks. Eventually, we did get ketonization to occur with water at higher temperatures (80-100 °C) in solvents such as nitrobenzene or benzene. Since ketonization does occur, we were also able to monitor the rate under different conditions in nitrobenzene or benzene. Quantification of ketonization rates was not possible by direct methods using ! H N M R spectroscopy due to the formation of carceplex hydrates, which yielded complicated spectra. A procedure was developed to determine the rates: reaction solutions containing 56«(139«140) (~2.5 mg/0.5 mL) and appropriate amounts of water (and acid or base) were sealed in Pyrex glass tubes (5 x 80 mm, ~1 m L total volume) at room temperature prior to heating. Each reaction was heated for a period of time, and then rapidly cooled to room temperature. Reaction vessels were opened, and the product mixture was subjected to a brief workup (including drying o f sample in vacuo), which was followed by ( ' H N M R ) spectroscopic analysis (see experimental section 4.7.3). ' H N M R spectra (in CDCI3) before, half-way, and after full ketonization are shown in Figures 4.5a-c. Ketonization is marked by the disappearance of the enol guest signals (Figure 4.5a) and the emergence o f new signals for the ketone (Figures 4.5b and c). 248 56«(139«140) 56.141 56»(142«140) Figure 4.5 *H N M R spectra (400 M H z , C D C 1 3 , 300 K ) of carceplexes 56»(139«140)/56»141 after various stages o f tautomerization. (a) before tautomerization, (b) half-way, and (c) after complete tautomeration. 249 4.4.3 Ketonization in Nitrobenzene 4.4.3.1 The Rate of Ketonization with H 2 0 Rate constants for the ketonization of 56»(139«140) to 56«(142«140) by H 2 0 and by D 2 0 (section 4 . 4 . 3 . 2 ) were measured separately in nitrobenzene at 1 0 0 °C ( [ H 2 0 ] - 0 . 8 4 M ) . 2 0 In each case, ketonization appears to follow pseudo-first order kinetics, as indicated by the straight lines in the Arrhenius plots from which rate constants were extracted (see section 4 . 4 . 3 . 2 and experimental section 4 . 7 . 3 ) . The pseudo-first order observed rate constant ( £ 0 b s 1 0 0 ( H 2 0 ) ) for ketonization by H 2 0 in nitrobenzene at 1 0 0 °C is 1 .5 x 1 0 " 4 s"1 (tU2 = 7 8 min, A G * = 2 9 kcal/mole). 2 1 A t 2 5 °C, £ 0 b s 2 5 ( H 2 0 ) is estimated to be 7 . 4 x 1 0 " 9 s"1 (tm = 1 .6 x 1 0 6 min, see experimental section 4 . 7 . 3 ) . For the "uncatalyzed" reaction between free acetophenone enol (139) and water at 2 5 °C in aqueous media the rate constant (kuc) has been reported to be 0 . 1 8 s' 1. 2 3 Ketonization o f acetophenone enol (139) in 56«(139»140) with H 2 0 under the conditions for this study is over 2 . 4 x 1 0 7 times slower than in aqueous solution. Without H 2 0 , the incarcerated enol (139) is stable indefinitely (see Table 4 . 4 , experimental section). N o tautomerization was observed when solid 56»(139»140) was heated in vacuo between 1 1 0 - 1 3 0 °C (refluxing toluene) for several days. Also , no tautomerization was observed after 2 6 days when the reaction was conducted in sieve-dried nitrobenzene-fife at 1 0 0 °C. Therefore, ketonization with H 2 0 is estimated to be at least 7 . 5 x 1 0 3 times faster than in the absence o f H 2 0 . 2 4 2 5 0 4.4.3.2 The Rate of Ketonization with D 2 0 Ketonization o f 56«(139»140) with D 2 0 (nitrobenzene, 100 °C) is slightly more complicated than with H 2 0 . Examination o f the Arrhenius plot in Figure 4 .6 2 5 shows that the initial rate of ketonization does not appear to be constant. Within the first hour, the initial rate does not appear to be linear, which suggests that there is a change in mechanism during this stage of the reaction. After an hour, the rate slows down and appears to be constant, as indicated by the the straight line fitting the data . The slope o f the line fitting the data points at later times (t > 60 min) corresponds to the rate of ketonization of 56«(139-d«140) by D 2 0 , for which the rate constant £ 0 b S 1 0 0 ( D 2 0 ) is 2.3 x 10"5 s"1 (ty2 = 5.0 x 10 2 min, A G * = 30 kcal/mole, see experimental section 4.7.3). The change in the initial rate in Figure 4.7 is discussed in greater detail in the next three sections. Comparison o f the observed rates o f ketonization in the presence o f H 2 0 and D 2 0 reveals a strong primary isotope effect (& obs 1 0° (H 2 O)/^ o bs 1 0 ° (D2O) = 6.5). O - D bonds are broken (i.e., D O - D , H D O + - D , D 2 0 + - D , or enol O - D ) and/or C - D bonds are formed (i.e., at the |3-carbon of enol) during the key (rate-determining) step o f the reaction. 251 0.10 0.00 * 1 1 1 1 ! ! 1 — ' 0 50 100 150 200 250 300 350 400 time (min) Figure 4.6 Arrhenius plot (-ln(%e) versus time) for the ketonization of 56»(139«140) in D 2 0 saturated nitrobenzene at 100 ° C . 2 5 4.4.3.3 H/D Exchange of the Enol Hydroxyl Proton by D 2 0 in Nitrobenzene The rate of H / D exchange of the hydroxyl proton of enol 139 in 56«(139»140) was also determined from the same experiments with D 2 0 described in section 4.4.3.2 by measuring the disappearance o f the enol hydroxyl proton in ' H N M R spectra (Figure 4.7). The observed rate constant for H / D exchange o f the enol hydroxyl proton (fcobs'00 (H/D)) is 1.8 x 10"4 s"1 (ty2 = 64 min, A G * = 28 kcal/mole), which is approximately eight times faster than & 0bs 1 0 0 ( D 2 0 ) . This extrapolates to kohs25 (H/D) = 9.2 x 10"9 s"1 (tm = 1.3 x 10 6 min) at 25 ° C . (a) (b) (c) (d) (e) 5.58 5i6 554 5.52 5.50 5.48 5.46 5.44 5.42 (ppm) Figure 4.7 Disappearance o f the enol hydroxyl proton with time in the presence o f D 2 O . Expanded region of ' H N M R spectra (C^D^, 300 K ) of carceplex 56»(139»140) after various reaction times with D 2 O in nitrobenzene at 100 °C. (a) initial, 0 min, (b) 20 min, (c) 40 min, (d) 60 min, (e) 86 min and (f) after 180 min Note that (a)-(e) are from the same run (run 1, Table 4.6 in experimental section), while (f) was from a separate run (run 3, Table 4.6 in experimental section) under similar conditions. In the ketonization of enol 139 with H 2 O in aqueous media, deprotonation/protonation o f the enol/enolate is fast. To our knowledge, these rate constants have not been measured directly, however, the lower limit must be 0.18 s"1 (kac, vide supra).23 4.4.3.4 The Formation of Protio Acetophenone in Carceplex 56«(142»140) with D 2 O in Nitrobenzene at 100 °C A striking and unexpected feature in the ketonization o f 56»(139»140) with D 2 O (in nitrobenzene) is the observed formation of acetophenone (142 in 56*(142«140)) in addition to acetophenone-' (142-d in 56»(142-'»140). Two resonances, one sharp and one broad, appear 253 after ketonization with D2O, which have similar chemical shifts of 0.17 and 0.16 ppm, respectively (Figures 4.8b-d). The sharp signal at 0.17 ppm corresponds to the acetyl methyl protons (CH3) o f 142 2 6 (Figure 4.8a), while the broad signal at 0.16 ppm (believed to be an unresolved triplet) is from the acetyl methyl protons (C//2D) o f 142-d. 0.28 0.24 0.20 0.16 0.12 0.08 (ppm) Figure 4.8 Expanded region of ' H N M R spectra (CDCI3) o f carceplex of 56«(139«140) after heating at 100 °C under various conditions, (a) After 32 h in H2O saturated nitrobenzene, (b) After three hours in D2O saturated nitrobenzene-afe. (c) After 6.7 h in D2O saturated nitrobenzene, (d) After 32 h in D2O saturated nitrobenzene-ds. * = CH3 protons o f acetophenone (142). • = CH2D protons of acetophenone-d (142-d). The broad signal at 0.16 ppm is exclusively due to 142-d and not acetophenone-d2 (142-d2) or acetophenone-d3 (142-d3). The only possible way 142-d2 or 142-d3 could form under the reaction conditions (D2O, nitrobenzene, 100 ° C ) is via H / D exchange of the acetyl methyl protons o f pre-formed 142-d (or 142). In order to check the extent to which H / D exchange occurs with 142 (or 142-d), carceplex 56»(142»140) was subjected D2O conditions (nitrobenzene, 100 °C, 32 h). Integration of the *H N M R spectrum of the resulting mixture suggested that no detectable amount of H / D exchange had taken place. Therefore: (1) 142-d 254 must only form from the ketonization o f 139 with D 2 O , and (2) no 142-'2 or 142-03 form (the signal at 0 . 1 6 ppm is entirely due to 142-d). Over the course o f the reaction in the presence o f D 2 O , the following occurs. During the first hour, both 142 and 142-' form, with 142 forming at a faster rate (recall the isotope effect & o b s 1 0 0 ( H 2 O ) / ^ o b s 1 0 ° (D2O) = 6 . 5 ) and contributing more towards the overall rate o f formation of total ketone, 142 and 142-'. This is indicated by the steeper slope(s) in the first portion o f the graph in Figure 4 . 6 , and the greater amount of 142 observed to form within one hour. A l so , recall that JUS'00 (H./D)/kohsm ( D 2 0 ) ~8. A s more enol 139-' forms, enol 139 is depleted, and the rate o f formation o f 142 contributes less to the overall rate o f total ketone formed, which is evident in the change in the slope in the initial hour of the reaction. After an hour, H / D exchange o f the enol hydroxyl proton is approximately half complete and the observed rate o f ketonization slows down: the protonated water formed at this point contains a 5 :1 ratio o f D : H (i.e., D 3 0 + : D 2 0 H + = 1 : 1 , see section 4 . 4 . 5 ) . The overall rate of reaction is now approximately one-fifth of the rate o f £ 0 b s 1 0 0 (H2O) , which is nearly equal to & 0 b s 1 0 0 (D2O) , so no further change in rate is observed. After approximately five hours, production o f 142 ceases as H / D exchange is nearly complete ( > 9 5 %), and from then on only 142-' forms. 4.4.4 The Accepted Mechanism for the "Uncatalyzed" Ketonization of Acetophenone Enol in Aqueous Solution Tautomerization o f enols always occurs via intermolecular reactions; theoretical calculations have predicted that intramolecular 1,3-hydrogen shifts are unlikely to occur thermally as they are a symmetry forbidden process involving a highly strained transition state.1 2 5 5 Scheme 4.4 Proposed mechanisms for the "uncatalyzed" ketonization of acetophenone enol by water. The ketonization of acetophenone enol (139), generated from the Norrish II photocleavage o f 121 and substituted butyrophenones, or from the photohydration o f phenyl acetylene,1 has been studied extensively in acidic and aqueous media by Kresge and coworkers. 2 3 ' 2 7 Three basic mechanisms (equations 4.1-4.3, Scheme 4.4) have been considered for the "uncatalyzed" reaction (ketonization by H 2 O ) . 2 3 Two are stepwise mechanisms, while the third is concerted. In the first stepwise mechanism, enol 139 acts as an acid, protonating H 2 O to form the corresponding enolate (146) and H 3 0 + . H 3 0 + then delivers a proton to the o c a r b o n of the enolate (146) in the second and rate determining step, to form acetophenone (142). In the second stepwise mechanism (equation 4.2, Scheme 4.4), protonation of the P-carbon of the enol 139 by water to give the protonated acetophenone (147) and hydroxide ("OH) occurs first and is rate determining. Deprotonation o f 147 by hydroxide then follows with the rapid formation of acetophenone (142). Finally, the third mechanism (equation 4.3) is a concerted process involving a cyclic transition state, in which water acts as a solvent bridge (148), accepting the 256 enol hydroxyl proton while protonating the p-carbon of 139. Experimental data reported for the "uncatalyzed" ketonization of acetophenone enol was most consistent with the first stepwise mechanism (equation 4.1) involving enolate 146 and H3O formation. 4.4.5 The Mechanism for the Ketonization of Acetophenone Enol in 56»(139«140) by Water Ketonization of the encapsulated enol (139) by water (in nitrobenzene) in 56»(139«140) also occurs via equation 4.1 (Kresge's mechanism for the analogous process in aqueous solution). The formation of both 56«(142-<M40) and 56«(142«140) from 56»(139«140) when only D 2 O is present rules out the mechanisms in equations 4.2 and 4.3 (Scheme 4.4). These mechanisms would require H 2 O or H O D to be present to protonate the P-carbon o f 139 to form 142 in equations 4.2 and 4.3. H O D and H 2 O can only be generated via a secondary process involving D / H exchange from D 2 O , which is independent o f the mechanisms in equations 4.2 and 4.3. Once formed, complexation/decomplexation of H O D / H 2 O at 100 °C is probably on the order of seconds or less, which is much faster than the observed rate of ketonization. Therefore, any H 2 O or H O D formed inside carceplex 56«(139»140) would be immediately replaced with D 2 O , which is i n at least five hundred-fold excess, before protons could be incorporated onto the p-carbon o f the enol. Also , £ 0 b S (H/D) > £ 0 b s ( D 2 0 ) for 56«(139»140) is most consistent with the accepted mechanism for ketonization of 139 (equation 4.1, Scheme 4.4). For the mechanisms in equations 4.2 and 4.3, &0bs (H/D) would be expected to be equal to &0bs (D2O) , which is clearly not the case. In addition to being able to establish the mechanism o f ketonization for entrapped enol 139, we were also able to determine an additional wrinkle. In the first step, the reversible 257 complexation/decomplexation of D2O in the formation of II is on the order of seconds or less at 100 °C (Scheme 4.5a). This is much faster the formation of III, which is only eight times faster than the rate determining step (minutes at 100 °C). Once D 2 0 H + is formed, the enolate oxygen can be protonated or deuterated; or the (3-carbon o f the enolate can be protonated (III - » I V ) or deuterated (III - I V or III' - IV") to give V and V (Scheme 4.5). The proton delivered to the (3-carbon o f enolate 146 must come from the same D2O molecule that deprotonates enol 139. 2 8 Egress of D 2 0 H + from the inner phase must occur at a negligible rate compared to ketonization, otherwise 142 would not form (only 142-d). Scheme 4.5 Ketonization o f 56»(139»140) with water. Note that D2O can be interchanged with H2O and in principle, H O D . I I I I I I I V V I I I I I 1 I I " I I I ' I V " The remaining sections o f this chapter w i l l discuss experiments designed to test the effect o f acid or base on ketonization o f the enol in 56»(139»140). If catalysis is observed, it would certainly suggest that H + , H 3 0 + , and " O H can exist for a significant period of time within the 258 inner phase of carceplex 56*guests. Cationic guests (e.g., protonated amines and ferrocenium ions) have been reported to form complexes with hemicarcerands that are stable enough to be observed by ! H N M R spectroscopy. 2 9 Complexation of anionic species to neutral carcerand/hemicarcerand hosts has never been reported. 4.4.6 Ketonization in Benzene 4.4.6.1 Ketonization by H 2 0 Trichloroacetic acid ( T C A ) and tetrabutylammonium hydroxide ( T B A O H ) were added to 56»(139»140) in benzene (with and without water) to investigate the potential effects o f acid or base on the ketonization of encapsulated enol 139. T C A was used because o f its low pATa (0.70), 3 0 high boiling point, and good solubility in organic solvents such as benzene. Hydroxide was chosen as the base because o f its similar size to water. T B A O H was chosen over other forms of hydroxide (e.g., sodium or potassium salts) because o f better solubility in nonpolar organic solvents. The solvent was switched to benzene for these experiments because nitrobenzene reacts with hydroxide. Ketonization rates with water for nitrobenzene and benzene measured from single-point experiments are compared in Table 4.2. The rate is faster in benzene than in nitrobenzene: with similar amounts of water (110 m M ) , ketonization is estimated to be - 5 0 times faster in benzene (entries 5 and 6) than in nitrobenzene (entry 2). Clearly, the change in solvent has a significant effect on the rate, although the nature of the effect is not clearly understood. Ketonization o f 56»(139»140) by 110 m M water in benzene is estimated to be at least 190 times faster than in dry benzene. 2 4 ' 259 Table 4.2 The results of ketonization experiments under various conditions. E n t r y Solvent T (°Q [H20] (mM) [H+] (mM) [OH] (mM) t (h) % yield of ketone (Jfcobsxl08)/sa 1 nitrobenzene 1 0 0 8 4 0 b - - 3 8 2 1 6 0 0 0 ( 2 3 0 0 ) c 2 nitrobenzene 1 0 0 1 1 0 - - 9 8 2 6 0 ( 3 1 ) ° 3 benzene 8 2 . 5 1 8 0 b - - 3 3 5 4 0 0 0 0 4 benzene 8 2 . 5 1 8 0 b ~ 2 6 d - 3 3 1 3 9 0 0 0 5 benzene 8 2 . 5 1 1 0 - - 3 1 5 1 5 0 0 6 benzene 8 2 . 5 1 1 0 - - 8 3 6 1 5 0 0 7 benzene 8 2 . 5 1 1 0 1 1 0 e - 3 < 5 < 4 7 0 8 benzene 8 2 . 5 1 1 0 1 1 0 e - 2 4 2 3 3 0 0 9 benzene 8 2 . 5 1 1 0 5 5 e - 2 4 4 1 6 1 0 10 benzene 8 2 . 5 - 1 1 0 e - 2 4 < 5 < 5 9 11 benzene 8 2 . 5 1 1 0 - 1 0 0 3 1 0 9 8 0 12 f benzene 8 2 . 5 - - 1 0 0 2 4 < 5 < 5 9 13 f ' g benzene-ofc 8 2 . 5 - - - 1 6 8 < 5 < 8 aPseudo-first order observed rate constant (& 0 bs ) estimated from a single point calculation, using the Arrhenius equation, ln(%e) = kobst- b Based on the reported solubility o f water at the temperature, T.20 cBracketed values correspond to kobs extrapolated to 8 2 . 5 °C (see experimental section 4 . 7 . 3 ) . dTrifluoroacetic acid was used. "Trichloroacetic acid was used. fCrushed 4 A molecular sieves were added. gReaction was performed in a sealed N M R tube. 4.4.6.2 The Effect of A c i d In Chapter 3, the formation of water complexes with carceplex 56»guests were inhibited upon addition o f acid (or base) to water-saturated solutions of C6D6 and/or CDCI3. This was said to occur because hydronium ions are poorly stabilized within the inner phase o f 56«guests. Therefore, for the entrapped enol in 56»(139»140), the presence of H + (H30+) in the external solution is expected to have the reverse o f the catalytic effect on ketonization o f the entrapped enol compared to ketonization normally observed for the free enol in acidic aqueous 2 6 0 m e d i a . 2 3 ' 2 ™ Ketonization o f the incarcerated acetophenone enol (139) by water is definitely slowed in the presence o f acid (entries 7-10, Table 4.2). The rate in dry benzene with acid (110 m M ) does not appear to be any faster than in dry benzene without acid (entries 10 and 13, respectively, Table 4.2). 2 4 Wi th an equal amount of acid and water, the reaction is approximately five times faster than with just acid in dry benzene, but is slower five times than the reaction with just water (compare entries 5, 6, and 8, Table 4.2). When the ratio o f water to acid is doubled, the rate o f ketonization also doubles. Protons (H + ) apparently compete for water with the trimer carceplex cavity, reducing the concentration of bound water available for reaction with the enol. 4.4.6.3 The Effect of Base Hydroxide had very little effect on the rate of ketonization o f the enol in carceplex 56«(139«140). No differences in rates are distinguishable between reactions with and without hydroxide in dry benzene (entries 11 and 12, respectively, Table 4.2). When an equal amount o f water is present with hydroxide (110 m M each), the reaction is only slightly retarded (entry 11, Table 4.2) in comparison to that for a similar reaction without base (entry 5, Table 4.2). Hydroxide ions either do not diffuse into the inner phase of 56«guests at all (because o f their instability in such an environment or because they prefer to associate tightly with the bulky tetrabutylammonium counter ions), or do, but cannot remain long enough to react with the residing enol. These results are consistent with the findings in Chapter 3: hydroxide, generated in situ from l,8-diazabicyclo[5.4.0]undec-7-ene ( D B U ) base and water, appears to compete with the carceplex for water, as does acid, but is less effective in this regard. Nonetheless, this 261 provides further support that charged species cannot be sufficiently stabilized within the inner 90 phase o f carceplexes and hemicarceplexes. 4.5 Attempts To Generate Other Enols Inside Carceplex 56»Gues ts Another method for generating aryl enols (1-phenylethenols) is by photohydration o f aryl acetylenes in water or dilute aqueous acids. 3 1 Recall that carceplex 56»116 had already been prepared from the templation studies reported in Chapter 3. Water also forms complexes with this carceplex, and therefore the photoreactivity o f the complexes 5 6 « [ 1 1 6 » ( H 2 0 ) J was investigated to see i f the tris-enol 146 (Scheme 4.6) could be generated inside 56»guest. In separate experiments, degassed water saturated and dry CeD6 solutions were photolyzed in sealed N M R tubes for up to sixteen hours (k > 300 nm), followed by analysis by ! H N M R spectroscopy. Although the recorded spectra showed evidence o f a reaction, both "wet" and "dry" experiments essentially gave the same results. This suggested that the acetylenic pi-bonds o f the guest 116 probably underwent cycloadditions with the arenes o f cavitand subunits of the host. N o further attempts were made to characterize the product mixtures, as it was evident that some type of undesired innermolecular cycloaddition had taken place. 262 Scheme 4.6 Photohydration of 116. 116 146 4.6 Summary The first use o f a carceplex for the stabilization of a reactive species as a permanently entrapped guest has been demonstrated in this Chapter. Acetophenone enol (139) was generated photolytically from an aryl ketone precursor (121) within the interior of carcerand 56 in a similar fashion to how Okazaki's group produced the "introverted" aryl enol functionality i n the cavitand derivative 139.14 The main difference between carceplex 56»(139»140) and 144 is that the enol in 56«(139»140) is not covalently attached to the host. In addition, the ketonization process in 56»(139«140) has been studied in more detail. The incarcerated enol in 56»(139»140) was found to be remarkably stable in organic solutions at ambient conditions, even in the presence of saturating amounts o f water. In addition, the hydroxyl group of the entrapped enol does not readily exchange its proton, even when in close proximity to water in the carceplex cavity. However, no information regarding the relative positions o f the enol 139 and bound water in the host was obtained experimentally. Clearly, the inner phase of the trimer carceplex 56«guests provides a unique reaction environment for its entrapped guests. Ketonization o f the entrapped enol 139 to acetophenone (142) by water was also demonstrated to occur by the same accepted mechanism for free 139 in aqueous buffer solutions. Water enters the carceplex and removes a proton from the enol to form the corresponding enolate 263 and a hydronium ion ( H 3 0 + , D 2 H O + , or D 3 0 + ) . The same molecule o f water ( H 2 0 or D 2 0 ) that deprotonates the hydroxyl o f the enol, delivers the proton (or deuteron) to the p-carbon o f the enolate (in the rate determining step) to give the thermodynamically more stable acetophenone (142 or 142-d). The work presented in this Chapter also demonstrated that the presence o f acid retards ketonization o f 56«(139»140) due to the acid-driven removal of water from the cavity, whereas acid accelerates ketonization of free 139 in aqueous media. The enol in Okazaki 's compound 144 was observed to slowly (three days) ketonize to the acetyl derivative 145 in the presence o f T F A in C D C 1 3 at room temperature.1 4 It was not clear how much ( i f any) water was involved in their experiments, 1 4 so comparison with our system is limited. Finally, hydroxide also was found to slightly retard ketonization of enol 139 in 56«(139«140). These results support previous conclusions regarding to the poor ability of the inner phase of cavitand derived hosts, namely carceplexes and hemicarceplexes, at stabilizing (solvating) charged guests. 264 4.7 Exper imenta l 4.7.1 Genera l A l l chemicals were purchased from the Aldr ich Chemical Company. Butyrophenone, hexanophenone and nitrobenzene were distilled separately under reduced pressure prior to use. Benzene was distilled over C a H prior to use. Nitrobenzene and benzene were both stored over activated 4 A molecular sieves under N 2 in separate sealed flasks immediately after distillation. Deionized H 2 0 was used for all experiments involving H 2 0 . 100 % D 2 0 was used for all experiments involving D 2 0 . A l l other chemicals were used as purchased unless otherwise stated. N M R spectra were recorded on Bruker A M X 500 and Avance 400 series spectrometers. Mass spectra were recorded on a Kratos Concept II H Q (DCI) and a V G Tofspec in reflectron mode ( M A L D I ) . 2,5-dihydroxybenzene ( D H B ) was used as the matrix for all carceplexes prepared in this Chapter. Note that the mass reported for the carceplex mixture 56*(139*140):56*138 is for the corresponding silver adduct, generated by doping the D H B matrix with silver triflate (-100 equiv. per mole of analyte). See the experimental section o f Chapter 3 for more details about M A L D I . Computer modeling on 56*121, 56*123, and 56*(139*140) was done using the software CS Chem 3D Pro (ver. 3.5). A Hanovia 450W medium pressure Hg-arc lamp equipped with a pyrex filter (>300nm, cutoff < 290 nm) was used for all photolysis reactions. I would like to thank the Scheffer group for their assistance and the use of their photolysis apparatus; and Vishnumurthy Kodumoru for his preliminary work on this project, and for the materials that he prepared (carceplex 56*121). 265 4.7.2 Synthesis and Characterizat ion 56, R1 = CH3, R = Ch^Ch^Cgh^ Carceplex mixture 56»(139«140)/56»141. 56«butyrophenone (56»121) was dissolved in either CDCI3 or C(,D(, (2-3 mg/0.5 mL) and transferred to a 5 mm diameter Nore l l HP-100 borosilicate glass N M R tube. The sample was thoroughly degassed via the freeze-pump-thaw method, sealed under vacuum, and then photolyzed (k > 300 nm) for a duration of 5-6 hours. The resulting product mixture was determined to contain an 85:15 ratio of carceplexes 56«(139»140):56«141 according to integration of ' H N M R spectra. Larger scale reactions (i.e., 30-50 mg) were conducted in 10 m L Pyrex tubes. Also , product mixtures from larger scale reactions required column chromatography on silica gel ( T L C grade, 40 ^im) eluting with C H C ^ h e x a n e s (2:1). Precipitation from CHCl 3 /hexanes gave 56»(139»140)/56»141 as a white solid. Carceplex 56«(139«140). ' H N M R (500 M H z , CDCI3) 8 7.34-7.13 (m, 60H, A r H (feet)), 7.01 (s, 6 H , H p i ) , 6.77 (s, 6 H , H p 2 ) , 6.34 (d, 2H , Hi ' ) , 5.88 (d, 12H, H 0 ) , 5.69 (s, 6 H , H a c ) , 5.31 (s, 1H, OH) , 5.24 (s, 12H, H x ) , 5.19 (t, 1H, H 3 ' ) , 5.04 (dd, 2 H , H 2 ' ) , 4.91 (t, 12H, H m ) , 4.10 (d, 12H, Hi), 2.86 (s, 18H, CH3 266 (cap)), 2.70 (t, 24H, CH2CH2C6]i5 (feet)), 2.52.(m, 24H, C/feCFfeCeHs (feet)), 2.01 (s, 4 H , Hj), 1.73 (s, 1H, H t a ) , 0.94 (s, 1H, H c a ) . 2D C O S Y , N O E S Y , and H M Q C spectra were also recorded. 139 140 M S ( M A L D I ) m/z (rel intensity) 3656 ((M»C 22iH208O 37 + A g + ) + ; 100), calcd for C 22iH208O 37»Ag + = 3656. Carceplex 56*141. ' H N M R (500 M H z , CDC1 3 ) 8 7.34-7.13 (m, 60H, A r H (feet)), 7.01 (s, 6 H , H p i ) , 6.79 (s, 6 H , H p 2 ) , 6.30 (d, 2 H , Hi"), 5.88 (d, 12H, H 0 ) , 5.69 (s, 6 H , H a c ) , 5.21 (s, 12H, H x ) , 4.91 (m, 14H, H m and H 2 " ) , 4.52 (t, 1H, H 3 " ) , 4.05 (d, 12H, HO, 2.84 (s, 18H, CH3 (cap)), 2.70 (t, 24H, CH2C//2C6H5 (feet)), 2.52 (m, 24H, C//2CH2C6H5 (feet)), 2.14 (s, 1H, OH) , 0.47 (m, 3H, H f and H h or Hg), 0.14 (m, 2 H , H e ) , -0.13 (m, 1H, H h or Hg). 2D C O S Y , N O E S Y , and H M Q C spectra were also recorded. M S ( M A L D I ) m/z (rel intensity) 3656 ( (M»C 2 2iH 2 08O 3 7 + A g + ) + ; 100), calcd for C 2 2 iH 2 08O 3 7«Ag + = 3656. 267 Table 4.3 2D N O E S Y data for carceplexes 56»(139»140) and 56-141. Carceplex Protons NOEs 56«(139»140) H , H 3 ' H 2 ' H i H t a O H (139) H 0 , O H (w), H x (w), H 3 (w), H 2 ' , C H 3 (cap), H i , H t a , H ca H i ' (w), H 0 , H i , H t a (w), H c a H 2 ' , H i , C H 3 (cap) H i ' , H 0 , H 3 ' , H i , C H 3 (cap) H i ' (w), H 2 ' , H 3 ' , O H , H o , H i H i ' , H 0 , O H (w), H i , H c a H i ' (w), H o (w), O H , H t a 56*141 O H (141) H e H f , H g or H h H g or H h H e (w) O H ( w ) , H f , H g or H h H e , Hh or H g Hf, H g or Hh Carceplex 56«(142«140). Procedure A . Carceplex mixture 56»(139«140)/56«141 (-2.5 mg, 0.7 | imol , ~1.4 m M ) was dissolved in dry nitrobenzene (500 p:L) and transferred into a 80 x 5 mm pyrex tube. H 2 0 (1-25 |xL) was carefully added via syringe, and the tube was flame-sealed (Note for safety, the reaction mixtures were frozen by dipping the reaction vessel in l iquid N 2 before flame sealing). The solution was shaken vigorously before completely submerging horizontally in a bath of boiling water (100 °C). The vessel was initially tipped several times in the boiling solvent to ensure good mixing. After heating overnight, the mixture was immediately cooled with cold tap water, upon which the solution became cloudy. The tube was opened and the mixture was added to 3 m L o f M e O H , which gave a white precipitate. The white solid was centrifuged into a pellet and the filtrate was decanted away and discarded. Precipitation was then repeated using CHCl 3 /hexanes. The remaining white solid was dried for at least 3 h at 110-130 °C in vacuo to quantitatively yield carceplex 56»(142«140) as a mixture with the unchanged carceplex 56*141. 268 Carceplex 5 6 » ( 1 4 2 « 1 4 0 ) *H N M R (400 M H z , CDC1 3 ) 5 7.34-7.13 (m, 60H, A r H (feet)), 7.00 (s, 6H , H p l ) , 6.78 (s, 6H , H p 2 ) , 6.42 (d, 2 H , H , m ) , 5.90 (d, 12H, H 0 ) , 5.72 (s, 6H , H a c ) , 5.41 (dd, 2 H , H 2 m ) , 5.20 (s, 12H, H x ) , 4.98 (t, 1H, Ha'"), 4.91 (t, 12H, H m ) , 4.11 (d, 12H, HO, 2.84 (s, 18H, CH3 (cap)), 2.70 (m, 24H, C H 2 C 7 7 2 C 6 H 5 (feet)), 2.51 (m, 24H, C i / 2 C H 2 C 6 H 5 (feet)), 1.96 (s, 4 H , Hj'), 0.18 (s, 3H , H c ' ) . 2D C O S Y and N O E S Y spectra were also recorded. 142 143 M S ( M A L D I ) m/z (rel intensity) 3571 ( ( M « C 2 2 i H 2 0 8 O 3 7 + N a + ) + ; 100), calcd for C 2 2 i H 2 0 8 O 3 7 » N a + = 3571. 3587 ( ( M » C 2 2 1 H 2 0 8 O 3 7 + K + ) + ; 100), calcd for C 2 2 I H 2 0 8 O 3 7 « K + = 3587. 4.7.3 Ketonization Experiments Single-time kinetic runs with H 2 0 . A single sample o f 5 6 » ( 1 3 9 « 1 4 0 ) / 5 6 « 1 4 1 was subjected to procedure A once, for the specified times, temperatures, and concentrations o f H 2 0 in Table 4.4 (entries 3-7). The relative amounts o f enol and ketone were measured by integration o f each set o f unique guest signals in the ! H N M R spectra o f the product mixtures containing carceplexes 5 6 » ( 1 3 9 » 1 4 0 ) , 5 6 » ( 1 4 2 » 1 4 0 ) , and 5 6 « 1 4 1 . Psuedo-first order rate constants (£ 0bs (min"1)) were estimated from single data points using the Arrhenius equation, -ln(Xe) = K ^ t , where % e is the mole fraction of enol 139. At e ne and tik are the relative moles of enol and ketone, respectively. Note that the units for £ 0 D S (min 1 ) were converted to s"1 by dividing by 60. Experiments were performed in both nitrobenzene (Table 4.4) and benzene (entries 3, 5, and 6, Table 4.2). Free energies of activation A G * (kcal/mole) were calculated from measuring rate constants (k) using equation 3.15 (see experimental section 3.8.4.3). Single-time kinetic runs with D 2 0. Same as above, except D 2 0 was used instead of F f 2 0 (runs 1, 8, and 9 in Table 4.4). Table 4.4 Additional data for single time (t) ketonization experiments with H 2 0 in nitrobenzene at 100 °C. Entry [H20]/M [D20]/M [H+]/M t Temp % yield o^bs (s ) xlO8 (u,L added) (p:L added) (h) ( ° Q ketone 1 >1.96 a ' b ' c(25) 1 167 66 30000 - - - 614 100 <5 <2 3 0.84 a (25) - - 3 100 82 16000 4 0.55 (5) - - 3 100 54 7200 5 0.44 (4) - - 3 100 40 4700 6 0.33 (3) - - 3 100 18 1800 7 0.11(1) - - 9 100 8 260 8 - 0.84 a , c (25) - 2.5 100 18 2200 9d - 0.84 a ' c (25) - 3 100 17 1700 10e - - 0.08 f 3 100 <5 <470 l l e - - 0.08 f 14.5 100 <5 <99 12e _ - 1.30f 0.5 100 X -13e _ - 0.08 g 26 100 <5 <55 14e - - 0.08 g 15.5 100 <5 <92 a M a x i m u m concentration of H 2 0 is based on calculations using reported solubility data. "Reaction was performed in a sealed N M R tube. C D 2 0 solubility in nitrobenzene was taken to be the same as for H 2 0 . d Nitrobenzene- ' 5 was used. eCrushed 4 A molecular sieves were added. Trifiuoroacetic acid was used. ^Trichloroacetic acid was used. % = sample decomposed. 270 Experiments with acid (in benzene). Procedure A , except that the specified concentrations of H 2 O , trifluoroacetic, and trichloroacetic acid were present (entries 4, 7-10, Table 4.2). Experiments with base (in benzene). Procedure A , except the specified concentration o f H 2 O and tetrabutylammonium hydroxide solid were present (entries 11 and 12, Table 4.2). Tetrabutylammonium hydroxide solid was prepared as follows. 40 % tetrabutylammonium hydroxide solution in water was concentrated by rotary evaporation in vacuo at 40 °C until a clear and colorless viscous solution was obtained. The viscous material was dried in vacuo for at least 12 h to afford a fluffy white solid. The white solid (-134.1 mg, -515 umol) was dissolved in dry benzene (5 m L , -103 m M ) . The solution was stored over activated 4 A molecular sieves in a dessicator for at least 6 h prior to use. Multiple-time kinetic runs with H 2 0. For each run, a single sample was subjected to procedure A , for the specified times in Table 4.5. Pseudo first-order observed rate constants were obtained from the slopes o f the Arrhenius plots in Figure 4.9. 271 Table 4.5 Additional data for multiple time (0 ketonization experiments with H 2 0 in nitrobenzene* at 100 °C. R u n i (min) l a 0 1.00 0.00 0.00 20 0.80 0.20 0.22 82 0.47 0.53 0.76 113 0.35 0.65 1.05 144 0.30 0.70 1.20 2 b 0 1.00 0.00 0.00 59 0.60 0.40 0.51 75 0.50 0.50 0.69 108 0.38 0.62 0.97 123 0.36 0.64 1.02 143 0.25 0.75 1.39 3 C 0 1.00 0.00 0.00 30 0.78 0.22 0.25 62 0.66 0.34 0.42 92 0.55 0.45 0.60 122 0.43 0.57 0.84 154 0.33 0.77 1.11 *[H 2 0] = 0.84 m M (25 uX H 2 O/500 u L nitrobenzene) based on calculations using reported solubility data. 2 0 aTrifluoroacetic acid (0.078 M ) was present. b25 (XL o f tetrabutylammonium hydroxide in H 2 0 (40 % v/v) was added instead o f 25 u L deionized H 2 0 . °Sample used was 56«(139-d»140), after subjecting run 1 in Table 4.5. 272 1.6 1.4 160 time (min) Figure 4.9 Arrhenius plots (-ln(Xe) versus time) for the ketonization o f 56«(139»140) with H 2 O in nitrobenzene at 100 °C. • = run 1 (Table 4.5), slope = kohs ( H 2 0 ) = (0.0085 ± 0.0014) min" 1 , r 2 = 0.992. - = run 2 (Table 4.5), slope = kohs ( H 2 0 ) = (0.0092 ± 0.0016) min" 1 , r 2 = 0.984. Error limits for k0bs are based on 95 % confidence limits from regression analysis. Although acid and base were present in runs 1 and 2, respectively, the concentrations o f each were not significant enough to effect rates. This was determined by comparing the rates in each run with the data for entry 3 in Table 4.4 ( £ 0 b s 1 0 0 ( H 2 0 ) = 0.0095 min"1). Therefore, these rates are taken to be the rates just due to H 2 0 . Mul t ip le- t ime kinetic runs wi th D 2 0 . Same as the above experiments for H 2 0 , except D 2 0 was used instead o f H 2 0 . The data for each run (three total) is shown in Table 4.6. Arrhenius plots are shown in Figures 4.6 (run 1) and Figure 4.10 (runs 2 and 3). Calcula t ion of 56»(142«140):56»(142-rf»140). The ratio 56»(142»140):56»(142-</«140) (i.e., % CHy.% C / / 2 D ) was measured from the integration in ' H N M R spectra o f samples from the reactions conducted in Table 4.6. This was done by comparing the total combined intensity 273 o f acetyl methyl proton signals at 0.17 and 0.16 ppm (m) relative to the intensity o f the signal for H i 1 " (n, for 2Hs). The ratio of 142 (x, 3H):142-d (y, 2H) was calculated from min, which is equal to: (3jc-2.y)/2 = min Since x + y = \,x andy were solved by substitutingy = x -1 into the above equation. Table 4.6 Additional data for ketonization experiments with D 2 0 in nitrobenzene* at 100 °C. R u n time (min) %e %k -lnxe % CH3 % CH2D 0 1.00 0.00 0.00 0 0 20 0.91 0.09 0.09 n/a n/a 40 0.88 0.12 0.13 n/a n/a 61 0.84 0.16 0.17 n/a n/a 86 0.82 0.18 0.20 44 56 176 0.74 0.26 0.30 52 48 266 0.67 0.33 0.40 76 24 311 0.63 0.37 0.46 34 66 401 0.58 0.42 0.54 52 48 150 0.80 0.20 0.22 32 68 304 0.65 0.35 0.43 71 29 455 0.55 0.45 0.60 40 60 575 0.43 0.57 0.84 33 67 726 0.37 0.63 0.99 29 71 180 0.71 0.29 0.34 64 36 305 0.57 0.43 0.56 48 52 455 0.50 0.50 0.69 24 76 635 0.38 0.62 0.97 10 90 180 0.71 0.29 0.34 26 74 1740 0.13 0.87 2.04 10 90 *Solubility o f D 2 0 was assumed to be the same as that for H 2 0 ( [D 2 0] -0.84 m M , 25 u L D 2 O/500 \iL nitrobenzene)). 2 0 aReaction contained TFA-d (0.078 M ) . ''Nitrobenzene-cis was used instead of nitrobenzene. 274 1.20 400 time (min) 500 600 700 800 Figure 4.10 Arrhenius plots (-ln(%e) versus time) for the ketonization o f 56»(139»140) with D 2 0 in nitrobenzene at 100 °C. - = run 2 (Table 4.6), slope = A : o b s 1 0 0 ( D 2 0 ) = (0.0015 ± 0.0003) min" 1 , r 2 = 0.992. • = run 3 (Table 4.6), slope = £ O D S 1 0 0 ( D 2 0 ) = (0.0013 ± 0.0002) min" 1 , r 2 = 0.987). Error limits for £ 0bs are based on 95 % confidence limits from regression analysis. H/D Exchange. H / D exchange of the enol hydroxyl proton in 56«(139»140) was measured from the integration intensity of the O H signal relative to the intensity of the H t a signal (1H) in *H N M R spectra ( C 6 D 6 ) . %OH (mole fraction of enol with O H ) was calculated from « O H and n0D, the relative amounts of 139 and 139-0?, respectively, from: n OH " O H + " O D « O H and « O D were measured from integration of the ' H N M R spectra in C 6 D 6 , where « 0 H and « 0 D was taken as nHta- The rate constant, £ o bs 1 0 ° (H/D), was obtained from the slope of the Arrhenius plot in Figure 4.11. 275 Table 4.7 Additional data for H / D exchange o f the enol hydroxyl proton in 56»(139»140) with D 2 0 in nitrobenzene at 100 ° C * R u n 3 t (min) XOH -InOCoH) l b 0 1.00 0.00 20 74 0.30 40 57 0.56 61 51 0.67 2 150 0.10 2.30 3 180 0.11 2.21 * [H 2 0] = 0.84 m M . a Data from the same experiments as runs 1, 2, and 3 in Table 4.6. bContained -80 m M T F A - d . 2.50 T 2.00 1.50 1.00 0.50 0.00 i 1 1 1 1 1 1 1 1 1 1 0 20 40 60 80 100 120 140 160 180 200 time (min.) Figure 4.11 Arrhenius plot for the H / D exchange o f the enol hydroxyl proton in 56«(139«140) in D 2 0 saturated nitrobenzene at 100 °C. • = run 1, Table 4.7. • = run 3, Table 4.7. Slope = £ o b s 1 0 ° (H/D) = (0.0112 ± 0.0070) min" 1 , .y-intercept = 0.0451, r 2 = 0.959. Error limits for kohs are based on 95 % confidence limits from regression analysis. 276 D r y experiments (in benzene and nitrobenzene). Procedure A , except no water ( H 2 O or D 2 O ) was added. One spatula tip of crushed activated 4 A molecular sieves were also added prior to sealing the reaction vessel. After the reaction was complete, the sieves were removed from the sample by dissolving the carceplex solid in the centrifuged pellet in chloroform and passing the solution through a pipette with a Kimwipe plug inside o f it. To show how the presence o f carceplex hydrates complicate ! H N M R spectra o f 56»(139»140), spectra of 56*121 in FfiO-saturated C6D5 before and after photolysis are shown below (Figure4.12). The spectra in Figure 4.12 also show the signals for bound water (labeled with h). (a) (b) r T r i ] 1 1 i 1 1 J i 1 1 i i 1 i 1 1 T ' l T ' i 1 1 1 1 ; i i 1 1 1 1 1 i 1 " 1 ••[—|—j r 1 1 1 1 1 ] 1 1 1 t t T i 1 1 1 1 1 1 1 1 i 1 1 i 1 1 1 r r n i i 1 [ i 1 1 1 r r r i 1 [ i 1 i 1 1 n 7 6 5 4 3 2 1 0 -1 -2 (ppm) Figure 4.12 ' H N M R spectra (400 M H z ) before and after photolysis in H 2 0 saturated C 6 D 6 . (a) 56»(139»140)/56»141 (after photolysis), (b) 56*141 before photolysis, h = several different bound H 2 O protons. 277 4.8 References 1. Rappoport, Z . The Chemistry of Enols, Wiley: Chichester, 1990. 2. Keefe, J. R.; Kresge, A . J. Chapter 7; Kinetics and Mechanism of Enolization and Ketonization; in The Chemistry of Enols; Rappoport, Z . (Ed.); Wi ley : Chichester, 1990; 399-480. 3. Richard, J. P. Chapter 11; The Biochemistry of Enols; in The Chemistry of Enols; Rappoport, Z . (Ed.); Wiley: Chichester, 1990; 651-689. 4. Milstein, D . Chapter 12; Organometallic Chemistry of Enols; in The Chemistry of Enols; Rappoport, Z . (Ed.); Wiley: Chichester, 1990; 691-711. 5. Elschenbroich, C ; Salzer, A . Organometallics: A Concise Introduction, V C H : N e w York , 1992; 425 6. (a) Hart, H . ; Rappoport, Z . ; B ia l i , S. E . Chapter 8; Isolable and Relatively Stable Simple Enols; in The Chemistry of Enols; Rappoport, Z . (Ed.); Wiley: Chichester, 1990; 481-589. (b) Hart, H . Chem. Rev. 1979, 79, 515-528. (c) Rappoport, Z ; B i a l i , S. Acc. Chem. Res. 1988, 21, 442-449. (d) Hart, H . ; Sasaoka, M . J. Chem. Ed. 1980, 57, 685-688. 7. Capon, B . ; Rycroft, D . S.; Watson, T. W . / . Chem. Soc, Chem. Commun. 1979, 724-725. 8. Henne, A . ; Fischer, H . Angew. Chem. Int. Ed. Engl. 1976,15, 435. 9. Laroff, B . P.; Fischer, H . Helv. Chim. Acta 1973, 56, 2011-2020. 10. Rosenfeld, S. M ; Lawler, R. G . ; Ward, H . R. J. Am. Chem. Soc. 1973, 95, 947-948. 11. Lepley, A . R.; Closs, G . L . Chemically Induced Magnetic Polarization; Wiley: N e w York , 1973. 12. Hoorspool, W . H . ; Song, P, -S. CRC Handbook of Photochemistry and Photobiology; C R C press, 1995,449-483. 278 13. (a) Brett, T.J.; Stezowksi, J.J. J. Chem. Soc., Chem. Commun. 2000, 857-858. (b) Reddy, G . D . ; Jayasree, B . ; Ramamurthy, V . J. Org. Chem. 1987, 52, 3107-3113. (c) Ramamurthy, V . ; Corbin, D . R.; Eaton, D.F . J. Chem. Soc, Chem. Commun. 1989, 1213-1214. (d) Turro, N . J . ; Wan, P. Tetrahedron Lett. 1984, 25, 3655-3658. (e) Goswami, P .C. ; De Mayo, P.; Ramnath, N . ; Bernard, G . ; Omkaram, N . ; Scheffer, J. R.; Wong, Y - F . Can. J. Chem. 1985, 63, 2719-2725. (f) Casal, H . L . , de Mayo , P.; Miranda, J.F.; Scaiano, J. C . J. Am. Chem. Soc. 1983,105, 5155-5156. 14. (a) Watanabe, S.; Goto, K . ; Kawashima, T.; Okazaki, R. J. Am. Chem. Soc. 1997,119, 3195-3196. (b) Goto, K . ; Okazaki, R. Liebigs Ann./Recrueil 1997, 2393-2407. 15. Attempts to entrap other aliphatic ketones (2-pentanone, 3-hexanone, 2-heptanone, 3-heptanone, and 2-octanone) in 56«guests were made by other members o f our group, but were unsuccessful. Kodumoru, V . ; unpublished results. 16. This ratio of cleavagexyclization products is in agreement with other reported values. Ratios o f 6.5-9:1 have been reported for 121 and 5.7-7.3:1 (142:l-phenyl-2-ethylcyclobutanol) for hexanophenone. See reference 13b-f. 17. In one experiment, separate solutions o f free 121 and 56*121 were photolyzed at the same time for 70 min, and the mixtures were analyzed by ' H N M R spectroscopy. Whi le the former reaction was essentially complete (no 121 was detected in the product mixture), the latter was estimated to contain ~10 % unreacted 56»121. 18. El ie l , E . L . ; Wilen, S. H . Stereochemistry of Organic Compounds. Wiley-Interscience: New York, 1994,617-618. 19. M M 2 minimized structures o f 56*123 with 123 in geometry A showed significant distortions in the host structure. 20. Stephen, H . ; Stephen, J. Solubilities of Inorganic and Organic Compounds, Volume 1, Binary Systems, Part I; Permagon Press: New York, 1963. 21. It is important to point out that £ O b s 1 0 ° (H2O) = 1.5 x 10"4 s'1 is the average value o f two different runs. In one run, T F A (~80 m M ) was present, while in the other 25 ( iL o f a 40 % tetrabutylammonium hydroxide solution in water was added instead o f 25 |aL o f H 2 O . The presence o f acid or base in either case had negligible effects on the observed rate (with H 2 O ) , as determined by comparison with other data obtained in the absence o f acid and base. See experimental section for more details. 279 22. For various enols in aqueous solution, the ketonization p H rate profile (i.e., logos'1)) vs. pH) curve is usually flat at hydronium ion concentrations in between pHs where catalysis by hydronium and hydroxide ions are found to be prevalent. The flat portion o f the curve corresponds to the "uncatalyzed" reaction between the enol and a water molecule, during which the rate is independent o f the hydronium ion concentration. See reference 2. 23. Chiang, Y . ; Kresge, A . J . ; Santaballa, J . A . ; Wirz , J. J. Am. Chem. Soc. 1988,110, 5506-5510. 24. - 5 % components in carceplex 56»guests mixtures (for the templation study in Chapter 3) have been detected i n ! H N M R spectra recorded using a number o f scans (128-256) that would allow for good signal-to-noise to be acquired within a reasonable amount o f time (5-15 min), given the total sample concentration (1.7 m M ) . Therefore, a conservative 5 % estimate was used for a minimum observable amount of 56»(142«140) that could be detected in ketonization reaction product mixtures. 25. T F A - d was present in these reactions, but at concentrations (-80 m M ) too low to have any noticeable effect on the values for £ 0 b S 1 0 0 (D2O). This was determined by comparison with other data generated from similar experiments without T F A - d (see experimental section). A c i d (D + ) also appears to have a negligible effect on & 0 b s 1 0 0 (D2O) compared to similar experiments without acid, and is thus interpreted as & 0 b s 1 0 0 (D2O) from D 2 O only. The data shown in Figure 4.6 is presented because it was the only run in which measurements were recorded at times before 150 min. 26. The identity o f the of the signal at 0.17 ppm was confirmed to be due to the acetyl methyl protons of 142 upon comparison to the chemical shift o f 142 in 56»(142»140) formed from the reaction with H 2 O (Figure 4.9a). Also , an increase in the intensity of the signal at 0.17 ppm relative to that at 0.16 ppm was observed in ' H N M R spectra, after a sample containing 56»(142»140)/56«(142-d»140) (after 6.7 h in D 2 O saturated nitrobenzene, Figure 4.9b) was reacted with H 2 O in nitrobenzene for 1 hour. 27. (a) Haspra, P.; Sutter, A . ; Wirz , J. Angew. Chem. Int. Ed. Engl. 1979,18, 617-619. (b) Chiang, Y . ; Kresge, A . J.; Wirz , J. J. Am. Chem. Soc. 1984,106, 6392-6395. (c) Keefe, J. R.; Kresge, A . J.; Toullec, J. Can. J. Chem. 1986, 64, 1224-1227. (d) Andraos, J. ; Kresge, A . J.; Obraztsov, P. A . J. Phys. Org. Chem. 1992, 5, 322-326. 28. This assumes that only one water molecule is involved in ketonization. We cannot determine how many waters are in the carceplex cavity nor what percentage o f the 280 carceplex cavities are hydrated. Moreover, there are several hydrated species, and it is not clear which one effects ketonization. 29. (a) Cram, D . J.; Tanner, M . E . ; Knobler, C. B . J. Am. Chem. Soc. 1991,113,11X1-1121. (b) Mendoza, S.; Davidov, P. D . ; Kaifer, A . E . Chem. Eur. J. 1998, 4, 864-870. 30. Weast, K . C. CRC Handbook of Chemistry and Physics, C R C Press: Boca Raton, 1985. 31. Lankin, D . C ; Chihal, E . M . ; Griffin, G . W . Tettrahedron Lett. 1973, 41, 4009-4012. 281 5. A Hexa-Cavitand Derived Carceplex That Entraps Seven Guest Molecules 5.1 Introduction The design and synthesis of nanoscale molecular hosts has attracted much attention in recent years. ' ' 2 ' 3 Large, symmetrical hosts have been synthesized that approach the size o f some natural systems (i.e., protein shells) 4 used to store, protect, and transport neutral and charged inorganic and organic molecules. In chemistry, nanoscale objects are typically arranged via the oligomerization of relatively small subunits held together by coordinative bonds to transition metals, 1 weak noncovalent interactions,2 or covalent bonds. 3 Although many reported structures possess cavities with large volumes, relatively few reversibly or permanently entrap multiple smaller guest molecules. 2 ' 5 ' 6 ' 7 One current aim of the Sherman group is to synthesize larger host molecules with internal cavities large enough to incarcerate large and esoteric guest molecules as well as multiple copies o f smaller guests. The design and synthesis of increasingly larger hosts that do not possess large holes in their shells is a challenge. 8 Likewise for creating hosts with larger enforced cavities. 5 d ' e Our group has recently succeeded in creating larger host molecules capable o f entrapping suitable guest molecules by covalently l inking multiple [4]cavitands 5 d" g and/or two wider [5]cavitand subunits (58»guests). 5 ' Recall that in Chapters 3 and 4 of this thesis, the formation of 56»guests from a cyclic trimer of [4]cavitands was studied, where single, two, and three-molecule guests could be incarcerated. 282 56«guests 58«guests 149«guests (82a) 6 (H 2 0) 8 (82b) 6 O = H 2 0 A few spectacular examples o f recently reported nanoscale capsular hosts are Atwood's resorcin[4] arene ((82a)6(H20)g)7a and Mattay's pyrogallol[4]arene (82b)67b'c hexamers, which both make use of similar, but less rigid building blocks than used for carceplexes/ hemicarceplexes. Hexamers ((82a)6(H20)g and (82b)6 are held together by numerous hydrogen-bonds and possess large cavities (1375 and -1515 A 3 , respectively) that can potentially be occupied by a single large guest or multiple small molecules (e.g., eighteen M e O H s in (82b)6).7a"° Initially, these supramolecules were formed exclusively in the solid state, but remained intact even in highly polar solvents.7 8"0 Rebek and coworkers also showed that resorcin[4]arenes can self-assemble into discrete hexamers in the presence o f suitable templates in solution. 7 d" f Resorcin[4] arene hexamers form in the presence o f the correct ratio o f resorcin[4]arene:H20 in organic solvents in the presence o f tetraalkylammonium/ 283 tetraalkylphosphonium salts or covalent antimony(V) bromides. 7 d ' e Hexameric capsules formed with covalent antimony(V) bromides as the principle guests also contain enough empty space to be co-inhabited by simple aromatics such as benzene, toluene, or /?ara-xylene. 7 e In this chapter, the synthesis and characterization of a new carceplex, 149»guests, is discussed. Carceplex 149»guests is the largest carceplex synthesized to date. It has a similar topology to the hexameric capsules (82a)6(H20)s and (82b)6. Six [4]cavitands (derived from resorcin[4]arenes) are linked together in a (pseudo)octahedral arrangement to form a large enclosed space that is occupied by seven guest molecules, the most for any carceplex/hemicarceplex. The numerous (12) interbowl methylene linkages impart some flexibility to the host shell that gives rise to unusual dynamic behavior. Investigations into the conformational dynamics o f the host, complexation o f water, and mobility o f the entrapped D M S O s in 149«(DMSO) 7 is also presented. 5.2 Synthesis of the A,B-trimer 152 The approach we took to synthesize 149»guests has some similarity to the procedure for 56«guests. A , B - d i o l 150 was isolated (26 %) from the same reaction as A , C - d i o l 41, from tetrol 8a (Scheme 5.1). Cyclization of A , B - d i o l 150 in D M S O with base and the bis-electrophile bromochloromethane gave A,B-trimer 151 (16 %, Scheme 5.1). Note that A,B-trimer 151 forms in nearly half the yield of A,C-trimer 41 under similar reaction conditions. This is l ikely due to the undesired A,B-trimer 153 that also forms (Scheme 5.1), according to M A L D I mass and ' H N M R spectra. Debenzylation o f A,B-trimer 151 (58 %) does not proceed as easily as for A , C -284 trimer 41: several days are required for 151 as opposed to a few hours for 41, and the reaction does not go to completion starting from 151. 9 Scheme 5.1 Synthesis o f A,B-trimers. H 2 , Pd/C I 151 R'= C H 2 C 6 H 5 153 (1atm) (16%) C 6 D 6 / M e O H U- 152R' = H (58 %) ' r i N M R spectra o f A,B-protected [4]cavitands show a characteristic 1:1:2 splitting o f the protons for H 0 , H m , and Hj, which is shown in the spectrum for the C3V-symmetric A,B-trimer 152 in DMSO-fife (Figure 5.1a). Spectra o f 152 recorded in CDCI3, C^De, toluene-fife, acetone-fife, and nitrobenzene-fife are broad and complicated due to the formation o f hydrogen-bonded aggregates (Figure 5.1b). To see i f monomelic, dimeric, or higher order oligomeric capsules could be generated, binding studies were performed with potential small-molecule guests. 285 Figure 5.1 *H N M R spectra (400 M H z , 300 K ) of A,B-trimer 152. (a) DMSO-</ 6 . (b) Nitrobenzene-c/5. 5.3 Complexation Experiments Involving A,B-Trimer 152 Recall that dimeric capsules (i.e., 12a«guest) can form reversibly from tetrol 8a in organic solution in the presence of base and a suitable template. The dimer held together by charged hydrogen-bonds. 1 0 This concept has also been applied in the formation o f b i s 5 e and tris-capsules, 5 8 which are hosts that have multiple separate chambers that are interconnected. A,B-286 trimer 152 can also potentially form hydrogen-bonded capsules. C P K models suggest that at least four topologically different capsules can form, three of which are shown in Figure 5.2: (1) an intramolecularly hydrogen-bonded capsule encapsulating a single guest, mono-capsule 152»guest; (2) intermolecular hydrogen-bonded dimers with eclipsing bowls, tris-capsule 152•(guest)*; (3) intermolecular hydrogen-bonded dimers with staggered bowls, capsule 152«guests; as well as (4) higher-order oligomeric capsules. 152«guest 1522«(guest)3 1522«guests (intramolecular) (eclipsing) (staggered) Figure 5.2 Hydrogen-bonded capsules from A,B-trimer 152. For clarity, the phenol OHs on each bowl and hydrogen-bonding interactions are not shown. In nitrobenzene-fife, binding o f D M S O (10 equiv.) was observed by the immediate appearance of abroad resonance at -0.23, and two sharper ones upfield at -0.34 and -0.39 ppm (Figure 5.3b). When more D M S O (>10 equiv.) is added, only two signals are observed, one very broad signal centered at -0.17 ppm and a considerably sharper one at -0.32 ppm (Figure 5.3c). These signals correspond to bound D M S O , which are in exchange with free D M S O , as determined by I D E X S Y ( N O E S Y ) experiments. Clearly, at least one type o f capsular species o f unknown structure does form in solution, which unfortunately could not be fully characterized by N M R . Similar results were observed when D M A and methyl acetate were added to 151 in nitrobenzene-fife.11 Other conditions were also investigated to observe discrete complexes between A,B-trimer 152 and D M S O in solution. D M S O binding to 152 in other solvents such as 287 C D C 1 3 , C 6 D 6 , toluene-fife, and acetone-fife only gave similar results to the experiments in nitrobenzene-^. Base ( D B U ) was also added to 152 and D M S O in nitrobenzene-fife. *H N M R spectra did change, but were still complicated and difficult to interpret. 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 i ' i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 8 7 6 5 4 3 2 1 0 (ppm) Figure 5.3 ' H N M R (500 M H z , nitrobenzene-afe, 300 K ) spectra o f A,B-tr imer 142. (a) 142 (1.86 m M ) . (b) 142 plus D M S O (-10 equiv.). (c) 142 plus D M S O (-50 equiv.). * = D M S O guest signals. Free D M S O in nitrobenzene-ofe occurs at -2.60 ppm. 5.4 The Format ion of 1 4 9 » ( D M S O ) 7 Reaction between A,B-trimer 152 and bis-electrophile bromochloromethane with base could conceivably give three structurally different nonpolymeric products. C P K models suggest that tris-carceplex 154«guests and structure 155 can form, but are highly strained (Scheme 5.2). In contrast, when linked in a staggered arrangement the two A,B-trimers 152 are highly complementary to one another, as in 149»guests. Models of 149»guests suggest that the host shell is relatively strain free, possesses some flexibility, and is somewhat porous due to the large 288 30-membered rings, through which small molecules (i.e., H2O) could pass. Perhaps suitable templates could be found that would selectively give one structure in favor o f the others. Scheme 5.2. A,B-trimer 152 coupling reaction. 154.(DMSO)3 155 Carceplex 149»(DMSO) 7 was exclusively synthesized in ~35 % yield from a reaction involving A,B-trimer 152 in the presence of CS2CO3 (or K 2 C 0 3 ) and B r C H 2 C l (10 equiv.) in D M S O at 60 °C. ' H N M R and M A L D I M S were both consistent with the structure o f 149«(DMSO) 7 (Scheme 5.2). In the M A L D I mass spectrum (Figure 5.4), only two signals appear at m/z = 6817 and 6834, which correspond to the N a + and K + adducts o f 149»(DMSO) 7 , respectively. Neither compound 154 nor 155 were isolated from the coupling reaction; M A L D I mass spectra o f other reaction byproducts suggested that only incompletely or incorrectly bridged intermediates to 149»(DMSO) 7 formed. 289 I 1 1 1 1 I 1 1 1 1 I ' ' 1 ' I 1 1 1 ' I 1 ' ' I I ' ' • ' I • ' 4000 5000 6000 7000 8000 9000 m/z Figure 5.4 M A L D I M S spectrum of carceplex 149«(DMSO) 7 . The 35 % yield obtained for 149»(DMSO) 7 is considered very good, as 15 discreet components are brought into a confined region of space and 12 bonds are formed (and broken) to form a single large entity. 1 2 This is quite striking considering the flexibility o f the A,B-tr imer subunits, and the number of incorrect linkages that are possible for each intermediate leading to the formation of 149«(DMSO) 7 . Complete characterization of 149»(DMSO) 7 in solution by N M R was challenging due to low solubility and unusual fluxional behavior observed for the host shell, which is discussed in detail in the next section. 290 5.5 NMR Spectroscopy of Carceplex 149»(DMSO) 7 5.5.1 Conformation of 149«(DMSO) 7 in Dry CDC13 Recall that ' H N M R spectra for trimer carceplex 56«guests in dry solution are relatively simple compared to spectra of 56»guests in the presence of water. For 149»(DMSO) 7 , the exact opposite is true. ' H N M R spectra of 149«(DMSO) 7 in sieve-dried CDCI3 (Figure 5.5), show a single resonance for bound D M S O (0.59 ppm) and all H p s , but each of the protons H 0 , H a c , H m , and H ; show three resonances in a 1:1:2 ratio. In contrast, in water-saturated C D C I 3 (or CD2CI2) , H 0 , H a c , H m , and H i each yield single resonances (see section 5.5.3). Integration o f *H N M R spectra in both sieve-dried and water-saturated CDCI3 give a 1:7 hostguest ratio, which is in accord with the mass for 149»(DMSO) 7 determined by M A L D I M S . The structure of 149»(DMSO) 7 in sieve-dried CDCI3 is discussed below, while the structure observed in water-saturated CDCI3 is deferred to section 5.5.3. The pattern of host signals in Figure 5.5 is more complicated than expected for the highly (Oh) symmetrical structure o f 149»(DMSO) 7 , according to molecular ( C P K ) models. In dry CDCI3, carceplex 149«(DMSO) 7 appears to adopt a flattened conformation (Figure 5.6), which features upper and lower A,B-trimers connected by a seam o f six equatorial, inter-trimer methylene bridges. The upper and lower A,B-trimer subunits are equivalent and both possess three identical pairs of diastereotopic geminal acetal protons H a c i / H a c 3 , which are nonequivalent to the enantiotopic inter-trimer pairs (H a C 2) located near the equatorial region. Note the similarity in the appearance o f the host signals in the ' H N M R spectrum o f 149»(DMSO) 7 in dry CDCI3 (Figure 5.5) with the signals in the spectrum for A,B-trimer 152 (Figure 5.1). From the pattern o f host signals in Figure 5.5a, one might initially interpret the 291 spectrum as structure 154»(DMSO) 3 (Scheme 5.2). However, extensive characterization using various N M R techniques in addition to M A L D I M S data led us to conclude that the structure is indeed 149«(DMSO) 7 . ac2 I r I I i | I I ' l l 1 I 1 1 I i i i i I i I I I I I i i | | 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 (ppm) R = Ch^ Ch^ CgHs Figure 5.5 ' H N M R spectrum (400 M H z , dry CDCI3, 300 K ) o f 149»(DMSO) 7 . * = D M S O C7/ 3 s. o, o' = Hoi or H 0 3 . o2 = H 0 2 , acz = H a c z (z = 1 , 2, 3). m, m' = H m i or H m 3 . m2 = H , ^ . i , i ' = H i , or H i 3 , i2 = H i 2 . 292 Figure 5.6 MM2 minimized space filling representations of the C 3 v conformation of 149«(DMSO) 7. (a) Viewing down the C3-axis. (b) Viewing from the side perpendicular to the C 3 axis (b). To simplify calculations, the phenylethyl groups on each of the six [4]cavitands were replaced with hydrogens. For clarity, the entrapped DMSOs are omitted 293 Evidence supporting a flattened conformation for 149»(DMSO)7 in dry solution (Figure 5.6) was also obtained from a series of 2D N M R experiments. These experiments were performed to rule out the possibility that the three sets of host signals in the spectrum in Figure 5.5 represent three different species in solution. One key piece o f information was the weak coupling observed between the pair of minor interbowl acetals, H a c i and FfaC3, which do not show splitting in the I D ' H spectrum, nor do they show coupling by 2D C O S Y . However, coupling was observed in long range C O S Y experiments. The small coupling constant (J) observed between H a c i / H a c 3 suggests that the H - C - H bond angles are large (0 > 109°), which l ikely results from ring strain < 109°) of the 30-membered rings connecting each trimer in the hexamer. ' H - 1 3 C H M Q C spectra also confirmed the geminal relationship between the signals for H a c i and H a C 3, showing one-bond coupling correlations to the same carbon atom. Intra-bowl acetal pairs, H 0 i / H a , H02/HJ2, and H 03/HJ3, were connected by 2D C O S Y correlations. N O E data in support of the flattened conformation and against separate species was acquired by 2D R O E S Y at 250 K . A t 250 K , the host signals in the ! H N M R spectrum only appear a bit broader than the signals observed at 300 K . The N O E s observed at 250 K (Figure 5.8b) are summarized in Figure 5.7. The most important N O E s are observed between H a c i / H a C 3 , H a C 3 / H o 2 , and H a C3/Hj2, 1 4 which connect the major and two minor sets of signals to the same structure, and suggest a possible orientation for the intra-trimer, inter-bowl acetals H a c i / H a C 3 . The N O E signal between HaC3/Hi2 (and H02) suggests that these protons on average are oriented more towards the outside the 30-membered rings of the upper and lower trimers. In addition, the chemical shifts o f H a c i (more downfield) and H a C 3 (more upfield) suggest that H a C 3 is oriented more inside the shielding carceplex cavity than H a c i . H H 294 H a c 2 Figure 5.7 Schematic representation o f the flat conformation of 149»(DMSO)7 (top view down the C3V axis) dry CDCI3 solution. Solid circles indicate the cavitand subunits o f the upper trimer and dotted bowls indicate the bottom trimer. Red arrows are N O E s observed. H 0 , H 0 ' = H 0 i or H03, Hi, Hi' = Hii or HJ3. 295 I I , i 6.4 (ppm) 6.4 6.0 5.6 5.2 4.8 4.4 (ppm) 6.4 6.0 5.6 5.2 4.8 4.4 Figure 5.8 2D R O E S Y spectrum of 149«(DMSO) 7 at 250 K in dry CDCI3. (a) Signals in phase (negative) with the diagonal indicating exchanging nuclei, (b) Signals in opposite phase (positive) to the diagonal indicating N O E s . Note that the N O E signal between HaC3/Hj2 is not very intense in (b), but it is clearly observed by 2D N O E S Y at 250 K (data not shown), o, o' = H o i or H03. o2 = H 02, acz = H a c z (z = 1, 2, 3). m, m' = H m i or H m 3 . m2 = H ^ . i , i ' = Hn or H i 3 , i2 = H i 2 . 296 (ppm) 6.0 5.6 5.2 4.8 4.4 (ppm) 6.0 5.6 5.2 4.8 4.4 Figure 5.9 2D R O E S Y spectrum of 149»(DMSO) 7 at 300 K in dry C D C 1 3 . (a) Signals in phase (negative) with the diagonal indicating exchanging nuclei, (b) Signals in opposite phase (positive) to the diagonal indicating N O E s . o, o' = H 0 i or H 0 3 . o2 = H 0 2 , acz = H A C Z (z= 1,2, 3). m, m ' = H m i or H M 3 . m2 = H , ^ . i , i ! = HJI or H I 3 , i2 = H I 2 . N o negative phase cross peak signals are observed in the R O E S Y spectrum at 250 K , (Figure 5.8a), indicating that exchange between the three sets of host proton signals at this temperature is too slow to detect (rate constants < 0.01 s"1).1 5 A t 300 K , exchange is detected between the three sets of signals for each type o f host proton. A l l exchanging host protons for 149»(DMSO) 7 in dry CDCI3 are clearly mapped out in the 2D spectrum as strong negative off-diagonal peaks observed between H0i-3, H a c i - 3 , Hmi-3, and Hn-3 (Figure 5.9a). Note that positive N O E signals (Figure 5.9b) between HaC3/Hj2, H 0 I / H J I , f W F f o , and H03/H3 are observed in addition to transfer N O E s due to chemical exchange between H0i/Hi2 and H 0 i /Hi3 at 300 K ; these signals are absent at 250 K (Figure 5.8b). 5.5.2 Host Dynamics 5.5.2.1 Host Dynamics in Dry CDC13 Rate constants for the three site exchange observed at 300 K for 149»(DMSO)7 in sieve-dried CDCI3 were measured by I D E X S Y . 1 5 ' 1 6 Hn, H2, and H3 were selectively irradiated at several mixing times at 300 K and the relative intensities of the response signals were recorded and measured (see experimental section 5.10.5). Selective irradiations of all H 0 s, H a c s , and H m s were not possible due to poor signal dispersion (Figure 5.5). The average chemical rate constant (£Chem) for the observed dynamic process interconverting Hi 1.3 in 149«(DMSO)7 was calculated to be 6.4 s"1, which corresponds to an energy barrier (AG^oo) of 16.5 kcal/mol (see experimental section). This energy barrier is interpreted as the physical barrier that has to be overcome to interconvert any two o f the four degenerate flattened conformers A-D (Scheme 5.3) in a process analogous to flattening a ball along different dimensions. For example, flattening along the pseudo C3-axis (I) passing through bowls 1, 2, 3 and 4, 5, 6 in structure A gives C, where Hn (labeled with *, bowl 1, structure A ) becomes HJ3 (bowl 1, structure C). Similarly, flattening along the other pseudo C3-axes (II), through bowls 1, 3, 5 and 2, 4, 6 to give D, and (III) through 1, 2, 4 and 3, 5, 6 to give B, 298 transforms Hn into H;2. Each interconversion probably involves concerted rotations about the many A r - 0 bonds (24 total) o f the 12 interbowl methylene linkages. Scheme 5.3 Interconversion between degenerate flattened conformations ( A - D ) for 149«(DMSO)7. The perspective shown is looking down the C3-axis o f the flatten structures A - D . Solid, shaded circles are the top bowls and non-shaded, dotted circles are the bottom bowls. The bowl subunits (circles) are arbitrarily labeled 1-6. To clearly show the fate o f the Hi protons after each exchange event, one of the H; protons is labeled with *. I, II, and III are pseudo C 3 -axes. II 299 Interconversion between any two flattened conformers A - D could occur through a transition state that resembles the Oh-symmetric structure of 149«(DMSO)7 (Figure 5.11). Support for this was obtained by heating 149«(DMSO)7 in pyridine-d?5 (Figure 5.10). A t 300 K , the ' f l N M R spectrum in pyridine-ds displays three sets of broad signals for each o f the host protons, H 0 , H a c , H m , and Hj (Figure 5.10a). Wi th heating, these signals broaden and coalesce into four single resonances whose multiplicity can be determined by 383 K (Figure 5.10e). 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 ' 7.8 7.6 7.4 7.2 7.0 6.8 6.6 6.4 6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8 (ppm) Figure 5.10 Sections of ' H N M R spectra (500 M H z ) o f 149«(DMSO) 7 in pyridine-d 5 . (a) 383 K . ( b ) 3 4 0 K . ( c ) 3 2 0 K . ( d ) 3 1 0 K . ( e ) 3 0 0 K . h = H 2 0(f ree) . (Note that broadening of the signals in Figure 5.10e is due to the presence of trace amounts o f H 2 0 (see section 5.5.3)) 300 Figure 5.11 M M 2 minimized space filling representations of the (9 h-symmetric conformation of 149(DMSO) 7 . (a) Viewing down the C 3 -axis. (b) Viewing down the C 4 -axis . To simplify calculations, the phenylethyl groups on each of [4]cavitand were replaced with hydrogens. For clarity, the entrapped guests are omitted 301 M M 2 calculations performed on the minimized structures of both flat C3 and Oh conformations of 149»(DMSO)7 are in agreement with the observations that the former is more thermodynamically stable conformation at 300 K . Wi th the seven D M S O guests included, the C3 host conformation (G° = 360 kcal/mole) is predicted to be -15 kcal/mole more stable than its highly symmetric Oh counterpart (G° = 375 kcal/mole). Without the D M S O guests, both conformations are predicted to be equal in energy (G° = 470 kcal/mole). Therefore, we conclude that the main driving force behind the host adopting the flat conformation is the maximization o f favorable host-guest (van der Waals) interactions. The cavity o f the flat conformer appears to be more complementary to the overall size and shape of the cluster of seven D M S O guests, than the Oh conformer. 5.5.2.2 Host Dynamics i n D r y CD2CI2 *H N M R