UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Supramolecular schiff base coordination chemistry : blueprints for self-assembling metallocavitands and… Frischmann, Peter David 2010

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2010_fall_frischmann_peter.pdf [ 8.85MB ]
Metadata
JSON: 24-1.0060428.json
JSON-LD: 24-1.0060428-ld.json
RDF/XML (Pretty): 24-1.0060428-rdf.xml
RDF/JSON: 24-1.0060428-rdf.json
Turtle: 24-1.0060428-turtle.txt
N-Triples: 24-1.0060428-rdf-ntriples.txt
Original Record: 24-1.0060428-source.json
Full Text
24-1.0060428-fulltext.txt
Citation
24-1.0060428.ris

Full Text

Supramolecular Schiff Base Coordination Chemistry: Blueprints for the Self-Assembly of Metallocavitands and Nanotubes by Peter David Frischmann B.Sc., Idaho State University, 2005  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Chemistry) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July 2010  © Peter David Frischmann, 2010  ABSTRACT Heptametallic zinc(II) and cadmium(II) clusters have been isolated after reacting the metal-acetate salts with large diameter [3+3] Schiff base macrocycles. Two tetrazinc complexes have been characterized and identified as intermediates in the formation of the heptazinc complexes. The heptametallic complexes are, in fact, templated by the Schiff base macrocycles, a process that has been investigated with 1H NMR spectroscopy and single-crystal X-ray diffraction. In the solid-state the heptametallic complexes have a bowl-shaped geometry, reminiscent of organic cavitands, leading to them being called metallocavitands. Solid-state investigation of the heptazinc and heptacadmium metallocavitands showed they organize into capsules with a cavity volume of 150 and 215 Å3, respectively. Solution dimerization was also observed in aromatic solvents and N,N-dimethylformamide (DMF). The thermodynamics of dimerization have been quantified by van’t Hoff analyses of association constants measured with variabletemperature, variable-concentration  1  H NMR spectroscopy. Both metallocavitands  exhibit entropy-driven dimerization in all solvents in which dimerization occurs. Unusual for dimerization of cavitands, this entropy-driven process can be attributed to the expulsion of solvent from the monomeric cavity upon dimerization. Inside the cavity of heptacadmium metallocavitands is a μ3-OH ligand where the proton is located at the base of the cavity and is capable of hydrogen bonding with guest molecules. The μ3-OH proton resonance is observable in low temperature 1H NMR spectra and exhibits two-bond J-coupling with three cadmium ions. Within capsules of the heptacadmium metallocavitands there are eight Lewis-acidic sites accessible to guest molecules, six unsaturated cadmium(II) centers and two μ3-OH ligands. Solid-state analysis shows that two DMF molecules are encapsulated in the heptacadmium capsule where they each simultaneously exhibit a host-guest hydrogen-bond and a dative metalligand interaction. New methodology has been developed that facilitates synthesis of polydentate [2+2] Schiff base macrocycles with unsymmetrical salphen pockets. Also a [3+3] macrocycle with triptycenyl substituents has been synthesized to prohibit alkali-metal induced solution aggregation. ii  The one-pot twelve component head-to-tail self-assembly of Pt4 rings directed by chelating imine-pyridyl donors has been demonstrated. These supramolecules exhibit extensive columnar organization in both solution and the solid-state, a phenomenon that imparts liquid crystalline properties on the macrocycles.  iii  TABLE OF CONTENTS Abstract ............................................................................................................................. ii Table of Contents ............................................................................................................. iv List of Tables .................................................................................................................. vii List of Figures ................................................................................................................ viii List of Schemes ............................................................................................................... xx List of Symbols and Abbreviations.............................................................................. xxiv Acknowledgements ..................................................................................................... xxvii Dedication .................................................................................................................... xxiv Co-Authorship Statement.............................................................................................. xxx CHAPTER 1 Introduction................................................................................................. 1 1.1 Metallocavitands and Supramolecular Coordination Chemistry .................... 1 1.1.1 Introduction to Supramolecular Chemistry ...................................... 1 1.1.2 Supramolecular Coordination Chemistry ........................................ 5 1.1.3 Traditional Cavitands and Their Metal Complexes ....................... 15 1.1.4 Metallocavitands: Coordination Induced Curvature ...................... 20 1.1.5 Aggregation of N Supported – Pd(II) and Pt(II) Metallocavitands 23 1.1.6 Aggregation of Organometallic Half-Sandwich Metallocavitands 34 1.1.7 Aggregation of Miscellaneous Metallocavitands........................... 44 1.1.8 Macrocycle-Templated Metallocavitands ...................................... 52 1.2 Summary ....................................................................................................... 60 1.3 Goals and Scope of this Thesis ..................................................................... 62 1.4 References ..................................................................................................... 65 CHAPTER 2 Heptazinc Clusters Templated by [3+3] Schiff Base Macrocycles .......... 79 2.1 Introduction ................................................................................................... 79 2.1.1 Abstract .......................................................................................... 79 2.1.2 Background .................................................................................... 80 2.2 Discussion ..................................................................................................... 83 2.2.1 Macrocycle Synthesis .................................................................... 83 2.2.2 Heptazinc Metallocavitands: Synthesis and Structure ................... 84 2.2.3 Mechanistic Insight ........................................................................ 90 2.2.4 Metal Ion Ring Walking ................................................................ 97 2.2.5 Zinc Methacrylate Complexes ..................................................... 100 2.2.6 Mixed Metal Clusters ................................................................... 102 2.3 Conclusions ................................................................................................. 103 2.4 Experimental ............................................................................................... 104 2.4.1 General ......................................................................................... 104 2.4.2 Procedures and Data .................................................................... 105 2.4.3 Crystallography ............................................................................ 109 2.5 References ................................................................................................... 111 iv  CHAPTER 3 Dimerization of Heptazinc Metallocavitands ......................................... 115 3.1 Introduction ................................................................................................. 115 3.1.1 Abstract ........................................................................................ 115 3.1.2 Background .................................................................................. 115 3.2 Discussion ................................................................................................... 117 3.2.1 Synthesis and Characterization .................................................... 117 3.2.2 Solution Dimerization .................................................................. 118 3.2.3 Solid-State Dimerization .............................................................. 123 3.3 Conclusions ................................................................................................. 124 3.4 Experimental ............................................................................................... 125 3.4.1 General ......................................................................................... 125 3.4.2 Procedures and Data .................................................................... 126 3.4.3 Thermodynamic Studies .............................................................. 126 3.5 References ................................................................................................... 134 CHAPTER 4 Cadmium Cluster Metallocavitands: Highly Dynamic Supramolecules 136 4.1 Introduction ................................................................................................. 136 4.1.1 Abstract ........................................................................................ 136 4.1.2 Background .................................................................................. 137 4.2 Discussion ................................................................................................... 140 4.2.1 Synthesis and Characterization .................................................... 140 4.2.2 SCXRD Analysis ......................................................................... 144 4.2.3 DFT Computations....................................................................... 148 4.2.4 Metallocavitand Capsules in Solution: Thermodynamics ........... 153 4.2.5 Capsule Occupancy and the μ3-OH Proton NMR Resonance ..... 158 4.2.6 Metallocavitand Capsules in Solution: Kinetics .......................... 163 4.2.7 Cadmium Cluster Dynamics: A Molecular Beehive ................... 164 4.2.8 Guest Dynamics Probed by Solid-State NMR Spectroscopy ...... 169 4.3 Conclusions ................................................................................................. 176 4.4 Experimental ............................................................................................... 176 4.4.1 General ......................................................................................... 176 4.4.2 Procedures and Data .................................................................... 177 4.4.3 Thermodynamic Studies .............................................................. 179 4.4.4 Kinetic Studies ............................................................................. 186 4.4.5 Crystallography ............................................................................ 188 4.4.6 Solid-State 2H NMR Spectroscopy .............................................. 192 4.4.7 Simulated 2JHCd Coupling ............................................................ 192 4.4.8 Computational Details ................................................................. 193 4.5 References ................................................................................................... 195  v  CHAPTER 5 Design and Synthesis of New Schiff Base Macrocycles ........................ 206 5.1 Introduction ................................................................................................. 206 5.1.1 Abstract ........................................................................................ 206 5.1.2 Background .................................................................................. 208 5.2 Discussion ................................................................................................... 209 5.2.1 Model Compounds and Macrocycle Precursors .......................... 209 5.2.2 Synthesis and Characterization of [2+2] Macrocycles ................ 213 5.2.3 Synthesis of Triptycene-Based [3+3] Macrocycles ..................... 215 5.2.4 Fiber Formation Inhibited by Triptycene Substituents ................ 217 5.3 Conclusions ................................................................................................. 219 5.4 Experimental ............................................................................................... 219 5.4.1 General ......................................................................................... 219 5.4.2 Procedures and Data .................................................................... 221 5.4.3 Kinetic Studies ............................................................................. 231 5.4.4 Crystallography ............................................................................ 232 5.5 References ................................................................................................... 233 CHAPTER 6 Columnar Organization of Head-to-Tail Self-Assembled Pt4 Rings ...... 237 6.1 Introduction ................................................................................................. 237 6.1.1 Abstract ........................................................................................ 237 6.1.2 Background .................................................................................. 238 6.2 Discussion ................................................................................................... 241 6.2.1 Model Complex Synthesis, Solid-State Structure, and Pyridine Binding .................................................................................................. 241 6.2.2 Synthesis of Head-to-Tail Directing N-ONO Pro-Ligands ......... 244 6.2.3 Self-Assembly of Pt4 Rings ......................................................... 247 6.2.4 Evidence for Columnar Organization of Pt4 Rings ...................... 253 6.3 Conclusions ................................................................................................. 261 6.4 Experimental ............................................................................................... 261 6.4.1 General ......................................................................................... 261 6.4.2 Procedures and Data .................................................................... 263 6.4.3 Crystallography ............................................................................ 275 6.4.4 Computational Details ................................................................. 276 6.5 References ................................................................................................... 277 CHAPTER 7 Conclusions and Future Directions ......................................................... 285 7.1 Conclusions ................................................................................................. 285 7.2 Reflection .................................................................................................... 288 7.3 Future Direction .......................................................................................... 289 7.3.1 Metallocavitands .......................................................................... 289 7.3.2 Pt4 Rings....................................................................................... 294 7.4 References ................................................................................................... 298  vi  List of Tables Table 2.1. Bond lengths pertaining to the zinc acetate cluster of complexes 52a, 52b, 52d, and basic zinc acetate ............................................................................. 88 Table 2.2. Crystallographic parameters for compounds 52d, 58, and 60 .................... 110 Table 3.1. Thermodynamic parameters for dimerization of 52c (R = C6H13) and 52e (R = C8H17) in different solvents............................................................................ 120 Table 3.2. Imine chemical shift (ppm) and association constants for 52c (R = OC6H13) in toluene-d8. .......................................................................................... 128 Table 3.3. Imine chemical shift (ppm) and association constants of 52e (R = OC8H17) in benzene-d6.......................................................................................... 129 Table 3.4. Imine chemical shift (ppm) and association constants of 52e (R = OC8H17) in toluene-d8. .......................................................................................... 131 Table 3.5. Imine chemical shift (ppm) and association constants of 52e (R = OC8H17) in p-xylene-d10 (Kdim at 0.45 mM and 352 K was erroneously large). .. 132 Table 4.1. Thermodynamics of dimerization for metallocavitands 61a and 61e in various deuterated solvents. The packing coefficient is also included as a percentage. .................................................................................................................... 155 Table 4.2. Imine chemical shift (ppm) and association constants of 61e in benzene-d6. .................................................................................................................... 182 Table 4.3. Imine chemical shift (ppm) and association constants of 61e in toluene-d8. ..................................................................................................................... 183 Table 4.4. Imine chemical shift (ppm) and association constants of 61e in p-xylene-d10................................................................................................................... 185 Table 4.5. Parameters for calculating the dimerization rate of 61a in DMF-d7. .......... 187  vii  List of Figures Figure 1.1. Space-filling solid-state structure of a zinc(II)-templated molecular Borromean rings and the retrosynthetic analysis from diformylpyridyl and diaminobipyridyl units. The three separate macrocycles (rings) are colored red, blue, and green and the zinc(II) ions are yellow (protons are omitted for clarity). .......... 9 Figure 1.2. Solid-state structure of a Pd6L4 coordination cage. a) Cage framework. b) With encapsulated tryptophan-tryptophan-alanine oligopeptide. The oligopeptide is space filling (C = green, N = blue, O = red, Pd = pink, H = white). Counter ions have been omitted for clarity. ......................................................................................... 12 Figure 1.3. Diagram and solid-state structure of chalice shaped coordination calix[4]arene stabilized by rhenium(I)-N bonds. Solvent in the cavity has been removed for clarity. The position of bromide and carbonyl ligands is crystallographically disordered(C = green, N = blue, O = red, H = white, Re = yellow, Br = brown). ................................................................................... 19 Figure 1.4. Solid-state structure of metallocavitand 2 in the 1,3-alternate conformation commonly adopted by calix[4]arenes in solution. a) Top-down view of the cavity. b) Side-on view showing 1,3-alternate conformation. Nitrate counter ions and protons are omitted for clarity (C = green, N = blue, O = red, Pt = yellow). ... 24 Figure 1.5. Coordination of gadolinium(III) yields closed bowl-shaped metallocavitand 4. In the solid state, two nitrato ligands are coordinated to gadolinium(III) in an exo fashion and inside the bowl is a coordinated aqua ligand. a) Diagram of the structure. b) Solid-state structure looking into the bowl. c) Side-on view showing the opening of the cavity. Uncoordinated counter ions and co-crystallized solvent have been omitted for clarity (C = green, N = blue, O = red, H = white, Pd = pink, Gd = yellow). ................................................................ 25 Figure 1.6. Solid-state structures of metallocavitands 5-7 represented as wireframe models with bound guest space filling. a) 5 with acetonitrile guest. b) Capsule assembly of 62 with four water molecules encapsulated. c) Triflate guest bound in the cavity of 7 (protons and unbound counter ions omitted for clarity). For all the structures: C = green, N = blue, O = red, H = white, Pd = pink, Pt = yellow, S = purple, F = brown. ....................................................................................... 28 Figure 1.7. Solid-state structure of double-bowl metallocavitand 10 represented as a wireframe models with bound PF6- and NO3- space filling. a) Top-down looking into the larger cavity. b) Side-on view showing selective recognition of NO3- and PF6- in the smaller and larger cavities, respectively. Additional counter ions and water molecules found outside the cavity are omitted for clarity (C = green, N = blue, O = red, H = white, Pd = pink, Pt = yellow, P = orange, F = brown). ..................................................................................................................... 30  viii  Figure 1.8. Solid-state structure of double-bowl metallocavitand 11b represented as a wireframe model with bound ether space filling. a) Top-down looking into the shallow cavity formed by 4,7-phenanthroline walls. b) Side-on view showing ether bound in the shallow cavity. Crystallographic disorder in the deep cavity prevented guest molecules from being identified. Counter ions have been omitted for clarity (C = green, N = blue, O = red, H = white, Pd = pink)........................................................................................................................ 32 Figure 1.9. Solid-state structure of metallocavitand 13. a) Space-filling model looking directly into the cavity. b) Wireframe model of capsular assembly 132 with six cis-stilbene guest molecules trapped inside, each a different color and modeled as space filling. Counter ions and protons are omitted for clarity in part b) (C = green, N = blue, H = white, Pd = pink). ...................................................... 33 Figure 1.10 Solid-state structure of metallocavitand 15 in a wireframe representation with protons and counter ions omitted for clarity. a) Side-on view of the assembly. b) Top-down view looking into the cavity (C = green, N = blue, Rh = yellow). ........... 35 Figure 1.11. Solid-state structure of metallocavitand 20. Counter ions, protons, and co-crystallized solvent molecules have been omitted for clarity. a) Top-down view showing the hexagonal arrangement of RhCp* units. b) Side-on view with the 6-purinethione metallocavitand walls represented space filling. The sugar substituents cap the cavity from the top and bottom (C = green, N = blue, O = red, S = purple, Rh = yellow).......................................................................................... 38 Figure 1.12. Solid-state structures depicted as wireframe models with cocrystallized solvent molecules and protons omitted for clarity (C = green, N = blue, O = red, Ru = yellow, Li = violet, Cl = brown). a) Molecular triangle 22. b) Top-down view of metallocavitand 24 looking into the cavity. c) Top-down view of 24-LiCl with a lithium ion chelated by three O-donors of the metallacrown. d) Side-on view of 24-LiCl showing the tetrahedral geometry of the lithium ion and capping chloride ligand. .............................................................. 41 Figure 1.13. Solid-state structures of metallocavitands: a) 26 with cymene guest of another metallocavitand represented space filling. b) Side-on view of 27 with chloroform guest space filling. c) Top-down view of 27. For each structure the metallocavitand is modeled as a wireframe and co-crystallized solvent molecules have been omitted for clarity (C = green, N = blue, O = red, H = white, Ru = yellow, Cl = brown). ....................................................................................................... 42  ix  Figure 1.14. Solid-state structures of tantalum(V) metallocavitands 28 and 32 represented as wireframe models with guest molecules space filling. a) Side-on view of 28 with a guest THF molecule simultaneously interacting with a Lewis-acidic boron center and H-bonding to the central μ3-OH ligand. b) Top-down view looking into the bowl of 28 with bound THF. c) Sideon view of metallocavitand 32 with a guest acetone molecule simultaneously interacting with a Lewis-acidic boron center and H-bonding to the central μ3-OH ligand. In each representation co-crystallized solvent molecules have been omitted for clarity (C = green, O = Red, H = white, Ta = yellow, B = pink, F = brown). .......... 44 Figure 1.15. Solid-state structures of metallocavitands 33 and 34. a) Top-down view of silver(I) stoppered 33. No guest molecules were located in the cavity by crystallography. The BF4- counter ion has been omitted for clarity. b) Side-on view of 33. c) Top-down view of 34 with host-guest interactions exhibited with the pyridyl arm of another metallocavitand represented as space filling. d) Side-on view of 34 (C = green, N = blue, O = red, H = white, Re = yellow, Ag = purple). ............................................................. 46 Figure 1.16. Solid-state structures of copper(I) metallocavitands represented as wireframe models with space-filling guest molecules. a) Side-on view of 35a with a PF6- ion in the cavity. b) Top-down view of 35a. c) Side-on view of 35b where two PF6- ions are bound, one in the larger cavity made with aromatic walls and one in the smaller cavity formed by CO ligands. d) Side-on view of 35c with encapsulated ClO4- ions trapped by the small and large cavity of two separate metallocavitands. In each instance, co-crystallized solvent molecules and free counter ions have been omitted for clarity (C = green, N = blue, O = red, H = white, Cu = yellow, F = brown, Cl = cyan, P = orange). .......... 48 Figure 1.17. Solid-state structure of double-bowl shaped tetrazinc(II) metallocavitand 36 represented as a wireframe model with guest ClO4anions space filling. a) Top-down view of the cavity. b) Side-on view showing the double-bowl geometry and the μ3-OH ligand H-bonding with an O atom of ClO4- in the top cavity. Free counter ions and co-crystallized solvent molecules have been omitted for clarity (C = green, N = blue, O = red, H = white, Zn = yellow, Cl = brown). ....................................................................................................... 50 Figure 1.18. Solid-state structures showing the all-up conformation adopted by metallocavitands 37 and 38. a) Side-on view of 37 with ethylacetate bound to rhodium(II) in the cavity. b) Top-down view of 37. c) Side-on view of 38 with THF molecules bound to both rhodium(II) centers in an exo/endo fashion. d) Top-down view of 38. Each model is wireframe with bound guest molecules space filling (C = green, N = blue, O = red, H = white, Rh = yellow, Cl = brown). .............................................................................................. 52  x  Figure 1.19. Solid-state structures of metallocavitands 41 and 42 highlighting the deeper and smaller diameter cavity of 42 due to the sulfonato bridging unit compared to thiol. a) Side-on view of 41 with bound methylcarbonate. b) Top-down view of 41. c) Side-on view of 42 with bound 3-chlorobenzoate. d) Top-down view of 41. For both structures counter ions and co-crystallized Solvent molecules have been omitted for clarity. The host is represented as a wireframe model and the metal bound guest molecule is space filling (C = green, N = blue, O = red, H = white, Ni = yellow, S = purple, Cl = brown). ......... 55 Figure 1.20. Solid-state structure of 412 dimer bridged by terephthalate. The capsule halves are rotated by 70° from one another. a) Side-on view of dimeric capsule with bound terephthalate (C = green, N = blue, O = red, H = white, Ni = yellow, S = purple). b) Side-on view of the capsule. c) Space-filling representation showing how isolated the bridging guest molecule is from solution. In b) and c) the capsule halves are colored red and blue and terephthalate is green. ..................................................................................................... 58 Figure 1.21. Solid-state structures highlighting the curvature of larger diameter Schiff base macrocycle metal complexes. a) Dizinc(II) complex of 46-2H+. b) Tetranickel(II) complex of 47-4H+. c) Heptacopper(II) complex of 48-6H+. This structure is actually a dimer bridged in the center by four azide ligands that have been omitted from this representation. Counter ions, co-crystallized solvent, and some coordinated ligands have been omitted for clarity (C = green, N = blue, O = red, H = white, Zn = purple, Ni = yellow, Cu = cyan). ...................................................................................................................... 60 Figure 1.22. Depiction of [3+3] Schiff base macrocycle 51 highlighting the a) three N2O2 salphen-like pockets and b) central crown ether-like cavity. ....................... 63 Figure 2.1. ORTEP depictions of heptazinc metallocavitand 52d crystallized from DMF. a) Top-down view of the complex. b) Side-on view of the complex. c) Side-on view of the complex with the Zn–O bonds of the central [Zn4O]6+ cluster highlighted in orange (some bonds have been removed for clarity). All ellipsoids are at 50% probability with hydrogen atoms omitted (C = black, N = blue, O = red, Zn = green). ...................................................................................... 85 Figure 2.2. a) The tetrahedral basic zinc acetate cluster, [Zn4O(OAc)6]. b) Distorted tetrahedral [Zn4O]6+ cluster from complex 52d (the macrocycle portion of 52d has been omitted). Thermal ellipsoids are at 50% probability with hydrogen atoms removed for clarity. (C = black, N = blue, O = red, Zn =green). c) Schematic of the heptazinc cluster templated by macrocycles 51a-d with individual zinc ions and acetate ligands labeled for bond length comparison......................................... 87  xi  Figure 2.3. Two possible mechanisms for the formation of heptazinc complexes 52a-d. a) Basic zinc acetate forms in situ and coordinates to a trimetallated macrocycle, or b) the trimetallated macrocycle templates the cluster formation. Acetate ligands, coordinated solvent, and the central μ4-O ligand are omitted for clarity. ........................................................................................................................ 91 Figure 2.4. 1H NMR spectrum of tetrazinc complex 58 in DMF-d7 (400 MHz). The equivalent integration (6H) of three imine resonances, 5-6 aromatic resonances, and one acetate resonance is evidence for a single σh symmetry element and supports the inset structure (acetates have been omitted and the zinc ions are labeled analogous to complexes 52a-d in Figure 2.2c). DMF resonances are labeled with (*) and non-bonded water is labeled with (°). The magnetic environments about the OCH2 and C(CH3)3 protons are only slightly perturbed by the broken symmetry and therefore appear as broad singlet resonances rather than as multiple resonances. ........................................................................................... 93 Figure 2.5. Solid-state structure of tetrazinc metallomacrocycle 58 determined by aSCXRD experiment. a) Side-on view of complex 58 revealing the central aqua ligand (the zinc ions have been labeled according to Figure 2.2c). b) View from above: Zn4 is projected out of the page. c) Side-on view of Zn4 bridged by acetates to Zn1 and Zn2. An aqua ligand completes the coordination sphere of Zn3. Hydrogen atoms are omitted and thermal ellipsoids are at 50% probability. Peripheral alkoxy chains have been omitted from a) and b) for clarity (C = black, N = blue, O = red, Zn = green). ........................................................ 95 Figure 2.6. 1H NMR spectra of the downfield and acetate regions during the stepwise addition of Zn(OAc)2 in DMF-d7 to a solution of tetrazinc complex 58, resulting in the formation of heptazinc metallocavitand 52d. The numbers on the left indicate the molar equivalents of zinc ions to macrocycle, (*) indicates DMF, and (°) indicates free acetic acid which accumulates as complex 52d forms. ...................... 96 Figure 2.7. Variable temperature 1H NMR spectra of tetrazinc complex 58 in DMF-d7 (400 MHz). As temperature is increased, the downfield imine and aromatic resonances coalesce, consistent with a shift from Cs to average 3-fold symmetry. This higher symmetry may be a result of the out-of-plane Zn2+ ion exchanging between three equivalent environments. ..................................................... 97 Figure 2.8. Illustration of Zn2+ ring walking in polydentate macrocycles. a) Zn2+ exchanges between four equivalent sites in a sulfonylcalixarene. b) Proposed exchange of Zn2+ between three equivalent sites in tetrazinc complex 58. .... 98 Figure 2.9. 1H NMR spectra in DMF-d7 of complex 58, 58 + 1 equivalent of NaOAc, and 52d. ............................................................................................................ 99  xii  Figure 2.10. Solid-state structure of complex 60 as obtained by SCXRD analysis. a) View from the front. b) Top-down view c) Side-on view clearly showing the DMSO coordinated to Zn3. Thermal ellipsoids are shown at 50% probability. Hydrogen atoms and peripheral alkoxy groups have been omitted for clarity (C = black, N = blue, O = red, Zn = green, S = yellow). ................................... 101 Figure 2.11. a) MALDI-TOF isotopic distribution pattern of [(51d6H)Zn3Co3O(OAc)3]+. b) Simulated isotopic distribution of the same ion. ................. 103 Figure 3.1. 1H NMR spectra of 52e (5.00 mM) a) in selected solvents at room temperature and b) in p-xylene-d10 at the indicated temperatures. .............................. 119 Figure 3.2. Solid-state representations of dimer 52d2 encapsulating a DMF molecule. Co-crystallized DMF molecules are omitted for clarity. Only one orientation of the encapsulated DMF is modeled. a) Side-on view of the capsule (C = green, N = blue, O = red, H = white, Zn = yellow). b) Side-on view 90° from part a) with capsule halves colored red and blue. c) Top-down view of the dimer represented as space filling. d) Side-on view of the dimer represented as space filling. .......................................................................................... 124 Figure 3.3. VTVC imine 1H NMR chemical shift dependence of 52c (R = OC6H13) in toluene-d8. .......................................................................................... 127 Figure 3.4. van’t Hoff plot of 52c (R = OC6H13) in a toluene-d8. ............................... 128 Figure 3.5. VTVC imine 1H NMR chemical shift dependence of 52e (R = OC8H17) in benzene-d6.......................................................................................... 129 Figure 3.6. van’t Hoff plot of 52e (R = OC8H17) in benzene-d6. ................................ 130 Figure 3.7. VTVC imine 1H NMR chemical shift dependence of 52e (R = OC8H17) in toluene-d8. .......................................................................................... 130 Figure 3.8. van’t Hoff plot of 52e (R = OC8H17) in toluene-d8.................................... 131 Figure 3.9. VTVC imine 1H NMR chemical shift dependence of 52e (R = OC8H17) in p-xylene-d10. ...................................................................................... 132 Figure 3.10. van’t Hoff plot of 52e (R = OC8H17) in p-xylene-d10. ............................. 133 Figure 4.1. Depiction of heptacadmium complex 61 with the two distinct acetate environments labeled A (red) or B (blue). a) Side-on representation. Cd1, the third dialkoxyphenylenediimine unit, and one type- A acetate are obscured by the cluster. b) Top-down representation of 61 highlighting the acetate-Cd connectivity... 140  xiii  Figure 4.2. 1H NMR spectra before and after heptacadmium metallocavitand Formation. a) Macrocycle 51c with the triplet OCH2 resonance inset. b) Metallocavitand 61c showing the imine resonance with 34.1 Hz 3JCd-H satellites and the ABX2 coupling observed for the OCH2 resonance inset (400 MHz, CDCl3; residual CHCl3 is at 7.27 ppm). ....................................................................... 141 Figure 4.3. 113Cd NMR spectrum of 61c in CDCl3 (88.7 MHz). ................................. 142 Figure 4.4. Capsular assembly adopted by metallocavitand 61a (R = CH2CH3) in the solid state. a) Top-down view of the dimer. b) Side-on view of the dimer. c) Hexagonal long-range packing (view down the c-axis of the unit cell). (C = green, N = blue, O = red, H = white, Cd = cyan) ................................................. 145 Figure 4.5. Solid-state structure of 61a focusing on the Cd7(μ3-OH)(OAc)6 (H2O)2OH cluster with the peripheral OEt substituents omitted. The acetate colors and cadmium labels are from Figure 4.1. a) Side-on view. b) Top-down view. (C, O, and N of the macrocyclic scaffold = green, Cd = cyan, O aqua/hydroxo ligands = brown, central μ3-OH ligand = purple, type-A acetate = red, type-B acetate = blue). ........................................................................................... 146 Figure 4.6. Crystal structure of metallocavitand 61d (R = CH2C(CH3)3) with the cluster-capping aqua and hydroxo ligands replaced by an acetate ligand and DMF. The homo-dimer that is isostructural to 61a is also present in the unit cell, but is omitted from this figure. a) Capsular assembly with two guest DMF molecules space filling. b) Close up view of the cluster with the capping hydroxo and two aqua ligands replaced by a DMF (O-bound) and a bidentate acetate ligand. c) Close-up view of the encapsulated DMF molecules with the carbonyl O-Cd bond and a potential H-bond to the μ3-OH ligand highlighted. The neopentyloxy chains have been omitted for clarity (C = green, N = blue, O = red, H = grey, Cd = cyan). ... 147 Figure 4.7. Prototropic tautomers of metallocavitand 61. a) The μ3-oxo complex with three capping aqua ligands is unstable. b) The μ3-hydroxo species with two capping aqua ligands and a capping hydroxo ligand is the species observed. .............. 149 Figure 4.8. Overlay of geometry-optimized (at ZORA/TZP level) metallocavitand 61f (red), solid-state structure of 61a (green), and solid-state structure of 61d (blue). a) Side-on view. b) Top-down view. ................................................................. 150 Figure 4.9. Space filling representations of model metallocavitand 61f from ZORA/TZP DFT optimization. a) Top-down view highlighting the W-shaped hydrogen-bonding network of one hydroxo and two aqua ligands facially coordinated to Cd7 (Figure 4.1). b) Bottom-up view of the cavity with a central μ3-OH ligand bridging Cd4-6. (C = green, H = white, N = blue, O = red, Cd = cyan). ................................................................................................................... 151  xiv  Figure 4.10. Relative energies of calculated structures for scenario 1) one DMF molecule H-bonded to the μ3-OH ligand; 2) one DMF molecule coordinated to Cd; 3) two DMF molecules both coordinated to Cd ions; and 4) three DMF molecules all coordinated to Cd ions. ........................................................................... 153 Figure 4.11. Variable temperature 1H NMR spectra in DMF-d7 (400 MHz). a) Protons Ha-c are assigned to the imine and aromatic resonances of metallocavitand 61, primes (’) belong to the dimer. b) Dimerization of 61a results in coalescence of each resonance as the temperature is increased from -26 to 57 °C. c) No dimerization is observed for 61d at low temperature. The DMF resonance, calibrated to 8.03 ppm, is indicated with an *. ............................................................. 154 Figure 4.12. Downfield region in the 1H NMR spectrum of metallocavitand 61a at -24 °C (400 MHz, DMF-d7). Three sets of resonances belonging to DMF 61a, (DMF)2 61a·61a, and DMF 61a·61a are observed and labeled with a half capsule (red), whole capsule (red), and unsymmetrical whole capsule (red and blue), respectively. Resonances of the unsymmetrical capsule halves in DMF 61a·61a are perfectly distributed around the (DMF)2 61a·61a resonances. The DMF resonance, calibrated to 8.03 ppm, is indicated with an *. ....... 160 Figure 4.13. a) 1H NMR spectrum for the μ3-OH proton resonance belonging to monomeric metallocavitand 61a (DMF-d7, -24 °C, 400 MHz). Inset depicts the environment that results in the observed spin system. Cadmium labels are from Figure 4.1. b) Simulation of the spin system with JH-Cd = 13 Hz and 4 Hz of line broadening. ........................................................................................................ 162 Figure 4.14. Eyring plot for dimerization of 61a in DMF-d7 constructed with the coalescence method (ΔH‡ = 69 ± 13 kJ mol-1 and ΔS‡ = -410 ± 60 J K-1 mol-1). ........................................................................................ 164 Figure 4.15. Dynamic ligand exchange in DMF-d7 is confirmed by coalescence of the acetate resonances in variable temperature 1H NMR spectra. a) Upon cooling a solution of metallocavitand 61a, two sets of two acetate resonances emerge with each set corresponding to monomer or dimer. b) In the absence of dimerization, only a single set of two acetate resonances is observed for metallocavitand 61d. Rates of exchange were calculated from the simulated resonances shown in blue. In each spectrum, the additional broad resonance is assigned to the coordinated aqua/hydroxo capping ligands. c) Eyring plot of acetate exchange for metallocavitand 61d (r2 = 0.992). ................................................................................. 165  xv  Figure 4.16. a) Proposed mechanism for formation of the tri-DMF intermediate that initiates both acetate exchange (top right) and dimerization (bottom right). The rate determining step involves formation of two Cd-O (DMF) coordination bonds after the μ3-OH----ODMF hydrogen bond is broken. In the top right, rapid cleavage of the Cd-OAc bond is promoted by guest DMF molecules trans to the acetate ligands and followed by exchange between type- A and B acetates, depicted in red and blue, respectively (much of the cluster has been omitted for clarity). In the bottom right, dimerization proceeds first by formation of a pre-capsule assembly followed by loss of coordinated DMF and fast dimerization of the unsaturated metallocavitands. b) Proposed reaction coordinate diagram leading to the tri-DMF coordinated intermediate. ......................................................................... 167 Figure 4.17. DFT (D95V(d,p)/SDD/B98) energy-minimized structure of tri-solvated intermediate where all DMF guest molecules are coordinated to cadmium ions 1-3 (Figure 4.1) in 61f and none is interacting with the μ3-OH ligand. a) Top-down view. b) Side-on view.................................................................................. 168 Figure 4.18. DFT (D95V(d,p)/SDD/B98) energy-minimized geometry of DMF hydrogen bonding inside the cavity of 61f. a) Top-down view looking down the 3-fold axis of rotation. b) Side-on view showing the H-bond. DMF is depicted as space filling............................................................................................................... 170 Figure 4.19. 2H NMR powder lineshapes for: a) a rigid C-D bond and b) a methyl group rotating at a rate which is fast with respect to the static line width. ................... 171 Figure 4.20. a) The experimental and, b) simulated solid-state 2H NMR spectra of DMF-d7 trapped in the capsules of 61a at 184 K. There are two overlapping powder patterns contributing to the spectrum. The more intense contribution is due to the rotating -CD3 groups and the broader, less intense contribution is due to the amide deuteron. ................................................................................................... 172 Figure 4.21. The experimental and simulated solid-state 2H NMR spectra for DMF-d7 trapped in the capsules of 61a measured as a function of temperature. Model 1 is a 3-fold rotation of the N-(CD3)2 moiety about the bisector of the (CD3)-N-(CD3) bond angle. Model 2 is an overall 3-fold molecular rotation about an axis defined by the DFT calculations. ..................................................................... 174 Figure 4.22. Arrhenius plot for the rotation of DMF-d7 in the capsules of 61a (r2 = 0.995). ............................................................................................................ 175 Figure 4.23. Variable temperature 1H NMR of 61d in DMF-d7 (400 MHz). .............. 180 Figure 4.24. VTVC imine 1H NMR chemical shift dependence of 61e in benzene-d6. .................................................................................................................... 181 Figure 4.25. van’t Hoff plot of 61e in benzene-d6. ...................................................... 182  xvi  Figure 4.26. VTVC imine 1H NMR chemical shift dependence of 61e in toluene-d8. ..................................................................................................................... 183 Figure 4.27. van’t Hoff plot of 61e in toluene-d8......................................................... 184 Figure 4.28. VTVC imine 1H NMR chemical shift dependence of 61e in p-xylene-d10................................................................................................................... 184 Figure 4.29. van’t Hoff plot of 61e in p-xylene-d10. .................................................... 185 Figure 4.30. Variable temperature 1H NMR spectra of 61a in DMF-d7 (400 MHz). .. 186 Figure 4.31. Eyring plot of OAc exchange on metallocavitand 61d in DMF-d7. ........ 188 Figure 4.32. ORTEP depictions of 61a – 1st crystal. Ellipsoids are at 50% probability and hydrogen atoms are omitted for clarity (C = black, N = blue, O = red, Cd = yellow). ................................................................................................. 189 Figure 4.33. ORTEP depictions of 61a – 2nd Crystal. Ellipsoids are at 50% probability and hydrogen atoms are omitted for clarity (C = green, N = blue, O = red, Cd = cyan). .................................................................................................... 190 Figure 4.34. Thermal ellipsoid plot of 61d. Ellipsoids are at 50% probability and hydrogen atoms are omitted for clarity (C = green, N = blue, O = red, Cd = cyan). .... 191 Figure 5.1. a,b) TEM, c,d) SEM, and e,f) AFM micorgraphs of [Na·51c]BF4. All samples were prepared by drop-casting a chloroform solution of [Na·51c]BF4 onto formvar carbon-coated grids (TEM and AFM) or aluminum stubs (SEM) and dried at ambient condition. ..................................................................................................... 218 Figure 5.2. Concentration of free p-anisidine vs. time upon introduction of two equivalents of 3,5-dimethylaniline to N-salicylidene-p-anisidine, 62, at 57 °C. .......... 232 Figure 6.1. Reported N2O2 Pt2+ Schiff base monomer and conceptual evolution of the monomer into a head-to-tail self-assembling metallocycle. .................................. 240 Figure 6.2. Solid-state structure of complex 80a. a) Top-down view highlighting the 90° angle that directs tetrameric self-assembly when the peripheral phenyl group is replaced by a 3-pyridyl group. b) Side-on view. Hydrogen atoms have been omitted (C = green, O = red, N = blue, Pt = yellow, S = purple). ............... 242  xvii  Figure 6.3. a) Equilibrium between DMSO-bound model complex 80a and pyridine-bound model complex 80b. b) 1H NMR spectra of model complex 80a in DMSO-d6 when titrated with pyridine (25 mmol L-1, 400 MHz). Equivalents of pyridine are given on the left and the resonances integrated for thermodynamic analysis are color coded on top. .................................................................................... 243 Figure 6.4. Aggregates of Pt4 ring 90a observed with MALDI-TOF MS. .................. 248 Figure 6.5. a) 1H NMR spectrum of 90d in CDCl3 (400 MHz) and an inset of the chemical structure. Resonances of the peripheral, R = tris(4-tBuPh), aromatic rings overlap with the residual CHCl3 resonance calibrated to 7.27 ppm. b) Variable temperature 1H NMR spectra of 90d in CDCl3 (2 mmol L-1) from -45 to 25 °C. Monomer-dimer equilibrium is inset and resonances are color coded (monomer = red, dimer = blue). c) Van’t Hoff plot for dimerization of 90d in CDCl3. ................... 250 Figure 6.6. 2D ROESY 1H NMR spectrum of 90d in CDCl3 with some of the most intense exchange cross peaks circled in red (400 MHz). .............................................. 251 Figure 6.7. a) DFT optimized geometry of Pt4 ring 90a. b) Computer model of two possible orientations for columnar aggregation. Each ring is stacked directly on the other in the syn orientation whereas the anti orientation exhibits alternating AB type stacking. .......................................................................................................... 252 Figure 6.8. MALLS analysis of 90b in CHCl3 a) Kratky plot (0.6 mg mL-1). b) Zimm plot (0.4, 0.6, and 0.8 mg mL-1). ........................................................................ 254 Figure 6.9. POM images observed under crossed polarizers of growing LC textures for 90b from a) CHCl3, and b,c) PhCl. Black indicates an isotropic phase. ................. 255 Figure 6.10. Normalized wide-angle PXRD patterns of as-prepared Pt4 rings: a) 90a, b) 90b, c) 90c, d) 90d, e) dried mesophase of 90b drop-cast from CHCl3 onto amorphous silicon, and f) dried mesophase of 90b dropcast from PhCl onto amorphous silicon with assigned indices for hexagonal ordering. All data is depicted from 2º to 30º 2θ. .......................................................................... 258 Figure 6.11. Low magnification TEM images of Pt4 ring 90b dried from various solvents. a) “Pill-shaped” oblate aggregates from cyclohexanone. b) Rigid rod-shaped bundles from CHCl3. c) Micron length flexible fibers from C6H6. High magnification TEM images of Pt4 ring 90b from cyclohexanone. d) Individual columnar arrays are visible with periodic spacing of roughly 4 nm. e) Model of columnar aggregates inset. ........................................................................ 260 Figure 6.12. 2D COSY NMR spectrum of 90d in CDCl3 (400 MHz)......................... 274 Figure 6.13. 1H NMR spectroscopic assignment from 2D COSY and ROESY of 90d2 in CDCl3 (400 MHz, 7.27 ppm) ........................................................................... 274  xviii  Figure 6.14. Thermal ellipsoid plot of 80a top and side view (C = green, O = red, Pt = yellow, N = blue, S = beige, H = white)................................................................ 275 Figure 7.1. Computer model of a heptacadmium metallocavitand, 97, templated by macrocycle 96 (R = CH2CH3). a) Side-on view. b) Looking into the cavity (C = green, N = blue, O = red, H = white, Cd = yellow). ..................................................... 293 Figure 7.2. Proposed supramolecular sensor composed of anthracene “loaded” non-covalent nanotubes. A stimulus that triggers disaggregation of the nanotubes would free encapsulated anthracene, turning on or altering the luminescence. ............ 296  xix  List of Schemes Scheme 1.1. Metal-ligand ring closing of coordination catenanes. Formation of a a) palladium(II)-catenane and b) platinum(II)-catenane. The platinum(II)catenane is considered a photoswitchable molecular lock because the platinum(II)-pyridine bond is only labile under forcing conditions. ................................ 6 Scheme 1.2. Copper(I) templated synthesis of a molecular trefoil knot. ......................... 8 Scheme 1.3. Self-assembly of a Pd6L4 coordination cage. ............................................. 11 Scheme 1.4. Self-assembly of a tetrahedral M4L6 coordination cage. ........................... 13 Scheme 1.5. Synthesis of pyrogallol[4]arene coordination cages. Alkyl chains and co-crystallized solvent have been omitted for clarity (C = green, O = red, H = white, Zn = cyan, Cu = yellow). ............................................. 17 Scheme 1.6. Synthesis of a Pd6L8 stella octahedron from tris(isonicotinoyl)cyclotriguaiacylene and palladium(II). Blue lines highlight the octahedral arrangement of palladium(II) ions, each palladium(II) is depicted space filling, and the organic ligands are wireframe (C = green, N = blue, O = red, Pd = purple). Counter ions and hydrogen atoms are omitted for clarity. ............................. 18 Scheme 1.7. General strategies for synthesizing metallocavitands. Metal ions are represented as blue spheres. a) Aggregation of aromatic or bulky ligands (green piece) with metals into double bowl- or bowl-shaped geometries. Depending on the diameter of the openings and angles of the ligand walls, ring or closed bowl geometries may be adopted. b) Rigid or flexible macrocycles (yellow) coordinate metal ions in the interior resulting in a bowl configuration. .................................................................................................. 22 Scheme 1.8. Synthesis of Pt4(uracil)4 metallocavitand 2 and the structure of a classic calix[4]arene for comparison. ...................................................................... 23 Scheme 1.9. Equilibrium between all-up cone and 1,3-alternate conformations of calix[4]arene. ...................................................................................... 24 Scheme 1.10. Synthesis of metallocavitands 5-7. .......................................................... 27 Scheme 1.11. Synthesis of hexanuclear double-bowl metallocavitands 9 and 10. ........ 30 Scheme 1.12. Synthesis of double-bowl metallocavitands 11a-c and 12a-c. ................ 31 Scheme 1.13. Synthesis of metallocavitands 13 and 14. ................................................ 33 Scheme 1.14. Synthesis of rhodium(III) metallocavitand 15. ........................................ 35 xx  Scheme 1.15. By reacting a tridentate ligand with piano-stool or dimeric complexes of η6-arene ruthenium(II) or η5-pentamethylcyclopentadienyl rhodium(III)/iridium(III), cyclic half-sandwich metallocavitands are isolated. Most often cationic or neutral cyclic trimers (n = 3) are formed; however, larger cycles have been reported (n = 4 or 6). X may be solvent molecules or anions, and R indicates alkyl substituents. ................................................................. 36 Scheme 1.16. Synthesis of molecular triangles 21 and 22, metallocavitands 23, 24, and 25, and their host-guest ion pair complexes. ................................................ 40 Scheme 1.17. Synthesis of tricyclic half-sandwich tantalum(V) boronate complexes 28-32. ............................................................................................................ 43 Scheme 1.18. Synthesis of tri-rhenium(I) silver(I) metallocavitand 33 and the photoinduced loss of silver(I) yielding metallocavitand 34. Metallation/demetallalation of silver(I) is reversible. ............................................... 45 Scheme 1.19. Synthesis of metallocavitands 35a-c........................................................ 47 Scheme 1.20. Synthesis of double-bowl shaped metallocavitand 36. ............................ 49 Scheme 1.21. Synthesis of metallocavitands 37 and 38 depicted in the stable all-up conformation. S is solvent and occupies the active site for metal-substrate binding. .................................................................................................. 51 Scheme 1.22. Synthesis of bimetallic metallocavitands 41-45 from flexible macrocycles 39 or 40. ..................................................................................................... 53 Scheme 1.23. Regioselective cis-bromination of cinnamic acid using dicobalt metallocavitands 44 and 45. Steric constraints inside the cavity prevent a backside bromide attack on the bromonium intermediate resulting in the observed syn-addition. Reduction of cobalt(III) to cobalt(II) followed by HCl treatment releases the substrate. .............................................................................. 57 Scheme 2.1. The reaction of shape-persistent, conjugated Schiff base macrocycles 51a-d with zinc acetate yields heptazinc cluster complexes 52a-d. The seventh zinc ion and sixth acetate ligand are obscured by the depiction of the cluster. ................................................................................................... 82 Scheme 2.2. Preparation of 4,5-diamino-1,2-dineopentyloxybenzene 55d and macrocycles 51a-d. ........................................................................................... 84 Scheme 3.1. Intermolecular hydrogen bonding drives dimerization of a modified resorcin[4]arene yielding a capsule. .............................................................. 116  xxi  Scheme 3.2. Synthesis of heptazinc metallocavitand 52e from octyloxy substituted macrocycle 51e. .......................................................................................... 118 Scheme 3.3. Loss of solvent molecules from the cavity of a monomeric metallocavitand is responsible for the observed entropy-driven dimerization. The actual number of solvent molecules interacting with the cavity is unknown. ....................................................................................................................... 122 Scheme 4.1. Synthesis of heptacadmium metallocavitands 61a-e. .............................. 139 Scheme 4.2. Model for entropy-driven dimerization of metallocavitands, rationalized by solvent expulsion in the monomer-dimer equilibrium (S = generic solvent molecules). Multiple solvent molecules interact with the cavity through H-bonds, Cd-coordination, and/or van der Waal’s interactions (represented by dashed lines). ................................................................... 155 Scheme 4.3. Equilibria between DMF 61a, (DMF)2 61a·61a, and DMF 61a·61a. Encapsulated DMF is not able to invert making the capsule halves unsymmetrical in DMF 61a·61a (red and blue). .............................. 159 Scheme 5.1. Aldimine and ketimine exchange equilibrium. ........................................ 209 Scheme 5.2. Synthesis of dihydroxydibenzoyl compounds 67 and 68. ....................... 210 Scheme 5.3. Synthesis and thermal ellipsoid plot of model compound 69. Ellipsoids are at 50 % probability (C = grey, N = blue, O = red, H = yellow). ............ 211 Scheme 5.4. Synthesis of 70 and 71a-b. ...................................................................... 212 Scheme 5.5. Synthesis of macrocycle 72. .................................................................... 213 Scheme 5.6. Synthesis of macrocycle 73. .................................................................... 214 Scheme 5.7. Synthesis of Macrocycles 74. .................................................................. 215 Scheme 5.8. Synthesis of 1,4-diformyl-2,3-dihydroxytriptycene 77. .......................... 216 Scheme 5.9. Synthesis of triptycene substituted macrocycle 78. ................................ 216 Scheme 6.1. Synthesis of model complex 80a. ............................................................ 241 Scheme 6.2. Synthesis of N-ONO imines 85a-d. ......................................................... 244 Scheme 6.3. Synthesis of 2-amino-4-(2-hexydecyl)phenol, 82. .................................. 245 Scheme 6.4. Synthesis of tris(phenyl)(3-amino-4-hydroxyphenyl)methane, 83. ......... 245  xxii  Scheme 6.5. Synthesis of tris(4-tbutylphenyl)(3-amino-4-hydroxyphenyl) methane, 84. .................................................................................................................. 246 Scheme 6.6. Head-to-tail self-assembly of Pt4 rings 90a-d. i) N-ONO salicylaldimine pro-ligand 85a-d (or corresponding amine and aldehyde), K2CO3, and K2PtCl4, are heated in DMSO at 150 °C for 2-4 h. ................................... 247 Scheme 6.7. Dynamic assembly of Pt4 ring 90b into randomly oriented oligomers, elongated and oriented columns, and finally columnar nematic or disordered hexagonal columnar mesophases upon concentration. ............................... 256 Scheme 7.1. Reported reactions catalyzed by Zn4(OCOCF3)6O. a) Oxazoline formation from methylbenzoate and (S)-valinol. b) Chemoselective acylation of an alcohol in the presence of a primary amine. ........................................................ 290 Scheme 7.2. Proposed synthesis of heptazinc-hexatrifluoroacetate metallocavitand 91. ....................................................................................................... 291 Scheme 7.3. Proposed synthesis of 1,4-diformyl-2,3-dihydroxytriphenylene, 95. ...... 292 Scheme 7.4. Proposed synthesis of triphenylene containing macrocycle 96. .............. 293 Scheme 7.5. Proposed synthesis of water soluble Pt4 ring 98. ..................................... 295 Scheme 7.6. Proposed Synthesis of ethynyl extended Pt4 ring 100. ............................ 297  xxiii  List of Symbols and Abbreviations  Abbreviation  Definition  Å  angstrom (10-10 meter)  acac  acetylacetonato  AFM  atomic force microscopy (microscopy)  AMP  adenosine monophosphate  BSSE  basis set superimposition error  Bu  butyl  CCD  charge coupled device  CD  circular dichroism  CP  Boys and Bernardi counterpoise  Cp  cyclopentadienyl  Cp*  pentamethylcyclopentadienyl  CSD  Cambridge Structural Database  DCM  dichloromethane  dec.  decomposed  DFT  density functional theory  DLS  dynamic light scattering  DMF  N,N-dimethylformamide  DMSO  dimethylsulfoxide  DNA  deoxyribonucleic acid  en  ethylenediamine  equiv  equivalent(s)  ESI-MS  electrospray ionization mass spectrometry  EtOH  ethanol  FT  Fourier transform  h  hour(s)  HMPA  hexamethylphosphoramide  HRMS  high resolution mass spectrometry  xxiv  in situ  in the reaction mixture  iPr  iso-propyl  IR  infrared  Kassoc  association constant  Kdim  dimerization constant  LC  liquid crystal or liquid chromatography  LD50  median lethal dose to kill 50% of the tested population  LMCT  ligand-to-metal charge-transfer  M  molarity  MALDI-TOF  matrix-assisted laser-desorption ionization – time of flight  MALLS  multi-angle laser light scattering  Me  methyl  MeOH  methanol  min  minute(s)  mM  millimolar  mol  mole  MS  mass spectrometry  NAD+  nicotinamide adenine dinucleotide  nm  nanometers  NMR  nuclear magnetic resonance  NOE  Nuclear Overhauser Effect  NOESY  Nuclear Overhauser Effect Spectroscopy  OAc  acetate  ORTEP  Oak Ridge Thermal Ellipsoid Plot  OTf  triflate  Ph  phenyl  POM  polarized optical microscope  ppt  precipitate  PXRD  powder X-ray diffraction  ROESY  Rotational Frame Nuclear Overhauser Effect Spectroscopy  RT  room temperature  xxv  salphen  salicylidinephenylenediimine  SCXRD  single-crystal X-ray diffraction  SEM  scanning electron microscopy (microscopy)  tBu  tert-butyl  TD-DFT  time dependant density functional theory  TEM  transmission electron microscopy (microscopy)  terpy  2,2’:6’,2’’-tertpyridyl  THF  tetrahydrofuran  UV  ultra violet  UV-vis  ultra violet-visible  VTVC  variable-temperature variable-concentration  X Y  X is hosted by Y  XRD  X-ray diffraction  ZORA  Zero Order Regular Approximation  xxvi  Acknowledgements I am deeply indebted to my outstanding supervisor, Prof. Mark MacLachlan. His unique combination of wisdom, energy, and humor has made my years as a graduate student thoroughly enjoyable and educational. Without his guidance this thesis would not have been possible. Also, I have greatly appreciated the opportunities he has given me to explore some of my own ideas. This freedom has enhanced my graduate experience at UBC and encouraged me to pursue a research oriented career in the future. I owe much gratitude to my parents, Pete and Cathy. They have instilled in me a thirst for knowledge and a contagious optimism that has helped me through many of the frustrating moments of my PhD studies. Their unconditional love and support have allowed me to pursue my own interests and be the man I am today. Thanks to Chimpy for sharing many great adventures in the Pacific Northwest and always taking time to visit her brother. I want to thank Dr. Brian O. Patrick and Anita Lam of the UBC Chemistry Department X-ray facility for training me to run and interpret powder and single-crystal X-ray diffraction techniques. Their assistance over the years has made my research at UBC much more manageable. Marshall Lapawa, David Wong, Derek Smith, and Yun Ling of the microanalytical staff at the UBC Chemistry Department have been very helpful collecting my mass spectra and elemental analyses. In particular, Marshall and Derek have logged many hours on the matrix-assisted laser-desorption time-of-flight mass spectrometer to tease out peaks of the supramolecules presented throughout this thesis.  xxvii  I would like to thank all of the staff at the UBC Chemistry Department High Resolution NMR facility for their guidance over the years, especially Maria Ezhova for the many helpful discussions about variable-temperature and two-dimensional NMR spectroscopic experiments. Thanks to my fellow MacLachlan group members for good times both inside and outside the lab. I owe Dr. Amanda J. Gallant extra thanks for leaving me a great research project. I appreciate research contributions made to this thesis by group members Samuel Guieu, Raymond Tabeshi, Joseph K.-H. Hui, Jian Jiang, Angela Crane, and Jonathan H. Chong. I also want to recognize Xavier Roy for the many interesting discussions/debates we have had, for taking me where no ski was meant to go, and for being a great friend since the beginning of our studies. Thanks to Prof. Francesco Lelj for the many thoughtful discussion about zinc and cadmium metallocavitands. I am also appreciative of Prof. David L. Bryce, Dr. Phuong Y. Ghi, and Dr. Glenn A. Facey for the solid-state NMR spectra and interpretations. Without the contributions of these collaborators far less would be known about the cadmium metallocavitand system and Chapter four of this thesis would not be what it is. Thanks to Dr. Johan Janzen for his help running light scattering experiments. I am indebted to Prof. Peter Legzdins for proof reading parts of this thesis and offering constructive criticism. Lastly, thanks goes to bright pink cherry blossoms for making me smile while compiling this thesis.  xxviii  Dedication To my loving wife Meggan, Thanks for being the amazing person you are and for loving me despite my shortcomings. Your support over the past five years has been unwavering and the sacrifices you have made for us will not be forgotten. Wherever life takes us, together we will make it a memorable adventure.  xxix  Co-authorship Statement The work in this thesis was carried out under the guidance of Prof. Mark J. MacLachlan. Mass spectra and elemental analyses were performed by Marshall Lapawa, David Wong, and Derek Smith in the UBC Microanalytical Facility at the Department of Chemistry, University of British Columbia. Unless otherwise noted, I performed the single-crystal X-ray diffraction experiments, solved the structure, and refined the data with occasional assistance from Dr. Brian O. Patrick (Crystallographic Services, Department of Chemistry, University of British Columbia). Chapter 1: A version of this chapter will be submitted for publication: Frischmann, P. D.; MacLachlan, M. J. “Metallocavitands as Supramolecular Hosts” 2010. I wrote this chapter with input from Prof. MacLachlan and Prof. Peter Legzdins. Chapter 2: A version of this chapter has been published: Frischmann, P. D.; Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. “Zinc Carboxylate Cluster Formation in Conjugated Metallomacrocycles” Inorg. Chem. 2008, 47, 101-112. Dr. Amanda J. Gallant first reported the synthesis of 51a-c and 52a-c and initiated investigation of the template mechanism. Dr. Jonathan H. Chong synthesized, characterized, and performed the single-crystal X-ray diffraction/refinement of 60. I performed all other experiments and wrote the manuscript with input from Prof. MacLachlan. Chapter 3: A version of this chapter will be submitted for publication: Frischmann, P. D.; Gallant, A. J.; MacLachlan, M. J. “Entropy-Driven Dimerization of Zn7 Metallocavitands” 2010. Dr. Amanda J. Gallant synthesized and characterized 52e. I performed all other experiments and wrote the chapter with input from Prof. MacLachlan and Prof. Legzdins. Chapter 4: Versions of this chapter have been published: a) Frischmann, P. D.; Facey, G. A.; Ghi, P. Y.; Gallant, A. J.; Bryce, D. L.; Lelj, F.; MacLachlan, M. J. “Capsule Formation, Carboxylate Exchange, and DFT Exploration of Cadmium Cluster Metallocavitands: Highly Dynamic Supramolecules” J. Am. Chem. Soc. 2010, 132, 38933908. b) Frischmann, P. D.; MacLachlan, M. J. “Capsule Formation in Novel Cadmium Cluster Metallocavitands” Chem. Commun. 2007, 4480-4482. Dr. Amanda J. Gallant synthesized and characterized 61c. All solid-state NMR spectroscopy was carried out and xxx  interpreted by Prof. David L. Bryce, Dr. Phuong Y. Ghi, and Dr. Glenn A. Facey (University of Ottawa NMR Facility). All DFT calculations were performed by Prof. Francesco Lelj (Universita della Basilicata). I performed all other experiments and wrote each manuscript with input from Prof. MacLachlan. Chapter 5: Versions of this chapter have been published: a) Frischmann, P. D.; Jiang, J.; Hui, J. K.-H.; Grzybowski, J. J.; MacLachlan, M. J. “Reversible-Irreversible Approach to Schiff Base Macrocycles: Access to Isomeric Macrocycles with Multiple Salphen Pockets” Org. Lett. 2008, 10, 1255-1258. b) Hui, J. K.-H.; Frischmann, P. D.; Tso, C.-H.; Michal, C. A.; MacLachlan, M. J. “Spontaneous Hierarchical Assembly of Crown Ether-like Macrocycles into Nanofibers and Microfibers Induced by Alkali-Metal and Ammonium Salts” Chem. Eur. J. 2010, 16, 2453-2460. For publication a), Jian Jiang synthesized and characterized 71a-b, 72, and 73. Joseph K.-H Hui synthesized and characterized 74. Prof. Joseph J. Grzybowski isolated 68. I performed all other experiments and wrote the manuscript with input from Prof. MacLachlan. For publication b), I synthesized and characterized 75-78. All other experiments were performed by Joseph K.-H. Hui, Chien-Hsin Tso, and Prof. Carl A. Michal. Chapter 6: A version of this chapter has been published: Frischmann, P. D.; Guieu, S.; Tabeshi, R.; MacLachlan, M. J. “Columnar Organization of Head-to-Tail SelfAssembled Pt4 Rings” J. Am. Chem. Soc. 2010, 132, 7668-7675. Raymond Tabeshi synthesized and characterized 80a and performed some of the preliminary investigations under my guidance. Samuel Guieu synthesized and characterized 82 and 86-89. Joseph K.-H. Hui performed the transmission electron microscopy analysis and Angela Crane calculated the structure of 90a. Dr. Johan Janzen from the Department of Pathology and Laboratory Medicine helped with operation of the multi-angle laser-light scattering instrument in the Centre for Blood Research at the University of British Columbia. I performed all other experiments and wrote the manuscript with input from Prof. MacLachlan. Chapter 7: I wrote this chapter with input from Prof. MacLachlan.  xxxi  Chapter 1 Introduction† 1.1  Metallocavitands  and  Supramolecular  Coordination  Chemistry  1.1.1 Introduction to Supramolecular Chemistry In 1961, Charles Pedersen was employed at the Elastomer Chemicals Department of du Pont de Nemours and Co. in Wilmington, Delaware, when a serendipitous result caught his attention. A small impurity, lacking hydroxyl functionality, from the successful synthesis of a pentadentate bis-phenoxy ether was soluble in methanol. Furthermore, the dissolution of the impurity was enhanced by addition of soluble sodium salts. Realizing the novelty of an organic system that complexes sodium, Pedersen went on to design and characterize a family of similar compounds which he dubbed “macrocyclic polyethers”, now known as the ubiquitous crown ethers.1,2 The selective binding of alkali metals based on the dimensions of individual crown ethers marked the advent of supramolecular chemistry. Synchronously, across the Atlantic, Jean-Marie Lehn was finishing his Docteur ès Sciences at Centre National de la Recherche Scientifique. After graduating, Lehn traveled to Harvard were he applied his organic talents toward the total synthesis of vitamin B12 and witnessed the birth of the Woodward-Hoffman rules. These experiences fueled his interest in physical organic chemistry and upon relocation to the University of Strasbourg  †  A version of this chapter will be submitted for publication: Frischmann, P. D.;  MacLachlan, M. J. “Metallocavitands as Supramolecular Hosts” 2010.  1  he pondered how a chemist could impact biologists’ understanding of the nervous system. In particular, Lehn noticed that natural antibiotics facilitated the permeability of sodium and potassium cations across cellular membranes. With this concept in mind, Lehn went on to design the first cation binding cryptates, systems that played an instrumental role in guiding a rapidly expanding field that he later called “Supramolecular Chemistry”, chemistry of the intermolecular bond.3 Lehn’s research, at the boundary of biology and chemistry, significantly contributed to the conceptual development of self-assembly – the thermodynamically driven organization of simple components into complex architectures. Integral to self-assembling systems, molecular recognition was the central theme in the career of another supramolecular chemistry pioneer, Donald J. Cram. After twenty years investigating stereochemical reactivity and cyclophane development, Cram took Pedersen’s crown ether breakthrough and expanded upon it. At UCLA he developed spherands, locked calixarenes, and rigidified resorcinarenes, all concave host molecules that maintain their structural integrity in the absence of guests.4,5 Cram’s carcerands and hemicarcerands, derived from resorcinarenes, show strong affinity for many guests and display amazing synthetic versatility to bind specific guest molecules.6 The wealth of molecular recognition applications developed with these supramolecules has kept carcerands and analogues at the forefront of research even today. For the birth and impetus behind the development of synthetic supramolecular chemistry, Pedersen, Lehn, and Cram were awarded the Nobel Prize in 1987.  2  Look no further than your cup of coffee for a “taste” of supramolecular chemistry. Caffeine self-assembles as dimers in aqueous solution driven by π-π and solvophobic interactions.7,8 Another simple supramolecular construct often overlooked is the selfassembly of carboxylic acids into dimers via hydrogen bonding in non-polar solvents. The geometry of intermolecular interactions dictates the outcome of self-assembly and these two examples represent the simplest structural unit – a linear link.  Nature is a true master of supramolecular chemistry, regularly demonstrating the power of self-assembly. Inside each cell a balance of hydrogen bonding, π-π, electrostatic, and/or metal-ligand interactions causes massive proteins to fold into exact conformations, making life possible. Nucleic acid segments rich in guanine, especially telomeres, self-assemble into tetramers via eight hydrogen bonds and four carbonyl Oalkali cation electrostatic interactions yielding what is known as a G-quartet.9 These quartets block telomere extension, and prevent genomic instability. An extensive network of complementary hydrogen bonds between base pairs is responsible for zipping DNA  3  into its quintessential double helix. Nature’s self-assembly lessons are frequently adapted to synthetic systems where elaborate architectures have been conceived and generated with great success.10,11 Melamine and cyanuric acid are both nitrogen rich aromatic heterocycles with multiple hydrogen bond donors and acceptors. As industrial chemicals manufactured on a massive scale for multiple uses including fire retardants, herbicide precursors, bleaches, thermosetting plastics, etc., now they are best known for their central role in Chinese protein adulteration scandals. Due to their high nitrogen content and low cost these compounds were added to vegetable protein extracts, dairy products, and animal feed to boost the apparent protein concentration, determined by testing the nitrogen content under the assumption the only nitrogen source is protein. Although alone each of these molecules is considered non-toxic with LD50s of 3.0 and 7.7 g per kg of body weight for melamine and cyanuric acid, respectively, when combined in high concentration they self-assemble in a 1:1 ratio forming co-crystals of a two-dimensional hydrogen bonded network known as melamine cyanurate. Crystallization of melamine cyanurate in urinefilled renal microtubules results in acute renal failure and this supramolecular poisoning has affected hundreds of thousands of people worldwide.12 By functionalizing melamine and cyanuric acid derivatives, materials scientists have used the same hydrogen bonding motif to self-assemble disk-shaped molecules, rosettes, molecular “tapes”, fibers, columnar liquid crystals, and template linear arrays of silver and gold atoms onto surfaces.13-19 Supramolecular chemistry plays a dramatic but often unrecognized role in society today. The field has expanded rapidly, encompasses many diverse disciplines, and is now broadly defined as “chemistry beyond the molecule”.20 Most applicable to this thesis is the use of supramolecular chemistry for synthesis of complex molecules geared toward host-guest chemistry and material science applications. By relying on weak, reversible intermolecular interactions, supramolecular chemists are able to design systems that selferror check en route to a thermodynamic minimum, often eliminating kinetic byproducts. This thermodynamic sorting provides a dramatic advantage over traditional bond-by-bond covalent synthesis and has opened doors to extremely complex architectures often constructed in a few simple steps.  4  1.1.2 Supramolecular Coordination Chemistry Transition-metal-templated  self-assembly  has  become  a  paradigm  of  supramolecular chemistry and dramatically simplifies the synthesis of complex molecules. Metals generally act as nodes, linking multiple organic donors in a predictable geometry. Perhaps the most distinct advantage of using a metal-ligand self-assembly strategy is the variety of geometries available for transition metals. Coordination numbers commonly range from two to six and metals predictably adopt linear, tetrahedral, square planar, square pyramidal, or octahedral geometry, pushing self-assembly from one to three dimensions by simply changing the metal. A second advantage of supramolecular coordination chemistry is the relative strength of metal-ligand bonds compared to other intermolecular forces. Stronger dative interactions, but still non-covalent, enhance the thermodynamic stability of assembled architectures without sacrificing the reversibility intrinsic to self-assembling systems. Interaction strength and exchange kinetics may be optimized for a specific assembly by changing metals and/or ligand. A few examples demonstrating the versatility and synthetic power of supramolecular coordination chemistry are presented here.  Catenanes are mechanically interlocked macrocycles that exhibit connectivity similar to a chain link. First isolated in very low yield from organic ring closing reactions under high dilution,21 now the conventional synthesis involves using a supramolecular template followed by ring closing and gives catenanes in excess of 90% yield.22-24 One method devised by Fujita and coworkers to ring close or “lock” catenanes relies on platinum(II)-pyridine bonding and is an excellent example of tuning the thermodynamics and kinetics of a supramolecular system by changing metals.  5  Initially Fujita and coworkers reported the synthesis of a palladium(II)-pyridine ring and found that solvophobic interactions drove catenation after exposing the palladium(II)-pyridine ring to an aqueous environment as outlined in Scheme 1.1a.25 Because palladium(II)-pyridine bonds are labile at room temperature and exhibit rapid exchange, catenation was easily induced.22 In a separate study, an analogous platinum(II)-pyridine ring was isolated that did not undergo catenation in an aqueous environment. Unlike palladium(II)-pyridine bonds, platinum(II)-pyridine bonds are inert except when irradiated with UV light or heated in the presence of nitrate salts.26 After exposing the platinum(II)-pyridine ring to 330 nm irradiation for 15 minutes in water the platinum(II)-pyridine catenane shown in Scheme 1.1b was obtained quantitatively.27 In the absence of UV light, this catenane does not equilibrate back to the platinum(II)pyridine ring in non-polar solvents making it a photoswitchable molecular lock.  Scheme 1.1. Metal-ligand ring closing of coordination catenanes. Formation of a a) palladium(II)-catenane and b) platinum(II)-catenane. The platinum(II)-catenane is considered a photoswitchable molecular lock because the platinum(II)-pyridine bond is only labile under forcing conditions. 6  Building on the catenane concept, synthesis of more complex molecular architectures composed of multiple interlocking rings has also been facilitated by supramolecular coordination chemistry. Molecular trefoil knots and Borromean rings are two excellent examples of complex interlocking structures. A trefoil knot is a single stranded knot that may not be untied without cutting it and exhibits topological chirality. A Borromean link exists when three individual rings are locked in a manner where removal of one ring frees the other two. Borromean rings are topologically achiral. In the molecular synthesis of each, the crossover points or intersections between thread or ring components are templated by a metal ion followed by a ring closing reaction. In the absence of metals, complex polymeric mixtures are obtained upon ring closing.  A molecular trefoil knot was first reported by J.-P. Sauvage’s group using a substituted bis(1,10-phenanthroline) ligand templated by copper(I) as shown in Scheme 2.28,29 Crucial to the supramolecular trefoil knot template is the helical conformation adopted by the ligand as a result of the tetrahedral geometry of the copper(I) center when coordinated to 1,10-phenanthroline. Ring closing was performed on the helicate followed by quantitative demetallation with KCN to give the uncoordinated molecular trefoil knot in 1% yield. Advances in the synthesis of copper(I) templated trefoil knots have opened avenues to the stereoselective synthesis of a single topological enantiomer in excess of 70% yield.30  7  Scheme 1.2. Copper(I) templated synthesis of a molecular trefoil knot. Borromean rings were first expressed molecularly by Mao and Seeman using single stranded DNA.31 Since their pioneering work, Stoddart, Atwood, and coworkers devised a supramolecular coordination chemistry approach to afford the second successful synthesis of molecular Borromean rings in 2004 via self-assembly of eighteen components in one-pot.32 In their method, six five-coordinate zinc(II) centers with kinetically labile coordination bonds to endo-diiminopyridyl and exo-bipyridyl donors geometrically direct formation of twelve imine bonds yielding three interlocked macrocycles – Borromean rings. A solid-state structure and retrosynthetic analysis of the Borromean rings is shown in Figure 1. The coordination template was found to be essential; in the absence of a zinc(II) template only a complex mixture of oligomers and macrocycles was isolated.  8  Figure 1.1. Space-filling solid-state structure of a zinc(II)-templated molecular Borromean  rings  and  the  retrosynthetic  analysis  from  diformylpyridyl  and  diaminobipyridyl units. The three separate macrocycles (rings) are colored red, blue, and green and the zinc(II) ions are yellow (protons are omitted for clarity). Another exploration of molecular topology that has gathered a tremendous amount of interest is metal-directed self-assembly of 2D and 3D polygons and polyhedra. One archetypical 2D coordination supramolecule that takes advantage of the square planar geometry common for d8 transition metals is a palladium(II) cornered metallocycle self-assembled from cis protected ethylenediaminedinitratopalladium(II) and 4,4’bipyridine.33 Similar metallocycles have been assembled from platinum(II), and this general supramolecular coordination strategy using directional pyridyl donors and platinum(II)/palladium(II) acceptors has been employed to construct an impressive library of 2D supramolecular polygons.34,35 Due to reversibility of the metal-ligand interaction, thermodynamic sorting frequently yields multi-component assemblies in excess of 90%.  9  Increasing the number of donors or acceptors per building block beyond two provides access to 3D supramolecular assemblies.36,37 These assemblies, also known as coordination cages, generally have an internal void space that is occupied by small guest molecules; together, these are called a host-guest complex. Inside this unique environment, host-guest interactions may dramatically alter reactivity, conformation, photochemistry, and other properties of the encapsulated guest molecule(s). One particular octahedral shaped coordination cage that has been extensively studied is selfassembled from four rigid aromatic 2,4,6-tris(4-pyridyl)-1,3,5-triazine donors and six ethylenediaminedinitratopalladium(II) acceptors as shown in Scheme 1.3.38 It is common to abbreviate coordination cages based on the number of metals and ligands involved in the self-assembly, e.g., MxLy.  10  Scheme 1.3. Self-assembly of a Pd6L4 coordination cage. Tashiro et al. have shown that the octahedral Pd6L4 coordination cage depicted in Scheme 1.3 has a large internal volume that may host multiple guest molecules. In one study, encapsulation of tri-, tetra-, and hexa-peptide sequences was investigated and binding constants determined by UV-vis titrations were in the range of 103 to >106 mol-1 L suggesting future host systems may be tuned to bind specific oligopeptides.39 The result of a single-crystal X-ray diffraction (SCXRD) analysis of an encapsulated tryptophantryptophan-alanine residue is depicted in Figure 1.2 and shows that close π-π contacts between the triazine ligands and indole units of the oligopeptide may be partially responsible for the strong host-guest interaction. Inside the same Pd6L4 coordination cage, host-guest interactions activate naphthalene for Diels-Alder cycloadditions,40 direct catalytic regioselective Diels-Alder cycloadditions to anthracene,41 photooxidize alkanes,42 stabilize radical additions to o-quinones,43 and suppress kinetically favored photocleavage of α-diketones en route to thermodynamically favored cyclization products.44 The selectivity and, in some cases, reactivity of these transformations is unprecedented in bulk solution.  11  Figure 1.2. Solid-state structure of a Pd6L4 coordination cage. a) Cage framework. b) With encapsulated tryptophan-tryptophan-alanine oligopeptide. The oligopeptide is space filling (C = green, N = blue, O = red, Pd = pink, H = white). Counter ions have been omitted for clarity. Demonstrating the nearly limitless opportunities to construct supramolecular architectures with coordination chemistry, Raymond and coworkers simultaneously developed an entirely different strategy than Fujita for self-assembling tetrahedral M4L4 and M4L6 coordination cages using octahedral metals.45-47 Due to the vast amount of information collected on these systems, only some self-assembly and host-guest aspects of the M4L6 tetrahedron will be discussed here. Stirring four equivalents of an iron(III), gallium(III),  tin(IV),  or  titanium(IV)  salt  and  six  equivalents  of  a  bis(catecholamide)naphthalene ligand in basic methanolic solution gives M4L6 coordination cages in nearly quantitative yield (Scheme 1.4).48 Self-assembly from achiral components yields a racemic mixture of four Δ,Δ,Δ,Δ- or Λ,Λ,Λ,Λ- trischelated metal ions at the vertices of a tetrahedron with  six bis(catecholamide)naphthalene  ligands making the edges of the tetrahedron. The stereochemistry adopted by the first trischelated metal directs the stereochemical outcome of the assembly and racemization may occur at high temperature post-assembly.49 It is worth noting that changing from a  12  phenyl to naphthyl spacer in the bis(catecholamide) ligand system yields the M4L6 tetrahedron, a subtle change that avoids formation of the entropy favored M2L3 helicate.50  Scheme 1.4. Self-assembly of a tetrahedral M4L6 coordination cage. These M4L6 coordination cages have a large hydrophobic cavity roughly 500 Å3 in volume and a high negative charge enabling their dissolution in water. Optimal guest recognition is thus achieved by matching size and polarity of the guest to the host cavity. Initial  studies  involved  selective  encapsulation  of  tetraethylammonium  over  tetrabutylammonium and tetramethylammonium cations, an entropy driven process due to release of encapsulated solvent upon ammonium ion uptake.51 Interestingly, the guest “senses” the chirality of the host and the methylene resonance of encapsulated tetraethylammonium is a complex multiplet in the 1H NMR spectrum instead of the expected quartet indicating the protons of the tetraethylammonium ion trapped inside the chiral host are diastereotopic.47 In another experiment, Caulder et al. demonstrated their M4L6 host is capable of encapsulating alkali metal ion 12-crown-4 ether complexes, resulting in a host-guest-guest assembly reminiscent of Russian Matryoshka dolls.52 Kinetic analysis of guest exchange rates coupled with molecular mechanics calculations  13  revealed guest exchange proceeds through a host-deformation mechanism as opposed to a ligand-dissociation mechanism, highlighting the dynamic nature of this coordination cage.53 Recently Bergman, Raymond, and coworkers reported the size selective supramolecular hydrolysis of orthoformates, acetals, and ketals catalyzed within a M4L6 coordination cage in basic media.54-56 Most remarkably, these reactions were previously only known to proceed under acidic conditions; however, the M4L6 host bypasses this requirement because it thermodynamically stabilizes protonation of the encapsulated substrate (in basic conditions!), making the observed catalysis possible. This process obeys Michaelis-Menten kinetics and exhibits enzyme-like competitive inhibition.57 Encapsulated within the same system, catalytic intermediates of ruthenium have been stabilized and studied,58 iridium organometallic complexes exhibit diastereoselective substrate CH activation,59 and enammonium substrates undergo catalytic enantioselective aza-Cope rearrangements.60-62 Unique reactivity facilitated by supramolecular host-guest catalysts, also known as “molecular flasks”, is still being discovered and is a rapidly growing area of research.63 The focus of this introduction to supramolecular coordination chemistry is primarily on examples relevant to this thesis and merely scratches the surface of the field. A variety of structures has been reported with unique architectures and equally interesting properties that are beyond the scope of this discussion. Helicates,64,65 metal-DNA superstructures,10 metal-organic polyhedra,66 porphyrin and phthalocyanine assemblies,67 polyoxometallates,68,69 and metallocrowns70,71 are just a few other research areas pushing the boundaries of supramolecular coordination chemistry in new directions.  14  1.1.3 Traditional Cavitands and Their Metal Complexes “We propose the class name cavitand for synthetic organic compounds that contain enforced cavities large enough to accommodate simple molecules or ions.” - John R. Moran, Stefan Karbach, and Donald J. Cram5  α  When the term cavitand was introduced in 1982 there were primarily five classes of synthetic organic cavitands: calixarenes, resorcinarenes, cyclotriveratrylenes, 15  cucurbiturils, and cyclodextrins. Today, the same five cavitands dominate host-guest chemistry research and a wealth of information has been collected on their ability to complex guest molecules. Intrinsic to a cavitand, each possesses either a concave bowlor ring-shaped cavity that binds guest molecules by matching size and electrostatic interactions between host and guest. Generally the periphery, or rim, of a cavitand is easier to synthetically manipulate while simultaneously maintaining the enforced cavity necessary for guest complexation. Peripheral modifications are common to improve solubility in desired solvents, increase the depth of a cavity, and/or alter host-guest interactions near the portal of the cavity. Much like enzymes, by maximizing complementary electronic, steric, and chemical interactions between host and guest, discretion amongst potential guests is achieved in synthetic host-guest systems.72,73 Traditional cavitands have been extensively modified over the last 30 years altering their molecular recognition properties for use in host-guest catalysis,74-77 stabilizing and studying reactive intermediates,78,79 controlling photochemical reactivity,80-84 gas separation,85,86 fullerene binding and purification,87-94 molecular switching,95 molecular sensing,96-98 caging  129  Xe for NMR contrast agents,99-101 and molecular machines102,103  amongst other applications. By marrying organic cavitands with coordination chemistry, elaborate architectures and novel host-guest systems with catalytic, sensing, and/or optical properties of metals are accessible. Coordination cages assembled with cavitands as ligands may have massive dimensions, so large that some cages constructed in this fashion are capable of encapsulating traditional cavitands. Scheme 1.5 depicts the synthesis of a hexameric pyrogallol[4]arene assembly with octahedral symmetry stitched together by 24 copper(II) ions and multiple H-bonding interactions.104 Pyrogallol[4]arene is an analogue of resorcin[4]arene with an additional hydroxyl group on each aromatic ring. This design strategy was also feasible with gallium(III) ions and both supramolecular coordination cages have internal volumes greater than 1200 Å3.105 When pyrogallol[4]arene is reacted with zinc(II) ions, a face-to-face cavitand dimer is isolated, sealed together by an octametallic coordination belt (Scheme 1.5).106 These coordination dimers exhibit selective guest encapsulation as well as exohedral ligand substitution, promising features for future molecular recognition applications.107  16  Scheme 1.5. Synthesis of pyrogallol[4]arene coordination cages. Alkyl chains and cocrystallized solvent have been omitted for clarity (C = green, O = red, H = white, Zn = cyan, Cu = yellow). Hardie and coworkers have created equally elegant architectures with peripheral functionalized cyclotriveratrylenes. When their tripyridyl functionalized cavitand, tris(isonicotinoyl)cyclotriguaiacylene, was reacted with palladium(II) ions, a hollow Pd6L8 stella octahedron structure was obtained with palladium(II) ion vertices and tris(isonicotinoyl)cyclotriguaiacylene polyhedral faces as shown in Scheme 1.6.108 Deep coordination calix[4]arenes have been accessed via a similar peripheral functionalization route as demonstrated by de Mendoza’s research group.109 Depicted in Figure 1.3, coordination of four rhenium(I) ions to the 3,8-phenanthroline units installed on the  17  calixarene’s upper rim provided a facile route to a structurally sound deep cavitand, far superior to its H-bonded predecessor.110 The enhanced cavity size allowed for encapsulation of a traditional calix[4]arene.  Scheme 1.6. Synthesis of a Pd6L8 stella octahedron from tris(isonicotinoyl)cyclotriguaiacylene and palladium(II). Blue lines highlight the octahedral arrangement of palladium(II) ions, each palladium(II) is depicted space filling, and the organic ligands are wireframe (C = green, N = blue, O = red, Pd = purple). Counter ions and hydrogen atoms are omitted for clarity.  18  Figure 1.3. Diagram and solid-state structure of chalice shaped coordination calix[4]arene stabilized by rhenium(I)-N bonds. Solvent in the cavity has been removed for clarity. The position of bromide and carbonyl ligands is crystallographically disordered (C = green, N = blue, O = red, H = white, Re = yellow, Br = brown). Coordination of metals to pre-existing cavitands not only facilitates the formation of unique architectures for supramolecular study, it also provides unique environments for catalytic studies. Richeter and Rebek have modified a deep resorcin[4]arene to include a zinc(II) salphen (salicylidinephenylenediimine) unit near the periphery of the cavitand.111 This complex binds ammonium functional groups in the cavity and catalyzes various reactions at the metal active site.112 Acting harmoniously, the cavity sorts substrates by size and electrostatics which in turn locks a portion of the substrate in close proximity to the Lewis acidic zinc(II) salphen center. Specifically, this metallated cavitand system catalyzes the esterification of choline to acetylcholine and catalytically hydrolyzes select carbonates. Enzyme-like selectivity and rate enhancements are achieved due to synergy between the cavity and the metal active site.74 There are many other interesting examples of cavitand-metal complexes as host-guest receptors and/or catalysts.113-117  19  As the examples above illustrate, coordinating metals to the periphery or core of a traditional organic cavitand is the synthetic paradigm for constructing metal-containing host molecules; however, that is not the only route.  1.1.4 Metallocavitands: Coordination Induced Curvature Scattered throughout inorganic and supramolecular literature are examples where in the absence of traditional cavitands, multimetallic coordination complexes adopt a bowl- or ring-like geometry, the essence of a cavitand. Although many of these systems exhibit host-guest properties similar to traditional cavitand coordination complexes, they are intrinsically different as their concave geometry is dependant upon metal ligand interactions and not C-C bonds. In an effort to distinguish these two archetypes, we proposed adopting the loosely defined term metallocavitand to describe multimetallic complexes where metal coordination is necessary for cavity formation.118 While organic chemistry has provided foolproof routes to traditional cavitands, the often unpredictable nature of metal-ligand interactions has limited the adoption of metallocavitands by the supramolecular community. Also, a majority of structures that fit our definition of 20  metallocavitand are serendipitously discovered by materials scientists interested in solidstate properties (e.g., magnetism or gas adsorption), hence the solution integrity and hostguest  chemistry  are  never  investigated.  Paramagnetic  metal  centers  hinder  metallocavitand research as NMR spectroscopy is the technique of choice for investigating host-guest phenomena. This section focuses on successful strategies for isolation of metallocavitands, reviews some relevant structures, and highlights metallocavitands that exhibit supramolecular host-guest chemistry. Reliable methods developed for the synthesis of metallocavitands are based on aggregation or macrocycle-templated strategies and are outlined in Scheme 1.7. The aggregation approach relies on polydentate ligands acting as geometric directors being “glued” together by metal-ligand interactions, generally into a metallacrown configuration.70 Rigid aromatic or sterically bulky substituents on the polydentate ligands provide the walls of the cavity and self-assembly of the metallacrown linkage yields a cylindrical shape. Macrocycle templates without a pre-formed cavity typically coordinate multiple metals in their interior resulting in a conformation change from planar or flexible to a rigid cavity. These systems also rely on rigid aromatic or bulky substituents to form the walls of the cavity and the metal cluster stabilized near the bottom closes one end of the cavity. Clusters templated by macrocycles often have coordination sites accessible to guest molecules, a valuable molecular recognition feature of metallocavitands constructed in this fashion. The aggregation method frequently yields bowl-, double bowl-, or ring-shaped metallocavitands whereas only bowl-shaped cavities have been reported using a macrocycle template.  21  Scheme 1.7. General strategies for synthesizing metallocavitands. Metal ions are represented as blue spheres. a) Aggregation of aromatic or bulky ligands (green piece) with metals into double bowl- or bowl-shaped geometries. Depending on the diameter of the openings and angles of the ligand walls, ring or closed bowl geometries may be adopted. b) Rigid or flexible macrocycles (yellow) coordinate metal ions in the interior resulting in a bowl configuration. Most metallocavitands reported to date have been isolated by the aggregation method and two systems in particular have produced a variety of structures: 1) aggregation of nitrogen donor-supported palladium(II) and platinum(II) ions, and 2) aggregation of organometallic half-sandwich complexes. Metallocavitands synthesized by the aggregation method with other metal-ligand building blocks are discussed in a miscellaneous section and finally macrocycle-templated metallocavitands are covered.  22  1.1.5 Aggregation of N Supported – Pd(II) and Pt(II) Metallocavitands While investigating mixed-valence “platinum pyrimidine blues”, Lippert and coworkers reacted a chloro ethylenediamine uracil complex of platinum(II), 1, with AgNO3 and serendipitously isolated the tetranuclear metallocavitand 2 (Scheme 1.8).119 As a geometric analogue of calix[4]arene, complex 2 bridges four aromatic rings with Pt2+ ions instead of CH2 groups and exhibits dynamic solution behavior; in particular, 1H NMR spectroscopy reveals complex 2 exists in equilibrium between the 1,3-alternate and all-up cone conformations commonly adopted by calix[4]arene as shown in Scheme 1.9.120 Although the synthesis of 2 was first described in 1981, the solid-state structure of 2 was not reported until 1992 (Figure 1.4).121  Scheme 1.8. Synthesis of Pt4(uracil)4 metallocavitand 2 and the structure of a classic calix[4]arene for comparison.  23  Scheme 1.9. Equilibrium between all-up cone and 1,3-alternate conformations of calix[4]arene.  Figure 1.4. Solid-state structure of metallocavitand 2 in the 1,3-alternate conformation commonly adopted by calix[4]arenes in solution. a) Top-down view of the cavity. b) Side-on view showing 1,3-alternate conformation. Nitrate counter ions and protons are omitted for clarity (C = green, N = blue, O = red, Pt = yellow). Metallocavitand 2 crystallizes with the aromatic walls in the 1,3-alternate conformation where the aromatic units alternate pointing up or down. No guest molecules are found inside the cavity of the 1,3-alternate conformation due to its small diameter of approximately 3 Å; however, deprotonation of the four uracil hydroxyl units (-4H+) with Mg(OH)2 followed by coordination of zinc(II), beryllium(II), or lanthanum(III) to 2-4H+  24  results in a 100% conversion to the all-up cone conformation. This change to the cone conformation upon metal chelation results in a bowl-shaped cavity with an approximate 8 Å diameter at the opening. Both the zinc(II) and beryllium(II) complexes of 2-4H+ act as hosts for organic sulfonate anions, p-toluenesulfonate and 3-(trimethylsilyl)-1propanesulfonate, in solution confirmed by 2D- NOESY and ROESY  1  H NMR  spectroscopy.122 Metallocavitand 2 also interacts with calf-thymus DNA in a noncovalent manner leading to uncoiling and formation of long inflexible DNA strands.123,124  Figure 1.5. Coordination of gadolinium(III) yields closed bowl-shaped metallocavitand 4. In the solid state, two nitrato ligands are coordinated to gadolinium(III) in an exo fashion and inside the bowl is a coordinated aqua ligand. a) Diagram of the structure. b) Solid-state structure looking into the bowl. c) Side-on view showing the opening of the cavity. Uncoordinated counter ions and co-crystallized solvent have been omitted for clarity (C = green, N = blue, O = red, H = white, Pd = pink, Gd = yellow).  25  Metallocavitand  3,  constructed  with  4,6-dimethyl-3-hydroxypyrimidine,  palladium(II), and (R,R)- or (S,S)-diaminocyclohexane, has a geometry much like 2 and selectively recognizes adenosine 5’-monophosphate over guanosine 5’-monophosphate, thymidine 5’-monophosphate, and cytidine 5’-monophsophate in D2O.125 Coordination of gadolinium(III) to a similar metallocavitand, 4, constructed with 4,6-dimethyl-3hydroxypyrimidine, palladium(II), and ethylenediamine gave a bowl-shaped structure that was characterized in the solid state by SCXRD and is depicted in Figure 1.5.126 An aqua ligand is coordinated to gadolinium(III) inside the bowl. This coordination site located inside the cavity is a promising feature for molecular recognition and host-guest catalysis. Trimetallic metallocavitands have also been isolated with the aggregation method and the synthesis of complexes 5, 6, and 7 is shown in Scheme 1.10. Metallocavitand 5 was synthesized by stirring a bis(benzimidazole) pro-ligand in acetic acid with Pd(OAc)2 followed by crystallization from acetonitrile.127 The solid-state structure, depicted in Figure 1.6a shows a single acetonitrile guest molecule inside the cavity. Although 5 does not act as a host molecule in solution, Carina et al. demonstrated that by substituting ligands so that the C-Pd bond is replaced with a N-Pd bond, the metallocavitand does bind non-coordinating anions such as mesylate and perchlorate in solution.128 Another organometallic metallocavitand, 6, is synthesized by dehalogenating a platinum(IV) complex, fac-(PtMe3I)4, in acetone with silver acetate then reacting with 9methyladenine (Scheme 1.10).129 The solution host-guest chemistry of metallocavitand 6 was not investigated; however, crystals obtained from hydrated ether show capsular assemblies of 62 (subscript denotes dimer) with four water molecules trapped inside (Figure 1.6b). A hydrogen bonding network organizes the guest water molecules into a tetrahedron. When crystallized from acetone, monomeric metallocavitands are obtained with an acetone molecule bound in the cavity.  26  Scheme 1.10. Synthesis of metallocavitands 5-7.  27  Figure 1.6. Solid-state structures of metallocavitands 5-7 represented as wireframe models with bound guest space filling. a) 5 with acetonitrile guest. b) Capsule assembly of 62 with four water molecules encapsulated. c) Triflate guest bound in the cavity of 7 (protons and unbound counter ions omitted for clarity). For all the structures: C = green, N = blue, O = red, H = white, Pd = pink, Pt = yellow, S = purple, F = brown. Aggregation based synthesis of metallocavitand 7 was achieved by stirring 4(3H)-pyrimidine and cis-protected 5,5’-di(tert-butyl)-2,2’bipyridinedichloropalladium(II) in a 1:1 mixture of dry THF:DCM after treatment with AgOTf (OTf = CF3SO3-).130 Multiple SCXRD experiments were performed on 7 and the results show OTf, NO3-, and ClO4counter ions may be found inside the cavity. The triflate bound structure of 7 is depicted in Figure 1.6c and reveals a wide, shallow cavity with a triflate ion found in the center.  28  Solution studies revealed that weak host-guest interactions exist between 7 and OTf, NO3-, SO4-, and ClO4- in dichloromethane. Cavity walls of the previously discussed metallocavitands are composed of the rigid aromatic donors that bridge each metal center. Metallocavitand 7 represents a divergence from that motif because, although it exhibits a small cavity formed by the bridging ligands, the wide, shallow cavity where guest molecules are bound is due to the cis-protecting bipyridyl ligands. Conceptually, the cis-protecting ligands may be altered to tune the dimensions of the cavity without impacting the aggregation based assembly. This represents a distinct advantage metallocavitands have over classic organic cavitands. Due to the self-assembling nature of metallocavitands, simple changes in ligand type may dramatically affect the final cavity geometry whereas for classic organic cavitands, often multiple complex organic transformations are required to tune cavity dimensions. Aggregation based self-assembly of metals using 2,2’-bipyrazine yields metallocavitands with increased cavity dimensions like those shown in Scheme 1.11. Molecular  triangle  8  was  synthesized  by  treating  cis-protected  dichloroethylenediamineplatinum(II) with silver(I) followed by addition of 2,2’bipyrazine in water and reacting at 45 °C for 3 days.131 In Scheme 1.11, 8 is depicted as the all cis isomer; however, it is dynamic, and the 2,2’-bipyrazine ligands are in equilibrium between cis- and trans- conformations that are both detected in low temperature 1H NMR spectra. In D2O, 8 binds anions PF6-, ClO4-, BF4-, and SO42- with association constants ranging from 13-42 mol-1 L.132 To construct a deeper cavity, Schnebeck et al. reacted 8 with three equivalents of dichloroethylenediamineplatinum(II) or dichloroethylenediaminepalladium(II) locking 8 into the all cis-isomer and giving double-bowl hexanuclear metallocavitands 9 and 10, respectively.133 From the solid-state structure of 10, shown in Figure 1.7, it is apparent the two different size cavities (approximate diameters 15 and 10 Å) selectively recognize different anions; NO3- is bound to the smaller cavity with O-donors pointed toward platinum(II) centers and PF6resides in the larger cavity. The same anion selectivity is observed for 9 in the solid state and both metallocavitands 9 and 10 recognize anions in D2O with the greatest affinity for SO42-, Kassoc = 260 ± 60 mol-1 L. Notably, addition of silver(I) to 8 yields nonametallic coordination cages that encapsulate a variety of guests.  29  Scheme 1.11. Synthesis of hexanuclear double-bowl metallocavitands 9 and 10.  Figure 1.7. Solid-state structure of double-bowl metallocavitand 10 represented as a wireframe models with bound PF6- and NO3- space filling. a) Top-down looking into the larger cavity. b) Side-on view showing selective recognition of NO3- and PF6- in the smaller and larger cavities, respectively. Additional counter ions and water molecules found outside the cavity are omitted for clarity (C = green, N = blue, O = red, H = white, Pd = pink, Pt = yellow, P = orange, F = brown).  30  Using a similar aggregation strategy, Yu et al. isolated trinuclear double-bowl metallocavitands 11a-c and 12a-c, shown in Scheme 1.12, via self-assembly of cisprotected platinum(II) and palladium(II) complexes and 4,7-phenanthroline.134 The solidstate structure of metallocavitand 11b is depicted in Figure 1.8. In this system, one cavity is consistently composed of 4,7-phenanthroline walls, has a diameter of approximately 8.2 Å, and a depth of 3.6 Å. Dimensions of the second cavity are easily tuned by aggregating  different  metal  cis-protected  complexes  (a-c,  Scheme  1.12).  Metallocavitands 11c and 12c, constructed with 1,10-phenanthroline, have an estimated cavity diameter and depth of 11 and 6.5 Å, respectively. Solution recognition of SO42- in 11a was investigated, and an estimated Kassoc = 250 mol-1 L was calculated. In a separate report, 11a was reacted with 2-pyrimidinol resulting in the substitution of one 4,7phenanthroline ligand and conversion to a metallocavitand with a single mirror plane.135 Functionalizing the pyrimidinol unit creates electrochemical- or fluorescence-based metallocavitand  sensors  capable  of  detecting  adenine-5’-  and  uridine-5’-  136  monophosphate.  Scheme 1.12. Synthesis of double-bowl metallocavitands 11a-c and 12a-c.  31  Figure 1.8. Solid-state structure of double-bowl metallocavitand 11b represented as a wireframe model with bound ether space filling. a) Top-down looking into the shallow cavity formed by 4,7-phenanthroline walls. b) Side-on view showing ether bound in the shallow cavity. Crystallographic disorder in the deep cavity prevented guest molecules from being identified. Counter ions have been omitted for clarity (C = green, N = blue, O = red, H = white, Pd = pink). Massive metallocavitands 13 and 14, shown in Scheme 1.13, were reported from self-assembly of 2,4,6-tris(3-pyridyl)-1,3,5-triazine and cis-protected palladium(II) and platinum(II) complexes.137 These M6L4 bowl-shaped complexes have a maximum cavity diameter of approximately 18 Å at the rim and a depth of 10 Å allowing them to encapsulate multiple guest molecules. In solution o-, m-, and p-terphenyl are all bound to varying degrees with m-terphenyl exhibiting the strongest interaction and p-terphenyl the weakest.138 Host-guest interactions between 13 and m-terphenyl result in the formation of a dimeric capsule, 132, with four m-terphenyl molecules encapsulated. Metallocavitand 13 also hosts cis-stilbene and the structure, determined by a SCXRD experiment, is shown in Figure 1.9. This host-guest complex also assembles into a capsule with six cisstilbene molecules trapped inside. Guest cis-stilbene molecules exist as an organic cluster inside the capsule with close π-π and CH-π contacts.  32  Scheme 1.13. Synthesis of metallocavitands 13 and 14.  Figure 1.9. Solid-state structure of metallocavitand 13. a) Space-filling model looking directly into the cavity. b) Wireframe model of capsular assembly 132 with six cisstilbene guest molecules trapped inside, each a different color and modeled as space filling. Counter ions and protons are omitted for clarity in part b) (C = green, N = blue, H = white, Pd = pink). When two equivalents of 2-naphthyltrimethoxysilane are mixed with 13 in water, a single condensation product is observed: naphthyl-Si(OH)2OSi(OH)2-naphthyl.139 Confinement of 2-naphthyltrimethoxysilane in the cavity directs synthesis of the silanol 33  dimer and suppresses polycondensation, observed in the absence of 13. Interestingly, the results of a SCXRD experiment revealed metallocavitand 13 adopts a more open box-like geometry rather than bowl-shape to accommodate the silanol dimer. This flexibility is due to the unlocked cavity base where only two N-Pd-N links are found and not four (Figure 1.9a). Hexa-, octa-, nona-, and heptadeca-oligopeptides, containing tryptophan residues, exhibit α-helical folding due to interactions with the non-polar cavity of platinum(II) metallocavitand 14.140-142 Host-guest induced folding is a novel method for stabilizing and studying peptide secondary structure.  1.1.6 Aggregation of Organometallic Half-Sandwich Metallocavitands In the late 1980s and early 1990s a new field, bioorganometallic chemistry, began to blossom as many coordination chemists started exploring the structure and applications of coordination complexes formed by reacting amino acids, nucleobases, nucleosides, nucleotides, and nucleic acids with organometallic half-sandwich complexes.143 A halfsandwich complex is an organometallic complex with one cyclopentadienyl or arene ligand. Throughout this section, cyclopentadienyl (Cp), pentamethylcyclopentadienyl (Cp*), and arene (CnHm) ligands are abbreviated accordingly. In 1992 a report on the reaction between Cp*Rh(L-phenylalinate)Cl and AgBF4 described the formation of a cyclic trimer, [Cp*Rh(μ-L-phenylalinate)]33+, that was isolated and studied by SCXRD.144 Although this trimer is not really a metallocavitand due to the minute cavity, the strategy presented by Beck and coworkers marks the conceptual inception of half-sandwich metallocavitands. Within a few months the group of Richard Fish reported a similar trimeric cyclization product with an actual cavity; when [Cp*Rh(μ-OH)2RhCp*](OTf)2 was reacted with 9-methyladenine at pH 7.2 metallocavitand 15 was isolated as shown in Scheme 1.14.145 The solid-state structure of 15 is depicted in Figure 1.10 where the bowl-shaped cavity with 9-methyladenine walls is apparent. This structure is quite similar to [Pt(CH3)3(9-methyladenine)]3 metallocavitand 6 (Scheme 1.10 and Figure 1.6b).  34  Scheme 1.14. Synthesis of rhodium(III) metallocavitand 15.  Figure 1.10 Solid-state structure of metallocavitand 15 in a wireframe representation with protons and counter ions omitted for clarity. a) Side-on view of the assembly. b) Top-down view looking into the cavity (C = green, N = blue, Rh = yellow). Most half-sandwich metallocavitands are synthesized using the same general method outlined in Scheme 1.15. A “piano-stool” or dimeric half-sandwich complex is reacted with a tridentate bridging ligand resulting in a metallocavitand composed of n = 3, 4, or 6 repeat units with 3 being most common.146 A variety of half-sandwich complexes have been used to assemble metallocavitands such as ruthenium(II)-arene, rhodium(III)-Cp*, and iridium(III)-Cp* complexes. Adenine, 9-methyladenine, 9methylhypoxanthine,  adenosine,  2’-deoxyadenosine,  2’,3’-dideoxyadenosine,  5’-  methyladenosine monophosphate (5’-Methyl-AMP), nicotinamideadenine dinucleotide 35  (NAD+), 3-hydroxy-2-pyridone (and derivatives), 3-hydroxy-2-methyl-4-pyridone (and derivatives), and 3-acetamido-2-pyridone are all tridentate ligands used to aggregate halfsandwich complexes into tricyclic metallocavitands.146,147 Tricyclic metallocavitands assembled in this fashion share the general skeleton of metallocavitand 15 (Figure 1.10) and are isolated as either a racemic mixtures of enantiomers or a mixture of diastereomers depending on the chirality of the building units. The diversity of components that direct half-sandwich metallocavitand aggregation provides an easy way to tune the structure and function of metallocavitands isolated in this fashion, and the one-pot self-assembly process greatly simplifies access to bowl-shaped host molecules.  Scheme 1.15. By reacting a tridentate ligand with piano-stool or dimeric complexes of η6-arene ruthenium(II) or η5-pentamethylcyclopentadienyl rhodium(III)/iridium(III), cyclic half-sandwich metallocavitands are isolated. Most often cationic or neutral cyclic trimers (n = 3) are formed; however, larger cycles have been reported (n = 4 or 6). X may be solvent molecules or anions, and R indicates alkyl substituents.  36  Host-guest  interactions  were  reported  for  metallocavitands  [RhCp*(9-  methyladenine)]33+ (15), [RhCp*(adenosine)]33+ (16), [RhCp*(2’-deoxyadenosine)]33+ (17), and [RhCp*(5’-Methyl-AMP)]33+ (18) when mixed with L-tryptophan and Lphenylalanine in water at pH = 7.148 By monitoring concentration-induced changes in NMR chemical shifts, association constants in the range of 100-103 mol-1 L were measured with the strongest interactions reported for binding of L-tryptophan to metallocavitand 17. Protons on the phenyl ring of L-tryptophan exhibited the greatest change in chemical shift, and host-guest intermolecular NOE contacts with the 2’deoxyadenosine walls of 17 were observed, confirming the interaction. Metallocavitand 17 also hosts o-, m-, and p-aminobenzoic acid with the greatest association constant, 810 mol L-1, observed for o-aminobenzoic acid. Host-guest complexation is predominantly driven by favorable π-π interactions and the hydrophobic effect.149 Stability of these hostguest complexes was also investigated in the gas phase by electrospray ionization mass spectrometry (ESI-MS) to elucidate the role of the hydrophobic effect vs favorable hostguest interactions in molecular recognition.150 It was also proposed that metallocavitand 17 may act as a 1H NMR shift reagent in water for aromatic carboxylic acids and peptides containing terminal L-tryptophan or L-phenylalanine residues.151  Diastereomeric separation of metallocavitand 17 and its iridium(III) analogue, 19, was accomplished by fractional crystallization and confirmed by circular dichroism (CD).152 Over 10 days at room temperature the pure diastereomer of 17 converted to a 70:30 mixture of each. A hexameric metallocavitand, 20, was isolated upon self-assembly of [MCp*(H2O)3]2+ (M = Rh3+ or Ir3+) with 6-purinethione riboside, and the solid-state structure of the rhodium(III) metallocavitand is depicted in Figure 1.11.153 The structure consists of six metal centers in a hexagonal arrangement with 6-purinethione walls  37  making a cubic cavity capped by sugar units. Metallocavitand 20 may also be considered to be a coordination cage based on its closed-cavity conformation. No solution host-guest studies  were  reported  for  the  hexamer.  Reactions  of  6-purinethione  with  2+  [RhCp*(H2O)3] or [IrCp*Cl2]2 yielded tetrametallic metallocavitands that exhibit a 1,3alternate calix[4]arene-like conformation (Scheme 1.9) in the solid state with cavities too small for host-guest studies.154  Figure 1.11. Solid-state structure of metallocavitand 20. Counter ions, protons, and cocrystallized solvent molecules have been omitted for clarity. a) Top-down view showing the hexagonal arrangement of RhCp* units. b) Side-on view with the 6-purinethione metallocavitand walls represented space filling. The sugar substituents cap the cavity from the top and bottom (C = green, N = blue, O = red, S = purple, Rh = yellow).  38  Using synthetic pyridones and not bioligands, Severin and coworkers have selfassembled a large family of tricyclic half-sandwich metallocavitands that exhibit a wealth of host-guest chemistry, particularly toward ion pairs. An initial report involved the reaction of [IrCp*(Cl)2]2 or [Ru(η6-cymene)(Cl)2]2 (cymene = p-CH3PhCH(CH3)2) with 3-hydroxy-2-methyl-4-pyridone, shown in Scheme 1.16, to yield molecular triangles 21 and 22, respectively.155 Figure 1.12a displays the results of a SCXRD study of 22 that revealed the pyridinone walls are vertical and the central cavity is too small to act as a host. A nearly identical solid-state structure was reported for 21. Switching the regiochemistry of the catechol unit in the pyridinone linker to 3-hydroxy-2-pyridinone afforded diastereoselective synthesis of metallocavitands 23, 24, and 25 as shown in Scheme 1.16.156 The solid-state structure depicted in Figure 1.12b shows the concave geometry adopted by 24 and the central 12-crown-3 metallacrown configuration that is similar to a classic organic crown ether. To tune the cation affinity, many derivatives of 24 have been self-assembled by varying the substituents on the Ru(arene) unit. Addition of alkali metal salts to metallocavitands 23, 24, or 25 yields ion pair host-guest complexes in which the cation adopts a tetrahedral geometry, tris-chelated by the metallacrown and coordinated to an anion where the ion pair is mostly shielded from solution by the π-arene walls coordinated to Ir3+, Ru2+, or Rh3+. Exceptionally high association constants were measured for both Li+ and Na+ (> 105 mol-1 L), higher than classic crown ethers and comparable to cryptates. This binding affinity is explained by the high degree of pre-organization expressed by metallocavitand hosts 23-25, minimizing entropy loss upon ion complexation. Notice in Figure 1.12c-d the solid-state structure of 24-LiCl exhibits very little geometric reorganization from empty host 24 (Figure 1.12b). These host-guest systems are exceptionally selective for Li +. In particular, it has been demonstrated that LiCl may be extracted from water and detected at millimolar concentrations electrochemically, colorimetrically, and by fluorescence using derivatives of 24.157-161  39  Scheme 1.16. Synthesis of molecular triangles 21 and 22, metallocavitands 23, 24, and 25, and their host-guest ion pair complexes. Metallocavitand 23 binds LiF as an ion pair over Cl-, Br-, I-, or NO3- anions with a selectivity greater that 103, and the solid-state characterization of 23-LiF was the first ever structural report of a molecular LiF complex.162 All three metallocavitands 23-25 stabilize molecular LiF and LiFHF complexes despite the very high lattice energy of LiF.163 Na2SiF6, an extremely insoluble salt with a high lattice energy, is encapsulated in 242 and solubilized, allowing for dissolution in non-polar organic solvents. The hostguest complex 242-Na2SiF6 has been studied in the solid state as well.164  40  Figure 1.12. Solid-state structures depicted as wireframe models with co-crystallized solvent molecules and protons omitted for clarity (C = green, N = blue, O = red, Ru = yellow, Li = violet, Cl = brown). a) Molecular triangle 22. b) Top-down view of metallocavitand 24 looking into the cavity. c) Top-down view of 24-LiCl with a lithium ion chelated by three O-donors of the metallacrown. d) Side-on view of 24-LiCl showing the tetrahedral geometry of the lithium ion and capping chloride ligand. By switching the tridentate ligand from 3-hydroxy-2-pyridinone derivatives to polycyclic aromatics such as 2,3-dihydroxyquinoline, 2,3-dihydroxyquinoxaline, or 6methyl-2,3-phenazinediol much wider and deeper tricyclic metallocavitands may be isolated with the same core structure outlined in Scheme 1.16.165 Metallocavitands 26 and 27 were self-assembled from [Ru(arene)(Cl2)]2 and 2,3-dihydroxyquinoline or 6-methyl2,3-phenazinediol, respectively, and their solid-state structures are depicted in Figure 41  1.13. In the solid state 26 hosts the cymene ligand of another metallocavitand, and 27 hosts a chloroform guest molecule. A maximum cavity diameter of 11 and 13 Å is reported for 26 and 27, respectively. No solution host-guest studies were reported for these metallocavitands.  Figure 1.13. Solid-state structures of metallocavitands: a) 26 with cymene guest of another metallocavitand represented space filling. b) Side-on view of 27 with chloroform guest space filling. c) Top-down view of 27. For each structure the metallocavitand is modeled as a wireframe and co-crystallized solvent molecules have been omitted for clarity (C = green, N = blue, O = red, H = white, Ru = yellow, Cl = brown). Metallocavitands with Lewis-acidic sites accessible to guest molecules were reported by Fontaine and coworkers after reacting TaCp*Me4 with two equivalents of RB(OH)2 and one equivalent of H2O as shown in Scheme 1.17. Substituents on the boronic acid unit make up the walls of the metallocavitand and are easily changed by initiating self-assembly with different boronic acids. The metallocavitand core is composed of an aggregated trimetallic tantalum(V) cluster stabilized by two μ-O, one μOH, one μ3-OH, and three μ-O2BR ligands.166 The exact location of μ-OH and μ-O ligands was not determined crystallographically and low temperature  1  H NMR  42  spectroscopy showed the μ-OH proton to be labile and undergoing rapid exchange above -40 °C.  Scheme 1.17. Synthesis of tricyclic half-sandwich tantalum(V) boronate complexes 2832. Figure 1.14 depicts the solid-state structures of metallocavitands 28 and 32. The maximum cavity diameter is about 10 Å near the opening, and in the cavities of 28 and 32 are located guest molecules with a Lewis-basic site. All five complexes, 28-32, were studied in the solid state, however, only 28, 31, and 32 are metallocavitands as the R groups associated with 29 and 30 effectively fill the space for a cavity. Strong host-guest interactions are apparent for 28-THF, the shortest B-O(THF) distance is 2.99 Å, much shorter than the sum of van der Waals radii (3.5 Å) and the μ3-OH to THF O-O distance is 2.70 Å, clearly in the range of a hydrogen bond. Host-guest complexation of THF and acetone are observed for metallocavitands 28, 31, and 32 in toluene-d8 indicated by the upfield 1H chemical shift observed for the protons of bound guest molecules. Association constants for 28-acetone and 31-acetone were reported as 1.3 x 103 and 0.8 x 102 mol-1 L, respectively, and poor solubility prevented a similar determination for 32-acetone. A density functional theory (DFT) calculation predicted the strength of the host-guest interaction to be around 60 kJ mol-1 with a slightly stronger host-guest interaction for the more Lewis acidic CF3-substituted metallocavitand 32.167 By changing boronic acid R groups, Fontaine and coworkers envision facile tuning of cavity dimensions and functional groups to select for specific guest molecules with potential to develop the system as a host-guest catalyst. 43  Figure 1.14. Solid-state structures of tantalum(V) metallocavitands 28 and 32 represented as wireframe models with guest molecules space filling. a) Side-on view of 28 with a guest THF molecule simultaneously interacting with a Lewis-acidic boron center and H-bonding to the central μ3-OH ligand. b) Top-down view looking into the bowl of 28 with bound THF. c) Side-on view of metallocavitand 32 with a guest acetone molecule simultaneously interacting with a Lewis-acidic boron center and H-bonding to the central μ3-OH ligand. In each representation co-crystallized solvent molecules have been omitted for clarity (C = green, O = Red, H = white, Ta = yellow, B = pink, F = brown).  1.1.7 Aggregation of Miscellaneous Metallocavitands Curiosity breeds serendipity, and metallocavitands are often the offspring of this relationship. Interesting examples of aggregation-assembled metallocavitands that do not fit into the previous two sections are presented here. Scheme 1.18 outlines the synthesis of monocationic metallocavitand 33 that begins with the reaction of an oxo functionalized terpy (2,2’:6’,2’’-terpyridyl) ligand with Re(CO)5Br to give a monomeric species Re(4’oxo-η2-terpy)(CO)3Br.168 Halide abstraction with three equivalents of AgBF4 yields metallocavitand 33 that has an internal volume of approximately 150 Å3. Crystals of neutral metallocavitand 34 were grown when a solution of 33 was left in sunlight. The photo-demetallation process was investigated, and it was found that laser illumination at  44  405 nm also induced the loss of silver(I), a process found to be reversible after addition of AgBF4. Both 33 and 34 were studied with SCXRD and their solid-state structures are presented in Figure 1.15. No solution host-guest studies have been reported. However, both metallocavitands are luminescent, a promising feature for developing future hostguest sensing applications.  Scheme 1.18. Synthesis of  tri-rhenium(I) silver(I) metallocavitand 33 and the  photoinduced loss of silver(I) yielding metallocavitand 34. Metallation/demetallalation of silver(I) is reversible.  45  Figure 1.15. Solid-state structures of metallocavitands 33 and 34. a) Top-down view of silver(I) stoppered 33. No guest molecules were located in the cavity by crystallography. The BF4- counter ion has been omitted for clarity. b) Side-on view of 33. c) Top-down view of 34 with host-guest interactions exhibited with the pyridyl arm of another metallocavitand represented as space filling. d) Side-on view of 34 (C = green, N = blue, O = red, H = white, Re = yellow, Ag = purple).  46  Scheme 1.19. Synthesis of metallocavitands 35a-c. Slight variations in reaction conditions of [Cu(CH3CN)4]+ with 4-(2pyridyl)pyrimidine (pprd) generated three unique cationic metallocavitands, 35a-c, shown in Scheme 1.19.169 Each was characterized in the solid-state and exhibits a bowlshaped cavity that hosts either PF6- or ClO4- (Figure 1.16). By switching from an octahedral, PF6-, to a tetrahedral, ClO4-, counter ion the isolated metallocavitand is either tetrametallic or trimetallic, respectively. This example demonstrates a new route toward anion-templated metallocavitands and highlights the variety of structures accessible by subtle changes in the self-assembly conditions. Although, metallocavitands 35a-c are stable in acetone, no solution host-guest studies have been reported.  47  Figure 1.16. Solid-state structures of copper(I) metallocavitands represented as wireframe models with space-filling guest molecules. a) Side-on view of 35a with a PF6ion in the cavity. b) Top-down view of 35a. c) Side-on view of 35b where two PF6- ions are bound, one in the larger cavity made with aromatic walls and one in the smaller cavity formed by CO ligands. d) Side-on view of 35c with encapsulated ClO4- ions trapped by the small and large cavity of two separate metallocavitands. In each instance, co-crystallized solvent molecules and free counter ions have been omitted for clarity (C = green, N = blue, O = red, H = white, Cu = yellow, F = brown, Cl = cyan, P = orange). In situ aerobic oxidation of a bisoxazoline ligand in the presence of Zn(ClO4)2 and base serendipitously yielded the double-bowl shaped metallocavitand 36 shown in Scheme 1.20.170 In the absence of O2 only monomeric zinc(II) complexes were observed, and upon exposure of the monomers to O2 36 was isolated. One interesting feature of 36 is the central μ3-OH found in the base of one cavity, rendering the two cavities chemically distinct from each other. In the solid-state a 2.96 Å O-O distance separates the μ3-OH and O of ClO4-, evidence for a host-guest hydrogen bond. The zinc(II)-O(ClO4-) distances in this cavity are too long for any interaction (Figure 1.17). In the other cavity, 48  where there is no potential for H-bonding, Lewis acid-base host-guest interactions are apparent from the 2.43-2.91 Å zinc(II)-O(ClO4-) distances. The multiple H-bonding and Lewis acid-base host-guest interactions within the two cavities bind ClO4- ions tightly, and these remain bound in the gas phase as shown by ESI-MS.  Scheme 1.20. Synthesis of double-bowl shaped metallocavitand 36. Under anhydrous conditions the μ3-OH ligand may be removed upon addition of HBF4, and a solid-state structure of 36-OH also showed strong Lewis acid-base zinc(II)O(ClO4-) interactions inside both equivalent cavities. Preliminary investigations indicate metallocavitand 36 catalyzes the hydrolysis of phosphate esters. It is proposed that the zinc(II) ions direct phosphate esters into the cavity and the μ3-OH ligand nucleophile initiates hydrolysis. Metal-ligand host-guest synergy is a promising feature for the future development of metallocavitand catalysts.  49  Figure 1.17. Solid-state structure of double-bowl shaped tetrazinc(II) metallocavitand 36 represented as a wireframe model with guest ClO4- anions space filling. a) Top-down view of the cavity. b) Side-on view showing the double-bowl geometry and the μ3-OH ligand H-bonding with an O atom of ClO4- in the top cavity. Free counter ions and cocrystallized solvent molecules have been omitted for clarity (C = green, N = blue, O = red, H = white, Zn = yellow, Cl = brown). Rhodium(II) dimers supported by chiral carboxylate ligands have been recognized by the synthetic organic community as powerful enantioselective catalysts promoting metal-carbene transformations; however, the mechanism of chirality transfer from catalyst to substrate is not well understood. Catalysts of this type are easily synthesized in excess of 90% yield by reacting the desired carboxylic acid with Rh2(OAc)4 and the synthesis of two particular metallocavitands, 37 and 38, is shown in Scheme 1.21.171 Until recently it was thought that the carboxylate “arms” are most stable in a 1,3-alternate conformation (Scheme 1.9) and that this conformation was the active species during catalysis. Relying on evidence from SCXRD, NOESY NMR spectroscopy, DFT, and enantioinduction studies, recent reports refute the 1,3 alternate conformation as being most stable and provide convincing evidence for an all-up conformation as the catalytically active species in solution.172,173  50  Scheme 1.21. Synthesis of metallocavitands 37 and 38 depicted in the stable all-up conformation. S is solvent and occupies the active site for metal-substrate binding. The solid-state structures of metallocavitands 37 and 38 are shown in Figure 1.18 where solvent molecules are bound within the all-up cavity. Altering the N-phthaloyl-(S)phenylalaninate precursors may allow for tuning of the cavity dimensions and possibly alter the observed substrate selectivity. The free coordination site on rhodium(II) in the base of the cavity is the catalytic active site for carbene transformations and the aromatic walls hold the substrate in the proper orientation for enantioselective catalysis to occur. The metal mediated host-guest catalysis exhibited by metallocavitands 37 and 38 epitomizes the advantages of developing metal-containing host molecules.  51  Figure 1.18. Solid-state structures showing the all-up conformation adopted by metallocavitands 37 and 38. a) Side-on view of 37 with ethylacetate bound to rhodium(II) in the cavity. b) Top-down view of 37. c) Side-on view of 38 with THF molecules bound to both rhodium(II) centers in an exo/endo fashion. d) Top-down view of 38. Each model is wireframe with bound guest molecules space filling (C = green, N = blue, O = red, H = white, Rh = yellow, Cl = brown).  52  1.1.8 Macrocycle-Templated Metallocavitands There are many reports of multimetallic clusters stabilized in the center of polydentate macrocycles that adopt the geometry of a metallocavitand; however, the majority of these metallocavitands have been isolated to study magnetic properties and no solution studies have been reported. Most macrocycle-templated metallocavitands studied in solution and in the solid state for host-guest applications have been described by Kersting and coworkers using the flexible peralkylated amine-thiophenol macrocycle, 39, or phenylsulfonate macrocycle, 40, shown in Scheme 1.22.174,175 In their reports, complexes of these macrocycles are referred to as metallated container molecules.176,177 The polydentate nature of 39 and 40 stabilizes two transition metals in distorted octahedral geometries, and upon metal coordination the macrocycle adopts a rigid bowlor U-shaped conformation. At the base of the cavity are two Lewis-acidic and often redox-active metal centers capable of host-guest molecular recognition and catalysis. The cavity walls are made up of shallow N-Me and deep tBu-phenyl groups, both hydrophobic.  Scheme 1.22. Synthesis of bimetallic metallocavitands 41-45 from flexible macrocycles 39 or 40.  53  By switching from thiophenol macrocycle 39 to phenylsulfonate macrocycle 40, the metallocavitand cavity may be tuned as bimetallic complexes of 40 are deeper with a smaller cavity diameter at the opening due to the μ1,3-bridging mode of the sulfonato group. SCXRD experiments highlight this geometric effect, and the solid-state structures of nickel(II) metallocavitands 41 and 42 are depicted in Figure 1.19. In each structure a carboxylate containing guest molecule is bound to the nickel(II) ions in a μ1,3-fashion inside the cavity. Besides a wide variety of carboxylates, metallocavitand 41 has been isolated with L =  BH4-, RCO3-, CO32-, PO43-, NO3-, NO2-, N3-, N2H4, pyrazolate,  pyridazine, and phthalazine where the guest molecule, L, binds to the nickel(II) centers in either a μ1,2- or μ1,3-fashion.178,179 Ferro- or antiferromagnetic exchange between the nickel(II) centers is observed depending upon the nature of the coordinated guest. An NMR spectroscopic study with thiophenol zinc(II) analogue, 43, showed that carboxylate guest molecules with more hydrophobic character are bound by up to two orders of magnitude higher, likely due to intermolecular hydrophobic interactions between the aromatic cavity walls and the non-polar guest molecule.180  54  Figure 1.19. Solid-state structures of metallocavitands 41 and 42 highlighting the deeper and smaller diameter cavity of 42 due to the sulfonato bridging unit compared to thiol. a) Side-on view of 41 with bound methylcarbonate. b) Top-down view of 41. c) Side-on view of 42 with bound 3-chlorobenzoate. d) Top-down view of 41. For both structures counter ions and co-crystallized solvent molecules have been omitted for clarity. The host is represented as a wireframe model and the metal bound guest molecule is space filling (C = green, N = blue, O = red, H = white, Ni = yellow, S = purple, Cl = brown). In MeOH metallocavitand 41 with L = Cl- or OH- fixes carbon dioxide to give methylcarbonate (Figure 1.19a-b). Interestingly, this reaction is only known to take place in basic media where M-OR or M-OH attacks CO2 in a nucleophilic manner; with 41, however, the reaction occurs in neutral or acidic media.180 Size selectivity is also observed for this reaction – only primary alcohols MeOH and EtOH react, whereas iPrOH and tBuOH do not give the corresponding carbonate products, a feature attributed to the size restraints of the host-guest interaction. Similar reactivity was observed for the  55  dicobalt(II) metallocavitand 44, suggesting that the Lewis-acidic character of the metal ions and not the dn configuration dictates the reaction outcome.181 Cobalt(II) containing metallocavitand 44 also stabilizes formation of the unprecedented molybdate ester [MoO3(OMe)]- in its bowl-shaped cavity.182 An elegant metal-mediated host-guest strategy for cis-bromination of alkenes was demonstrated by Kersting and coworkers. With this method metallocavitands 44 and 45 act as molecular flasks for the regioselective cis-bromination of α,β-unsaturated carboxylates, the opposite regiochemical outcome of classic bulk-solution Br2 bromination.183 Starting with dicobalt(II) metallocavitand 44 and cinnamic acid in MeOH, a stoichiometric addition of Br2 oxidizes the host-guest complex to dicobalt(III) metallocavitand 45, effectively locking the L = μ1,3-cinnamato into the cavity, coordinated to the inert cobalt(III) centers. Following Scheme 1.23 from here, addition of more Br2 yields the bromonium ion intermediate; however, due to the steric restraints inside the cavity, a backside attack is impossible leaving only the front-side syn-addition pathway. After the bromination is complete (6 h), the reaction mixture is treated with NaBH4, reducing 45 to 44, and HCl is added to liberate the cis-addition threo-dibromo product. Kinetic studies show the reaction proceeds 100 to 1000 times slower and the ΔS‡ is 140 J mol-1 more negative than conventional alkene bromination. Both observations support a pathway with a highly ordered transition state as may be expected for the hostguest mediated syn-addition. Unique regioselectivity was also observed for host-guest Diels-Alder reactions facilitated by metallocavitands 41 and 43.184 Bridging two metallocavitands with dipyrazolyl or dicarboxylate linkers gave capsular assemblies of 412 in the solid-state.185,186 A terephthalate dimeric structure representative of other capsules is displayed in Figure 1.20. By incorporating redox active ferrocene or luminescent/redox active naphthalene diimide spacers into the dicarboxylate bridging ligand, capsular assemblies with interesting electro- and photo-active properties were isolated.187,188  56  Scheme 1.23. Regioselective cis-bromination of cinnamic acid using dicobalt metallocavitands 44 and 45. Steric constraints inside the cavity prevent a backside bromide attack on the bromonium intermediate resulting in the observed syn-addition. Reduction of cobalt(III) to cobalt(II) followed by HCl treatment releases the substrate.  57  Figure 1.20. Solid-state structure of 412 dimer bridged by terephthalate. The capsule halves are rotated by 70° from one another. a) Side-on view of dimeric capsule with bound terephthalate (C = green, N = blue, O = red, H = white, Ni = yellow, S = purple). b) Side-on view of the capsule. c) Space-filling representation showing how isolated the bridging guest molecule is from solution. In b) and c) the capsule halves are colored red and blue and terephthalate is green. Formed from multiple condensation reactions between primary amines and aldehydes or ketones, polydentate Schiff base macrocycles show untapped potential as metallocavitand templates. The reversible nature of imine condensation, shown in equation 1.1 (R and R’ may be aryl or alkyl but not hydrogen), allows for self-assembly of thermodynamic macrocyclic products in high yield despite numerous oligomeric kinetic products forming initially. What makes a Schiff base special compared to a standard imine is its stability; Schiff bases are generally stable to hydrolysis under standard atmospheric conditions.  A vast number of Schiff base macrocycles has been reported in the literature, one of the simplest being the Robson-type [2+2] macrocycle, 46, where [n+m] is a typical notation referring to the number of diamine, n, and dialdehyde, m, units participating in 58  the cyclization.189,190 Larger diameter [2+2] and [3+3] macrocycles, 47191-195 and 48,196-198 respectively, are depicted for comparison. Generally, as the diameter of the macrocycle increases, so does the denticity and the number of metals that may be coordinated in the center of the cycle. Metal complexes of larger diameter Schiff macrocycles frequently exhibit metallocavitand-like geometries in the solid state, and often the depth of the cavity increases with the macrocycle diameter. To demonstrate this concept Figure 1.21 depicts a dizinc(II) complex of 46-2H+,199 a tetranickel(II) complex of 47-4H+,193 and a heptacopper(II) complex of 48-6H+.196  59  Figure 1.21. Solid-state structures highlighting the curvature of larger diameter Schiff base macrocycle metal complexes. a) Dizinc(II) complex of 46-2H+. b) Tetranickel(II) complex of 47-4H+. c) Heptacopper(II) complex of 48-6H+. This structure is actually a dimer bridged in the center by four azide ligands that have been omitted from this representation. Counter ions, co-crystallized solvent, and some coordinated ligands have been omitted for clarity (C = green, N = blue, O = red, H = white, Zn = purple, Ni = yellow, Cu = cyan). Most polynuclear metal complexes templated by Schiff base macrocycles are synthesized for magnetic studies, and their potential for host-guest supramolecular chemistry goes unexplored.  1.2  Summary Supramolecular chemists have consistently demonstrated that systems relying on  the synergy of multiple non-covalent interactions may assemble into fantastic, often preconceived, 2D and 3D architectures. Thanks to the self-error checking intrinsic to reversible non-covalent interactions, numerous pieces may be brought together forming a single thermodynamic product in one-pot, making self-assembly the paradigm for  60  construction of molecular species expanding into the nanometer regime. A rising interest in the unique properties exhibited by nanomaterials has catalyzed the development and adoption of self-assembly as the primary technique for accessing previously unknown topologies. Directional hydrogen-bonding, π-π, electrostatic, and metal-ligand interactions have all proven to be versatile for self-assembling supramolecules. Aside from bioinspired self-assembly, which benefits from billions of years of evolution, the multiple geometries, interaction strengths, and exchange rates available with supramolecular coordination chemistry makes metal-ligand interactions the premier route for assembling highly diverse and complex molecules like those highlighted throughout this chapter. One such construct is the coordination cage in which metal-ligand interactions direct the selfassembly of supramolecules that exhibit an internal void space capable of hosting smaller guest molecules. Throughout the literature there are many interesting applications of coordination cages acting as “molecular flasks” or sensors. Despite numerous applications of coordination cages, almost always the metal ions are coordinatively saturated and act only as structure-directing units. The absence of available coordination sites in most coordination cages leaves something to be desired since inorganic chemistry is rich in catalytic and sensing applications. In pursuit of host molecules that contain metal ions capable of interacting with encapsulated guest molecules, researchers originally turned to traditional organic cavitands and outfitted them with sites capable of anchoring metal ions at the periphery or in the base of the cavity. This fruitful marriage has yielded many interesting cavitand coordination complexes and uncovered previously unknown molecular recognition properties. Despite the many applications of cavitand coordination complexes, there are two significant limitations: often their synthesis involves multiples steps and is almost always  limited  to  using  calix[n]arenes,  resorcin[n]arenes,  cyclotriveratrylenes,  cyclodextrins, or cucurbit[n]urils for the concave cavity necessary in a host-guest system. One fresh idea with potential to deliver an endless variety of host molecules with tunable cavities, free metal coordination sites accessible to encapsulated guest molecules, and isolable in high yield all in one-pot is the metallocavitand. Metallocavitands are a product of supramolecular coordination chemistry, often isolated serendipitously. This  61  chapter highlighted two principal routes to synthesize metallocavitands based on aggregation or macrocycle-templating strategies and covered some of the interesting supramolecular phenomena discovered for these novel host molecules. One avenue showing undeveloped promise is the use of polydentate Schiff base macrocycles to template metallocavitands.  1.3  Goals and Scope of this Thesis Design of new Schiff base macrocycles and study of coordination complexes  formed with macrocyclic ligands are both topics of interest to our research group. An early report that influenced the direction of research in the MacLachlan lab described the barium(II) templation of [3+3] Schiff base macrocycle 49.200 In the absence of barium, macrocycle 49 decomposes and no further work has been published on this system. Using a similar [3+3] self-assembly strategy, Gallant and MacLachlan reported the synthesis of metal-free macrocycle, 51, that has three N2O2 salphen-like pockets and a central crown ether-like cavity (Figure 1.22).201-204 In 2001, while our group was working on the synthesis of 51, an unsubstituted version of 51, macrocycle 50, was described.205 Isolation of [3+3] macrocycle 50 requires a two week reaction time, and 50 exhibits low solubility in most organic solvents. Installing alkoxy chains on the phenylenediamine unit allowed our group to explore a variety of solution-based supramolecular phenomena including alkali-metal induced self-assembly of 51 into tubular structures.201  62  Figure 1.22. Depiction of [3+3] Schiff base macrocycle 51 highlighting the a) three N2O2 salphen-like pockets and b) central crown ether-like cavity. Macrocycle 51 was originally targeted as a scaffold for assembling trimetallated macrocycles into supramolecular columns with potential for axial metal-metal interactions. Although this goal was never fully realized with transition metals, within months of my arrival at UBC, our group reported a heptazinc cluster complex coordinated by macrocycle 51.206 In the complex, each N2O2 salphen pocket binds a Zn2+ ion, and a tetrahedral Zn4 cluster, supported by acetates, is found in the central crown ether-like cavity. In the solid state the complex is bowl-shaped, and one solid-state structure shows potential for capsule formation. This discovery laid the framework for my thesis. The long-term goals of my research were to understand the fundamentals of zinc cluster formation in macrocycle 51, investigate new macrocycle-metal coordination chemistry, and explore supramolecular behavior of these complexes. Characterizing coordination complexes of 51 has been one of the biggest challenges of my research. Due to the polydentate nature of the macrocycle, nearly all first-row transition metals form multimetallic complexes when reacted with 51. Crystallography is essential to interpret results, and complexes of 51 are not easily crystallized. The unit cell of each 51-metal complex presented in this thesis is about 50% free, disordered solvent making 63  crystallography of these complexes very challenging. For this reason my research has focused on diamagnetic zinc(II) and cadmium(II) complexes of 51 where NMR spectroscopy has proven very useful for both characterization and investigation of various dynamic processes. Overall the objectives of my research have been met and I have opened new routes to previously unknown Schiff base macrocycles and metallocycles. The next chapter covers the mechanism of formation and structure of heptazinc metallocavitands templated by a family of [3+3] macrocycles, 51. Chapter three presents a thermodynamic study of heptazinc metallocavitand dimerization in solution and discusses a capsule structure isolated in the solid state. In Chapter four, the versatility of macrocycles 51 to coordinate other metals is demonstrated. Heptacadmium metallocavitands are templated by 51 with 50% larger cavities than their heptazinc counterparts due to the extended puckering of macrocycle 51 necessary to accommodate the larger radius of cadmium(II). Chapter four also discusses in detail heptacadmium metallocavitand dimerization, hostguest chemistry, and ligand exchange dynamics probed with a variety of techniques. Interestingly, both zinc and cadmium metallocavitands exhibit entropy-driven dimerization in solution that is discussed in detail. Chapter five highlights a new method for synthesizing [2+2] Schiff base macrocycles with four unsymmetrical N2O2 salphen pockets as well as [3+3] macrocycles with triptycene substituents used to study electrostatic aggregation of Schiff base macrocycles into fibers. Discussion of the synthesis, lyotropic liquid crystallinity, and solution and solid-state columnar aggregation observed for a novel Pt4 metallocycle is presented in Chapter six. Finally, Chapter seven draws conclusions based on my studies and proposes new directions for future investigations.  64  1.4  References  (1)  Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 7017-7036.  (2)  Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 2495-2596.  (3)  Dietrich, B.; Lehn, J. M.; Sauvage, J. P. Tetrahedron Lett. 1969, 2889-2892.  (4)  Cram, D. J.; Kaneda, T.; Helgeson, R. C.; Lein, G. M. J. Am. Chem. Soc. 1979, 101, 6752-6754.  (5)  Moran, J. R.; Karbach, S.; Cram, D. J. J. Am. Chem. Soc. 1982, 104, 5826-5828.  (6)  Cram, D. J.; Karbach, S.; Kim, Y. H.; Baczynskyj, L.; Kallemeyn, G. W. J. Am. Chem. Soc. 1985, 107, 2575-2576.  (7)  Poltev, V. I.; Grokhlina, T. I.; Gonzalez, E.; Deriabina, A.; Cruz, A.; Gorb, L.; Leszczynski, J.; Djimant, L. N.; Veselkov, A. N. Theochem 2004, 709, 123-128.  (8)  Cesaro, A.; Russo, E.; Crescenzi, V. J. Phys. Chem. 1976, 80, 335-339.  (9)  Davis, J. T. Angew. Chem., Int. Ed. 2004, 43, 668-698.  (10)  Aldaye, F. A.; Sleiman, H. F. Pure Appl. Chem. 2009, 81, 2157-2181.  (11)  Aldaye, F. A.; Palmer, A. L.; Sleiman, H. F. Science 2008, 321, 1795-1799.  (12)  Kobayashi, T.; Okada, A.; Fujii, Y.; Niimi, K.; Hamamoto, S.; Yasui, T.; Tozawa, K.; Kohri, K. Urol. Res. 2010, 38, 117-125.  (13)  Kinge, S.; Peter, M.; Crego-Calama, M.; Reinhoudt, D. N. Supramol. Chem. 2008, 20, 593-600.  (14)  Vazquez-Campos, S.; Peter, M.; Dong, M.; Xu, S.; Xu, W.; Gersen, H.; Linderoth, T. R.; Schoenherr, H.; Besenbacher, F.; Crego-Calama, M.; Reinhoudt, D. N. Langmuir 2007, 23, 10294-10298.  (15)  Piermattei, A.; Giesbers, M.; Marcelis, A. T. M.; Mendes, E.; Picken, S. J.; Crego-Calama, M.; Reinhoudt, D. N. Angew. Chem., Int. Ed. 2006, 45, 75437546.  65  (16)  Yagai, S.; Kinoshita, T.; Higashi, M.; Kishikawa, K.; Nakanishi, T.; Karatsu, T.; Kitamura, A. J. Am. Chem. Soc. 2007, 129, 13277-13287.  (17)  Seto, C. T.; Whitesides, G. M. J. Am. Chem. Soc. 1990, 112, 6409-6411.  (18)  Zerkowski, J. A.; Seto, C. T.; Whitesides, G. M. J. Am. Chem. Soc. 1992, 114, 5473-5475.  (19)  Zerkowski, J. A.; Seto, C. T.; Wierda, D. A.; Whitesides, G. M. J. Am. Chem. Soc. 1990, 112, 9025-9026.  (20)  Reinhoudt, D. N.; Crego-Calama, M. Science 2002, 295, 2403-2407.  (21)  Wasserman, E. J. Am. Chem. Soc. 1960, 82, 4433-4434.  (22)  Dietrich-Buchecker, C. O.; Sauvage, J. P.; Kintzinger, J. P. Tetrahedron Lett. 1983, 24, 5095-5098.  (23)  Stoddart, J. F. Chem. Soc. Rev. 2009, 38, 1521-1529.  (24)  Amabilino, D. B.; Stoddart, J. F. Chem Rev 1995, 95, 2725-2829.  (25)  Hori, A.; Akasaka, A.; Biradha, K.; Sakamoto, S.; Yamaguchi, K.; Fujita, M. Angew. Chem., Int. Ed. 2002, 41, 3269-3272.  (26)  Fujita, M.; Ibukuro, F.; Yamaguchi, K.; Ogura, K. J. Am. Chem. Soc. 1995, 117, 4175-4176.  (27)  Yamashita, K.-i.; Kawano, M.; Fujita, M. J. Am. Chem. Soc. 2007, 129, 18501851.  (28)  Dietrich-Buchecker, C. O.; Guilhem, J.; Pascard, C.; Sauvage, J. P. Angew. Chem., Int. Ed. 1990, 29, 1154-1156.  (29)  Dietrich-Buchecker, C. O.; Sauvage, J. P. Angew. Chem., Int. Ed. 1989, 28, 189192.  (30)  Perret-Aebi, L.-E.; von Zelewsky, A.; Dietrich-Buchecker, C.; Sauvage, J.-P. Angew. Chem., Int. Ed. 2004, 43, 4482-4485.  (31)  Mao, C.; Sun, W.; Seeman, N. C. Nature 1997, 386, 137-138.  66  (32)  Chichak, K. S.; Cantrill, S. J.; Pease, A. R.; Chiu, S.-H.; Cave, G. W. V.; Atwood, J. L.; Stoddart, J. F. Science 2004, 304, 1308-1312.  (33)  Fujita, M.; Yazaki, J.; Ogura, K. J. Am. Chem. Soc. 1990, 112, 5645-5647.  (34)  Li, S.-S.; Northrop, B. H.; Yuan, Q.-H.; Wan, L.-J.; Stang, P. J. Acc. Chem. Res. 2009, 42, 249-259.  (35)  Stang, P. J.; Olenyuk, B. Acc. Chem. Res. 1997, 30, 502-518.  (36)  Fujita, M.; Tominaga, M.; Hori, A.; Therrien, B. Acc. Chem. Res. 2005, 38, 369378.  (37)  Seidel, S. R.; Stang, P. J. Acc. Chem. Res. 2002, 35, 972-983.  (38)  Fujita, M.; Oguro, D.; Miyazawa, M.; Oka, H.; Yamaguchi, K.; Ogura, K. Nature 1995, 378, 469-471.  (39)  Tashiro, S.; Tominaga, M.; Kawano, M.; Therrien, B.; Ozeki, T.; Fujita, M. J. Am. Chem. Soc. 2005, 127, 4546-4547.  (40)  Murase, T.; Horiuchi, S.; Fujita, M. J. Am. Chem. Soc. 2010, 132, 2866-2867.  (41)  Yoshizawa, M.; Tamura, M.; Fujita, M. Science 2006, 312, 251-254.  (42)  Furutani, Y.; Kandori, H.; Kawano, M.; Nakabayashi, K.; Yoshizawa, M.; Fujita, M. J. Am. Chem. Soc. 2009, 131, 4764-4768.  (43)  Yamaguchi, T.; Fujita, M. Angew. Chem., Int. Ed. 2008, 47, 2067-2069.  (44)  Furusawa, T.; Kawano, M.; Fujita, M. Angew. Chem., Int. Ed. 2007, 46, 57175719.  (45)  Beissel, T.; Powers, R. E.; Raymond, K. N. Angew. Chem., Int. Ed. Engl. 1996, 35, 1084-1086.  (46)  Bruckner, C.; Powers, R. E.; Raymond, K. N. Angew. Chem., Int. Ed. 1998, 37, 1837-1839.  (47)  Caulder, D. L.; Powers, R. E.; Parac, T. N.; Raymond, K. N. Angew. Chem., Int. Ed. 1998, 37, 1840-1843.  67  (48)  Caulder, D. L.; Brueckner, C.; Powers, R. E.; Koenig, S.; Parac, T. N.; Leary, J. A.; Raymond, K. N. J. Am. Chem. Soc. 2001, 123, 8923-8938.  (49)  Caulder, D. L.; Raymond, K. N. Acc. Chem. Res. 1999, 32, 975-982.  (50)  Caulder, D. L.; Raymond, K. N. Angew. Chem., Int. Ed. Engl. 1997, 36, 14401442.  (51)  Parac, T. N.; Caulder, D. L.; Raymond, K. N. J. Am. Chem. Soc. 1998, 120, 80038004.  (52)  Parac, T. N.; Scherer, M.; Raymond, K. N. Angew. Chem., Int. Ed. 2000, 39, 1239-1242.  (53)  Davis, A. V.; Raymond, K. N. J. Am. Chem. Soc. 2005, 127, 7912-7919.  (54)  Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Acc. Chem. Res. 2009, 42, 16501659.  (55)  Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Angew. Chem., Int. Ed. 2007, 46, 8587-8589.  (56)  Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Science 2007, 316, 85-88.  (57)  Pluth, M. D.; Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc. 2008, 130, 11423-11429.  (58)  Fiedler, D.; Bergman, R. G.; Raymond, K. N. Angew. Chem., Int. Ed. 2006, 45, 745-748.  (59)  Leung, D. H.; Fiedler, D.; Bergman, R. G.; Raymond, K. N. Angew. Chem., Int. Ed. 2004, 43, 963-966.  (60)  Fiedler, D.; Leung, D. H.; Bergman, R. G.; Raymond, K. N. Acc. Chem. Res. 2005, 38, 349-358.  (61)  Fiedler, D.; Bergman, R. G.; Raymond, K. N. Angew. Chem., Int. Ed. 2004, 43, 6748-6751.  (62)  Brown, C. J.; Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc. 2009, 131, 17530-17531.  68  (63)  Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew. Chem., Int. Ed. 2009, 48, 3418-3438.  (64)  Albrecht, M. Chem. Rev. 2001, 101, 3457-3497.  (65)  Piguet, C.; Bernardinelli, G.; Hopfgartner, G. Chem. Rev. 1997, 97, 2005-2062.  (66)  Tranchemontagne, D. J.; Ni, Z.; O'Keeffe, M.; Yaghi, O. M. Angew. Chem., Int. Ed. 2008, 47, 5136-5147.  (67)  Drain, C. M.; Varotto, A.; Radivojevic, I. Chem. Rev. 2009, 109, 1630-1658.  (68)  Muller, A.; Serain, C. Acc. Chem. Res. 2000, 33, 2-10.  (69)  Muller, A.; Roy, S. Coord. Chem. Rev. 2003, 245, 153-166.  (70)  Mezei, G.; Zaleski, C. M.; Pecoraro, V. L. Chem. Rev. 2007, 107, 4933-5003.  (71)  Saalfrank, R. W.; Maid, H.; Scheurer, A. Angew. Chem., Int. Ed. 2008, 47, 87948824.  (72)  Hooley, R. J.; Shenoy, S. R.; Rebek, J., Jr. Org. Lett. 2008, 10, 5397-5400.  (73)  Hooley Richard, J.; Rebek, J., Jr. Chem. Biol. 2009, 16, 255-264.  (74)  Zelder, F. H.; Rebek, J., Jr. Chem. Commun. 2006, 753-754.  (75)  Pinacho Crisostomo, F. R.; Lledo, A.; Shenoy, S. R.; Iwasawa, T.; Rebek, J., Jr. J. Am. Chem. Soc. 2009, 131, 7402-7410.  (76)  Kamioka, S.; Ajami, D.; Rebek, J., Jr. Chem. Commun. 2009, 7324-7326.  (77)  Shenoy Siddhartha, R.; Pinacho Crisostomo Fernando, R.; Iwasawa, T.; Rebek, J., Jr. J. Am. Chem. Soc. 2008, 130, 5658-9.  (78)  Hooley, R. J.; Restorp, P.; Iwasawa, T.; Rebek, J., Jr. J. Am. Chem. Soc. 2007, 129, 15639-15643.  (79)  Hooley, R. J.; Iwasawa, T.; Rebek, J., Jr. J. Am. Chem. Soc. 2007, 129, 1533015339.  69  (80)  Kaanumalle, L. S.; Gibb, C. L. D.; Gibb, B. C.; Ramamurthy, V. J. Am. Chem. Soc. 2004, 126, 14366-14367.  (81)  Kaanumalle, L. S.; Gibb, C. L. D.; Gibb, B. C.; Ramamurthy, V. J. Am. Chem. Soc. 2005, 127, 3674-3675.  (82)  Gibb, C. L. D.; Sundaresan, A. K.; Ramamurthy, V.; Gibb, B. C. J. Am. Chem. Soc. 2008, 130, 4069-4080.  (83)  Natarajan, A.; Kaanumalle, L. S.; Jockusch, S.; Gibb, C. L. D.; Gibb, B. C.; Turro, N. J.; Ramamurthy, V. J. Am. Chem. Soc. 2007, 129, 4132-4133.  (84)  Kaanumalle, L. S.; Gibb, C. L. D.; Gibb, B. C.; Ramamurthy, V. Org. Biomol. Chem. 2007, 5, 236-238.  (85)  Gibb, C. L. D.; Gibb, B. C. J. Am. Chem. Soc. 2006, 128, 16498-16499.  (86)  Chaffee, K. E.; Fogarty, H. A.; Brotin, T.; Goodson, B. M.; Dutasta, J.-P. J. Phys. Chem. A 2009, 113, 13675-13684.  (87)  Atwood, J. L.; Barbour, L. J.; Raston, C. L.; Sudria, I. B. N. Angew. Chem., Int. Ed. 1998, 37, 981-983.  (88)  Park, S. J.; Kwon, O.-H.; Lee, K.-S.; Yamaguchi, K.; Jang, D.-J.; Hong, J.-I. Chem.--Eur. J. 2008, 14, 5353-5359.  (89)  Haino, T.; Yanase, M.; Fukunaga, C.; Fukazawa, Y. Tetrahedron 2006, 62, 20252035.  (90)  Balch, A. L.; Olmstead, M. M. Coord. Chem. Rev. 1999, 185-186, 601-617.  (91)  Iglesias-Sanchez, J. C.; Fragoso, A.; de Mendoza, J.; Prados, P. Org. Lett. 2006, 8, 2571-2574.  (92)  Huerta, E.; Metselaar, G. A.; Fragoso, A.; Santos, E.; Bo, C.; de Mendoza, J. Angew. Chem., Int. Ed. 2007, 46, 202-205.  (93)  Huerta, E.; Cequier, E.; de Mendoza, J. Chem. Commun. 2007, 5016-5018.  (94)  Makha, M.; Hardie, M. J.; Raston, C. L. Chem. Commun. 2002, 1446-1447.  70  (95)  Azov, V. A.; Beeby, A.; Cacciarini, M.; Cheetham, A. G.; Diederich, F.; Frei, M.; Gimzewski, J. K.; Gramlich, V.; Hecht, B.; Jaun, B.; Latychevskaia, T.; Lieb, A.; Lill, Y.; Marotti, F.; Schlegel, A.; Schlittler, R. R.; Skinner, P. J.; Seiler, P.; Yamakoshi, Y. Adv. Funct. Mater. 2006, 16, 147-156.  (96)  Kim, S. K.; Moon, B.-S.; Park, J. H.; Seo, Y. I.; Koh, H. S.; Yoon, Y. J.; Lee, K. D.; Yoon, J. Tetrahedron Lett. 2005, 46, 6617-6620.  (97)  Ballester, P.; Sarmentero, M. A. Org. Lett. 2006, 8, 3477-3480.  (98)  Melegari, M.; Suman, M.; Pirondini, L.; Moiani, D.; Massera, C.; Ugozzoli, F.; Kalenius, E.; Vainiotalo, P.; Mulatier, J.-C.; Dutasta, J.-P.; Dalcanale, E. Chem.-Eur. J. 2008, 14, 5772-5779.  (99)  Taratula, O.; Dmochowski, I. J. Curr. Opin. Chem. Biol. 2010, 14, 97-104.  (100) Chambers, J. M.; Hill, P. A.; Aaron, J. A.; Han, Z.; Christianson, D. W.; Kuzma, N. N.; Dmochowski, I. J. J. Am. Chem. Soc. 2009, 131, 563-569. (101) Schlundt, A.; Kilian, W.; Beyermann, M.; Sticht, J.; Guenther, S.; Hoepner, S.; Falk, K.; Roetzschke, O.; Mitschang, L.; Freund, C. Angew. Chem., Int. Ed. 2009, 48, 4142-4145. (102) Ajami, D.; Rebek, J., Jr. J. Am. Chem. Soc. 2006, 128, 15038-15039. (103) Scarso, A.; Onagi, H.; Rebek, J., Jr. J. Am. Chem. Soc. 2004, 126, 12728-12729. (104) McKinlay, R. M.; Cave, G. W. V.; Atwood, J. L. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 5944-5948. (105) McKinlay, R. M.; Thallapally, P. K.; Cave, G. W. V.; Atwood, J. L. Angew. Chem., Int. Ed. 2005, 44, 5733-5736. (106) Power, N. P.; Dalgarno, S. J.; Atwood, J. L. New J. Chem. 2007, 31, 17-20. (107) Power, N.; Dalgarno, S. J.; Atwood, J. L. Angew. Chem., Int. Ed. 2007, 46, 86018604. (108) Ronson, T. K.; Fisher, J.; Harding, L. P.; Hardie, M. J. Angew. Chem., Int. Ed. 2007, 46, 9086-9088.  71  (109) Botana, E.; Da Silva, E.; Benet-Buchholz, J.; Ballester, P.; de Mendoza, J. Angew. Chem., Int. Ed. 2007, 46, 198-201. (110) Botana, E.; Naettinen, K.; Prados, P.; Rissanen, K.; De Mendoza, J. Org. Lett. 2004, 6, 1091-1094. (111) Richeter, S.; Rebek, J., Jr. J. Am. Chem. Soc. 2004, 126, 16280-16281. (112) Zelder, F. H.; Salvio, R.; Rebek, J., Jr. Chem. Commun. 2006, 1280-1282. (113) Jeunesse, C.; Armspach, D.; Matt, D. Chem. Commun. 2005, 5603-5614. (114) Engeldinger, E.; Armspach, D.; Matt, D. Chem. Rev. 2003, 103, 4147-4173. (115) Armspach, D.; Matt, D.; Peruch, F.; Lutz, P. Eur. J. Inorg. Chem. 2003, 805-809. (116) Yang, J.; Gabriele, B.; Belvedere, S.; Huang, Y.; Breslow, R. J. Org. Chem. 2002, 67, 5057-5067. (117) Lenthall, J. T.; Steed, J. W. Coord. Chem. Rev. 2007, 251, 1747-1760. (118) Frischmann, P. D.; MacLachlan, M. J. Comments Inorg. Chem. 2008, 29, 26-45. (119) Lippert, B. Inorg. Chem. 1981, 20, 4326-4343. (120) Rauter, H.; Hillgeris, E. C.; Erxleben, A.; Lippert, B. J. Am. Chem. Soc. 1994, 116, 616-624. (121) Rauter, H.; Hillgeris, E. C.; Lippert, B. J. Chem. Soc., Chem. Commun. 1992, 1385-1386. (122) Navarro, J. A. R.; Janik, M. B. L.; Freisinger, E.; Lippert, B. Inorg. Chem. 1999, 38, 426-432. (123) Galindo, M. A.; Olea, D.; Romero, M. A.; Gomez, J.; del Castillo, P.; Hannon, M. J.; Rodger, A.; Zamora, F.; Navarro, J. A. R. Chem.--Eur. J. 2007, 13, 5075-5081. (124) Galindo, M. A.; Romero, M. A.; Navarro, J. A. R. Inorg. Chim. Acta 2009, 362, 1027-1030. (125) Barea, E.; Navarro, J. A. R.; Salas, J. M.; Quiros, M.; Willermann, M.; Lippert, B. Chem.--Eur. J. 2003, 9, 4414-4421. 72  (126) Navarro, J. A. R.; Salas, J. M. Chem. Commun. 2000, 235-236. (127) Ruettimann, S.; Bernardinelli, G.; Williams, A. F. Angew Chem Int Ed Engl 1993, 32, 392-394. (128) Carina, R. F.; Williams, A. F.; Bernardinelli, G. Inorg. Chem. 2001, 40, 18261832. (129) Zhu, X.; Rusanov, E.; Kluge, R.; Schmidt, H.; Steinborn, D. Inorg. Chem. 2002, 41, 2667-2671. (130) Qin, Z.; Jennings, M. C.; Puddephatt, R. J. Inorg. Chem. 2002, 41, 3967-3974. (131) Schnebeck, R.-D.; Randaccio, L.; Zangrando, E.; Lippert, B. Angew. Chem., Int. Ed. 1998, 37, 119-121. (132) Schnebeck, R.-D.; Freisinger, E.; Glahe, F.; Lippert, B. J. Am. Chem. Soc. 2000, 122, 1381-1390. (133) Schnebeck, R.-D.; Freisinger, E.; Lippert, B. Angew. Chem., Int. Ed. 1999, 38, 168-171. (134) Yu, S.-Y.; Huang, H.; Liu, H.-B.; Chen, Z.-N.; Zhang, R.; Fujita, M. Angew. Chem., Int. Ed. 2003, 42, 686-690. (135) Galindo, M. A.; Navarro, J. A. R.; Romero, M. A.; Quiros, M. Dalton Trans. 2004, 1563-1566. (136) Galindo, M. A.; Houlton, A.; Clegg, W.; Harrington, R. W.; Dobado, J.; SantoyoGonzalez, F.; Linares, F.; Romero, M. A.; Navarro, J. A. R. Chem. Commun. 2008, 3735-3737. (137) Fujita, M.; Su, S.-Y.; Kusukawa, T.; Funaki, H.; Ogura, K.; Yamaguchi, K. Angew. Chem., Int. Ed. 1998, 37, 2082-2085. (138) Yu, S.-Y.; Kusukawa, T.; Biradha, K.; Fujita, M. J. Am. Chem. Soc. 2000, 122, 2665-2666. (139) Yoshizawa, M.; Kusukawa, T.; Fujita, M.; Sakamoto, S.; Yamaguchi, K. J. Am. Chem. Soc. 2001, 123, 10454-10459.  73  (140) Tashiro, S.; Tominaga, M.; Yamaguchi, Y.; Kato, K.; Fujita, M. Angew. Chem., Int. Ed. 2006, 45, 241-244. (141) Tashiro, S.; Tominaga, M.; Yamaguchi, Y.; Kato, K.; Fujita, M. Chem.--Eur. J. 2006, 12, 3211-3217. (142) Dolain, C.; Hatakeyama, Y.; Sawada, T.; Tashiro, S.; Fujita, M. J. Am. Chem. Soc. 2010, 132, 5564-5565. (143) Fish, R. H. Coord. Chem. Rev. 1999, 185-186, 569-584. (144) Kraemer, R.; Polborn, K.; Robl, C.; Beck, W. Inorg. Chim. Acta 1992, 198-200, 415-420. (145) Smith, D. P.; Baralt, E.; Morales, B.; Olmstead, M. M.; Maestre, M. F.; Fish, R. H. J. Am. Chem. Soc. 1992, 114, 10647-10649. (146) Severin, K. Chem. Commun. 2006, 3859-3867. (147) Fish, R. H.; Jaouen, G. Organometallics 2003, 22, 2166-2177. (148) Chen, H.; Maestre, M. F.; Fish, R. H. J. Am. Chem. Soc. 1995, 117, 3631-3632. (149) Chen, H.; Ogo, S.; Fish, R. H. J. Am. Chem. Soc. 1996, 118, 4993-5001. (150) Bakhtiar, R.; Chen, H.; Ogo, S.; Fish, R. H. Chem. Commun. 1997, 2135-2136. (151) Ogo, S.; Nakamura, S.; Chen, H.; Isobe, K.; Watanabe, Y.; Fish, R. H. J. Org. Chem. 1998, 63, 7151-7156. (152) Yamanari, K.; Ito, R.; Yamamoto, S.; Fuyuhiro, A. Chem. Commun. 2001, 14141415. (153) Yamanari, K.; Yamamoto, S.; Ito, R.; Kushi, Y.; Fuyuhiro, A.; Kubota, N.; Fukuo, T.; Arakawa, R. Angew. Chem., Int. Ed. 2001, 40, 2268-2271. (154) Yamanari, K.; Ito, R.; Yamamoto, S.; Konno, T.; Fuyuhiro, A.; Fujioka, K.; Arakawa, R. Inorg. Chem. 2002, 41, 6824-6830. (155) Habereder, T.; Warchhold, M.; Noth, H.; Severin, K. Angew. Chem., Int. Ed. 1999, 38, 3225-3228.  74  (156) Piotrowski, H.; Polborn, K.; Hilt, G.; Severin, K. J. Am. Chem. Soc. 2001, 123, 2699-2700. (157) Piotrowski, H.; Severin, K. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 4997-5000. (158) Piotrowski, H.; Hilt, G.; Schulz, A.; Mayer, P.; Polborn, K.; Severin, K. Chem.-Eur. J. 2001, 7, 3196-3208. (159) Grote, Z.; Scopelliti, R.; Severin, K. J. Am. Chem. Soc. 2004, 126, 16959-16972. (160) Rochat, S.; Grote, Z.; Severin, K. Org. Biomol. Chem. 2009, 7, 1147-1153. (161) Grote, Z.; Lehaire, M.-L.; Scopelliti, R.; Severin, K. J. Am. Chem. Soc. 2003, 125, 13638-13639. (162) Lehaire, M.-L.; Scopelliti, R.; Piotrowski, H.; Severin, K. Angew. Chem., Int. Ed. 2002, 41, 1419-1422. (163) Lehaire, M.-L.; Scopelliti, R.; Severin, K. Inorg. Chem. 2002, 41, 5466-5474. (164) Lehaire, M.-L.; Scopelliti, R.; Severin, K. Chem. Commun. 2002, 2766-2767. (165) Lehaire, M.-L.; Scopelliti, R.; Herdeis, L.; Polborn, K.; Mayer, P.; Severin, K. Inorg. Chem. 2004, 43, 1609-1617. (166) Sigouin, O.; Garon, C. N.; Delaunis, G.; Yin, X.; Woo, T. K.; Decken, A.; Fontaine, F.-G. Angew. Chem., Int. Ed. 2007, 46, 4979-4982. (167) Garon, C. N.; Gorelsky, S. I.; Sigouin, O.; Woo, T. K.; Fontaine, F.-G. Inorg. Chem. 2009, 48, 1699-1710. (168) Coogan, M. P.; Fernandez-Moreira, V.; Kariuki, B. M.; Pope, S. J. A.; ThorpGreenwood, F. L. Angew. Chem., Int. Ed. 2009, 48, 4965-4968. (169) Maekawa, M.; Konaka, H.; Minematsu, T.; Kuroda-Sowa, T.; Munakata, M.; Kitagawa, S. Chem. Commun. 2007, 5179-5181. (170) Jacques, B.; Dro, C.; Bellemin-Laponnaz, S.; Wadepohl, H.; Gade, L. H. Angew. Chem., Int. Ed. 2008, 47, 4546-4550. (171) Tsutsui, H.; Abe, T.; Nakamura, S.; Anada, M.; Hashimoto, S. Chem. Pharm. Bull. 2005, 53, 1366-1368. 75  (172) Lindsay, V. N. G.; Lin, W.; Charette, A. B. J. Am. Chem. Soc. 2009, 131, 1638316385. (173) DeAngelis, A.; Dmitrenko, O.; Yap, G. P. A.; Fox, J. M. J. Am. Chem. Soc. 2009, 131, 7230-7231. (174) Kersting, B.; Steinfeld, G. Chem. Commun. 2001, 1376-1377. (175) Hausmann, J.; Kaess, S.; Klod, S.; Kleinpeter, E.; Kersting, B. Eur. J. Inorg. Chem. 2004, 4402-4411. (176) Kersting, B.; Lehmann, U. Adv. Inorg. Chem. 2009, 61, 407-470. (177) Kersting, B. Z. Anorg. Allg. Chem. 2004, 630, 765-780. (178) Journaux, Y.; Lozan, V.; Klingele, J.; Kersting, B. Chem. Commun. 2006, 83-84. (179) Hausmann, J.; Klingele, M. H.; Lozan, V.; Steinfeld, G.; Siebert, D.; Journaux, Y.; Girerd, J. J.; Kersting, B. Chem.--Eur. J. 2004, 10, 1716-1728. (180) Kersting, B. Angew. Chem., Int. Ed. 2001, 40, 3987-3990. (181) Kersting, B.; Steinfeld, G. Inorg. Chem. 2002, 41, 1140-1150. (182) Lozan, V.; Kersting, B. Inorg. Chem. 2006, 45, 5630-5634. (183) Steinfeld, G.; Lozan, V.; Kersting, B. Angew. Chem., Int. Ed. 2003, 42, 22612263. (184) Kaess, S.; Gregor, T.; Kersting, B. Angew. Chem., Int. Ed. 2006, 45, 101-104. (185) Klingele, J.; Klingele, M. H.; Baars, O.; Lozan, V.; Buchholz, A.; Leibeling, G.; Plass, W.; Meyer, F.; Kersting, B. Eur. J. Inorg. Chem. 2007, 5277-5285. (186) Lozan, V.; Solntsev, P. Y.; Leibeling, G.; Domasevitch, K. V.; Kersting, B. Eur. J. Inorg. Chem. 2007, 3217-3226. (187) Lozan, V.; Buchholz, A.; Plass, W.; Kersting, B. Chem.--Eur. J. 2007, 13, 73057316. (188) Lee, K. A.; Lozan, V.; Langford, S. J.; Kersting, B. Dalton Trans. 2009, 74817485. 76  (189) Borisova, N. E.; Reshetova, M. D.; Ustynyuk, Y. A. Chem. Rev. 2007, 107, 4679. (190) Pilkington, N. H.; Robson, R. Aust. J. Chem. 1970, 23, 2225-2236. (191) Edwards, A. J.; Hoskins, B. F.; Kachab, E. H.; Markiewicz, A.; Murray, K. S.; Robson, R. Inorg. Chem. 1992, 31, 3585-3591. (192) Edwards, A. J.; Hoskins, B. F.; Robson, R.; Wilson, J. C.; Moubaraki, B.; Murray, K. S. J. Chem. Soc., Dalton Trans. 1994, 1837-1842. (193) Bell, M.; Edwards, A. J.; Hoskins, B. F.; Kachab, E. H.; Robson, R. J. Chem. Soc., Chem. Commun. 1987, 1852-1854. (194) Bell, M.; Edwards, A. J.; Hoskins, B. F.; Kachab, E. H.; Robson, R. J. Am. Chem. Soc. 1989, 111, 3603-3610. (195) Grannas, M. J.; Hoskins, B. F.; Robson, R. J. Chem. Soc., Chem. Commun. 1990, 1644-1646. (196) Tandon, S. S.; Bunge, S. D.; Thompson, L. K. Chem. Commun. 2007, 798-800. (197) Tandon, S. S.; Thompson, L. K.; Bridson, J. N.; Benelli, C. Inorg. Chem. 1995, 34, 5507-5515. (198) Tandon, S. S.; Thompson, L. K.; Bridson, J. N. J. Chem. Soc., Chem. Commun. 1992, 911-913. (199) Dutta, B.; Bag, P.; Floerke, U.; Nag, K. Inorg. Chem. 2005, 44, 147-157. (200) Huck, W. T. S.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Recl. Trav. Chim. Pays-Bas 1995, 114, 273-276. (201) Gallant, A. J.; MacLachlan, M. J. Angew. Chem., Int. Ed. 2003, 42, 5307-5310. (202) Gallant, A. J.; Hui, J. K. H.; Zahariev, F. E.; Wang, Y. A.; MacLachlan, M. J. J. Org. Chem. 2005, 70, 7936-7946. (203) Gallant, A. J.; Patrick, B. O.; MacLachlan, M. J. J. Org. Chem. 2004, 69, 87398744.  77  (204) Gallant, A. J.; Yun, M.; Sauer, M.; Yeung, C. S.; MacLachlan, M. J. Org. Lett. 2005, 7, 4827-4830. (205) Akine, S.; Taniguchi, T.; Nabeshima, T. Tetrahedron Lett. 2001, 42, 8861-8864. (206) Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2006, 45, 52485250.  78  Chapter 2 Heptazinc Clusters Templated by [3+3] Schiff Base Macrocycles† 2.1  Introduction  2.1.1 Abstract The primary goal of this research was to better understand the mechanism of Zn7 cluster formation in conjugated Schiff base macrocycles 51a-d. A new [3+3] Schiff base macrocycle, 51d, incorporating three N2O2 salphen-type binding sites and peripheral neopentyloxy substituents was prepared. The incorporation of Zn2+ ions into 51a-d has been studied by NMR spectroscopy, mass spectrometry, and X-ray diffraction. When reacted with seven equivalents of zinc acetate, the macrocycles template the formation of heptanuclear bowl-shaped complexes, metallocavitands. Two tetranuclear Zn2+ complexes that are plausible intermediates in the assembly of the heptanuclear metallocavitands have been isolated and structurally characterized. These reactive intermediates are promising substrates for the synthesis of polynuclear, mixed metal clusters. I also demonstrate that this chemistry may be generalized to other bridging carboxylate ligands, such as methacrylate.  †A version of this chapter has been published: Reproduced in part with permission from Frischmann, P. D.; Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. “Zinc Carboxylate Cluster Formation in Conjugated Metallomacrocycles” Inorg. Chem. 2008, 47, 101-112. Copyright 2008 American Chemical Society. 79  2.1.2 Background Unusual optical and magnetic properties frequently arise when metal ions are confined to well-defined molecular clusters, making these multimetallic clusters attractive building blocks for functional materials.1-5 These complexes are also often of interest as models for the active sites of enzymes, such as that of nitrogenase, which contains an FeMo cluster. Controlling the nuclearity of cluster complexes presents a challenge that may be solved using supramolecular chemistry. Macrocyclic ligands with several donor atoms often coordinate polynuclear transition metal complexes in their interior and may serve as a scaffold for the synthesis of molecular metal clusters. Polynuclear complexes have been reported inside Robson-type macrocycles,6-8 sulfonylcalixarenes,9,10 thiacalixarenes,11 and a variety of other organic macrocycles. Recently a tetradecanuclear copper(II) cluster templated inside two multidentate macrocycles was reported by Tandon et al.12 In general, metal clusters may be coordinated within the preformed macrocycle, or, as is often the case, they may template the formation of the macrocycle. Metal cluster formation inside macrocycles is not well understood. These complexes are usually obtained by hydrothermal methods and it is assumed that the  80  thermodynamic product is obtained, but little thought is given to the mechanism of cluster formation. The synthesis of complex multimetallic cluster complexes, such as the FeMo cluster found in nitrogenase, will undoubtedly require a templating mechanism in which metal atoms are incorporated in a step-wise fashion. Mechanistic insight into metal cluster formation may aid in the development of new clusters and enhance researchers’ synthetic control over existing multimetallic clusters. Our research group and others have been investigating the chemistry of Schiff base macrocycles.13-18 In particular, the macrocycles relevant to this chapter are shapepersistent, conjugated Schiff base macrocycles such as 51a-d, which have three N2O2 coordination sites for metal complexation.19-27 Moreover, these rings have a crown-etherlike interior composed of six hydroxyl groups with the potential to further coordinate metals. It has been demonstrated that the metal-free, Schiff base macrocycles coordinate alkali metals in the central hexa(hydroxy) interior rather than in the N2O2 pockets.28 Recently, Nabeshima et al. coordinated a lanthanum ion in this same environment to template a related oxime-based macrocycle.29 In prior work from our group, it was reported that macrocycles 51a-c can coordinate [Zn4O]6+ clusters in their interiors.30,31 Macrocyclic complexes of zinc are attractive as models for the active sites of metalloenzymes such as alkaline phosphatase, P1 nuclease, and phospholipase C, the latter two both possessing trinuclear zinc clusters in the active site.32 Zacharias and coworkers reported DNA cleavage by a trinuclear zinc complex formed within a chiral macrocycle.33 The same complex was used by Gao et al. to catalyze asymmetric aldol and Henry reactions.34 Other catalytic reactions, including carboxy- and phosphate-ester hydrolysis and the direct catalytic conversion of lactones and esters to oxazolines have been demonstrated using high nuclearity zinc complexes.35,36 Reactions involving a zinc source and a carboxylate ligand most often yield threedimensional networks or layered structures but occasionally multinuclear molecular cluster compounds are isolated. The most common molecular zinc carboxylate cluster, basic zinc acetate, consists of a tetrahedral [Zn4(μ4-O)]6+ core with six acetate ligands bridging the zinc ions in a μ-1,3 fashion.37 Tetranuclear cubane,38-41 drum-like hexanuclear dicubane42 as well as double tetrahedral heptanuclear zinc clusters have also  81  been synthesized.43,44 Roesky et al. reacted tert-butylphosphonic acid in THF with ZnEt2 in toluene, resulting in the highest nuclearity molecular zinc oxo complex to date, a dodecanuclear zincophosphate aggregate that consists of a [Zn4(μ4-O)]6+ core and a zincophosphonate shell.45 In general, the clusters of zinc complexes that have been obtained are by chance and are usually not stable in solution. This chapter discusses my investigations of the incorporation of zinc into the Schiff base macrocycles 51a-d, affording heptanuclear metallocavitands 52a-d as illustrated in Scheme 2.1. The isolation and characterization of two intermediate tetranuclear complexes strongly suggests that macrocycles 51a-d template the formation of [Zn4O]6+ clusters and provides insight into the mechanism of metal cluster assembly. These tetranuclear complexes exhibit unusual metal-ion ring walking behavior and may function as scaffolds for the templation of mixed metal cluster complexes. Macrocycles 51a-d are promising hosts for the development of well-defined mixed metal clusters.  Scheme 2.1. The reaction of shape-persistent, conjugated Schiff base macrocycles 51a-d with zinc acetate yields heptazinc cluster complexes 52a-d. The seventh zinc ion and sixth acetate ligand are obscured by the depiction of the cluster.  82  2.2  Discussion  2.2.1 Macrocycle Synthesis While investigating the incorporation of zinc into macrocycles 51a-c I was frustrated by difficulties obtaining clear NMR and MS identification of multimetallic intermediates. Consequently, I have worked to obtain crystallographic evidence for the compounds, and this has required the synthesis of macrocycles with various substituents to promote crystallization. New macrocycle 51d was prepared with the notion that the tert-butyl groups on the neopentyl chains would assist crystallization of the large complexes. Macrocycle 51d was prepared in four steps from catechol via the route illustrated in Scheme 2.2. Compound 53 was prepared in 81% yield using a literature procedure.46 This reaction was conducted under several different conditions (e.g., K2CO3 / DMF; NaOH / EtOH) that are effective for the preparation of analogues with n-alkyl chains, but it was necessary to use hexamethylphosphoramide (HMPA) to achieve a decent yield ( < 5% yield with alternative procedures). Compound 53 was readily nitrated to 54 and the nitro groups subsequently reduced to give compound 55d in good overall yield. Schiff base condensation of 55d and 56 in CHCl3 / MeCN afforded macrocycle 51d in 84% yield. Macrocycle 51d is soluble in DMF-d7 and DMSO-d6, and shows a 1H NMR spectrum that is characteristic of average D3h symmetry. Matrix-assisted laser-desorption ionization time-of-flight (MALDI-TOF) mass spectrometry confirmed that the macrocycle was the [3+3] product, and not a larger cycle. Macrocycles 51a-c and precursors 55a-c were previously described by our group.22  83  Scheme  2.2.  Preparation  of  4,5-diamino-1,2-dineopentyloxybenzene  55d  and  macrocycles 51a-d.  2.2.2 Heptazinc Metallocavitands: Synthesis and Structure Previously our group reported the unexpected formation of heptanuclear zinc cluster complexes 52a-c after preparing the macrocycles 51a-c with an excess of zinc acetate as outlined in Scheme 2.1. Single-crystal X-ray diffraction (SCXRD) studies of 52a and 52b revealed the complexes consist of a trimetallic macrocycle with each N2O2 pocket coordinated to a zinc ion and a distorted tetrahedral [Zn4O]6+ cluster fastened to the macrocycle through bridging acetate ligands. I have since established by MALDITOF mass spectrometry, SCXRD, and NMR spectroscopy that reacting macrocycle 51d 84  with an excess of zinc acetate also yields a heptanuclear zinc cluster complex, 52d (Figure 2.1). As expected, cluster complex 52d is very similar to complexes 52a and 52b. The structural consistency supports the rigidity of the macrocycle-cluster complex, and suggests that the alkyl chains, which aid in solubility, have little influence on the coordination environment found in the interior of the macrocycles.  (a)  (b)  (c)  Figure 2.1. ORTEP depictions of heptazinc metallocavitand 52d crystallized from DMF. a) Top-down view of the complex. b) Side-on view of the complex. c) Side-on view of the complex with the Zn–O bonds of the central [Zn4O]6+ cluster highlighted in orange (some bonds have been removed for clarity). All ellipsoids are at 50% probability with hydrogen atoms omitted (C = black, N = blue, O = red, Zn = green).  Because of the three-fold rotational symmetry of 52a-d, only one imine, two aromatic, and two acetate resonances are observed in their 1H NMR spectra. The OCH2 resonance of the neopentyloxy substituents is a singlet for macrocycle 51d, but appears as two doublets for complex 52d. This indicates that the two methylene protons in 52d are diastereotopic and display geminal coupling of 8.8 Hz. Similarly, the OCH2 resonances in 51a-c are triplets, whereas 52a-c all show complex coupling patterns that were simulated as an ABX2 spin system. The protons on the OCH2 for 52a-c are inequivalent and display both geminal and vicinal coupling. Since macrocycles 51a-d clearly exhibit C3 rotational symmetry in solution, and  13  C NMR spectroscopy indicates that there is only a single  85  OCH2 carbon environment, the horizontal plane of symmetry in the macrocycle must be absent from 52a-d, leading to diastereotopic H atoms in the OCH2 group. These spectroscopic observations support the structural assignment. Peaks corresponding to the molecular ion or fragment ion [M-OAc]+ are observed by MALDI-TOF mass spectrometry for each heptazinc complex, further confirming their structures. In the case of 52d, peaks found by MALDI-TOF MS at m/z 1604.3, 1746.3, and 1811.3 are assigned to tetra-, penta-, and hexazinc fragments. Similar fragmentation patterns are observed for 52a-c using soft ionization techniques. Complexes 52a-d are isolated as deep red to bright orange powders but all appear deep red upon dissolution in CH2Cl2, CHCl3, DMSO, or DMF. The nearly superimposable electronic absorption spectra of the complexes are dominated by three peaks. Of the three, the highest intensity absorption is between 413 and 419 nm with a hypsochromic shoulder found between 346 and 348 nm. These peaks are most likely π or n to π* transitions since similar absorptions are observed for macrocycle 51a-d prior to metal coordination (406 and 343 nm for 51d). The final absorption exhibits similar intensity to the shoulder and is located between 240 and 242 nm. Basic zinc acetate exhibits a ligand-to-metal charge-transfer (LMCT) band at 216 nm, evidence that supports this final absorption also being a LMCT band,47 though it may be associated with the large conjugated macrocycle present in 52a-d. The near-tetrahedral [Zn4O]6+ acetate clusters within structures 52a-d are very similar to basic zinc acetate. Figure 2.2 depicts basic zinc acetate and the [Zn4O]6+ acetate cluster portion of complex 52d. For the purpose of comparing bond lengths, Figure 2.2c shows the arrangement of the metal ions and acetate ligands in complexes 52a-d. Selected bond lengths of the capping cluster are summarized in Table 2.1. The Zn2+ ions are designated as Zn1-Zn7 and the acetate bridging ligands as OAc1-OAc6 as shown in Figure 2.2c. The three Zn2+ ions within the N2O2 pockets of the macrocycle are Zn1, Zn2 and Zn3 (square pyramidal), those at the bottom of the tetrahedral capping cluster are Zn4, Zn5 and Zn6 (distorted octahedral), and the lone Zn2+ ion at the top of the capping cluster is Zn7 (tetrahedral). The acetate ligands that bridge the macrocycles to the capping clusters in a P-1,1,3 fashion are designated OAc1, OAc2 and OAc3 and those linking the bottom of the [Zn4O]6+ tetrahedron (Zn4-6) to the top (Zn7) are OAc4, OAc5 and OAc6.  86  Basic zinc acetate may be visualized as the tetrahedron made up of Zn4, Zn5, Zn6 and Zn7 with bridging P-1,3 acetate ligands OAc4, OAc5, and OAc6. The other three acetate ligands found within basic zinc acetate are comparable to the O–Zn(1-3)–O (catechol oxygen atoms) units in 52a-d.  (a)  (b)  (c)  Figure 2.2. a) The tetrahedral basic zinc acetate cluster, Zn4O(OAc)6. b) Distorted tetrahedral [Zn4O]6+ cluster from complex 52d (the macrocycle portion of 52d has been omitted). Thermal ellipsoids are at 50% probability with hydrogen atoms removed for clarity. (C = black, N = blue, O = red, Zn =green). c) Schematic of the heptazinc cluster templated by macrocycles 51a-d with individual zinc ions and acetate ligands labeled for bond length comparison.  87  Table 2.1. Bond lengths pertaining to the zinc acetate cluster of complexes 52a, 52b, 52d, and basic zinc acetate.a,b  52a (Å)  Bond  52b (Å)  52d (Å)  Zn1 – OAc1  1.987(4)  2.072(5)  2.008(4)  Zn2 – OAc2  1.995(4)  1.992(4)  2.009(4)  Zn3 – OAc3  2.056(4)  1.983(4)  2.009(4)  Zn4 – OAc1  2.507(4)  2.437(5)  2.449(4)  Zn4 – OAc2  2.572(4)  2.719(4)  2.618(4)  Zn5 – OAc2  2.592(4)  2.749(4)  2.506(4)  Zn5 – OAc3  2.462(4)  2.473(4)  2.594(4)  Zn6 – OAc3  2.400(4)  2.576(4)  2.595(4)  Zn6 – OAc1  2.667(4)  2.591(4)  2.476(3)  Zn4 – OAc4  1.978(4)  1.966(4)  1.983(3)  5  Zn – OAc  1.977(4)  1.968(4)  1.974(4)  Zn6 – OAc6  1.979(4)  1.964(5)  1.973(4)  Zn7 – OAc4  1.981(4)  1.951(4)  1.966(4)  Zn7 – OAc5  1.979(4)  1.961(4)  1.972(4)  Zn7 – OAc6  1.976(4)  1.973(5)  1.984(4)  Zn4 – μ4-Oc  1.916(4)  1.919(4)  1.913(3)  Zn5 – μ4-O  1.926(4)  1.921(4)  1.894(3)  Zn6 – μ4-O  1.932(4)  1.927(4)  1.913(3)  Zn7 – μ4-O  1.981(3)  1.975(4)  Zn1 --- Zn4  5  Basic Zinc  52a, 52b, &  Acetate  52d (Å)  (Å)  2.012  2.555  1.974  1.976  1.971  1.976  1.918  1.966  1.969(3)  1.975  1.966  3.536  3.512(5)  3.547  3.501  6  Zn --- Zn  3.602(5)  3.478  3.509  Zn2 --- Zn4  3.553(5)  3.497  3.559  Zn2 --- Zn5  3.568(5)  3.602  3.490  Zn3 --- Zn5  3.559(5)  3.573  3.498  1  Average of  88  Zn3 --- Zn6  3.523(5)  3.540  3.536  Zn4 --- Zn5  3.305(5)  3.249  3.256  Zn5 --- Zn6  3.233(5)  3.195  3.251  Zn6 --- Zn4  3.315(5)  3.326  3.200  Zn4 --- Zn7  2.969(5)  2.966(1)  3.043(1)  Zn5 --- Zn7  3.004(5)  3.078(1)  2.978(1)  Zn6 --- Zn7  2.987(5)  3.040(1)  2.966(1)  3.259  3.210  3.003  3.210  a  Zinc nuclei and acetate ligand labels correspond to Figure 2.2c.  b  The table is sectioned into bonds that are equivalent by symmetry in complexes 52a,  52b, and 52d. c  μ4-O is the central oxygen of the [Zn4O]6+ tetrahedron. The most striking feature of complexes 52a-d and basic zinc acetate is that each  zinc cluster is centered about a tetrahedral μ4-O atom with similar Zn-O bond lengths. The Zn - μ4-O distance is 1.966 Å in basic zinc acetate and 1.975 Å (average μ4-O - Zn7) and 1.918 Å (average of symmetry equivalent Zn4, Zn5, and Zn6 - μ4-O bond lengths) in complexes 52a, 52b, and 52d. The deviations from tetrahedral symmetry in the [Zn4O]6+ cluster found within 52a, 52b, and 52d may be attributed to electronic effects. Greater Lewis-acidic character exhibited by Zn4, Zn5, and Zn6 due to their proximity to Zn1, Zn2 and Zn3 in N2O2 pockets may account for the reduced average bond length between Zn4, Zn5, Zn6, and the central μ4-O. This enhanced bonding also limits the electron density available for the μ4-O - Zn7 bond in agreement with the observed longer bond. Steric effects may also play a role in the elongation of the μ4-O - Zn7 bond. Tetrahedral [Zn4O]6+ clusters are most often supported by carboxylate ligands; however, they have been reported with carbamato,48-50 phosphato,51 formamidinate,52 and N-heterocyclic53 ligands. Within this family of basic zinc acetate analogues, the Zn - μ4-O distances range from 1.84 to 2.01 Å, a span within which complexes 52a, 52b, and 52d comfortably lie. Unusual heptazinc clusters consisting of two [Zn4O]6+ units that share a single octahedral Zn2+ ion have also been reported and have slightly longer Zn - μ4-O distances.43,44  89  2.2.3 Mechanistic Insight Formation of 52 from 51 begins with trimetallation of the macrocycle in the N2O2 sites. There are two probable avenues for formation of the tetrahedral cluster above the plane of the macrocycle: 1) the trimetallated macrocycle coordinates to a preformed basic zinc acetate cluster in solution, or 2) the cluster is templated inside the metallomacrocycle. A templation mechanism is most plausible because basic zinc acetate is generally synthesized by heating zinc acetate under a vacuum, eliminating acetic anhydride, but complex 52 may be synthesized at room temperature (or even lower) and atmospheric pressure. Figure 2.3 shows a simple schematic comparing these two routes, neglecting the acetate and oxo ligands.  90  a)  RO  OR  RO  N  N  N  N  O  O  O  O  O  O  N  O N  O RO  OR  RO  RO  OR  N O  N  N  OR  RO  OR  OR  RO  OR  RO  OR  N  N  N  N  N  N  O  O  O  O  O  O  O  O  N  O N  O RO  O O  N  RO  = Zn2+  N  O  N  b)  OR  O  N  O  N  O N  OR RO  RO  OR  N  OR  N  N  O  O  O  O  O  OR  OR OR  N O  OR RO  N  O  N  O N  O  OR  N  O  N  RO  RO  N N  OR RO  N  O  RO  O  OR  O  N  O  N  N  O  N  RO  RO  RO  O N  N  RO  O N  OR OR  Figure 2.3. Two possible mechanisms for the formation of heptazinc complexes 52a-d. a) Basic zinc acetate forms in situ and coordinates to a trimetallated macrocycle, or b) the trimetallated macrocycle templates the cluster formation. Acetate ligands, coordinated solvent, and the central μ4-O ligand are omitted for clarity.  To investigate the possibility of a supramolecular templating mechanism such as that depicted in Figure 2.3b, a trizinc metallomacrocycle was first synthesized hoping it would act as a scaffold to monitor the formation of the [Zn4O]6+ acetate cluster within  91  macrocycles 51a-d. Reaction of 51c with three equivalents of Zn(OAc)2 gave a product with a very broad 1H NMR spectrum. The ESI mass spectrum of the product indicates that the trizinc metallomacrocycle, complex 57c, forms but also shows many species that correspond to aggregates (e.g., [57c2]2+, [57c2 + Zn]2+, [57c3]2+). These results suggest that the trizinc metallomacrocycle can be prepared, but that it aggregates strongly in solution. Strong aggregation has been observed in larger macrocycles complexed to zinc.24 The absence of larger metallomacrocycle complexes such as 52c in the MS and 1H NMR spectra of the products from the trizinc reactions and the absence of hydroxyl protons in the NMR spectra lead the conclusion that complexation within the N2O2 pockets occurs prior to any cluster formation. Attempts to isolate and characterize the trizinc complex using macrocycles with other substituents were also fraught with difficulties.  When macrocycle 51d was reacted with four equivalents of Zn(OAc)2, a new product, 58, was observed. Notably, in the 1H NMR spectrum of complex 58 the deshielded hydroxyl resonances are absent, indicative of zinc coordination to the N2O2 pockets. The equivalent integration of one acetate resonance, numerous aromatic resonances, and three distinct imine resonances suggests a mirror plane is the only symmetry element present in the compound. The experimental 1H NMR spectrum and proposed structure for complex 58 are shown in Figure 2.4. Two broad resonances hardly  92  distinguishable from the baseline were located around 12.1 and 10.0 ppm and assigned to aqua ligands after obtaining a SCXRD structure. The molecular ion of tetrazinc complex 58 was not detectable by MALDI-TOF mass spectrometry, but instead the spectrum was dominated by the [58-AcOH + Na]+ fragment.  Figure 2.4. 1H NMR spectrum of tetrazinc complex 58 in DMF-d7 (400 MHz). The equivalent integration (6H) of three imine resonances, 5-6 aromatic resonances, and one acetate resonance is evidence for a single σh symmetry element and supports the inset structure (acetates have been omitted and the zinc ions are labeled analogous to complexes 52a-d in Figure 2.2c). DMF resonances are labeled with (*) and non-bonded water is labeled with (°). The magnetic environments about the OCH2 and C(CH3)3 protons are only slightly perturbed by the broken symmetry and therefore appear as broad singlet resonances rather than as multiple resonances.  93  To confirm the structure of complex 58, single crystals were grown by diffusing ether into a DMF/pyridine solution and a SCXRD analysis was conducted. The solidstate structure is depicted in Figure 2.5. Using the labeling scheme from Figure 2.2c, Zn1, Zn2, and Zn3 are found within each N2O2 pocket. They adopt a square pyramidal geometry, distorting the macrocycle from nearly planar to concave. The Zn-O and Zn-N bond lengths of the N2O2 pockets range from 1.95 to 2.01 Å and 2.06 to 2.11 Å, respectively. Zn1 and Zn2 are equivalent by a mirror plane (σh) and each has an axially coordinated acetate ligand that bridges to Zn4. An aqua ligand was found bound to the axial position of Zn3. The final zinc ion, Zn4, exhibits square pyramidal geometry and is coordinated to the interior of the macrocycle. The coordination sphere of Zn4 is filled by a catechol group, two acetate ligands bridging from Zn1 and Zn2, and an aqua ligand directed at the center of the macrocycle. Most intriguing, the aqua ligand is perfectly poised to become the central μ4-oxo ligand of the [Zn4O]6+ acetate cluster found within complex 52d. The Zn4-Oaqua distance is 1.955 Å, very similar to the Zn-O distance in basic zinc acetate (1.966 Å).  94  (b)  (a) Zn4  Zn1  Zn2  Zn3  (c)  Figure 2.5. Solid-state structure of tetrazinc metallomacrocycle 58 determined by a SCXRD experiment. a) Side-on view of complex 58 revealing the central aqua ligand (the zinc ions have been labeled according to Figure 2.2c). b) View from above: Zn4 is projected out of the page. c) Side-on view of Zn4 bridged by acetates to Zn1 and Zn2. An aqua ligand completes the coordination sphere of Zn3. Hydrogen atoms are omitted and thermal ellipsoids are at 50% probability. Peripheral alkoxy chains have been omitted from a) and b) for clarity (C = black, N = blue, O = red, Zn = green).  The templation of the tetrahedral zinc acetate cluster found in heptazinc metallomacrocycle 52d begins with the coordination of Zn4 above the plane of the trizinc macrocycle. To further understand this templating, the addition of Zn(OAc)2 to complex 58 was monitored by 1H NMR spectroscopy. The cluster growth was best observed in the aromatic and acetate regions shown in Figure 2.6. As Zn(OAc)2 was incrementally added 95  to a solution of 58, new intermediate resonances arose but not in a smooth stepwise fashion. After exposure to a total of 5.5 equivalents of Zn2+ ions, the three imine and multiple aromatic resonances corresponding to complex 58 disappeared and resonances corresponding to the heptazinc complex 52d became prominent. Upon further addition of zinc acetate, the single acetate resonance observed for complex 58 also disappeared and two resonances assigned to complex 52d dominated the final spectrum.  Figure 2.6. 1H NMR spectra of the downfield and acetate regions during the stepwise addition of Zn(OAc)2 in DMF-d7 to a solution of tetrazinc complex 58, resulting in the formation of heptazinc metallocavitand 52d. The numbers on the left indicate the molar equivalents of zinc ions to macrocycle, (*) indicates DMF, and (°) indicates free acetic acid which accumulates as complex 52d forms.  Although the NMR titration presents evidence of other intermediates, possibly pentazinc or hexazinc complexes, during cluster templation, it was not possible to isolate these species. Crystals suitable for SCXRD could not be obtained from the addition of five or six equivalents of Zn(OAc)2 to any of macrocycles 51a-d and mass spectrometry provided little concrete evidence, as the heptametallic species is known to fragment to pentazinc and hexazinc species upon ionization.  96  2.2.4 Metal Ion Ring Walking High temperature 1H NMR studies of complexes 52a-d (in CDCl3 or DMF-d7) failed to show any intramolecular dynamic processes such as acetate exchange up to ~120 °C in DMF-d7; this is a testament to the robustness of the zinc complexes. High temperature 1H NMR spectroscopy was used to probe fluxional behavior of complex 58. When a sample of 58 in DMF-d7 was heated, coalescence of the downfield resonances was observed as shown in Figure 2.7. The spectral transformation is consistent with a change from Cs symmetry (observed in the solid-state and room temperature 1H NMR spectroscopic data) to three-fold rotational symmetry. Specifically, the three imine resonances and six aromatic resonances coalesce to one and two resonances, respectively, resulting in a spectrum nearly identical to the C3 symmetric heptazinc metallocavitands 52a-d. The process is reversible, and upon cooling back to room temperature, the original 1  H NMR spectrum of 58 showing Cs symmetry is obtained. This unexpected behavior  implies that the fourth zinc ion, located within the central ring of the macrocycle, is labile, leading complex 58 to reversibly adopt a higher symmetry at elevated temperatures.  385 K 357 330 300 10  10  9  9  8  8  7  7  6  6  5  5  4  4  3  3  2  2  1  1  0 ppm  Figure 2.7. Variable temperature 1H NMR spectra of tetrazinc complex 58 in DMF-d7 (400 MHz). As temperature is increased, the downfield imine and aromatic resonances coalesce, consistent with a shift from Cs to average 3-fold symmetry. This higher symmetry may be a result of the out-of-plane Zn2+ ion exchanging between three equivalent environments.  97  Kajiwara et al. observed similar behavior by variable temperature NMR spectroscopy  while  studying  a  mononuclear  Zn2+  complex  of  p-tert-  butylsulfonylcalix[4]arene.54 At room temperature the aromatic protons of the calixarene are equivalent and exhibited a single resonance, suggesting the zinc ion is walking between four equivalent tridentate sites. However, four different aromatic environments are observed in the 1H NMR spectrum taken at -60 °C (Ha-d in Figure 2.8a). At lower temperature, the Zn2+ ion is frozen in one environment on the NMR timescale but freely exchanges or “walks” between equivalent sites at higher temperature as depicted in Figure 2.8. Sekiguchi et al. also observed metal ion ring walking by multi nuclear NMR spectroscopy in a lithium salt of a [4]radialene dianion.55  Figure 2.8. Illustration of Zn2+ ring walking in polydentate macrocycles. a) Zn2+ exchanges between four equivalent sites in a sulfonylcalixarene. b) Proposed exchange of Zn2+ between three equivalent sites in tetrazinc complex 58.  The coalescence observed in the 1H NMR spectra of complex 58 upon heating can be explained by either Zn2+ ion ring walking as illustrated in Figure 2.8b or dissociation of the zinc ion and acetate ligands from the macrocycle. To test for dissociation, sodium acetate was added to a solution of 58 in DMF-d7, the sample was heated to 100 °C, and the number of acetate resonances in the 1H NMR spectrum was monitored. The presence of two different acetate resonances, one from free acetate and the other from complex 58, would confirm Zn2+ ring walking whereas a single resonance would be evidence for 98  dissociation (all the acetate is free in solution). Surprisingly, a reaction took place and the 1  H NMR spectrum showed a mixture of products including heptazinc complex 52d and  other unidentifiable species, possibly the trizinc macrocycle as shown in Figure 2.9.  Figure 2.9. 1H NMR spectra in DMF-d7 of complex 58, 58 + 1 equivalent of NaOAc, and 52d. We attempted to isolate a trizinc complex 57d of macrocycle 51d, the hypothesized dissociation product of 58, to compare the 1H NMR chemical shifts in DMF-d7 at 100 °C with those of complex 58 at 100 °C. Different chemical shifts would help confirm Zn2+ ring walking whereas identical shifts would be evidence for dissociation. Unfortunately, isolation of the trizinc complex of macrocycle 51d was not possible.  99  2.2.5 Zinc Methacrylate Complexes The investigation of zinc carboxylate cluster templation was expanded to include zinc methacrylate in an effort to alter the reactivity of the growing clusters allowing for study and isolation of possible intermediates. Reaction of macrocycle 51a with an excess of zinc methacrylate afforded heptanuclear metallocavitand 59 in 25% yield (after recrystallization). The 1H NMR spectrum of complex 59 showed one imine, two aromatic, and two methacrylate CH3 resonances, identical to the analogous heptazinc complex 52a. Four resonances were assigned to olefinic methacrylate protons, consistent with the existence of two carboxylate environments. Interestingly, the OCH2 protons are chemical shift equivalent in DMF-d7, and appear as a quartet rather than an ABX2 spin system as observed for the OCH2 protons in complex 52a. The molecular ion [59]+ and [59-methacrylate]+ were both observed by MALDI-TOF mass spectrometry.  When four equivalents of zinc methacrylate were reacted with macrocycle 51a, a tetrazinc complex 60 was formed. The 1H NMR spectrum of 60 is very similar to that of tetrazinc acetate complex 58. Single crystals suitable for X-ray diffraction were obtained from DMSO and the SCXRD analysis revealed a complex with four zinc centers, analogous to tetrazinc complex 58. Each zinc ion exhibits square pyramidal geometry as depicted in Figure 2.10. Three Zn2+ ions occupy the N2O2 binding sites, imparting a saucer shape to the macrocycle, with Zn–O and Zn–N distances ranging from 1.96 to 100  2.05 Å and 2.10 to 2.16 Å, respectively. Using the labeling scheme from Figure 2.2c, methacrylate ligands are coordinated to the axial position of both Zn1 and Zn2. The axial position of Zn3 is occupied by a DMSO molecule coordinated through a Zn-O interaction. One zinc ion, Zn4, is coordinated within the central ring of the macrocycle to the oxygen atoms of one catechol unit. Two methacrylate ligands bridging from Zn 1 and Zn2 and an aqua ligand complete the coordination sphere of the capping Zn4 ion. Much like complex 58, the aqua ligand is centrally located and in perfect position to template the formation of 59. The Zn4-Oaqua distance is 1.966 Å, identical to the Zn-O distance in basic zinc acetate.  Figure 2.10. Solid-state structure of complex 60 as obtained by SCXRD analysis. a) View from the front. b) Top-down view c) Side-on view clearly showing the DMSO coordinated to Zn3. Thermal ellipsoids are shown at 50% probability. Hydrogen atoms and peripheral alkoxy groups have been omitted for clarity (C = black, N = blue, O = red, Zn = green, S = yellow).  101  Tetranuclear complexes 58 and 60 represent trapped intermediates en route to heptanuclear metallocavitands 52d and 59, respectively. More studies are required to fully elucidate what drives the templation of tetrahedral metal clusters inside macrocycles 51a-d, but some conclusions may be drawn from the present work. Upon coordination of Zn2+ to the N2O2 pockets, the zinc ions are forced into a pseudo-square planar geometry due to the rigid nature of the conjugated macrocycles. Typically, four coordinate zinc ions prefer tetrahedral geometry, which is unachievable inside the N2O2 pockets, forcing them to adopt square pyramidal geometry by coordinating a carboxylate ligand in the apical position. This trizinc scaffold, with tethered carboxylate ligands, appears to be an optimal substrate to template clusters similar to basic zinc acetate.  2.2.6 Mixed Metal Clusters Appealing candidates for the study of metalloenzyme mimics, mixed metal clusters have been a longstanding target of synthetic inorganic chemists. Owing to the often unpredictable geometry and reactivity of metal clusters, the controlled synthesis of mixed metal clusters poses a significant challenge. Complex 58 is an isolable, reactive intermediate with potential for further metal complexation. To demonstrate this potential, four equivalents of cobalt(II) acetate were added to a suspension of 58 in EtOH and the mixture was heated to reflux. Cobalt(II) was chosen because it is known to form tetrahedral [Co4O]6+ clusters analogous to basic zinc acetate.56 A rusty orange, microcrystalline solid was isolated by filtration and subjected to a MALDI-TOF mass spectrometry study. The mass spectrum revealed overlapping isotopic distributions centered about m/z = 1975.0 and 1968.6 that may be assigned to the mixed metal clusters [(51d6H)Zn4Co3O(OAc)5]+ and [(51d-6H)Zn3Co4O(OAc)5]+, respectively. Another peak corresponding to the cluster [(51d-6H)Zn3Co3O(OAc)3]+ was resolved well enough for isotopic assignment as displayed in Figure 2.11. This peak is likely due to a fragment of the previous two clusters, suggesting that the parent cluster is a mixed metal complex that is structurally related to complex 52d. This preliminary result illustrates the potential of complex 58 for generating mixed metal clusters. 102  Figure  2.11.  a)  MALDI-TOF  isotopic  distribution  pattern  of  [(51d-  6H)Zn3Co3O(OAc)3]+. b) Simulated isotopic distribution of the same ion.  2.3  Conclusions This chapter described the synthesis of a new macrocycle with neopentyloxy  substituents and the investigation of zinc(II) incorporation into Schiff base macrocycles 51a-d. With an excess of zinc, a stable heptanuclear metallocavitand was obtained that may be described as a trizinc metallomacrocycle with a tetranuclear basic zinc acetate cluster in the macrocycle’s interior. Crystallographic and NMR studies revealed that these complexes exhibit C3v symmetry and have a closed-bowl shape that is retained in solution. As a first step to elucidating the mechanism of zinc carboxylate cluster templation, two intermediates with four Zn2+ ions were isolated and structurally characterized. I have also shown that heptanuclear metallocavitands were obtained by reacting the intermediate tetranuclear complexes with additional equivalents of zinc carboxylate. This provided evidence that a supramolecular templating mechanism may be active, with the macrocycle templating a molecular metal-oxo cluster in its interior. This system has potential for developing soluble cluster complexes that might find use as catalysts or molecular magnets. The ability to isolate reactive intermediates en route to larger metal clusters is a feature that has allowed for the synthesis of mixed ZnCo clusters and may aid in the development of polynuclear enzyme mimics. 103  2.4  Experimental  2.4.1 General Compounds 51a-c,22 52a-c,30 53,46 and 5621 were prepared according to literature procedures. All reactions were carried out under air unless otherwise noted. 1H and  13  C  NMR spectra were recorded on either a Bruker AV-300 or AV-400 spectrometer.  13  C  1  13  NMR spectra were recorded using a proton decoupled pulse sequence. H and C NMR spectra were calibrated to the residual protonated solvent at δ 7.27 and δ 77.23 ppm, respectively, in CDCl3 or at δ 8.03 and δ 163.15 for DMF-d7. UV-vis spectra were obtained in CH2Cl2 (ca. 5 x 10-5 M) on a Varian Cary 5000 UV-vis-near-IR spectrophotometer using a 1 cm quartz cuvette. FT-IR spectra were obtained as KBr discs with a Nicolet 4700 FT-IR spectrometer. MALDI-TOF mass spectra were obtained in a trans-2-[3-(4-tert-Butylphenyl)-2-methyl-2-propenylidene]-malonitrile (DCTB) matrix (solvent free) at the UBC Microanalytical Services Laboratory on a Bruker Biflex IV instrument. Electrospray ionization (ESI) mass spectra were obtained on a Bruker Esquire LC instrument from MeOH at the UBC Microanalytical Services Laboratory. Singlecrystal XRD was performed on Bruker X8 APEX CCD with Mo radiation. Elemental analyses (C,H,N) were performed at the UBC Microanalytical Services Laboratory. Melting points were obtained on a Fisher-John’s melting point apparatus.  104  2.4.2 Procedures and Data Synthesis of 1,2-dineopentyloxy-4,5-dinitrobenzene (54): 1,2-Dineopentyloxybenzene 53 (4.0 g, 16.0 mmol) was added to 50 mL of concentrated nitric acid and heated to 60 °C. A yellow precipitate formed after 30 min and after 12 h, the reaction mixture was poured into 400 mL of water. The yellow solid was isolated on a Buchner funnel, washed with water, and recrystallized from hot EtOH, affording large yellow needles of 54. Yield: 3.2 g (9.4 mmol, 59%). Data for 54. 13C NMR (100.6 MHz, CDCl3) δ 152.3, 136.6, 107.5, 79.5, 32.3, 26.6. 1H NMR (400 MHz, CDCl3) δ 7.24 (s, 2H, Ar), 3.72 (s, 4H, OCH2), 1.10 (s, 18H, CH3). UVvis (CH2Cl2) λmax (ε) = 343 (3.03 x 105), 274 (1.41 x 105), 247 (1.15 x 105) nm (L mol-1 cm-1). EI-MS m/z = 340 (54)+, 255 (54 - OCH2C(CH3)3)+. FT-IR (KBr) ῡ = 3432, 3071, 2971, 2959, 2872, 1586, 1530, 1478, 1378, 1366, 1336, 1292, 1266, 1228, 1045, 1014, 984, 941, 872, 818, 753, 637 cm-1. Mp = 168 °C. Anal. Calc’d for 54: C16H24N2O6: C, 56.46; H, 7.11; N, 8.23. Found: C, 56.67; H, 6.95; N, 8.60. Synthesis of 4,5-diamino-1,2-dineopentyloxybenzene (55d): In a Schlenk flask, compound 54 (4.5 g, 13.2 mmol) was suspended in 100 mL of EtOH and sparged with N2. After 30 min, the flask was charged with 4 mL of hydrazine monohydrate and approximately 750 mg of Pd/C. The Schlenk flask was then fit with a condenser and heated to 70 °C under N2. Raney Ni was added after 12 h to quench excess hydrazine and 4 h later the reaction was filtered through Celite using inert atmosphere techniques. Removal of solvent under vacuum gave compound 55d, an air sensitive white powder. Yield: 3.376 g (12.0 mmol, 91%). Data for 55d: 13C NMR (100.6 MHz, CDCl3) δ 144.0, 128.3, 106.5, 80.7, 32.4, 26.9; 1H NMR (300 MHz, CDCl3) δ 6.37 (s, 2H, Ar), 3.55 (s, 4H, OCH2), 3.17 (s, 4H, amine), 1.04 (s, 18H, CH3). EI MS m/z = 280 (55d)+, 210 (55d – C5H11)+, 140 (55d – (C5H11)2)+.  105  Synthesis of macrocycle 51d (R = CH2tBu): A Schlenk flask was charged with 55d (2.0 g, 7.1 mmol) and dissolved in 30 mL of degassed 1:2 CHCl3:MeCN. Addition of 3,6diformyl-1,2-dihydroxybenzene (56) (1.19 g, 7.1 mmol) to the clear, light yellow solution resulted in a deep red color. The flask was fit with a condenser and the reaction mixture was heated to reflux (80 °C) for 2 h under an atmosphere of nitrogen. After cooling to room temperature, the dark red microcrystalline precipitate of 51d was isolated on a Buchner funnel. Yield: 2.45 g (2.0 mmol, 84%). Macrocycle 51d was recrystallized from hot DMF for elemental analysis. The macrocycle was not sufficiently soluble to obtain a 13  C NMR spectrum.  Data for 51d (R = CH2tBu): 1H NMR (300 MHz, DMF-d7) δ 13.80 (s, 6H, OH), 9.16 (s, 6H, imine), 7.42 (s, 6H, Ar), 7.20 (s, 6H, Ar), 3.88 (s, 12H, OCH2), 1.13 (s, 54H, CH3). UV-vis (CH2Cl2) λmax (ε) = 343 (8.03 x 105), 406 (1.09 x 106) nm (L mol-1 cm-1). MALDI-TOF MS m/z = 1231 (51d)+, 1253 (51d + Na)+, 1268 (51d + K)+, 2506 (51d2 + Na)+. FT-IR (KBr) ῡ = 3440, 2955, 2904, 2869, 1612, 1514, 1495, 1476, 1456, 1301, 1259, 1221, 1190, 1009 cm-1. Mp > 270 °C. Anal. Calc’d for 51d: C72H90N6O12 · DMF: C, 69.37; H, 7.06; N, 7.55. Found: C, 69.06; H, 7.21; N, 7.33. Synthesis of the heptazinc metallocavitand 52d (R = CH2tBu): Zinc acetate dihydrate (66 mg, 0.30 mmol) was added to a suspension of 51d (52 mg, 42.3 μmol) in 5 mL of ethanol and the mixture was stirred. Upon addition of zinc acetate, the deep red solution rapidly turned bright orange. After being stirred at room temperature for 2 h, the reaction mixture was filtered and 52d was isolated as an orange powder. Yield: 61 mg (30 μmol, 70%). Recrystallization from hot DMF afforded analytically pure 52d. Single crystals suitable for X-ray diffraction were obtained from DMF.  106  Data for 52d (R = CH2tBu):  13  C NMR (100.6 MHz, DMF-d7) δ 174.8, 158.9, 135.1,  122.2, 121.0, 103.0, 100.9, 79.6, 33.0 , 27.2, 24.1, 22.5. 1H NMR (300 MHz, DMF-d7) δ 8.85 (s, 6H, imine), 7.55 (s, 6H, Ar), 6.95 (s, 6H, Ar), 3.93 (d, 2JHH = 8.8 Hz, 6H, OCH2), 3.89 (d, 2JHH = 8.8 Hz, 6H, OCH2), 1.90 (s, 9H, OAc), 1.74 (s, 9H, OAc), 1.14 (s, 54H, C(CH3)3). UV-vis (CH2Cl2) λmax (ε) = 242 (8.9 x 105), 348 (7.9 x 105), 419 (1.5 x 106) nm (L mol-1 cm-1). MALDI-TOF MS m/z = 1604.3 (52d-Zn3(OAc)4-O)+, 1746.3 (52dZn2(OAc)3+H)+ 1811.3 (52d-Zn(OAc)3)+, 1993.1 (52d-OAc)+, 2054.1 (52d)+. FT-IR (KBr) ῡ = 3440, 2956, 2906, 2870, 1612, 1508, 1463, 1456, 1321, 1262, 1221, 1182, 1112, 1045, 1018, 568, 493 cm-1. Mp > 270 °C. Anal. Calc’d for 52d: C84H102N6O25Zn7 · 2 DMF: C, 49.14; H, 5.32; N, 5.09. Found: C, 49.51; H, 5.39; N, 5.31. Synthesis of tetrazinc complex 58 (R = CH2tBu): Complex 58 was prepared following the procedure for heptazinc metallocavitand 52d but the stoichiometry of 51d to Zn(OAc)2 was 1:4 (50 mg, 41 μmol of macrocycle 51d and 35.6 mg, 0.16 mmol of zinc acetate dihydrate). Complex 58 was isolated as an orange powder by filtration and recrystallized from hot DMF. Yield: 40 mg (25 μmol, 61%). Low symmetry and poor solubility prevented analysis by 13C NMR spectroscopy. Single crystals suitable for X-ray diffraction were obtained from the diffusion of ether into a DMF/pyridine solution of 58. Data for 58 (R = CH2tBu): 1H NMR (400 MHz, DMF-d7) G12.13 bs, 2H, H2O), 9.97 (bs, 2H, H2O), 9.08 (s, 2H, imine), 8.97 (s, 2H, imine), 8.87 (s, 2H, imine), 7.67 (s, 2H, Ar), 7.58 (s, 4H, Ar), 6.86 (s, 2H, Ar), 6.60 (s, 4H, Ar), 3.89 (s, 12H, OCH2), 1.85 (s, 6 UV-vis (CH2Cl2) λmax (ε) = 241 (4.0 x 105) 360 (4.2 x 105), 421 (8.4 x 105) nm (L mol-1 cm-1). MALDI-TOF MS m/z = 1604.4 [58 - AcOH + Na]+. FT-IR (KBr) ῡ= 2957, 2907, 2869, 1608, 1507, 1450, 1386, 1325, 1261, 1220, 1179, 1112, 1045, 1017, 925, 842, 762, 669, 614 cm-1. Mp > 270 °C. Anal. Calc’d for 58: C76H92N6O17Zn4·2 DMF: C, 55.66; H, 6.04; N, 6.33. Found: C, 54.90; H, 5.92; N, 6.40.  107  Synthesis of the heptazinc methacrylate metallocavitand 59 (R = CH2CH3): Zinc methacrylate (171 mg, 0.73 mmol) was added to a stirred suspension of macrocycle 51a (100 mg, 0.10 mmol) in 15 mL of EtOH and heated to reflux for 11 h. The reaction was then cooled to room temperature and 160 mg of a bright orange powder was isolated by filtration. The crude product was recrystallized from DMF to give 59 as a red, microcrystalline solid. Yield: 50 mg (25 μmol, 25%). Low solubility prevented analysis by 13C NMR spectroscopy. Data for 59 (R = CH2CH3): 1H NMR (400 MHz, DMF-d7) δ 8.85 (s, 6H, imine), 7.57 (s, 6H, Ar), 6.95 (s, 6H, Ar), 5.84 (d, 2JHH = 2 Hz, 3H, C=CH2), 5.70 (d, 2JHH = 2 Hz, 3H, C=CH2), 5.22 (s, 3H, C=CH2), 5.10 (s, 3H, C=CH2), 4.32 (q, 3JHH = 7.2 Hz, 12H, OCH2), 1.74 (s, 9H, CH3 methacrylate), 1.64 (s, 9H, CH3 methacrylate), 1.47 (t, 3JHH = 7.2 Hz, 18H, OCH2CH3). UV-vis (CH2Cl2) λmax (ε) = 241 (4.9 x 105) 347 (4.4 x 105), 415 (8.9 x 105) nm (L mol-1 cm-1). MALDI-TOF MS m/z = 1955.6 [59]+, 1872.5 [59-methacrylate]+, 1722.1 [59-(methacrylate)2-Zn+H]+, 1637.3 [59-(methacrylate)3-Zn]+. FT-IR (KBr) ῡ= 2960, 2925, 2839, 1612, 1561, 1508, 1463, 1419, 1394, 1319, 1268, 1221, 1187, 1108, 1040, 940, 831, 753, 669, 616, 558 cm-1. Mp > 270 °C. Anal. Calc’d for 59: C96H114N6O25Zn7·2 DMF: C, 47.97; H, 4.41; N, 5.33. Found: C, 47.45; H, 4.49; N, 5.49. Tetrazinc methacrylate complex 60 (R = C2H5): A round bottom flask was charged with 51 mg (0.05 mmol) of macrocycle 51a and 49 mg (0.21 mmol) of zinc methacrylate. The mixture was cooled to 0 ºC in an ice bath, and 20 mL of cold ethanol was added. After being stirred at 0 ºC overnight, the precipitate from the cloudy dark red-orange solution was isolated by centrifugation. The powder was purified by a second centrifugation with fresh ethanol. Compound 60 was dried under vacuum and obtained as a deep red powder. Yield: 62 mg, (44 μmol, 80%). Single crystals were grown from DMSO. Low symmetry and poor solubility prevented analysis by  13  C NMR  spectroscopy.  108  Data for 60 (R = C2H5): 1H NMR (300 MHz, DMF-d7) G 10.0 (bs, 2H, H2O), 8.91 (s, 2H, imine), 8.83 (s, 2H, imine), 8.77 (s, 2H, imine), 7.60 (s, 2H, Ar), 7.57 (s, 2H, Ar), 7.51 (s, 2H, Ar), 6.67 (s, 2H, Ar), 6.52 (s, 4H, Ar), 5.91 (s, 2H, C=CH2), 5.29 (s, 2H, C=CH2), 4.31 (bs, 12H, OCH2), 1.81 (s, 6H, CH3 methacrylate), 1.47 (t, 3JHH = 7.8 Hz 18H, CH2CH3). UV-vis (CH2Cl2) Omax (H) = 421 (1.3 x 105), 350 (5.2 u 104) nm (L mol-1 cm-1). MALDI-TOF MS m/z = 1459.1 [60+(H2O)2+H]+. FT-IR (KBr) ῡ= 3433, 2978, 2929, 2899, 1607, 1553, 1505, 1450, 1419, 1390, 1365, 1326, 1264, 1211, 1183, 1106, 1040, 942, 903, 887, 847, 832, 753, 734 cm-1. Mp > 270 °C. Anal. Calc’d for 60·3H2O (C62H66N6O20Zn4): C, 50.42; H, 4.50; N, 5.69. Found: C, 50.43; H, 4.58; N, 5.74.  2.4.3 Crystallography Experimental details are summarized in Table 2.2. Single crystals of 52d, 58, and 60 were grown from DMF, DMF/pyridine ether diffusion, and DMSO, respectively. The structures were solved by direct methods with either SIR 9757 or SIR 200258 and further refined with SHELXL-97.59 Fifteen DMF molecules were modeled in the unit cell of 52d from the difference map with 8 geometric restraints placed on the DMF molecule at the special position. Complex 58 crystallized with five DMF molecules per asymmetric unit, one of which was too disordered to be resolved from the difference map. The electron density from the disordered DMF molecule was accounted for using the SQUEEZE function of the PLATON60 software package. Additionally, two aqua ligands were found coordinated in structure 58 but the protons were not modeled. Two DMSO molecules were modeled with complex 60 by difference maps; the remaining solvent could not be modeled due to disorder so the electron density was accounted for using the SQUEEZE function of PLATON. For each structure all non-hydrogen atoms attached to the macrocycle were refined anisotropically, whereas hydrogen atoms were added at geometrically expected positions and left isotropic.  109  Table 2.2. Crystallographic parameters for compounds 52d, 58, and 60. 52d  58  60  Formula  C106.5H154.5N13.5O32.5Zn7  C82H102N8O20Zn4  C66H70N6O18S2Zn4  M (g mol-1)  2601.53  1781.20  1560.98  Crystal system  Triclinic  Triclinic  Triclinic  Space group  P-1 (# 2)  P-1 (# 2)  P-1 (# 2)  a (Å)  16.1439(12)  14.9100(12)  13.0855(21)  b (Å)  19.9971(14)  20.1369(16)  17.6039(33)  c (Å)  20.2118(11)  21.2381(17)  21.7220(40)  α  108.279(3)  62.141(4)  70.3680(80)  β  93.599(2)  81.526(4)  81.4740(70)  γ  99.512(3)  84.554(4)  70.9570(80)  V (Å3)  6064.6(7)  5573.5(8)  4450.8(21)  Z  2  2  2  Temp (K)  173(2)  173(1)  173(2)  ρcalc (g/cm3)  1.425  1.061  1.165  μ (Mo Kα)  0.71073  0.71073  0.71073  F(000)  2716  1860  1608  Θ range (º)  1.5 – 25.3  1.8 – 26.9  1.7 – 28.3  reflns collected  96897  113091  81131  indep reflns  21651  23540  20789  obsd reflns [I >  14596  12780  8512  Rint  0.0530  0.0781  0.0981  parameters  1442  1017  878  R1  0.0518  0.0755  0.0642  wR2  0.1529  0.2166  0.1549  GOF  1.032  1.016  0.885  (mm-1)  2σ]  refined  110  2.5  References  (1)  Kong, X.-J.; Ren, Y.-P.; Long, L.-S.; Zheng, Z.; Huang, R.-B.; Zheng, L.-S. J. Am. Chem. Soc. 2007, 129, 7016-7017.  (2)  Bolink, H. J.; Cappelli, L.; Coronado, E.; Recalde, I. Adv. Mater. 2006, 18, 920923.  (3)  Ako, A. M.; Hewitt, I. J.; Mereacre, V.; Clerac, R.; Wernsdorfer, W.; Anson, C. E.; Powell, A. K. Angew. Chem., Int. Ed. 2006, 45, 4926-4929.  (4)  Hou, H.; Wei, Y.; Song, Y.; Mi, L.; Tang, M.; Li, L.; Fan, Y. Angew. Chem., Int. Ed. 2005, 44, 6067-6074.  (5)  Jian, F.; Xiao, H.; Bai, Z.; Zhao, P. J. Mater. Chem. 2006, 16, 3746-3752.  (6)  Tandon, S. S.; Thompson, L. K.; Bridson, J. N. J. Chem. Soc., Chem. Commun. 1992, 911-913.  (7)  Tandon, S. S.; Thompson, L. K.; Bridson, J. N.; Benelli, C. Inorg. Chem. 1995, 34, 5507-5515.  (8)  Bell, M.; Edwards, A. J.; Hoskins, B. F.; Kachab, E. H.; Robson, R. J. Am. Chem. Soc. 1989, 111, 3603-3610.  (9)  Kajiwara, T.; Katagiri, K.; Hasegawa, M.; Ishii, A.; Ferbinteanu, M.; Takaishi, S.; Ito, T.; Yamashita, M.; Iki, N. Inorg. Chem. 2006, 45, 4880-4882.  (10)  Kajiwara, T.; Kobashi, T.; Shinagawa, R.; Ito, T.; Takaishi, S.; Yamashita, M.; Iki, N. Eur. J. Inorg. Chem. 2006, 1765-1770.  (11)  Kajiwara, T.; Kon, N.; Yokozawa, S.; Ito, T.; Iki, N.; Miyano, S. J. Am. Chem. Soc. 2002, 124, 11274-11275.  (12)  Tandon, S. S.; Bunge, S. D.; Thompson, L. K. Chem. Commun. 2007, 798-800.  (13)  Vigato, P. A.; Tamburini, S.; Bertolo, L. Coord. Chem. Rev. 2007, 251, 13111492.  (14)  Volpe, M.; Reid, S. D.; Blake, A. J.; Wilson, C.; Love, J. B. Inorg. Chim. Acta 2007, 360, 273-280.  111  (15)  Wang, J.; Slater, B.; Alberola, A.; Stoeckli-Evans, H.; Razavi, F. S.; Pilkington, M. Inorg. Chem. 2007, 46, 4763-4765.  (16)  de Geest, D. J.; Noble, A.; Moubaraki, B.; Murray, K. S.; Larsen, D. S.; Brooker, S. Dalton Trans. 2007, 467-475.  (17)  Sessler, J. L.; Tomat, E.; Lynch, V. M. J. Am. Chem. Soc. 2006, 128, 4184-4185.  (18)  Hui, J. K. H.; MacLachlan, M. J. Chem. Commun. 2006, 2480-2482.  (19)  MacLachlan, M. J. Pure Appl. Chem. 2006, 78, 873-888.  (20)  Huck, W. T. S.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Recl. Trav. Chim. Pays-Bas 1995, 114, 273-276.  (21)  Akine, S.; Taniguchi, T.; Nabeshima, T. Tetrahedron Lett. 2001, 42, 8861-8864.  (22)  Gallant, A. J.; Hui, J. K. H.; Zahariev, F. E.; Wang, Y. A.; MacLachlan, M. J. J. Org. Chem. 2005, 70, 7936-7946.  (23)  Ma, C.; Lo, A.; Abdolmaleki, A.; MacLachlan, M. J. Org. Lett. 2004, 6, 38413844.  (24)  Ma, C. T. L.; MacLachlan, M. J. Angew. Chem., Int. Ed. 2005, 44, 4178-4182.  (25)  Gallant, A. J.; Yun, M.; Sauer, M.; Yeung, C. S.; MacLachlan, M. J. Org. Lett. 2005, 7, 4827-4830.  (26)  Nabeshima, T.; Miyazaki, H.; Iwasaki, A.; Akine, S.; Saiki, T.; Ikeda, C. Tetrahedron 2007, 63, 3328-3333.  (27)  Frischmann, P. D.; MacLachlan, M. J. Chem. Commun. 2007, 4480-4482.  (28)  Gallant, A. J.; MacLachlan, M. J. Angew. Chem., Int. Ed. 2003, 42, 5307-5310.  (29)  Akine, S.; Sunaga, S.; Taniguchi, T.; Miyazaki, H.; Nabeshima, T. Inorg. Chem. 2007, 46, 2959-2961.  (30)  Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2006, 45, 52485250.  112  (31)  Nabeshima, T.; Miyazaki, H.; Iwasaki, A.; Akine, S.; Saiki, T.; Ikeda, C.; Sato, S. Chem. Lett. 2006, 35, 1070-1071.  (32)  Parkin, G. Chem. Rev. 2004, 104, 699-767.  (33)  Korupoju, S. R.; Mangayarkarasi, N.; Zacharias, P. S.; Mizuthani, J.; Nishihara, H. Inorg. Chem. 2002, 41, 4099-4101.  (34)  Gao, J.; Zingaro, R. A.; Reibenspies, J. H.; Martell, A. E. Org. Lett. 2004, 6, 2453-2455.  (35)  Bazzicalupi, C.; Bencini, A.; Berni, E.; Giorgi, C.; Maoggi, S.; Valtancoli, B. Dalton Trans. 2003, 3574-3580.  (36)  Ohshima, T.; Iwasaki, T.; Mashima, K. Chem. Commun. 2006, 2711-2713.  (37)  Koyama, H.; Saito, Y. Bull. Chem. Soc. Jpn. 1954, 27, 112-14.  (38)  Driess, M.; Merz, K.; Schoenen, R. Organometallics 2007, 26, 2133-2136.  (39)  Murugavel, R.; Kuppuswamy, S.; Boomishankar, R.; Steiner, A. Angew. Chem., Int. Ed. 2006, 45, 5536-5540.  (40)  Hudson, T. A.; Berry, K. J.; Moubaraki, B.; Murray, K. S.; Robson, R. Inorg. Chem. 2006, 45, 3549-3556.  (41)  Polarz, S.; Orlov, A. V.; van den Berg, M. W. E.; Driess, M. Angew. Chem., Int. Ed. 2005, 44, 7892-7896.  (42)  Westerhausen, M.; Sapelza, G.; Goerls, H.; Mayer, P. Inorg. Chem. 2006, 45, 409-414.  (43)  Lalioti, N.; Perlepes, S. P.; Manessi-Zoupa, E.; Raptopoulou, C. P.; Terzis, A.; Aliev, A. E.; Gerothanassis, I. P. Chem. Commun. 1998, 1513-1514.  (44)  Waheed, A.; Jones, R. A.; McCarty, J.; Yang, X. Dalton Trans. 2004, 3840-3841.  (45)  Yang, Y.; Pinkas, J.; Noltemeyer, M.; Schmidt, H.-G.; Roesky, H. W. Angew. Chem., Int. Ed. 1999, 38, 664-666.  (46)  Khanamiryan, A. K.; Bhardwaj, N.; Leznoff, C. C. J. Porphyrins Phthalocyanines 2000, 4, 484-490. 113  (47)  Kunkely, H.; Vogler, A. J. Chem. Soc., Chem. Commun. 1990, 1204-1205.  (48)  Dell'Amico, D. B.; Calderazzo, F.; Labella, L.; Marchetti, F. Inorg. Chim. Acta 2003, 350, 661-664.  (49)  McCowan, C. S.; Groy, T. L.; Caudle, M. T. Inorg. Chem. 2002, 41, 1120-1127.  (50)  Belforte, A.; Calderazzo, F.; Englert, U.; Straehle, J. Inorg. Chem. 1991, 30, 3778-3781.  (51)  Lugmair, C. G.; Tilley, T. D.; Rheingold, A. L. Chem. Mater. 1997, 9, 339-348.  (52)  Cotton, F. A.; Daniels, L. M.; Falvello, L. R.; Matonic, J. H.; Murillo, C. A.; Wang, X.; Zhou, H. Inorg. Chim. Acta 1997, 266, 91-102.  (53)  Zheng, S.-L.; Zhang, J.-P.; Chen, X.-M.; Huang, Z.-L.; Lin, Z.-Y.; Wong, W.-T. Chem.--Eur. J. 2003, 9, 3888-3896.  (54)  Kajiwara, T.; Yokozawa, S.; Ito, T.; Iki, N.; Morohashi, N.; Miyano, S. Angew. Chem., Int. Ed. 2002, 41, 2076-2078.  (55)  Sekiguchi, A.; Matsuo, T.; Sakurai, H. Angew. Chem., Int. Ed. 1998, 37, 16621664.  (56)  Jaitner, P.; Rieker, C.; Wurst, K. Chem. Commun. 1997, 1245-1246.  (57)  Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 1999, 32, 115-119.  (58)  Burla, M. C.; Camalli, M.; Carrozzini, B.; Cascarano, G. L.; Giacovazzo, C.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 2003, 36, 1103.  (59)  Sheldrick, G. M.; SHELXL-97, University of Göttingen: Göttingen, Germany, 1997.  (60)  Spek, A. L.; PLATON, Utrecht University: Utrecht, The Netherlands, 2002.  114  Chapter 3 Dimerization of Heptazinc Metallocavitands† 3.1  Introduction  3.1.1 Abstract Entropy-driven dimerization of a new heptazinc metallocavitand with octyloxy substituents, 52e, is investigated in aromatic solvents. Thermodynamic parameters of self-association are elucidated by analyzing the imine chemical shift in variabletemperature variable-concentration (VTVC) 1H NMR spectroscopic experiments. The thermodynamics of dimerization are significantly different in benzene-d6, toluene-d8, and p-xylene-d10. Results of a single-crystal X-ray diffraction analysis show metallocavitand 52c dimerizes in the solid-state and encapsulates a single N,N-dimethylformamide (DMF) molecule. This system opens new avenues toward reversible, self-assembling, coordination capsules with potential for host-guest catalysis and molecular recognition.  3.1.2 Background Self-assembly of small components into elaborate architectures has become an attractive synthetic strategy to access functional materials. Cavitands are a class of bowlshaped supramolecules that are known to self-assemble into dimers, polyhedra, and polymers.1-4 Interesting host-guest properties have been observed in the cavity of these assemblies, often imparting catalytic, mechanical, and/or sensory functions to the assembly.5-10 Modified resorcin[4]arenes such as the one shown in Scheme 3.1 form  † A version of this chapter will be submitted for publication: Frischmann, P. D.; Gallant, A. J.; MacLachlan, M. J. “Entropy-Driven Dimerization of Zn7 Metallocavitands” 2010.  115  capsules in solution driven by complementary hydrogen bonding at the meridian of the capsule and favorable host-guest interactions.11,12 Self-assembly of cavitands into supramolecular dimers is generally an enthalpy-driven process and has been facilitated by H-bonding, electrostatic, solvophobic, and van der Waals interactions.13-19 If coordinatively unsaturated transition metals could be incorporated into the cavitands, a self-assembling metallocavitand capsule may be realized with potential to wed the selectivity and recognition of organic supramolecular chemistry with the catalytic, magnetic, and optical properties of metals. Reminiscent of Cram’s velcrands, heptazinc metallocavitands that reversibly self-assemble into interlocking dimers in aromatic solvents are described in this chapter.20 This process is entropy-driven and enthalpy opposed in all solvents in which dimerization is observed except p-xylene-d10 where dimerization is entropy and enthalpy favored.  Scheme 3.1. Intermolecular hydrogen bonding drives dimerization of a modified resorcin[4]arene yielding a capsule. 116  The previous chapter discussed the synthesis and structural characteristics of metallocavitands 52c and 52d. Dimerization of 52c in solution and 52d in the solid state is discussed in this chapter.  3.2  Discussion  3.2.1 Synthesis and Characterization In an effort to enhance the solubility range of heptazinc metallocavitands in nonpolar organic solvents, metallocavitand 52e was synthesized in 94% yield from octyloxy substituted macrocycle 51e as shown in Scheme 3.2. A single imine resonance at 8.31 ppm, two aromatic resonances found at 6.96 and 6.69 ppm, and one OCH2 methylene resonance at 4.06 ppm are observed in the 1H NMR spectrum in CDCl3 indicating 52e has three-fold rotational symmetry in solution. Much like metallocavitands 52c and 52d, the OCH2 methylene protons of 52e are diastereotopic, and a complex ABX2 coupling pattern is observed confirming the absence of a horizontal mirror plane. Although no solid-state structure was acquired for 52e,  13  C NMR spectroscopy, ESI-MS, FT-IR,  electronic absorption, and elemental analysis all support the assertion that 52e shares the same metallocavitand structure as 52a-d.  117  Scheme 3.2. Synthesis of heptazinc metallocavitand 52e from octyloxy substituted macrocycle 51e.  3.2.2 Solution Dimerization Unexpectedly, a preliminary 1H NMR spectrum of 52e in toluene-d8 displayed broad proton resonances, a result that encouraged further investigation into the selfassociation of 52e. Figure 3.1a displays the 1H NMR spectra of metallocavitand 52e in a variety of solvents at room temperature. Sharp resonances are observed in CD2Cl2 and CDCl3 (not depicted) whereas in aromatic solvents, broad resonances suggest that a dynamic process is occurring. In the last chapter I reported that the heptazinc cluster is stable at 25 °C in non-coordinating solvents so the observed resonance broadening may be attributed to monomer-dimer exchange and not a ligand rearrangement at the zinc cluster.21  118  Figure 3.1. 1H NMR spectra of 52e (5.00 mM) a) in selected solvents at room temperature and b) in p-xylene-d10 at the indicated temperatures. As the temperature of a solution of 52e in p-xylene-d10 is increased, the aromatic and imine peaks narrow and shift downfield while the resolution of the diastereotopic methylene protons adjacent to oxygen on the periphery improves (Figure 3.1b). Interestingly, there is a significant difference in the effects of temperature on the two diastereotopic protons; while both appear sharper at elevated temperature, J-coupling is only resolved for the downfield resonance. Dynamic behaviour is best analyzed when the process occurs at an intermediate exchange rate.22 This is defined as a rate at which resonance frequencies are similar to the exchange rate and both species involved in the dynamic process may be observed simultaneously by NMR spectroscopy. In an effort to slow the exchange rate of 52e below the coalescence temperature and reach an intermediate exchange regime, a 1H NMR spectrum was collected at -70 °C in toluene-d8; however, even at this temperature 119  the resonances remain coalesced. Rapid exchange, solvent freezing point, and metallocavitand solubility prevented observation of the coalescence temperature of 52e in C6D6, toluene-d8, and p-xylene-d10. To investigate the thermodynamic parameters of dimerization in a fast exchange regime, variable-temperature variable-concentration (VTVC) 1H NMR experiments were conducted in CD2Cl2, C6D6, toluene-d8, and pxylene-d10. A dimerization model fit well to the measured change in imine chemical shift over a range of VTVC 1H NMR spectra.23 The quality of fit enabled extraction of dimerization constants at a variety of temperatures. Thermodynamic parameters for selfassociation were determined from van’t Hoff plots. The experimental dimerization constants, enthalpies, and entropies of dimerization for compounds 52e and 52c are shown in Table 3.1. Table 3.1. Thermodynamic parameters for dimerization of 52c (R = C6H13) and 52e (R = C8H17) in different solvents. Kdim 25° C  ΔH°  ΔS°  (mol L-1)  (kJ mol-1)  (J mol-1 K-1)  toluene-d8  110 ± 10  19 ± 2  103 ± 8  CD2Cl2  <1  -  -  52e  C6D6  9±3  24 ± 12  100 ± 34  (R = C8H17)  toluene-d8  110 ± 10  10 ± 2  72 ± 5  p-xylene-d10  1300 ± 200  -2 ± 3  54 ± 9  Complex 52c (R = C6H13)  Solvent  Interesting trends are observed for the thermodynamic data. First, there is no dimerization in CD2Cl2 (or in CDCl3). Aromatic solvents, however, facilitate dimerization to varying degrees. The extent of dimerization is low in benzene, but increased by an order of magnitude with each additional methyl substituent (toluene and p-xylene). Dimerization in both toluene and p-xylene shows decreases in enthalpy and entropy from benzene of approximately 10 kJ mol-1 and 20 J K-1 mol-1 per methyl group,  120  respectively. Surprisingly, in aromatic solvents dimerization of these metallocavitands is primarily an entropy-driven process. Most self-association processes are enthalpy-driven and entropy-opposed since order is generally being imparted on the system and the dimers are held together by strong intermolecular interactions. Similar to Cram’s velcrands20 and Rebek’s molecular capsules,24 we attribute the entropy-driven process to the expulsion of solvent molecules from monomers upon dimerization as shown in Scheme 3.3. In aromatic solvents, the solvent forms strong π-π interactions with the bowl of the metallocavitand. In this semi-rigid solvent-monomer complex, the solvent molecule’s degrees of freedom are restricted. When dimerization of the metallocavitand occurs, solvent is expelled from the bowl, increasing the net entropy in the system. A reasonable explanation for the trend of decreasing ΔS° as solvent bulk is increased may be related to the steric requirements of the monomer-solvent and dimersolvent complexes. Bulkier solvents may be held tighter inside the capsule, decreasing the entropy gain upon dimerization. Also, in solvents composed of smaller molecules (e.g., C6D6) there may be more solvent molecules interacting with the monomeric cavity; and upon dimerization, liberation of more guest molecules results in a larger entropy change.  121  Scheme 3.3. Loss of solvent molecules from the cavity of a monomeric metallocavitand is responsible for the observed entropy-driven dimerization. The actual number of solvent molecules interacting with the cavity is unknown. The trends in ΔH° are also consistent with this explanation. Dimerization of the metallocavitand leads to disruption of the strong π-π interactions between the solvent and the cavity. With toluene, and particularly for p-xylene, there are fewer solvent-monomer interactions due to the steric bulk of the methyl substituents. With less average π surface area per guest solvent molecule, weaker interactions are expected. Dimerization was not observed in CDCl3 or CD2Cl2, solvents that are unable to participate in π-π interactions with the bowl-shaped metallocavitand. The absence of strong monomer-solvent interactions makes dimerization entropically unfavorable in these solvents. Although experimentally determined thermodynamic parameters strongly suggest that expulsion of solvent from the monomer’s bowl-like cavity promotes dimerization, I conducted VTVC 1H NMR spectroscopic experiments on 52c in toluene-d8 to gauge the impact of peripheral chain length on self-association. Entropy-driven, enthalpy-opposed dimerization was still observed but the magnitude of ΔH° and ΔS° differs slightly from those of 52e in toluene-d8. Enhanced alkyl chain interactions between capsule halves for 52e may account for the lower ΔH° and ΔS°.  122  3.2.3 Solid-State Dimerization Long peripheral alkyl chains present for solubility purposes hindered isolation of single crystals of 52c or 52e. Instead, crystals of neopentyloxy substituted 52d were grown from DMF and subjected to a SCXRD experiment (see Chapter two for refinement statistics). The solid-state structure of 52d is depicted in Figure 3.2 and shows capsules of the metallocavitand with a single DMF guest molecule inside. The cavity volume, without the guest DMF molecule, is roughly 150 Å3. Metallocavitand 52d crystallizes with two molecules in a face-to-face orientation in which one bowl is twisted 60° and shifted slightly off center forming a nearly perfect D3d symmetry capsule in the solid state. Between the two capsule halves there are 50 intermolecular atom-to-atom interactions less than the van der Waals distances (+0.2 Å). A single molecule of DMF, disordered about two positions, is located at the inversion center of the unit cell, which lies at the midpoint of the dimer. The zinc(II)-O(DMF) distance is 3.2 Å indicating the host-guest interaction is purely van der Waals and not Lewis acid-base. Although metallocavitand 52d packs into a dimer in the solid state, it does not exhibit solution dimerization, as investigated by 1H NMR in DMF-d7, likely due to rapid tumbling of the bulky neopentyloxy substituents.  123  Figure 3.2. Solid-state representations of dimer 52d2 encapsulating a DMF molecule. Cocrystallized DMF molecules are omitted for clarity. Only one orientation of the encapsulated DMF is modeled. a) Side-on view of the capsule (C = green, N = blue, O = red, H = white, Zn = yellow). b) Side-on view 90° from part a) with capsule halves colored red and blue. c) Top-down view of the dimer represented as space filling. d) Side-on view of the dimer represented as space filling.  3.3  Conclusions In summary, dimerization of metallocavitands 52c and 52e was observed and the  thermodynamics of dimerization were determined in solution. A solid-state structure of metallocavitand 52d provides a model for capsule 52c2 and 52e2 in solution. Unlike most cavitands, these metallocavitands dimerize in an entropy-driven process, promoted by loss of solvent from the bowl’s interior. Although entropy-driven aggregation is common 124  in aqueous media where it is important in biological systems (hydrophobic effect), it is rarely observed in non-polar organic solvents. Metallocavitand capsules with accessible metal sites on their interiors are alluring candidates for host-guest catalysis and molecular recognition.  3.4  Experimental  3.4.1 General Macrocycle 51e was prepared by the literature procedure.25 1H and  13  C NMR spectra  were recorded on a Bruker AV-300 spectrometer. The 13C NMR spectrum was recorded using a proton decoupled pulse sequence. 1H and 13C NMR spectra were calibrated to the residual protonated solvent at δ 7.27 and δ 77.23 ppm, respectively, in CDCl3. The UVvis spectrum was obtained in CH2Cl2 (ca. 5 x 10-5 M) on a Varian Cary 5000 UV-visnear-IR spectrophotometer using a 1 cm quartz cuvette. The FT-IR spectrum was obtained in a KBr disc with a Nicolet 4700 FT-IR spectrometer. The electrospray ionization (ESI) mass spectrum was obtained on a Bruker Esquire LC instrument at the UBC Microanalytical Services Laboratory. The sample was analyzed in MeOH:CHCl3 (1:1) at 100 μM. Elemental analysis (C,H,N) was performed at the UBC Microanalytical Services Laboratory. The melting point were obtained on a Fisher-John’s melting point apparatus. See Chapter two for SCXRD information.  125  3.4.2 Procedures and Data Synthesis of metallocavitand 52e (R = C8H17): Zinc acetate dihydrate (1.87 g, 8.5 mmol) was added to a solution of 51e (1.7 g, 1.2 mmol) in 150 mL of EtOH. The dark red solution turned bright orange upon heating to reflux (80 °C). After 2 h, the reaction mixture was cooled to room temperature and filtered through a frit. The crude solid was recrystallized from hot EtOH to obtain 52e. Yield: 2.5 g (1.1 mmol, 94%). Data for 52e (R = C8H17): 13C NMR (75.5 MHz, CDCl3) δ 180.1, 178.9, 161.9, 158.7, 150.2, 134.2, 120.8, 119.3, 102.0, 69.7, 31.8, 29.4, 29.3, 29.2, 26.0, 23.4, 22.7, 21.7, 14.1. 1  H NMR (300 MHz, CDCl3) δ 8.31 (s, 6H, imine), 6.96 (s, 6H, Ar), 6.69 (s, 6H, Ar), 4.06  (m, 12H, OCH2), 1.93 (s, 9H, O2CCH3), 1.85 (m, 12H, CH2), 1.84 (s, 9H, O2CCH3), 1.71 (s, H2O), 1.50 (m, 24H, CH2), 1.30 (m, 36H, CH2), 0.88 (t, 3JHH = 6.4 Hz, 18H, CH2CH3). UV-vis (CH2Cl2) λmax (ε) = 414 (1.2 x 105), 347 (6.2 x 104), 242 (6.6 x 104) nm (L mol-1 cm-1). ESI-MS: m/z = 2062 (52e-Zn2O(OAc)2 + Na)+, 2244 (52e -ZnO + H2O + H)+. FTIR (KBr) ῡ = 2959, 2927, 2856, 1617, 1506, 1459, 1447, 1391, 1324, 1264, 1216, 1185, 1105, 1022, 756, 669, 616 cm-1. Mp = dec. 270 ºC. Anal. Calc’d for 52e C102H138N6O25Zn7: C, 53.13; H, 6.03; N, 3.64. Found: C, 53.42; H, 6.30; N, 3.94.  3.4.3 Thermodynamic Studies Variable-Temperature Variable-Concentration 1H NMR Experiments. Due to low solubility of 52c and 52e in aromatic solvents, the concentration range was limited (0.1 to 5.0 mM). A standard 5.00 mM solution was prepared by first dissolving 0.005 mmol of metallocavitand (52c or 52e) with heat in 1.00 mL of the deuterated solvent of choice. Except for the 5.00 mM samples, NMR tubes were primed with 500 μL of selected deuterated solvent, and standard was then added via syringe to achieve the desired concentrations. The concentration and temperature dependence of the imine resonance was measured in CD2Cl2, benzene-d6, toluene-d8, and p-xylene-d10 for 52e and only in toluene-d8 for 52c. The data, displayed in Tables 3.2, 3.3, 3.4, and 3.5, was treated with the least-squares curve-fitting equation, 1 (monomer-dimer equilibrium model), as shown 126  in Figures 3.3, 3.5, 3.7, and 3.9, to find the association constants for dimerization.23 Figures 3.4, 3.6, 3.8, and 3.10 depict the van’t Hoff plots constructed to calculate the thermodynamic parameters of dimerization using equation 2.  (1)  (2)  G  Gm  § 1  8KdimCT  1 · ¸ G d  G m ¨1   ln Kdim  'H $ 1 'S  ˜  R T R  ¨ ©  4KdimCT  ¸ ¹  $  Chemical Shift of Imine (ppm)  8.20  8.15  357 K 339 K 321 K 284 K 274 K 265 K  8.10  8.05  8.00  7.95 0.000  0.002  0.004  0.006  Concentration (mol/L)  Figure 3.3. VTVC imine 1H NMR chemical shift dependence of 52c (R = OC6H13) in toluene-d8.  127  Table 3.2. Imine chemical shift (ppm) and association constants for 52c (R = OC6H13) in toluene-d8. Conc. (mM)  265 K  274 K  284 K  321 K  339 K  357 K  5.00  8.039  8.071  8.097  8.149  8.167  8.181  1.67  8.006  8.027  8.046  8.108  8.128  8.145  0.84  7.992  8.009  8.026  8.084  8.103  8.122  0.45  7.986  8.000  8.014  8.064  8.087  8.103  0.24  7.980  7.992  8.001  8.049  8.068  8.087  0.10  7.978  7.988  7.997  8.036  8.057  8.074  Kdim  52 ± 8  60 ± 6  85 ± 18  300 ± 9  321 ± 44  377 ±13  10  ln (Kdim)  8  6  4  2  0 0.0026  0.0028  0.0030  0.0032  0.0034  0.0036  0.0038  0.0040  -1  1/T (K )  Figure 3.4. van’t Hoff plot of 52c (R = OC6H13) in a toluene-d8.  128  8.06 311 K 321 K 330 K 339 K 348 K  Imine Chemical Shift  8.04  8.02  8.00  7.98  7.96 0.000  0.002  0.004  0.006  Concentration (mol/L)  Figure 3.5. VTVC imine 1H NMR chemical shift dependence of 52e (R = OC8H17) in benzene-d6. Table 3.3. Imine chemical shift (ppm) and association constants of 52e (R = OC8H17) in benzene-d6. Conc.  311 K  321 K  330 K  339 K  348 K  5.00  8.011  8.020  8.032  8.041  8.052  1.67  7.986  7.995  8.004  8.014  8.023  0.84  7.976  7.986  7.997  8.004  8.014  0.45  7.974  7.982  7.992  8.000  8.009  0.24  7.972  7.979  7.988  7.997  8.004  0.10  7.971  7.977  7.986  7.994  8.003  Kdim  10 ± 9  31 ± 3  27 ± 1  32 ± 6  33 ± 9  (mM)  129  10  ln (Kdim)  8  6  4  2  0 0.00285 0.00290 0.00295 0.00300 0.00305 0.00310 0.00315 0.00320 0.00325  1/T (K-1)  Figure 3.6. van’t Hoff plot of 52e (R = OC8H17) in benzene-d6.  8.18  Imine Chemical Shift (ppm)  8.16 8.14 363 K 347 K  8.12  331 K 314 K 298 K  8.10 8.08  276 K 260 K  8.06 8.04 8.02 8.00 0.000  0.002  0.004  0.006  Concentration (mol/L)  Figure 3.7. VTVC imine 1H NMR chemical shift dependence of 52e (R = OC8H17) in toluene-d8.  130  Table 3.4. Imine chemical shift (ppm) and association constants of 52e (R = OC8H17) in toluene-d8. Conc.  260 K  276 K  298 K  314 K  331 K  347 K  363 K  5.00  8.074  8.102  8.125  8.139  8.144  8.150  8.154  2.50  8.051  8.070  8.090  8.105  8.117  8.124  8.129  1.67  8.042  8.059  8.077  8.091  8.102  8.109  8.115  1.00  8.030  8.042  8.061  8.072  8.083  8.090  8.096  0.65  8.024  8.033  8.051  8.063  8.072  8.079  8.086  0.45  8.020  8.026  8.044  8.054  8.062  8.071  8.078  0.24  8.013  8.021  8.031  8.041  8.050  8.058  8.066  0.10  8.009  8.015  8.024  8.030  8.038  8.046  8.054  Kdim  80 ± 11  68 ± 10  (mM)  109 ± 22 146 ± 27 210 ± 15 212 ± 20 213 ± 23  10  ln (Kdim)  8  6  4  2  0 0.0026  0.0028  0.0030  0.0032  0.0034  0.0036  0.0038  0.0040  -1  1/T (K )  Figure 3.8. van’t Hoff plot of 52e (R = OC8H17) in toluene-d8.  131  8.38  Imine Chemical Shift (ppm)  8.36 8.34 8.32 8.30 8.28  380 K 369 K 352 K 336 K 320 K 303 K  8.26 8.24 8.22 8.20 8.18 0.000  0.002  0.004  0.006  Concentration (mol/L)  Figure 3.9. VTVC imine 1H NMR chemical shift dependence of 52e (R = OC8H17) in pxylene-d10. Table 3.5. Imine chemical shift (ppm) and association constants of 52e (R = OC8H17) in p-xylene-d10 (Kdim at 0.45 mM and 352 K was erroneously large). Conc.  303 K  320 K  336 K  352 K  369 K  380 K  5.00  8.311  8.327  8.335  8.344  8.349  8.353  2.50  8.294  8.309  8.320  8.327  8.333  8.336  1.67  8.286  8.301  8.309  8.317  8.324  8.326  1.00  8.263  8.275  8.291  8.298  8.304  8.307  0.65  8.254  8.269  8.286  8.291  8.297  8.300  0.45  8.245  8.260  8.272  -  8.285  8.289  0.24  8.227  8.240  8.253  8.259  8.265  8.269  0.10  8.201  8.215  8.227  8.234  8.243  8.248  1180 ±  1076 ±  1693 ±  1165 ±  1115 ±  175  192  213  276  216  (mM)  Kdim  984 ± 194  132  10  ln (Kdim)  8  6  4  2  0 0.0026  0.0027  0.0028  0.0029  0.0030  0.0031  0.0032  0.0033  0.0034  Figure 3.10. van’t Hoff plot of 52e (R = OC8H17) in p-xylene-d10.  133  3.5  References  (1)  MacGillivray, L. R.; Atwood, J. L. Nature 1997, 389, 469-472.  (2)  Atwood, J. L.; Barbour, L. J.; Jerga, A. Perspect. Supramol. Chem. 2003, 7, 153175.  (3)  Yebeutchou, R. M.; Tancini, F.; Demitri, N.; Geremia, S.; Mendichi, R.; Dalcanale, E. Angew. Chem., Int. Ed. 2008, 47, 4504-4508.  (4)  Ugono, O.; Moran, J. P.; Holman, K. T. Chem. Commun. 2008, 1404-1406.  (5)  Ajami, D.; Rebek, J., Jr. Angew. Chem., Int. Ed. 2007, 46, 9283-9286.  (6)  Dube, H.; Ajami, D.; Rebek, J., Jr. Angew. Chem., Int. Ed. 2010, 49, 3192-3195.  (7)  Rebek, J., Jr. Chem. Commun. 2007, 2777-2789.  (8)  Amaya, T.; Rebek, J., Jr. J. Am. Chem. Soc. 2004, 126, 14149-14156.  (9)  Barrett, E. S.; Dale, T. J.; Rebek, J., Jr. J. Am. Chem. Soc. 2007, 129, 8818-8824.  (10)  De Jong, M. R.; Engbersen, J. F. J.; Huskens, J.; Reinhoudt, D. N. Chem.--Eur. J. 2000, 6, 4034-4040.  (11)  Heinz, T.; Rudkevich, D. M.; Rebek, J., Jr. Nature 1998, 394, 764-766.  (12)  Craig, S. L.; Lin, S.; Chen, J.; Rebek, J., Jr. J. Am. Chem. Soc. 2002, 124, 87808781.  (13)  Gonzalez, J. J.; Prados, P.; De Mendoza, J. Angew. Chem., Int. Ed. 1999, 38, 525528.  (14)  Corbellini, F.; Knegtel, R. M. A.; Grootenhuis, P. D. J.; Crego-Calama, M.; Reinhoudt, D. N. Chem.--Eur. J. 2005, 11, 298-307.  (15)  Oshovsky, G. V.; Reinhoudt, D. N.; Verboom, W. Eur. J. Org. Chem. 2006, 2810-2816.  134  (16)  Oshovsky, G. V.; Reinhoudt, D. N.; Verboom, W. J. Am. Chem. Soc. 2006, 128, 5270-5278.  (17)  Corbellini, F.; van Leeuwen, F. W. B.; Beijleveld, H.; Kooijman, H.; Spek, A. L.; Verboom, W.; Crego-Calama, M.; Reinhoudt, D. N. New J. Chem. 2005, 29, 243248.  (18)  Liu, S.; Gibb, B. C. Chem. Commun. 2008, 3709-3716.  (19)  Gibb, C. L. D.; Gibb, B. C. J. Am. Chem. Soc. 2004, 126, 11408-11409.  (20)  Cram, D. J.; Choi, H. J.; Bryant, J. A.; Knobler, C. B. J. Am. Chem. Soc. 1992, 114, 7748-7765.  (21)  Frischmann, P. D.; Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2008, 47, 101-112.  (22)  Bain, A. D. Prog. Nucl. Magn. Reson. Spectrosc. 2003, 43, 63-103.  (23)  Horman, I.; Dreux, B. Helv. Chim. Acta 1984, 67, 754-764.  (24)  Kang, J.; Rebek, J., Jr. Nature 1996, 382, 239-241.  (25)  Gallant, A. J.; Hui, J. K. H.; Zahariev, F. E.; Wang, Y. A.; MacLachlan, M. J. J. Org. Chem. 2005, 70, 7936-7946.  135  Chapter 4 Cadmium  Cluster  Metallocavitands:  Highly  Dynamic Supramolecules† 4.1  Introduction  4.1.1 Abstract A family of molecular heptacadmium carboxylate clusters templated inside [3+3] Schiff base macrocycles has been isolated and studied by variable temperature solution and solid-state NMR spectroscopy, single-crystal X-ray diffraction (SCXRD), and density functional theory (DFT) calculations. These metallocavitand cluster complexes adopt bowl-shaped structures, induced by metal coordination, giving rise to interesting host-guest and supramolecular phenomena. Specifically, dimerization of these metallocavitands yields capsules with vacant coordination and hydrogen-bonding sites accessible to encapsulated guests. Strong host-guest interactions explain the exceptionally high packing coefficient (0.80) observed for encapsulated N,N-dimethylformamide (DMF). The guest-accessible hydrogen-bonding sites arise from an unusual μ3-OH ligand bridging three cadmium ions. Thermodynamic and kinetic studies show that dimerization is an entropy-driven process with a highly associative mechanism.  †Versions of this chapter have been published: a) Reproduced in part with permission from Frischmann, P. D.; Facey, G. A.; Ghi, P. Y.; Gallant, A. J.; Bryce, D. L.; Lelj, F.; MacLachlan, M. J. “Capsule Formation, Carboxylate Exchange, and DFT Exploration of Cadmium Cluster Metallocavitands: Highly Dynamic Supramolecules” J. Am. Chem. Soc. 2010, 132, 3893-3908. Copyright 2010 American Chemical Society. b) Frischmann, P. D.; MacLachlan, M. J. “Capsule Formation in Novel Cadmium Cluster Metallocavitands” Chem. Commun. 2007, 4480-4482. Reproduced by permission of the Royal Society of Chemistry http://dx.doi.org/10.1039/b710809e 136  In DMF the exchange rate of peripheral cluster supporting carboxylate ligands is intrinsically linked to the rate of dimerization and these two seemingly different events have a common rate-determining step. Investigation of guest dynamics with solid-state 2  H NMR spectroscopy revealed three-fold rotation of an encapsulated DMF molecule.  These studies provide a solid understanding of the host-guest and dynamic properties of a new family of metallocavitands and may help in designing new supramolecular catalysts and materials.  4.1.2 Background The ability to selectively confine metal clusters to desired size domains helped usher in the age of nanoscience, giving rise to materials with exceptional novelty.1,2 Besides numerous materials science applications,3-5 synthesis of molecular metal clusters is paramount in the design of multimetallic catalysts6 and enzyme mimics.7,8 Templatedirected or aggregation-based synthetic approaches are most often employed to access complex multimetallic clusters.9-20 For example, both approaches have been used to synthesize models of the Fe7Mo cofactor in nitrogenase enzymes, a highly sought after complex for elucidating the mechanism of nitrogen fixation.21 Metal clusters mimicking enzyme active sites as well as non-biological systems have proven to be powerful catalysts, allowing or accelerating a variety of organic transformations.22-24 Often synergy between metal ions imparts improved regioselectivity or rate enhancements compared to similar monometallic catalysts. To expedite the discovery of novel multimetallic catalysts and enzyme mimics, more reliable strategies for controlling geometry, composition, and nuclearity of metal clusters are needed. Developing template-assisted routes to hetero- and homo-polymetallic clusters is of interest to us. In particular, my research efforts have focused on the coordination chemistry of tris(salphen) Schiff base macrocycle 51.25-32 When 51 is reacted with an excess of zinc acetate, a heptazinc cluster supported by the macrocyclic scaffold is isolated.33 Experimental evidence shows macrocycle 51 acts as a template for the growth of heptazinc clusters and that isolable, reactive tetrazinc intermediates are good precursors for the formation of mixed-metal Zn4Co3 and Zn3Co4 clusters.34 Nabeshima  137  and coworkers demonstrated that the cluster-capping Zn4O tetrahedron, supported by six acetate ligands, may be released from the template upon addition of La3+ ions, suggesting these macrocyclic templates can steer polymetallic cluster formation then deliver the cluster upon addition of an external stimulus.35  Interestingly, the Zn4O cluster templated by macrocycle 51 closely resembles a tetrametallic Zn4(OCOCF3)6O catalyst discovered by Ohshima and coworkers that promotes the acylation of hydroxyl groups over amines with selectivity up to 99% in some cases.36,37 The same catalyst provides a facile route for the direct conversion of esters, lactones, and carboxylic acids to oxazolines in the presence of an amino alcohol. 38 Inspired by the catalytic utility of Zn4(OCOCF3)6O clusters and their structural similarity to heptazinc clusters templated by macrocycle 51, the versatility of macrocycle 51 as a template for novel molecular cadmium carboxylate clusters has been explored. Cadmium chalcogenide materials are well known for their semiconducting properties and molecular cadmium sulfide or selenide clusters have frequently been isolated.39-43 When carboxylate ligands are coordinated to Cd2+ ions, 2D and 3D metalorganic frameworks (MOFs) are obtained.44-53 In a few cases cadmium carboxylate MOFs are constructed of high nuclearity (>4 Cd2+ ions) secondary building units.54-61  138  However, to the best of my knowledge there is only one report of a high nuclearity molecular cadmium carboxylate cluster – a decacadmium wheel-shaped cluster that was only characterized in the solid state.48 Supported by phosphonates and not carboxylates, the molecular icosacadmium cluster reported by Roesky and coworkers is the largest Odonor supported molecular cadmium cluster reported to date.62 Reacting macrocycle 51 with cadmium acetate yields a heptacadmium cluster complex 61 (Scheme 4.1) that dimerizes, forming capsules both in solution and in the solid state.63 Since metal coordination induces the otherwise planar macrocycle to adopt a bowl-like conformation, we have dubbed it a metallocavitand.64,65 Crystallography, 1H NMR spectroscopy, and DFT calculations confirm an unusual μ3-hydroxo ligand is located deep inside each cavity and it is able to form hydrogen bonds with encapsulated guests. This chapter describes the supramolecular self-assembly, cadmium carboxylate metal  cluster  dynamics,  and  solid-state  guest  dynamics  of  heptacadmium  metallocavitands explored by DFT, X-ray crystallography, variable temperature NMR spectroscopy, and variable temperature solid-state NMR spectroscopy.  Scheme 4.1. Synthesis of heptacadmium metallocavitands 61a-e.  139  4.2  Discussion  4.2.1 Synthesis and Characterization Reacting seven equivalents of Cd(OAc)2•2H2O with a suspension of macrocycle 51a-e in MeOH or EtOH followed by filtration yields metallocavitands 61a-e in 57-75% yield with no further purification necessary. Metallocavitands 61a and 61d have been studied in the solid state by single-crystal X-ray diffraction (SCXRD) and experimental evidence strongly supports that metallocavitands 61a-e share the core structure depicted in Figure 4.1.  Figure 4.1. Depiction of heptacadmium complex 61 with the two distinct acetate environments labeled A (red) or B (blue). a) Side-on representation. Cd1, the third dialkoxyphenylenediimine unit, and one type- A acetate are obscured by the cluster. b) Top-down representation of 61 highlighting the acetate-Cd connectivity. NMR spectra of metallocavitands 61a and 61d were recorded in DMF-d7 whereas for the more soluble 61c and 61e, spectra were recorded in CDCl3. Like their parent macrocycles, 61a-e exhibit C3 rotational symmetry in solution resulting in greatly simplified NMR spectra. The most diagnostic resonances in the 1H NMR spectrum of 61c (Figure 4.2) are the imine and OCH2 methylene resonances found around 8.26 and 4.05  140  ppm, respectively. Upon metallation, three prominent changes from macrocycle 51c are noteworthy in the proton NMR spectrum: 1) downfield OH resonances are absent; 2) three bond  111/113  Cd-H J-coupling is observed for the imine resonance; and 3) OCH2  methylene resonances show a complex ABX2 coupling pattern. The first two changes confirm metallation of the N2O2 salphen pockets while the diastereotopic nature of the OCH2 methylene resonances indicates the horizontal mirror plane present in the parent macrocycle is lost, effectively switching the symmetry from D3h in the macrocycle to C3v in the cadmium product. Identical spectral changes are observed for metallocavitands 61a-e with one exception: the OCH2 methylene resonance of 61d is a broad singlet rather than two doublets due to a dynamic process discussed later in this chapter.  a)  b)  14  12  10  8  6  4  2  ppm 0  Figure 4.2. 1H NMR spectra before and after heptacadmium metallocavitand formation. a) Macrocycle 51c with the triplet OCH2 resonance inset. b) Metallocavitand 61c showing the imine resonance with 34.1 Hz 3JCd-H satellites and the ABX2 coupling observed for the OCH2 resonance inset (400 MHz, CDCl3; residual CHCl3 is at 7.27 ppm).  141  To complement 1H NMR spectroscopic investigations of these metallocavitands, metallocavitand 61c was further probed by cadmium NMR spectroscopy. Both 111Cd and 113  Cd isotopes have nearly identical NMR spectroscopic characteristics so  113  Cd was  arbitrarily chosen for this study and the spectrum is depicted in Figure 4.3. Three separate 113  Cd resonances were located at 139.5, 10.1, and 1.1 ppm in the 1H-coupled 113Cd NMR  spectrum of 61c, and integration of the resonances gave a 3:1:3 ratio, respectively. The resonance at 139.5 ppm is assigned to cadmium ions 1-3 because of the 34.1 Hz 3JCd-H coupling constant, in agreement with the coupling observed for the imine resonance in the 1H NMR spectrum. Since the spectrum was collected with a reasonably long delay between pulses (30 s) an accurate integration is obtained and the resonances at 10.1 and 1.1 ppm are assigned to Cd7 and Cd4-6, respectively.  Figure 4.3. 113Cd NMR spectrum of 61c in CDCl3 (88.7 MHz). Mass spectrometry of metallocavitands 61a-e was problematic with the best results being obtained by matrix-assisted laser-desorption ionization time of flight mass spectrometry (MALDI-TOF MS). From separate experiments, diagnostic peaks are found at m/z = 2383.5, 2130.9, and 2465.7 corresponding to [61d-(H2O)2OH]+, [61c(H2O)2(OH)2Cd2(OAc)2+Na]+, and [61c-(H2O)2OH]+, respectively. Peaks are found near  142  the expected position for the molecular ion in the spectra of 61a and 61e; however, none could be assigned. Both parent macrocycles 51a-e and metallocavitands 61a-e are isolated as orange to deep red powders that yield deep red solutions when dissolved in a variety of solvents. The electronic absorption spectra of 61a-e are very similar with two absorption maxima found around 350 and 412 nm in CH2Cl2. These absorptions are likely attributed to π-π* transitions since they are shared with their parent macrocycles and have extinction coefficients with magnitudes around 104 to 105 L mol-1 cm-1. In agreement with this hypothesis, the difference among the D95V(d,p)/SDD/PBE1PBE DFT computed energies of the three π highest occupied orbitals and the two lowest π* unoccupied orbitals, LUMO and LUMO+1, of metallocavitand 61f give a group of wavelengths in the range 403-407 nm. To reduce computational time, fictitious metallocavitand 61f, with peripheral hydroxyl groups, is used for DFT calculations throughout this chapter.  The most notable resonance in the IR spectra of 61a-e is the imine stretching mode found between 1606 and 1617 cm-1. Values within this range are common for metallated salphens.  143  4.2.2 SCXRD Analysis Solid-state structures of metallocavitands 61a and 61d are very similar with a few striking differences. Two separate crystals of 61a have been studied by SCXRD, the first was grown from hot DMF and the second from a mixture of DMF, CH2Cl2, CHCl3, and MeOH. Nearly identical results were obtained for the two SXCRD experiments. The structure of 61a, shown in Figure 4.4, has a high degree of symmetry, crystallizing in the rhombohedral space group R-3m. Each unit cell contains six metallocavitands organized into interlocking dimers with electron density corresponding to two DMF molecules located inside each capsule (PLATON/SQUEEZE results show ~80 electrons are located inside the capsule of 61a·61a).66 Disorder prevented the orientation of encapsulated DMF molecules from being determined by crystallography. I initially believed only one DMF could fit inside each capsule based on packing coefficients (0.40 for one vs. 0.80 for two); however, the crystallographic results strongly support the presence of two DMF molecules per capsule in the solid state (the packing coefficient is the volume ratio of guest compared to host cavity).67 Each face-to-face capsule is formed by rotating one 3fold symmetric metallocavitand by 60° resulting in interdigitation of the peripheral phenylenediimine units. To better describe an individual metallocavitand, the labeling scheme from Figure 4.1 is used in the following discussion.  144  a)  b)  c)  Figure 4.4. Capsular assembly adopted by metallocavitand 61a (R = CH2CH3) in the solid state. a) Top-down view of the dimer. b) Side-on view of the dimer. c) Hexagonal long-range packing (view down the c-axis of the unit cell). (C = green, N = blue, O = red, H = white, Cd = cyan). Figure  4.5  shows  the  connectivity of  the  neutral,  acetate-supported,  heptacadmium cluster in 61a. Cadmium ions 1-3 occupy the salphen-like N2O2 pockets and their geometry may be described as distorted trigonal prismatic or distorted pentagonal pyramidal geometry. The fifth and sixth coordination sites of cadmium ions 1-3 are occupied by one type-A acetate in a bidentate fashion where the meridional and axial Cd-OAc bonds are 2.43 and 2.28 Å, respectively. Above the macrocycle rest cadmium ions 4-6, each supported by a catechol unit, two type-A acetates, one type-B acetate, and bridged by a central μ3-hydroxo ligand to give a distorted pentagonal bipyramidal coordination environment. The Cd-O bond lengths for catechol Cd-O, axial Cd-O type-B acetate, and Cd-O (μ3-OH) range from 2.22 to 2.28 Å whereas the meridional Cd-O type-B acetate and both Cd-O type-A acetate bonds are longer at 2.47 and 2.53 Å, respectively. Cluster-capping cadmium ion 7 exhibits octahedral geometry imparted by one hydroxo and two aqua ligands facially coordinated and three Cd-O type145  B acetate bonds. The +14 charge of seven Cd2+ ions is balanced by the macrocyclic phenoxides (-6), six acetates (-6), one central μ3-hydroxo ligand (-1), and one capping hydroxo ligand (-1).  Figure 4.5. Solid-state structure of 61a focusing on the Cd7(μ3-OH)(OAc)6(H2O)2OH cluster with the peripheral OEt substituents omitted. The acetate colors and cadmium labels are from Figure 4.1. a) Side-on view. b) Top-down view. (C, O, and N of the macrocyclic scaffold = green, Cd = cyan, O aqua/hydroxo ligands = brown, central μ3OH ligand = purple, type-A acetate = red, type-B acetate = blue). As most of the hydrogen atoms were not located in the SCXRD structure of 61a, the μ3-hydroxo ligand was originally reported to be an oxo ligand and the three ligands capping Cd7 as aqua ligands.63 The μ3-oxo assignment was reconsidered after crystals of a crude sample of neopentyloxy substituted 61d were subjected to a SCXRD experiment. Grown in a mixture of DMF, CH2Cl2, CHCl3, and by-product AcOH, the crystal of 61d is triclinic with each unit cell consisting of four metallocavitands. Surprisingly, two different metallocavitand species are co-crystallized, each in duplicate, within the unit cell. One species is analogous to the structure of 61a; however, in the second species the capping hydroxo and two aqua ligands are replaced by a DMF (O-bound) and a bidentate acetate ligand as shown in Figure 4.6. Although the solid-state structure of 61d is relatively low resolution, the connectivity is certain. No cations are located within the unit cell to account for the negative charge of the capping acetate. A neutral complex is  146  obtained only if the previously assigned, central μ3-oxo ligand is in fact a μ3-hydroxo ligand.  a)  b)  c)  Figure 4.6. Solid-state structure of metallocavitand 61d (R = CH2C(CH3)3) with the cluster-capping aqua and hydroxo ligands replaced by an acetate ligand and DMF. The homo-dimer that is isostructural to 61a is also present in the unit cell, but is omitted from this figure. a) Capsular assembly with two guest DMF molecules space filling. b) Close up view of the cluster with the capping hydroxo and two aqua ligands replaced by a DMF (O-bound) and a bidentate acetate ligand. c) Close-up view of the encapsulated DMF molecules with the DMF carbonyl O-Cd bond and a potential H-bond to the μ3-OH ligand highlighted. The neopentyloxy chains have been omitted for clarity (C = green, N = blue, O = red, H = grey, Cd = cyan). The two co-crystallized species of metallocavitand 61d exist as homo-dimers, and each dimer encapsulates two DMF molecules related by an inversion center (one encapsulated DMF molecule is removed after heating under vacuum for 48 h). Inside each capsule the carbonyl oxygen of DMF is weakly coordinating to cadmium (2.5-2.7 Å Cd-O) and potentially hydrogen-bonding with the μ3-OH ligand (3.1-3.2 Å O-O). Eight Lewis-acidic sites, including six unsaturated cadmium ions and two μ3-OH ligands, are accessible to encapsulated guests. A 3.1-3.3 Å distance separates the two DMF molecules, bringing them into van der Waal's contact. In this Lewis-acidic and sterically demanding environment these metallocavitands may act as nano-sized reaction vessels if proper guests are encapsulated.68-71  147  While μ3-OH ligands are common in cubane clusters of the type M4(OH)4 where vertices of the cube alternate between metals and μ3-OH ligands,72-76 this coordination mode is rare for Cd2+ ions. From a search of the Cambridge Structural Database (CSD), thirty four structures have been reported that possess a tetrahedral μ3-oxygen ligand bridging three cadmium ions (CSD search conducted on May 7, 2009 with ConQuest V. 1.11); eighteen of those are μ3-OR,77-89 thirteen are μ3-OH,90-95 and three are μ3oxo.63,96,97 The μ3-O-Cd bond lengths in 61a and 61d, 2.22 Å and 2.19-2.27 Å, respectively, are essentially in the 2.20-2.41 Å range of both the hydroxo and oxo bridged structures reported in the literature. As the disorder in the crystal structures of 61a and 61d prevented accurate assignment of the proton positions from the electron density maps, DFT calculations were performed to confirm the location of any oxo, hydroxo, and/or aqua ligands in metallocavitands 61a-e.  4.2.3 DFT Computations Two possible cluster configurations were explored computationally, one with a μ3-oxo and the other with a μ3-hydroxo ligand; these prototropic tautomers are illustrated in Figure 4.7. DFT energy minimization at the 3-21g/SDD/PBE level of theory, without any constraints and assuming a C1 point group symmetry, of complex 61f (R = H) as depicted in Figure 4.7a (i.e., characterized by the central μ3-oxo ligand bonded to Cd1-3 atoms and three capping aqua ligands) resulted in dramatic geometrical reorganization of the cluster. In particular, two of the three water molecules dissociate and the μ3-oxo ligand moves toward Cd7 while at the same time bridging the Cd4-6 atoms. This rearrangement leads to a tetrahedral μ4-oxo coordinated to the Cd4-7 atoms akin to the zinc clusters, indicating the starting μ3-oxo configuration is unstable. This behavior is independent of the basis set and xc functional used. Calculations verified that the central μ3-oxo ligand must be a μ3-hydroxo ligand and that the capping ligands are two aqua and one hydroxo ligand, suggesting that the structure should be that reported in Figure 4.7b. Computations were further performed using the D95V(d,p) basis set and hybrid PB1PBE or B98 xc functionals. To account for relativistic contributions explicitly, and not by means of atom pseudopotentials, further DFT computations were performed at Zero  148  Order Regular Approximation (ZORA) using a TZV Slater basis set and pure PBE xc functional. The computed Hessian matrix at different levels of theory did not show any negative eigenvalues qualifying the C1 point group symmetry structure as a true minimum. Attempts to impose a Cs point group symmetry to the structure of Figure 4.7b with the aim to reduce the computational effort led to a Hessian matrix with negative eigenvalues associated with the rotation of methyl groups on A and B acetate ligands. The optimized structure at ZORA TZV/PBE level of theory depicted in Figure 4.8 is in close agreement with the solid-state structures of 61a and 61d and with the other DFT computations as well.  a)  H2O  OH2  OH2  O  μ3-O(H2O)3  b)  H2O  OH2  OH  O H  μ3-OH(H2O)2OH  Figure 4.7. Prototropic tautomers of metallocavitand 61. a) The μ3-oxo complex with three capping aqua ligands is unstable. b) The μ3-hydroxo species with two capping aqua ligands and a capping hydroxo ligand is the species observed.  149  a)  b)  Figure 4.8. Overlay of geometry-optimized (at ZORA/TZP level) metallocavitand 61f (red), solid-state structure of 61a (green), and solid-state structure of 61d (blue). a) Sideon view. b) Top-down view. Under all computational conditions explored, the most stable metallocavitand has a μ3-hydroxo ligand bridging cadmium ions 4-6 (Figure 4.1) with an O-H bond pointing directly into the center of the cavity as illustrated in Figure 4.9. This conformation leaves a W-shaped hydrogen-bonding network between two aqua and one hydroxo ligand coordinated to the cluster capping Cd7 characterized by a basis set superimposition error (BSSE) corrected interaction energy of the two water molecules with the cluster amounting to -118.5 kJ mol-1. A water molecule inside the cavity hydrogen bonding to the μ3-oxo ligand does not prevent the rearrangement of the cluster to a μ4-oxo species. This indicates that the μ3 oxygen atom in the field of the four 4-7 Cd atoms has a strong tendency to assume a  150  tetrahedral coordination with “sp3” hybridization and that only a strong interaction with a proton can overcome the alternative interaction with the capping Cd7 ion. The favored μ3-OH ligand configuration provides a hydrogen-bonding site inside the already Lewis-acidic cavity of metallocavitands 61a-e, similar to the tantalum boronate metallocavitands reported by Fontaine.98,99 The DFT computations show that inside the cavity the μ3-hydroxo ligand may hydrogen bond with a water molecule or the carbonyl oxygen of an encapsulated DMF. At 3-21g/SDD/PBE and D95V(d,p)/SDD/B98 level of theories the OH···O distance between the hydrogen atom of the P3-OH and the carbonyl oxygen of the DMF molecule is 1.78 and 1.81 Å, respectively. Taking into account the BSSE correction by the Boys and Bernardi counterpoise (CP) approach, the interaction energy of the DMF molecule hydrogen bonded with 61f by means of the P3OH at the two levels of theory is -42.05 and -30.59 kJ mol-1, respectively.  a)  b)  Figure 4.9. Space-filling representations of model metallocavitand 61f from ZORA/TZP DFT optimization. a) Top-down view highlighting the W-shaped hydrogen-bonding network of one hydroxo and two aqua ligands facially coordinated to Cd7 (Figure 4.1). b) Bottom-up view of the cavity with a central μ3-OH ligand bridging Cd4-6. (C = green, H = white, N = blue, O = red, Cd = cyan). A space-filling model suggests that up to three DMF molecules fit inside the cavity of monomeric metallocavitand 61f. Energy minimization was performed for  151  metallocavitand 61f containing one, two, and three DMF molecules in its cavity to determine the optimal occupancy. Four possible DMF-cavity interactions were computed: 1) one DMF molecule H-bonded to the μ3-OH ligand; 2) one DMF molecule coordinated to Cd; 3) two DMF molecules each coordinated to Cd; and 4) three DMF molecules all coordinated to Cd atoms. The (DMF)2  61f energy minimization revealed that the  situation where each DMF guest molecule is coordinated to cadmium is energetically favored over the case where one DMF is coordinated and one DMF is H-bonded to the μ3-OH ligand. In the tri-DMF case there is only sufficient space for each DMF if all are Cd-coordinated and none is H-bonded. The deformation energy100,101 of the metallocavitand and the DMF-cavity corrected BSSE interaction energy were both computed in the gas phase at D95V(d,p)/SDD/B98 level of theory and gave net interaction energies for scenarios 1-4 of -30.59, -24.10, -24.69, and -30.54 kJ mol-1, respectively. From these values it is safe to conclude the cavity of a monomeric metallocavitand in DMF is occupied by one DMF H-bonding to the μ3-OH ligand and a few DMF molecules solvating the metallocavitand-DMF complex (considering the solvation of free DMF, scenario 1 is stabilized relative to scenario 4). The tri-DMF occupied cavity is next lowest in energy followed by di- and mono- coordinated DMF configurations. Although these calculations do not include the solvation energy, they nicely complement the experimentally determined thermodynamic and kinetic parameters of dimerization and ligand exchange discussed later in the chapter. Figure 4.10 shows the relative energies of the computed structure of 61f with one, two, or three DMF molecules in the bowl.  152  Figure 4.10.  Relative energies of calculated structures for scenario 1) one DMF  molecule H-bonded to the μ3-OH ligand; 2) one DMF molecule coordinated to Cd; 3) two DMF molecules both coordinated to Cd ions; and 4) three DMF molecules all coordinated to Cd ions.  4.2.4 Metallocavitand Capsules in Solution: Thermodynamics Significant resonance broadening, characteristic of a dynamic process on the NMR timescale, is observed in the 1H NMR spectrum of metallocavitand 61a in DMF-d7 at room temperature. Coalescence of the resonances assigned to the imine CH, the OCH2, the OCH2CH3, and both aromatic CH protons occurs upon warming from -26 to 47 °C (Figure 4.11). Integrals of the separated resonances exhibit concentration dependence allowing for their assignment to either monomer or dimer. Dimerization constants were calculated (270 ± 10 L mol-1 at 25 °C) and a van’t Hoff plot was constructed that gave ΔH° = 19 ± 1 kJ mol-1 and ΔS° = 110 ± 2 J mol-1 K-1. This entropy-driven and enthalpyopposed process is attributed to expulsion of solvent from the monomeric cavity upon dimerization as shown in Scheme 4.2.102,103  153  a)  b) 57 °C 47 27 11 1 -26  Hb’  Hb *  Hb  *  Ha’  Hc’  Ha Hc  c) 52 °C 25 14 0 -35  9  Ha  8  Hc  7  6  ppm5  Figure 4.11. Variable-temperature 1H NMR spectra in DMF-d7 (400 MHz). a) Protons Ha-c are assigned to the imine and aromatic resonances of metallocavitand 61, primes (’) belong to the dimer. b) Dimerization of 61a results in coalescence of each resonance as the temperature is increased from -26 to 57 °C. c) No dimerization is observed for 61d at low temperature. The DMF resonance, calibrated to 8.03 ppm, is indicated with an *.  154  . Scheme 4.2. Model for entropy-driven dimerization of metallocavitands, rationalized by solvent expulsion in the monomer-dimer equilibrium (S = generic solvent molecules). Multiple solvent molecules interact with the cavity through H-bonds, Cd-coordination, and/or van der Waal’s interactions (represented by dashed lines). Expansion of the dimerization study to include benzene-d6, toluene-d8, and pxylene-d10 was made possible due to the enhanced solubility of octyloxy substituted metallocavitand 61e in non-polar solvents. In these solvents the rate of monomer-dimer exchange is significantly faster than for 61a in DMF-d7 and resonances assigned to monomer and dimer cannot be discerned by 1H NMR spectroscopy even after chilling 61e in toluene-d8 to -25 °C. Monitoring the imine chemical shift over a series of variabletemperature variable-concentration (VTVC) 1H NMR spectroscopic experiments enabled the calculation of association constants and thermodynamic parameters of dimerization for 61e in these solvents.104 Thermodynamic data are summarized in Table 4.1; in all cases, dimerization is an entropy-driven process.  155  Table 4.1. Thermodynamics of dimerization for metallocavitands 61a and 61e in various deuterated solvents. The packing coefficient is also included as a percentage. 61a Solvent  61e  DMF  benzene  toluene  p-xylene  Kdim (L mol-1) a  270 ± 10  800 ± 100  1500 ± 400  1000 ± 300  ΔH° (kJ mol-1)  19 ± 1  -7 ± 5  1±4  11 ± 8  ΔS° (J mol-1 K-1)  110 ± 2  32 ± 14  64 ± 14  94 ± 24  Vguest / Vhost %  40/80 b  46  54  63  a  Calculated at 25 °C from van't Hoff plots.  b  Calculated for single and double occupancy. Since metallocavitand dimerization involves solvent encapsulation, I first turned  to the “55% rule” for molecular recognition, developed by Mecozzi and Rebek, to rationalize the thermodynamic parameters.67 The empirical model predicts that optimal guest recognition occurs when the guest occupies 55 ± 9% of the host’s void space. In this system, each metallocavitand capsule has a void space of 215 Å3, and packing coefficients for DMF, benzene, toluene, and p-xylene are 0.40 (0.80 for 2 DMFs), 0.46, 0.54, and 0.63, respectively (capsule void space was calculated with PLATON and solvent volumes were calculated with SPARTAN 04).66,105 Dimerization constants for 61 follow the 55% rule quite well, with the highest being in toluene followed by p-xylene, benzene, and finally DMF. In this entropy-driven system, the 55% rule is not expected to exactly apply since solvent release from the monomer is probably the most important criterion for dimerization – the more restricted that solvent is in the cavity of the monomer, the greater the entropy increase when it is released during dimerization. Aromatic solvents (e.g., benzene, toluene, xylenes) or Lewis bases (DMF) that can participate in strong host-guest interactions with the cavity of monomeric 61 facilitate dimerization, while other solvents that only interact weakly with the monomer (e.g., CH2Cl2 and CHCl3) do not lead to dimerization despite packing coefficients similar to that for DMF. Dimerization is an enthalpy-opposed process in all solvents except benzene-d6, suggesting the face-to-face van der Waal’s interactions between metallocavitands in the  156  dimer are negligible relative to solvent-cavity interactions. Since the cavity of each monomeric metallocavitand is solvated by multiple guests, most of these cavity-guest interactions must be eradicated upon dimerization to accommodate the volume restraints imposed by capsule formation. The large positive enthalpy contribution for DMF results from cleavage of μ3-OH----ODMF hydrogen bonds upon dimerization; however, the enthalpy differences between aromatic solvents are harder to rationalize. Less favorable packing coefficients and perhaps more favorable cavity-methyl group interactions may explain the enthalpy increase from benzene to p-xylene. In benzene, capsule assembly is spontaneous regardless of temperature suggesting the solvent-cavity interactions in the monomeric state are quite weak or the smaller guest allows for closer/stronger contacts between metallocavitand in the dimer. Regardless of solvent type, entropy-driven dimerization is observed due to expulsion of solvent from the monomer’s cavity upon capsule formation. Dimerization of 61a in DMF shows the highest entropy contribution followed by 61e in p-xylene, toluene, and finally benzene. This trend may be attributed to stronger cavity-DMF interactions binding DMF tighter than cavity-aromatic solvent interactions. Steric effects may be responsible for the reduced entropic contribution from p-xylene to benzene; bulkier solvent has less “wiggle” room inside the monomeric cavity therefore gains more entropy upon expulsion from the cavity. It seemed likely that dimerization of metallocavitand 61d in DMF-d7 would proceed similarly to 61a, however, there was no evidence for dimerization in the low temperature 1H NMR spectrum of 61d (Figure 4.11c). Aromatic and imine resonances of 61d exhibit a slight sharpening as the temperature decreases and their chemical shift is static below 14 °C. Despite packing into a dimer in the solid state, rapid spinning/tumbling of the bulky neopentyloxy chains prohibits dimerization in solution.  157  4.2.5 Capsule Occupancy and the μ3-OH Proton NMR Resonance An incongruity arises between my solution-state 55% rule analysis and the solidstate structures of 61a and 61d – in the solid state, two DMF molecules are found in each capsule. The resulting exceptionally high packing coefficient, 0.80, may be explained by metal coordination and weak hydrogen-bond synergy between host and guests. Complementary interactions between host and guest result in enhanced packing coefficients, effectively “increasing” the cavity volume.106 Low temperature 1H NMR spectroscopy in DMF-d7 confirmed doubly occupied capsules, (DMF)2  61a·61a, are  prominent in solution; however, a small fraction of capsules are singly occupied, DMF 61a·61a. The capsule (DMF)2 guest DMF, giving DMF  61a·61a may be sufficiently dynamic to expel one  61a·61a + DMF, without fully dissociating as shown in  Scheme 4.3.  158  Scheme 4.3. Equilibria between DMF  61a, (DMF)2  61a·61a, and DMF  61a·61a.  Encapsulated DMF is not able to invert making the capsule halves unsymmetrical in DMF  61a·61a (red and blue). Perturbation of a host’s symmetry due to guest orientation is frequently observed  by 1H NMR spectroscopy in host-guest systems.107-110 Thorough examination of the 1H NMR spectrum of 61a obtained at -24 °C in DMF-d7 reveals three different sets of resonances that are highlighted in Figure 4.12. Resonances found at 8.6, 7.4, and 6.6 ppm confirm monomer 61a has C3 rotational symmetry and suggest that guest DMF molecules  159  are labile or dynamic on the NMR timescale. Two explanations arise for the D3d symmetry observed for the dominant 61a dimer resonances found at 8.7, 7.6, and 5.8 ppm: 1) one DMF molecule is encapsulated that may rotate about the C3 axis of the capsule and invert freely around an axis perpendicular to the C3 one; or 2) two DMF molecules are encapsulated and rotate simultaneously about the axis of the capsule without inversion. In both cases, the capsule halves remain equivalent and C3 rotational symmetry is maintained. The latter explanation is supported by X-ray crystallography that revealed two DMF molecules inside each capsule and steric constraints inside the capsule suggest inversion is significantly hindered.  *  9.0  8.5  8.0  7.5 7.0 Chemical Shift (ppm)  6.5  6.0  5.5  5.0  Figure 4.12. Downfield region in the 1H NMR spectrum of metallocavitand 61a at -24 °C (400 MHz, DMF-d7). Three sets of resonances belonging to DMF 61a·61a, and DMF  61a, (DMF)2  61a·61a are observed and labeled with a half capsule (red), whole  capsule (red), and unsymmetrical whole capsule (red and blue), respectively. Resonances of the unsymmetrical capsule halves in DMF the (DMF)2  61a·61a are perfectly distributed around  61a·61a resonances. The DMF resonance, calibrated to 8.03 ppm, is  indicated with an *. Simultaneous rotation of both encapsulated DMF molecules should require some expansion of the capsule and/or breathing of the cavity walls. In fact the very low frequency vibration modes Q1 = 18 cm-1,Q2 = 20 cm-1 and Q4 = 25 cm-1 calculated at D95V(d,p)/SDD/B98 level of theory are related to the breathing motion of the macrocyclic ligand suggesting that the energy requirement for this type of distortion is very small. It is likely that the conformational flexibility that allows for rotation may also 160  lead to the escape of one encapsulated DMF yielding the unsymmetrical capsule DMF 61a·61a.111-113 Resonances of DMF (DMF)2  61a·61a are evenly distributed around resonances of  61a·61a and integration of the 1H NMR spectrum obtained at -24 °C gives a  global equilibrium constant between (DMF)2  61a·61a and DMF  61a·61a of Kexchange  = 0.32 ± 0.06. This equilibrium constant also accounts for dimerization of monomers yielding the unsymmetrical capsule DMF  61a·61a (the low intensity of these  resonances prevented a full thermodynamic analysis). Despite dimerization being entropy-driven, the entropy gained by expelling all but one DMF molecule is compensated by the enthalpy loss associated with cleaving the hydrogen and coordination bonds between metallocavitand and DMF, making the capsule (DMF)2 thermodynamically favored over DMF  61a·61a  61a·61a.  Besides observation of unsymmetrical capsules in low temperature 1H NMR spectra, the suppression of dimerization and guest exchange kinetics enables resolution of the proton resonance assigned to the μ3-OH ligand bridging cadmium ions 4-6 (Figure 4.1). At 15 °C a low-intensity, broad resonance emerges from the baseline at 2.35 ppm and sharpens considerably upon chilling the sample to -24 °C. Initially it appeared to be an impurity; however, the resonance, depicted in Figure 4.13, has an exceptionally diagnostic splitting pattern that was successfully simulated as two-bond coupling of the μ3-OH proton to three chemically equivalent cadmium ions. Both  111  Cd and  113  Cd are  roughly 12% abundant, have a spin of 1/2, and couple to μ3-OH proton with a J = 13 Hz. The simulation was performed with all ten possible permutations, a J value of 13 Hz, and 4 Hz of line broadening (See Experimental section 4.4.7 on page 192 for simulated permutations).  161  a)  Cd5 Cd 4  Cd6 O H  b)  2.40  2.35 Chemical Shift (ppm)  2.30  Figure 4.13. a) 1H NMR spectrum for the μ3-OH proton resonance belonging to monomeric metallocavitand 61a (DMF-d7, -24 °C, 400 MHz). Inset depicts the environment that results in the observed spin system. Cadmium labels are from Figure 4.1. b) Simulation of the spin system with 2JH-Cd = 13 Hz and 4 Hz of line broadening. Integration of the μ3-OH proton resonance at 2.35 ppm shows this particular resonance belongs to the monomer and another resonance found at 2.51 ppm with the same splitting pattern belongs to the μ3-OH proton resonance of the symmetric capsule, (DMF)2  61a·61a (integral ratio 1.0/0.7 respectively). Both μ3-OH proton resonances  are lost upon addition of D2O, confirming a proton exchange pathway is active. Depressed guest exchange, dimerization, and/or H-bonding rates may all be responsible for the sharpening of these resonances at low temperature. This unequivocal piece of evidence eliminates any remaining doubt about the proton assignment. I believe this is the first report of 2J (111/113Cd/1H) J-coupling in a cadmium-hydroxo complex. In most  162  other complexes, the NMR spectra are obtained in water or in other solvents where the proton can rapidly exchange. The sheltered environment of the metallocavitands prevents rapid exchange and offers a unique opportunity to observe this coupling.  4.2.6 Metallocavitand Capsules in Solution: Kinetics Although metallocavitand dimerization is a complex event, involving multiple guest-cavity, cavity-cavity, and guest-guest interactions, some mechanistic information may be gained from kinetic measurements. Since dimerization of 61a in DMF-d7 occurs on the NMR timescale, this system was studied instead of 61e in aromatic solvents where very rapid exchange is observed. Coalescence temperatures of the resonances assigned to the imine, downfield aromatic, upfield aromatic, methylene, and methyl protons of 61a were recorded in DMF-d7 from variable temperature 1H NMR spectroscopic experiments and then fit to an unequally populated two-site exchange model.114 Rates were calculated and an Eyring plot was constructed (Figure 4.14) giving a kdim of 170 s-1 at 25 °C, ΔH‡ = 69 ± 13 kJ mol-1, and ΔS‡ = -410 ± 60 J mol-1 K-1. A large negative entropy contribution indicates the mechanism is associative in nature and a high degree of order must be imparted upon the system. As the temperature drops below -26 °C, the dimerization rate approaches zero and a static mixture of monomer and dimer results. The accelerated dimerization rate qualitatively observed for 61e in non-polar organic solvents relative to 61a in DMF-d7 suggests the mechanism of dimerization may be different when a coordinating solvent (e.g., DMF) is present.  163  1.0 0.5  y = -8283.82x + 27.25 ln(kT-1)  0.0  r ² = 0.9971639797 -0.5 -1.0 -1.5 -2.0 0.032  0.033  0.033  0.034  0.034  0.035  0.035  0.036  1/T (K-1)  Figure 4.14. Eyring plot for dimerization of 61a in DMF-d7 constructed with the coalescence method (ΔH‡ = 69 ± 13 kJ mol-1 and ΔS‡ = -410 ± 60 J K-1 mol-1).  4.2.7 Cadmium Cluster Dynamics: A Molecular Beehive Besides dimerization, careful inspection of the acetate resonances of 61a and 61d in the 1H NMR spectra obtained at different temperatures revealed more dynamic behavior. In DMF-d7, at room temperature, the acetate, aqua, and capping hydroxo ligands are rapidly exchanging or “buzzing” around the cadmium cluster. This rapid exchange is manifested in a single broad resonance located around 2 ppm and assigned by integration of the proton resonances to both acetate environments and the capping aqua/hydroxo ligands. Although the rate of Cd-aqua/hydroxo exchange is known to be very fast (> 106 s-1),115 the rate of acetate exchange was suppressed to an intermediate exchange regime by chilling the sample. At low temperature, the broad resonance at 2 ppm separates into two acetate resonances at 2.05 and 1.95 ppm and one broad aqua/hydroxo proton resonance at 1.94 ppm belonging to monomeric metallocavitand 61d. These same monomer resonances are found in the low temperature spectrum of 61a along with two acetate resonances belonging to dimers of 61a found at 1.93 and 1.86 ppm. Figure 4.15 shows coalescence of the acetate region in variable temperature 1H NMR spectra. The acetate lability is highly solvent dependent; in CDCl3, distinct type- A and B acetate resonances (Figure 4.1) are evident in the 1H NMR spectra of metallocavitands 61c and 61e at room  164  temperature suggesting exchange is very slow or non-existent. When DMF-d7 is added to a solution of 61e in CDCl3, coalescence of type- A and B acetate resonances is observed, linking DMF to the mechanism of acetate exchange.  Figure 4.15. Dynamic ligand exchange in DMF-d7 is confirmed by coalescence of the acetate resonances in variable temperature 1H NMR spectra. a) Upon cooling a solution of metallocavitand 61a, two sets of two acetate resonances emerge with each set corresponding to monomer or dimer. b) In the absence of dimerization, only a single set of two acetate resonances is observed for metallocavitand 61d. Rates of exchange were calculated from the simulated resonances shown in blue. In each spectrum, the additional broad resonance is assigned to the coordinated aqua/hydroxo capping ligands. c) Eyring plot of acetate exchange for metallocavitand 61d (r2 = 0.992). 165  In the case of complex 61d, where the bulky neopentyloxy substituents prevent dimerization, the acetate region of the 1H NMR spectrum was easily simulated providing quantitative kinetic information about the rate of acetate exchange.116 From the Eyring plot shown in Figure 4.15c at 25 °C the calculated kexchange = 160 s-1, ΔH‡ = 47 ± 13 kJ mol-1, and ΔS‡ = -490 ± 120 J mol-1 K-1. Within experimental error these values are intriguingly similar to those obtained for dimerization of 61a in DMF-d7. A reasonable explanation for this kinetic congruency is that both mechanisms share a common ratedetermining step. Transition state solvation requires a high degree of order resulting in large negative entropy terms similar to those observed for both acetate exchange and dimerization of these metallocavitands. Intuitively, solvent-cavity interactions are involved in the mechanism of dimerization and may be invoked to explain the ΔS‡ = -410 ± 60 J mol-1 K-1 observed for dimerization of 61a; it is less obvious, however, how cavity solvation plays a role in the mechanism of acetate exchange. I propose that these two processes share a rate-determining transition state that involves the coordination of two DMF molecules to available cadmium centers inside the cavity of monomeric metallocavitands as outlined in Figure 4.16. To accommodate two DMF molecules inside the cavity, the μ3-OH----ODMF hydrogen bond with one DMF must be broken as the DMF forms a bond to one of the N2O2 chelated cadmium ions 1-3 (Figure 4.1), and a second nearby solvating DMF molecule must bond to a cadmium ion in the cavity contributing to the negative entropy. Cleavage of the stronger hydrogen bond and formation of the weaker coordination bonds are responsible for the positive enthalpy observed for this intermediate, as calculated earlier (Figure 4.10). After this rate-determining step, a third molecule of DMF may enter the bowl and coordinate to the last vacant site on cadmium, forming a more stable intermediate. The computed model (DFT) of this intermediate shows there is sufficient space inside the cavity for a third DMF guest molecule to coordinate to a cadmium ion only if none of the DMF molecules is simultaneously hydrogen bonding with the μ3-OH ligand (Figure 4.17). After formation of the tri-DMF intermediate, the mechanisms of acetate exchange and dimerization diverge.  166  Figure 4.16. a) Proposed mechanism for formation of the tri-DMF intermediate that initiates both acetate exchange (top right) and dimerization (bottom right). The rate determining step involves formation of two Cd-O (DMF) coordination bonds after the μ3OH----ODMF hydrogen bond is broken. In the top right, rapid cleavage of the Cd-OAc bond is promoted by guest DMF molecules trans to the acetate ligands and followed by exchange between type- A and B acetates, depicted in red and blue, respectively (much of the cluster has been omitted for clarity). In the bottom right, dimerization proceeds first by formation of a pre-capsule assembly followed by loss of coordinated DMF and fast dimerization of the unsaturated metallocavitands. b) Proposed reaction coordinate diagram leading to the tri-DMF coordinated intermediate.  167  a)  b)  Figure 4.17. DFT (D95V(d,p)/SDD/B98) energy-minimized structure of tri-solvated intermediate where all DMF guest molecules are coordinated to cadmium ions 1-3 (Figure 4.1) in 61f and none is interacting with the μ3-OH ligand. a) Top-down view. b) Side-on view. Coordination of a third DMF molecule saturates all of the N2O2 chelated cadmium ions (1-3, Figure 4.1) and destabilizes each type-A OAc-Cd bond trans to the DMF ligands. When three DMF molecules are coordinated inside the cavity, the absence of a μ3-OH----ODMF hydrogen bond localizes all of the carbonyl (DMF) electron density on cadmium ions 1-3, lowering the activation barrier to OAc-Cd bond scission. Indeed, the calculations of the complex with three coordinated DMF molecules in the bowl shows that the Cd-OAc bonds are lengthened by as much as ~0.1 Å. Cleavage of the Cd-OAc bonds and subsequent type- A to B acetate exchange must happen rapidly compared to the initial association of a second DMF inside the cavity. DMF coordination is essential to  168  promote acetate exchange as exchange is not observed in non-coordinating solvents such as CDCl3 until small quantities of DMF are introduced. Acetate exchange also has long-range effects on the rigidity of the cavity. Broadening of the proton resonance of the catechol unit pointing out of the cavity is observed at room temperature by 1H NMR spectroscopy of metallocavitand 61d (Hc in Figure 4.11c). Upon cooling the sample, the proton resonances of the catechol (CH) and acetate sharpened simultaneously suggesting the metallocavitand “breathes” when the acetates are labile but as their exchange is slowed, the cavity rigidifies. The tri-DMF reactive intermediate is actually more compact than the ground state hydrogen-bonded DMF assembly and allows for two metallocavitands to enter in close proximity forming a 6 DMF + 2 metallocavitand pre-capsule assembly. Expulsion of coordinated DMF from the compact pre-capsule leads to dimerization yielding a capsule with two encapsulated DMF molecules both coordinated and hydrogen bonded as observed in the solid state. Competition between DMF guests to form a hydrogen bond with the μ3-OH ligand and subsequent dimerization of the unsaturated metallocavitands must proceed quickly. The proposed acetate exchange and dimerization mechanisms fit reasonably well to the experimental and computational evidence. I acknowledge, however, that these events are more intricate than my observations indicate – it is not feasible to calculate the exact transition state for such an enormous cluster. Nevertheless, from the presented data, it is apparent that dimerization requires association followed by solvent expulsion from the bowl rather than the other way around and that the rate-determining step involves coordination of DMF to the metal center inside the bowl.  4.2.8 Guest Dynamics Probed by Solid-State NMR Spectroscopy In solution, fast dimerization kinetics hindered the investigation of guest molecules encapsulated in metallocavitands 61a-e. To overcome this setback, solid-state NMR spectroscopic experiments were conducted to study guest dynamics in metallocavitand capsules.117,118 Interestingly, after heating capsules of 61a under vacuum for 48 h one encapsulated DMF molecule is removed as confirmed by 1H NMR  169  spectroscopy in DMSO-d6. For solid-state 2H NMR spectroscopic investigations, samples of 61a were recrystallized from DMF-d7 followed by heating under vacuum for 48 h to give capsules of 61a with one DMF-d7 molecule encapsulated. Figure 4.18 depicts the DFT-optimized orientation of DMF hydrogen-bonding with the μ3-OH ligand inside the cavity.  a)  b)  Figure 4.18.  DFT (D95V(d,p)/SDD/B98) energy-minimized geometry of DMF  hydrogen-bonding inside the cavity of 61f. a) Top-down view looking down the 3-fold axis of rotation. b) Side-on view showing the H-bond. DMF is depicted as space filling. Solid-state 2H NMR powder lineshapes depend on the interaction between the electric quadrupole moment of the deuterium nucleus and the electric field gradient  170  tensor at the nuclear site. Although the quadrupole moment is invariant, the observed electric field gradient tensor depends on the degree of motional averaging. The 2H NMR lineshapes are therefore very sensitive to and characteristic of both the mechanism and rate of molecular motion.119 In the absence of molecular motion, solid-state 2H NMR spectra take on a lineshape like that shown in Figure 4.19, described by splittings, Δυxx , Δυyy and Δυzz and characterized by the quadrupolar coupling constant, χ = e2Qq/h and asymmetry parameter, η, where: Δυxx = (3χ/4)(1 – η) Δυyy = (3χ/4)(1 + η) Δυzz = 3χ/2 η = (Δυyy – Δυxx ) / Δυzz  0≤η≤1  Figure 4.19. 2H NMR powder lineshapes for: a) a rigid C-D bond and b) a methyl group rotating at a rate which is fast with respect to the static line width. When a molecular motion occurs at a rate that is fast with respect to the reciprocal of the overall static line width, Δυzz , χ is averaged by the motion and the lineshape is as described above with χ replaced by χ’, the reduced quadrupolar coupling constant defined by the molecular motion. An example of what one would expect for fast methyl group  171  rotation is shown in Figure 4.19b. When the molecular motion occurs at a rate of the same order as the reciprocal line width, the lineshape is more complicated and very sensitive to the rate of the motion.120 The solid state 2H NMR spectrum of DMF-d7 trapped within the cavities of 61a at 184 K is shown in Figure 4.20a and simulated in Figure 4.20b. The spectrum consists of two overlapping powder patterns: one from the methyl deuterons (χ’ = 49 ± 2 kHz, η = 0, relative intensity = 6) and one from the amide deuteron (χ = 155 ± 4 kHz, η = 0, relative intensity = 0.31). The expected intensity ratio of 6:1 is not observed as the T1 for the amide deuteron is longer than that for the methyl deuterons. The 5 second recycle delay used in the measurement was insufficiently long to allow for full relaxation of the amide deuteron. The quadrupolar coupling constant for the amide deuteron of DMF-d7 has been reported to be between 154 and 181 kHz and those for rigid methyl deuterons between 156 kHz and 167 kHz.121 The experimentally determined χ = 155 kHz for the amide deuteron indicates that the amide C-D bond is essentially rigid and is consistent with the hydrogen bonding observed in metallocavitand 61a.  Simulation  (b)  184 K  Experiment  (a) 100 kHz  Figure 4.20. a) The experimental and, b) simulated solid-state 2H NMR spectra of DMFd7 trapped in the capsules of 61a at 184 K. There are two overlapping powder patterns contributing to the spectrum. The more intense contribution is due to the rotating -CD3 groups and the broader, less intense contribution is due to the amide deuteron. The reduced quadrupolar coupling constant of 49 kHz observed for the methyl deuterons is typical for fast three-fold methyl group rotation122 in which case χ’ is 172  expected to be χ/3 ≈ 54 kHz. The additional reduction of χ’ to 49 kHz may be due to an additional small scale librational motion. The 2H NMR spectra measured as a function of temperature are shown in Figure 4.21 along with simulations. As the contribution from the amide deuteron is much broader than that for the methyl deuterons and has only a minor contribution to the overall spectrum, it was neglected in the simulations. The changes in lineshape observed as a function of temperature indicate that in addition to the fast methyl group rotation, a large scale molecular motion also occurs. Furthermore, since experiments show only one powder spectrum and not two different overlapping powder spectra in a 1:1 ratio for the CD3 groups, each methyl group must be treated identically by the motion. The spectra are consistent with an n-fold rotation (n ≥ 3) of the methyl groups about an axis 60° from each of the N-(CD3) bonds. One model that would explain the spectra (Model 1 in Figure 4.21) is a rotation of the N-(CD3)2 moiety about the bisector of the (CD3)-N-(CD3) bond angle. For such a motion, both methyl groups precess on the surface of a cone of half-angle 60°. It may be reasonable to assume that n = 3, in accordance with the approximate 3-fold symmetry of the void space of the capsule. The DFT calculations discussed earlier, however, give reason to believe that this is not the case. Calculations indicate that the DMF molecule is hydrogen bonded with the μ 3-OH ligand inside the capsule and any motion of the DMF molecule is likely to be a rotation about the μ3-OH----ODMF hydrogen bond (Model 2 in Figure 4.21). If this axis of rotation is translated such that it passes through the nitrogen atom of the DMF molecule, the methyl groups lie on cones of half-angles 61.2° and 64.6°. These angles are reasonably close to 60° and the effects on the 2H NMR spectrum resulting from a 3-fold rotation about this axis, in accordance with the approximate 3-fold symmetry of the void space, will be similar to those for a simple 3-fold internal rotation of the N-(CD3)2 moiety about the bisector of the (CD3)-N-(CD3) bond angle. For Model 2 in Figure 4.21, spectra were simulated independently for each of the methyl groups and the two spectra were added together in a 1:1 intensity ratio. The contribution from the amide deuteron was neglected in both models.  173  Figure 4.21. The experimental and simulated solid-state 2H NMR spectra for DMF-d7 trapped in the capsules of 61a measured as a function of temperature. Model 1 is a 3-fold rotation of the N-(CD3)2 moiety about the bisector of the (CD3)-N-(CD3) bond angle. Model 2 is an overall 3-fold molecular rotation about an axis defined by the DFT calculations. The agreement between the experimental and simulated spectra is good; however, the spectra at the higher temperatures are somewhat narrower than either model predicts. This additional narrowing is likely due to small scale librational motion. The spectra are very sensitive to temperature and therefore to the rate of the molecular motion. An Arrhenius plot was constructed with the rates of the rotation used for the simulations assuming n = 3 (Figure 4.22). The activation energy for the rotation was found to be 14.4 ± 1 kJ mol-1, which agrees well with the 14.3 kJ mol-1 obtained from 3-21G/SDD/PBE DFT calculations.  174  Figure 4.22. Arrhenius plot for the rotation of DMF-d7 in the capsules of 61a (r2 = 0.995).  4.3  Conclusions To drive discovery of useful multimetallic clusters with unique catalytic and  materials science applications improved methods for their synthesis must be pursued. In this chapter the propensity for macrocycle 51 to template formation of high nuclearity metal clusters was emphasized and a family of heptacadmium metallocavitands was described. These metallocavitands exhibit a wide variety of interesting dynamic processes ranging from dimerization, to ligand exchange, to guest exchange. SCXRD experiments and low temperature 1H NMR spectroscopy revealed the previously assigned μ3-O ligand is actually a μ3-OH ligand and DFT calculations confirmed an energetic preference for the μ3-OH ligand. Although the difference is structurally subtle, the μ3-OH ligand dramatically altered the host-guest chemistry of these metallocavitands, resulting in a μ3-OH ligand capable of H-bonding to encapsulated guest molecules deep in the cavity. Hydrogen bonding and metal coordination synergy inside metallocavitand capsules resulted in an exceptionally high packing coefficient, 0.80, observed for encapsulated DMF. Variable temperature 1H NMR spectroscopy revealed dimerization to  175  be an entropy-driven event that is highly solvent dependant. Formation of two DMFcadmium coordination bonds inside the monomer’s cavity was proposed as the rate determining step for both dimerization and acetate exchange in DMF-d7. Three-fold rotation of an encapsulated DMF molecule was also quantified by variable-temperature solid-state 2H NMR spectroscopy.  4.4  Experimental  4.4.1 General Macrocycles 51a-e were prepared as previously reported.27,34 All reactions were carried out under air unless otherwise noted. 1H and 13C NMR spectra were recorded on either a Bruker AV-300 or AV-400 spectrometer. 13C NMR spectra were recorded using a proton decoupled pulse sequence. 1H and  13  C NMR spectra were calibrated to the residual  protonated solvent at δ 7.27 and δ 77.0 ppm, respectively, in CDCl3, or at δ 8.03 and δ 163.15 for DMF-d7. The  113  Cd NMR spectrum was recorded in CDCl3 on a Bruker AV-  400 spectrometer with a pulse delay of 30 seconds to allow for sufficient relaxation. Calibration was performed with a 0.1 mol L-1 solution of Cd(ClO4)2 in D2O set to 0 ppm. UV-vis spectra were obtained in CH2Cl2 (ca. 1 x 10-6 M) on a Varian Cary 5000 UV-visnear-IR spectrophotometer using a 1 cm quartz cuvette. FT-IR spectra were collected neat in the solid state on a Thermo Nicolet 6700 FT-IR spectrometer or in a KBr pellet on a Nicolet 4700 FT-IR spectrometer. MALDI-TOF mass spectra were obtained with either a trans-2-[3-(4-tert-Butylphenyl)-2-methyl-2-propenylidene]-malonitrile (DCTB) matrix (solvent free) or 1,8-dihydroxy-9,10-dihydroanthracen-9-one (dithranol) matrix (from EtOH) at the UBC Microanalytical Services Laboratory on a Bruker Biflex IV instrument. Elemental analyses (C,H,N) were performed at the UBC Microanalytical Services Laboratory. Melting points were obtained on a Fisher-John’s melting point apparatus. Single-crystal X-ray diffraction experiments were performed on a Bruker X8 Apex CCD instrument.  176  4.4.2 Procedures and Data Synthesis of heptacadmium metallocavitand 61a (R = C2H5): Cadmium acetate dihydrate (408 mg, 1.53 mmol) was added to a suspension of macrocycle 51a (200 mg, 0.20 mmol) in 8 mL of EtOH and heated to reflux for 1.5 h. The solution was cooled to room temperature precipitating a red solid that was then isolated on a glass frit. Recrystallization from hot DMF gave 61a as a dark red microcrystalline solid. Yield: 247 mg (0.11 mmol, 57%). Low solubility of 61a prevented observation of all of the resonances in the 13C NMR spectrum. Data for 61a (R = C2H5): 13C NMR (100.6 MHz, DMF-d7) δ 165.9, 163.3, 150.6, 136.4, 120.5, 105.6, 66.3, 21.9, 15.6. 1H NMR (400 MHz, DMF-d7) δ 8.64 (s, 6H, imine), 7.50 (s, 6H, Ar), 6.52 (s, 3H, monomer Ar), 5.83 (s, 3H, dimer Ar), 4.33 (s, 12H, OCH2CH3), 1.96 (s, 18H, acetate), 1.49 (s, 18H, CH2CH3). UV-vis (CH2Cl2) λmax (ε) = 349 (2.14 x 105), 412 (5.36 x 105) nm (L mol-1 cm-1). MALDI-TOF MS: many peaks around the molecular ion but none unambiguously assignable. FT-IR (KBr): ῡ = 2979, 2931, 2883, 1608, 1565, 1503, 1454, 1415, 1395, 1331, 1264, 1182. Mp > 270 °C. Anal. Calc’d for 61a: C66H72Cd7N6O28·2DMF: C, 37.11; H, 3.72; N, 4.81. Found: C, 37.12; H, 3.81; N, 4.59. Synthesis of heptacadmium metallocavitand 61c (R = nC6H13): Prepared the same as above for 61a with 86 mg of cadmium acetate dihydrate (0.32 mmol) and 60 mg of 51c (0.046 mmol) in 60% yield (0.04 mmol).  177  Data for 61c (R = nC6H13): 113Cd NMR (84.9 MHz, CDCl3) δ 139.5 (t, 3JCd-H = 34.1 Hz), 10.1, 1.1. 13C NMR (75.5 MHz, CDCl3) δ 181.0, 164.4, 161.3, 149.8, 134.6, 120.4, 119.2, 103.5, 69.9, 31.5, 29.2, 25.6, 22.6, 21.3, 14.0. 1H NMR (300 MHz, CDCl3) δ 8.26 (d, 3JCdH  = 34.1 Hz, 6H, imine) 6.87 (s, 6H, Ar), 6.56 (s, 6H, Ar), 4.05 (m, 12H, OCH2), 2.15 (s,  9H, OAc), 2.00 (s, 9H, OAc), 1.84 (m, 12H, CH2), 1.50 (m, 24H, CH2), 1.37 (m, 12H, CH2), 0.91 (t, 3JHH = 6.9 Hz, 18H CH2CH3). UV-vis (CH2Cl2) λmax (ε) = 410 (1.1 x 105), 348 (6.0 x 104), 247 (6.6 x 104) nm (L mol-1 cm-1). MALDI-TOF MS (DCTB matrix) m/z = 2130.9 [61c-Cd2O(OAc)2+Na]+, 2465.7 [61c-(H2O)OH]+. FT-IR (KBr): ῡ = 3443, 2956, 2932, 2860, 1608, 1572, 1499, 1451, 1415, 1399, 1334, 1266, 1181, 1156, 1100, 1012, 932, 903, 751, 698 cm-1. Mp > 250 ºC. Synthesis of heptacadmium metallocavitand 61d (R = CH2C(CH3)3): Cd(OAc)2 dihydrate (121 mg, 0.45 mmol) was dissolved in 5 mL of MeOH and added to a stirring suspension of macrocycle 51d (73 mg, 0.06 mmol) in 2 mL MeOH. After stirring for 12 h, the deep red suspension was filtered through a frit yielding 100 mg of product as a red powder. (0.04 mmol, 68%). Recrystallization from hot DMF was necessary for elemental analysis. Data for 61d (R = CH2C(CH3)3): 13C NMR (100 MHz, DMF-d7) δ 165.1, 162.9, 150.0, 135.4, 119.9, 119.8, 103.7, 79.2, 32.6, 26.8, 21.6. 1H NMR (300 MHz, DMF-d7) δ 8.63 (d, 3JCd-H 31.4 Hz, 6H, imine), 7.38 (s, 6H, Ar), 6.60 (bs, 6H, Ar), 3.90 (s, 12H, OCH2), 2.00 (bs, 24H, OAc & aqua ligands), 1.13 (s, 54H, C(CH3)3) ppm. UV-vis (CH2Cl2) λmax (ε) = 413 (1.6 x 105), 347 (8.2 x 104) nm (L mol-1 cm-1). MALDI-TOF MS (dithranol matrix) m/z = 2383.5 [61d-(H2O)2OH]+. FT-IR (neat): ῡ = 2953, 2866, 1606, 1560, 1501, 1449, 1414, 1395, 1318, 1258, 1219, 1241, 1176, 1151, 1116, 1043, 1010, 919, 840, 759, 666, 609, 597. Mp > 250 ºC. Anal. Calc'd for: C185H249Cd14N17O60 (calculated from heterocapsules found in the solid state): C, 42.37; H, 4.79; N, 4.54. Found: C, 42.89; H, 4.75; N, 3.53.  178  Synthesis of heptacadmium metallocavitand 61e (R = C8H17): Cadmium acetate dihydrate (382 mg, 1.44 mmol) was added to a suspension of macrocycle 51e (300 mg, 0.20 mmol) in 5 mL of EtOH and reacted at room temperature for 3 h. The solution was chilled to 0 °C for 16 h to precipitate a red powder that was isolated on a glass frit. The product was washed with hexanes yielding 61e as a red microcrystalline powder. Yield: 410 mg (0.15 mmol, 75%).  Data for 61e (R = C8H17):  13  C NMR (100.6 MHz, CDCl3) δ 183.0 (t, JCd-C 21.9 Hz) ,  164.67, 164.61, 161.56, 161.51, 150.1, 134.8, 120.7, 119.5 (t, JCd-C 18.5 Hz), 103.8, 70.2, 32.0, 29.6, 29.5, 26.2, 22.9, 14.3. 1H NMR (300 MHz, CDCl3) δ 8.29 (t, JCd-H 34.2 Hz, 6H, imine), 6.89 (s, 6H, Ar), 6.59 (s, 6H, Ar), 4.08 (m, 12H, OCH2CH2), 2.19 (s, 9H, OAc), 2.03 (s, 9H, OAc), 1.87 (m, 12H, OCH2CH2), 1.52 (m, 12H, CH2), 1.32 (bs, 48H, CH2), 0.91 (t, 3JHH 6.6 Hz, 18H, CH2CH3). UV-vis (CH2Cl2) λmax (ε) = 353 (3.28 x 105), 412 (5.75 x 105) nm (L mol-1 cm-1). MALDI-TOF MS: many peaks around the molecular ion but none were unambiguously assignable. FT-IR (KBr): ῡ = 2927, 2855, 1608, 1558, 1503, 1454, 1417, 1333, 1267, 1179. Mp > 270 °C. Anal. Calc’d for 61e: C102H144Cd7N6O28·DMF: C, 45.66; H, 5.51; N, 3.55. Found: C, 47.29; H, 5.54; N, 3.47.  4.4.3 Thermodynamic Studies Variable Temperature 1H NMR Experiments for 61a in DMF-d7:  A standard  solution was prepared with 109.0 mg of 61a dissolved in 1.75 mL of DMF-d7 (28.5 mmol L-1). A 50 μL aliquot of the standard was then diluted to 550 μL yielding a 2.6 mmol L-1 solution. The 2.6 mmol L-1 solution was chilled to 238, 247, 256, and 265 K and a 1H NMR spectrum was taken at each temperature after equilibrating for 7 minutes. The upfield and downfield aromatic proton peaks corresponding to monomer and dimer were then integrated and association constants were calculated using a standard monomerdimer equilibrium equation. The association constants for dimerization (Kdim), calculated from Ha and Hc, were averaged, and the average association constants were used to make a van’t Hoff plot.  An estimated 5% error in integration was applied to the  thermodynamic parameters.  179  Metallocavitand 61d (6 mg, 0.0025 mmol) was dissolved in 0.500 mL of DMF-d7 and subjected to 1H NMR spectroscopic analysis at 52, 25, 15, 0.5, -12, -24, -35, and -45 °C. No dimerization was observed as seen in Figure 4.23. Slight broadening of the catechol C-H aromatic resonance, found at 6.6 ppm, coincides with the coalescence of the acetate resonances. As the acetate exchange slows below the NMR timescale the catechol resonance sharpens.  52 °C 25 14 0.5 -12 -24 -35 -45 ppm 9  8  7  6  5  4  3  2  1  0  Figure 4.23. Variable temperature 1H NMR of 61d in DMF-d7 (400 MHz). VTVC 1H NMR Experiments for 61e in aromatic solvents: Due to solubility constraints, the concentration range of 61e was limited in aromatic solvents (0.1 to 5.0 mM). A standard 5.00 mM solution was prepared by first dissolving 0.005 mmol of metallocavitand (61e) with heat in 1.00 mL of the deuterated solvent of choice. Except for the 5.00 mM samples, NMR tubes were primed with 500 μL of selected deuterated solvent and standard was then added via syringe to achieve the desired concentrations. The concentration and temperature dependence of the imine resonance was measured in CD2Cl2, benzene-d6, toluene-d8, and p-xylene-d10. The data, displayed in Tables 4.2, 4.3,  180  and 4.4, was treated with the least-squares curve-fitting equation, 1 (monomer-dimer equilibrium model), as shown in Figures 4.24, 4.26, and 4.28, to find the association constants for dimerization.104 Figures 4.25, 4.27, and 4.29 show the van’t Hoff plots constructed to calculate the thermodynamic parameters of dimerization using equation 2.  + (1)  G  Gm  § 1  8K dimCT  1 · ¸ G d  G m ¨1   (2)  ln K dim    ¨ ©  ¸ ¹  4 K dimCT  'H 1 'S ˜  R T R  8.24  Imine Chemical Shift (ppm)  8.22 8.20 8.18 298 K 311 K 321 K 330 K 339 K  8.16 8.14 8.12 8.10 8.08 0.000  0.002  0.004  0.006  0.008  0.010  0.012  Concentration (M)  Figure 4.24. VTVC imine 1H NMR chemical shift dependence of 61e in benzene-d6.  181  Table 4.2. Imine chemical shift (ppm) and association constants of 61e in benzene-d6.  Conc. (mM)  298 K  311 K  321 K  330 K  339 K  10.00  8.211  8.218  8.222  8.225  8.227  3.33  8.189  8.193  8.196  8.201  8.204  1.67  8.169  8.173  8.180  8.183  8.186  0.48  8.131  8.131  8.138  8.143  8.146  0.20  8.105  8.108  8.114  8.119  8.123  0.10  8.090  8.100  8.106  8.111  8.116  Kdim  900 ± 100  500 ± 100  600 ± 100  600 ± 100  600 ± 100  10  ln Kdim  8  6  4  2  0 2.9e-3  3.0e-3  3.1e-3  3.2e-3  3.3e-3  3.4e-3  1/T (K-1)  Figure 4.25. van’t Hoff plot of 61e in benzene-d6.  182  8.26  ImineChemical Shift (ppm)  8.24  8.22  8.20 283 K 321 K 339 K 357 K  8.18  8.16  8.14 0.000  0.002  0.004  0.006  0.008  0.010  0.012  Concentration (M)  Figure 4.26. VTVC imine 1H NMR chemical shift dependence of 61e in toluene-d8. Table 4.3. Imine chemical shift (ppm) and association constants of 61e in toluene-d8.  Conc. (mM)  283 K  321 K  339 K  357 K  10.00  8.250  8.245  8.243  8.236  3.33  8.231  8.228  8.226  8.223  1.67  8.221  8.216  8.214  8.211  0.91  8.209  8.206  8.204  8.196  0.48  8.196  8.194  8.189  8.182  0.20  8.172  8.174  8.167  8.160  Kdim  1500 ± 700  1000 ± 300  1500 ± 400  1800 ± 300  183  14 12  ln Kdim  10 8 6 4 2 0 0.0026  0.0028  0.0030  0.0032  0.0034  0.0036  -1  1/T (K )  Figure 4.27. van’t Hoff plot of 61e in toluene-d8.  Imine Chemical Shift (ppm)  8.36  8.34  8.32 307 K 321 K 334 K 348 K 362 K 376 K  8.30  8.28  8.26 0.000  0.002  0.004  0.006  0.008  0.010  0.012  Concetration (M)  Figure 4.28. VTVC imine 1H NMR chemical shift dependence of 61e in p-xylene-d10.  184  Table 4.4. Imine chemical shift (ppm) and association constants of 61e in p-xylene-d10. Conc.  307 K  321 K  334 K  348 K  362 K  376 K  10.00  8.342  8.342  8.342  8.344  8.349  8.352  3.33  8.328  8.329  8.333  8.333  8.336  8.338  1.67  8.32  8.321  8.323  8.324  8.326  8.327  0.48  8.313  8.313  8.313  8.313  8.317  8.318  0.20  8.301  8.301  8.303  8.303  8.304  8.304  0.10  8.289  8.286  8.282  8.283  8.285  8.283  Kdim  800 ± 300  1300 ±  3700 ±  2500 ±  1900 ±  2000 ±  400  1400  700  400  600  (mM)  14 12  ln Kdim  10 8 6 4 2 0 0.0026  0.0027  0.0028  0.0029  0.0030  0.0031  0.0032  0.0033  -1  1/T (K )  Figure 4.29. van’t Hoff plot of 61e in p-xylene-d10.  185  4.4.4 Kinetic Studies Kinetics of Dimerization: Dimerization rates of metallocavitand 61a at various temperatures in DMF-d7 were calculated with a coalescence method. A 0.005 mol L-1 solution of 61a in DMF-d7 was prepared and subjected to a variable temperature 1H NMR spectroscopic experiment (Figure 4.30). The coalescence point of the imine, downfield aromatic, upfield aromatic, methylene, and methyl resonance was monitored and fit to an Eyring plot to establish kinetic parameters.  112° C 98 84 57 47 38 27 20 11 1 -8 -17 -26  8  6  4  2  0 ppm  Figure 4.30. Variable temperature 1H NMR spectra of 61a in DMF-d7 (400 MHz).  186  Unequally populated exchange such as the monomer-dimer equilibrium of 61a may be analyzed with the coalescence method as outlined by Pons and Millet.114 The rate constants at each coalescence temperature are determined with: k = π(Δv)/X Where Δv is the chemical shift difference between monomer and dimer in the absence of exchange (calculated at -26° C for 61a) and X is calculated by solving the polynomial: 0 = X6 - 6X4 + X2(12-27(Δp)2) – 8. The population difference between monomer (pm) and dimer (pd) is represented as Δp = pm - pd where: pm = (# monomer molecules) / (# monomer molecules + # dimer molecules) and pd = (# dimer molecules) / (# monomer molecules + # dimer molecules). The polynomial is necessary because dimerization is an equilibrium process with unequal populations of monomer and dimer. The population difference (Δp) at different temperatures was determined from equilibrium constants calculated with the van’t Hoff plot. Table 4.5. Parameters for calculating the dimerization rate of 61a in DMF-d7.a Resonance  Upfield  aromatic  aromatic  289  297  204.95  methylene  methyl  311  287  285  253.60  358.57  193.96  183.42  2.48E-03  2.31E-03  2.03E-03  2.53E-03  2.57E-03  [Dimer] (mol L )  1.26E-03  1.35E-03  1.48E-03  1.24E-03  1.21E-03  Population of Monomer  7.47E+17  6.94E+17  6.12E+17  7.60E+17  7.75E+17  Population of Dimer  3.79E+17  4.06E+17  4.47E+17  3.73E+17  3.65E+17  Δp  0.3261  0.2622  0.1566  0.3424  0.3588  X value  2.0769  1.9853  1.8158  2.0994  2.1216  42.60  100.26  336.00  39.61  32.77  64.4  158.7  581.3  59.3  48.5  Coalescence T (K) -1  Keq (L mol ) -1  [Monomer] (mol L ) -1  Δv at 247 K (Hz) -1  -1  Rate k = π(Δv)X (s ) a  downfield  Imine  See Figure 4.14 for the dimerization Eyring plot.  187  OAc Exchange: Upon raising the temperature, 1H NMR spectra of metallocavitand 61d (R = CH2C(CH3)3) show coalescence of the broad acetate/aqua/hydroxo region into three resonances, type-A acetates, type-B acetates, and aqua/hydroxo ligands (Figure 4.1). The variable temperature 1H NMR spectra were simulated with Spinworks116 to give rate constants of 3, 11, 30, and 69 s-1 at 249, 261, 263, and 287 K respectively. Construction of an Eyring plot (Figure 4.15c and Figure 4.31) gave kexchange = 160 s-1 at 25 °C, ΔH‡ = 47 ± 13 kJ mol-1, and ΔS‡ = -490 ± 120 J mol-1 K-1.  0  ln (kT-1)  y = -5628x + 18  -2  r ² = 0.9920514424  -4 -6 0.0034  0.0037  0.0040  1/T (K-1) Figure 4.31. Eyring plot of OAc exchange on metallocavitand 61d in DMF-d7.  4.4.5 Crystallography Crystallography of 61a – 1st Crystal - (R = CH2CH3): C66H72Cd7N6O28, Mw = 2184.18, red plate (0.25 x 0.25 x 0.10 mm3), rhombohedral, space group R-3m, a = b = 28.712(3), c = 26.128(3) Å, V = 18654(4) Å3, Z = 6, ρcalcd = 1.867 g cm-3, F000 = 6420, MoKα radiation, λ = 0.71069 Å, T = 173(2) K, 2θmax = 45.0°, 46300 reflections collected, 2884 unique (Rint = 0.1841). Final GoF = 1.005, R1 = 0.0665, wR2 = 0.1845, R indices based on 1816 reflections with I>2σ(I). The structure was solved by direct methods123 and refined by full-matrix least-squares methods on F2. All non-hydrogen atoms were refined anisotropically, whereas hydrogen atoms were added at geometrically expected positions and left isotropic. Hydrogen atoms from the capping aqua ligands were not modeled. The  188  intermolecular Oaqua-Oaqua distance is 4.504 Å ruling hydrogen bonding. Electron density from disordered solvent located both inside and outside the capsule, was accounted for using the PLATON/SQUEEZE program.124 The ORTEP depiction is found in Figure 4.32. Disordered solvent was found inside and outside the capsule. I was able to model with some certainty an oxygen atom inside the capsule, approximately 3 Å from a cadmium ion in the N2O2 pockets. This oxygen atom had two possible binding orientations to each cadmium ion. Within each capsule (two metallocavitands), there are twelve possible orientations for the disordered solvent, rendering it nearly impossible to model the encapsulated solvent with certainty. It was necessary to use the PLATON/SQUEEZE software package to account for the disorder. Reassessment of the SQUEEZE analysis gave a void space inside each capsule of 194 Å3 that contains 75 electrons, almost exactly 2 DMF molecules (40 electrons per DMF).  Figure 4.32. ORTEP depictions of 61a – 1st crystal. Ellipsoids are at 50% probability and hydrogen atoms are omitted for clarity (C = black, N = blue, O = red, Cd = yellow).  189  Crystallography of 61a – 2nd Crystal - (R = CH2CH3): C33H33Cd3.5N3O14, Mw = 1089.2 g mol-1, red plate (0.35 x 0.30 x 0.20 mm3), rhombohedral, space group R-3m, a = b = 28.712(5), c = 26.128(5) Å, V = 18654(6) Å3, Z = 12, ρcalcd = 1.163 g cm-3, F000 = 6384, MoKα radiation, λ = 0.71069 Å, T = 173(2) K, 2θmax = 53.70°, 78323 reflections collected, 9612 unique (Rint = 0.0618). Final GoF = 1.011, R1 = 0.0911, wR2 = 0.2992, R indices based on 2850 reflections with I>2σ(I). The structure was solved by direct methods123 and refined by full-matrix least-squares methods on F2. All non-hydrogen atoms were refined anisotropically, whereas hydrogen atoms were added at geometrically expected positions and left isotropic. Hydrogen atoms from the aqua and hydroxo ligands were not modeled. Electron density from disordered solvent located both inside and outside the capsule, was accounted for using the PLATON/SQUEEZE program.124 The ORTEP depiction is found in Figure 4.33. From the SQUEEZE analysis, the void space inside each capsule (239 Å3) contains 82 electrons, corresponding to almost exactly 2 DMF molecules (40 electrons per DMF).  Figure 4.33. ORTEP depictions of 61a – 2nd Crystal. Ellipsoids are at 50% probability and hydrogen atoms are omitted for clarity (C = green, N = blue, O = red, Cd = cyan).  190  Crystallography of 61d (R = CH2C(CH3)3: C390H515Cd28N42O128, Mw = 10986.97, red block (0.70 x 0.45 x 0.20 mm3), triclinic, space group P-1, a = 20.083(3) b = 22.576(4) c = 34.101(6) Å, α = 81.279(9) β = 88.262(9) γ = 82.512(9)°, V = 15152(4) Å3, Z = 1, ρcalcd = 1.206 g cm-3, F000 = 5533, MoKα radiation, λ = 0.71073 Å, T = 173(2) K, 2θmax = 44.4°, 31117 reflections collected, 31117 unique (Rint = 0.000). Final GoF = 1.113, R1 = 0.1158, wR2 = 0.3266, R indices based on 19376 reflections with I>2σ(I). The structure was solved by direct methods123 and refined by full-matrix least-squares methods on F2. The cluster capping DMF molecule was modeled with 50% occupancy over two conformations. Disordered neopentyl substituents, free DMF molecules, and two aromatic rings required a total of 388 restraints to model sufficiently. Roughly half of the non-hydrogen atoms were modeled anisotropic with the other half being left isotropic. All C-H hydrogen atoms were added to geometrically expected positions. Aqua and hydroxo ligand protons were not modeled. Although 6 free DMF molecules were located and modeled from the difference map, electron density corresponding to 40 more disordered DMF molecules (1612 electrons, 4783 Å3) located outside the capsule was accounted for using the PLATON/SQUEEZE program.124 A thermal ellipsoid plot is depicted in Figure 4.34  Figure 4.34. Thermal ellipsoid plot of 61d. Ellipsoids are at 50% probability and hydrogen atoms are omitted for clarity (C = green, N = blue, O = red, Cd = cyan). 191  4.4.6 Solid-State 2H NMR Spectroscopy The 2H NMR spectra were measured by Professor David Bryce and coworkers at the University of Ottawa on a Bruker ASX 200 NMR spectrometer equipped with a 5 mm wideline solenoid probe operating at 30.7 MHz for 2H.  The quadrupolar echo  sequence125 was used to collect the data with 2 K complex points in the FID using a spectral width of 1 MHz. 600-800 transients were signal averaged for each spectrum. The 90° pulses were 3.5 μs and the recycle delay was 5 seconds. The first and second echo delays were set to 35 μs and 33 μs, respectively. The time domain data were leftshifted to locate the top of the echo and an exponential line broadening of 200 Hz was applied prior to Fourier transformation. The temperature was lowered by passing dry nitrogen gas through a heat exchanger coil. The cold gas was heated to the desired temperature with the heater in the NMR probe and controlled by the variable temperature control unit of the spectrometer which uses a thermocouple in the vicinity of the coil to regulate the temperature. The precise temperature of the sample was measured after the NMR measurement by placing a thermocouple directly on the coil of the NMR probe. The reported temperatures are thought to be accurate and precise to within 1 K. The simulations were conducted using NMR-WEBLAB V4.1.2.126  4.4.7 Simulated 2JHCd Coupling The observed coupling was between the μ3-OH and three Cd2+ ions of 61a in DMF-d7 at 24 °C. Simulated permutations: 1H 112Cd 112Cd 112Cd singlet 0.4217 probability, 1H 112Cd 112 1  Cd 111Cd doublet 0.2151 probability, 1H 112Cd 112Cd 113Cd doublet 0.2068 probability,  H  112  Cd  111  Cd  probability, 1H  111  112  Cd triplet 0.0366 probability, 1H  Cd  111  Cd  113  112  Cd  113  Cd triplet 0.0703 probability, 1H  Cd  111  113  Cd  Cd triplet 0.0338  113  Cd  113  Cd quartet  0.0057 probability, 1H 113Cd 111Cd 111Cd quartet 0.0060 probability, 1H 111Cd 111Cd 111Cd quartet 0.0021 probability, 1H 113Cd 113Cd 113Cd quartet 0.0018 probability.  192  4.4.8 Computational Details To support the experimental results, density functional theory (DFT) calculations were performed by Professor Francesco Lelj (Universita della Basilicata) using the ADF2007 (ADF2007.01, SCM, Theoretical Chemistry, Vrije Universiteit, Amsterdam, The Netherlands, http://www.scm.com)127,128 and Gaussian03 (revision D02)129 packages. The ADF2007 program allowed explicit consideration of relativistic effects using the Zero Order Regular Approximation (ZORA) approach.130-133 This variationally stable scheme is based on a regular-potential expansion and produces a Hamiltonian that in zeroth order reproduces all important relativistic effects including the spin-orbit interaction. A valence triple-zeta STO basis set with one polarization function (ZORA/TZP without frozen cores) was used for the relativistic calculations. Gradient corrected exchange-correlation (xc) functionals Perdew, Burke and Ernzerhof (PBE)134 have been used. The local part of the xc functionals was the default Vosko, Wilk and Nusair (VWN)135 one. Integrals accuracy has been increased to 10-8 (using keyword “INTEGRATION 8.0”) and Self Consistence Field convergence criterion has been chosen equal to 10-9 (keyword “SCF”, subkey “converge 1e-9”). Structures have been fully optimized136,137 without using symmetry constraints and applying default convergence criteria apart from the maximum gradient which has been set to 10-4 hartree/bohr. Delocalized coordinates have been used ("delocal" option). In case of Gaussian03 computations the DFT has been applied by using both pure 134  PBE  and the hybrid B98138 or PBE1PBE139 xc functionals. Geometry optimizations  have been performed by using both the 3-21g140-144 and Dunning/Huzinaga valence double-] (D95V(d,p)) basis sets,145 adding a set of polarization functions to the same basis set in case of H, C, N, and O. The Stuttgart/Dresden ECP basis set146 and pseudopotential (SDD) including relativistic effects have been used for Cd atoms. In the study of the rotation of the DMF molecule when H-bonded to the P3-OH ligand of 61f, the 6-31g(d,p)147-151 basis set was used for the H, C, N and O atoms of DMF and P3-OH in order to better describe the H-bonding interaction. The 3-21g basis set was used for all remaining H, C, N and O atoms. Basis set superimposition error (BSSE)152,153 contribution to the interaction energies have been corrected by means of the 193  Boys and Bernardi counterpoise (CP) approach. These corrections ranged from 28% to 39% of the value of the interaction energy. Default gradient and displacement thresholds have been used for the geometry optimization convergence criteria. To confirm that the obtained geometries are relative minima of the potential energy surface (electronic energy in the Born-Oppenheimer approximation + internuclear repulsion energy), analytical computation of the Hessian matrix with respect to the nuclear coordinates at the same level of theory was performed.  194  4.5  References  (1)  Ozin, G. A.; Arsenault, A. C. Nanochemistry; Royal Society of Chemistry: Cambridge, 2005.  (2)  Ozin, G. A. Adv. Mater. 1992, 4, 612-649.  (3)  Mezei, G.; Zaleski, C. M.; Pecoraro, V. L. Chem. Rev. 2007, 107, 4933-5003.  (4)  Brechin, E. K. Chem. Commun. 2005, 5141-5153.  (5)  Christou, G.; Gatteschi, D.; Hendrickson, D. N.; Sessoli, R. MRS Bull. 2000, 25, 66-71.  (6)  Shibasaki, M.; Yamamoto, Y.; Eds. Multimetallic Catalysts in Organic Synthesis; Wiley-VCH: Weinheim, 2004.  (7)  Ohki, Y.; Ikagawa, Y.; Tatsumi, K. J. Am. Chem. Soc. 2007, 129, 10457-10465.  (8)  Laplaza, C. E.; Holm, R. H. J. Am. Chem. Soc. 2001, 123, 10255-10264.  (9)  Feltham, H. L. C.; Brooker, S. Coord. Chem. Rev. 2009, 253, 1458-1475.  (10)  Lan, Y.; Novitchi, G.; Clerac, R.; Tang, J.-K.; Madhu, N. T.; Hewitt, I. J.; Anson, C. E.; Brooker, S.; Powell, A. K. Dalton Trans. 2009, 1721-1727.  (11)  Kleij, A. W. Chem.--Eur. J. 2008, 14, 10520-10529.  (12)  Niel, V.; Milway, V. A.; Dawe, L. N.; Grove, H.; Tandon, S. S.; Abedin, T. S. M.; Kelly, T. L.; Spencer, E. C.; Howard, J. A. K.; Collins, J. L.; Miller, D. O.; Thompson, L. K. Inorg. Chem. 2008, 47, 176-189.  (13)  Schemberg, J.; Schneider, K.; Demmer, U.; Warkentin, E.; Mueller, A.; Ermler, U. Angew. Chem., Int. Ed. 2007, 46, 2408-2413.  (14)  Nabeshima, T.; Miyazaki, H.; Iwasaki, A.; Akine, S.; Saiki, T.; Ikeda, C. Tetrahedron 2007, 63, 3328-3333.  (15)  Zheng, N.; Lu, H.; Bu, X.; Feng, P. J. Am. Chem. Soc. 2006, 128, 4528-4529.  195  (16)  Artuso, F.; D'Archivio, A. A.; Lora, S.; Jerabek, K.; Kralik, M.; Corain, B. Chem.--Eur. J. 2003, 9, 5292-5296.  (17)  Aromi, G.; Gamez, P.; Roubeau, O.; Kooijman, H.; Spek, A. L.; Driessen, W. L.; Reedijk, J. Angew. Chem., Int. Ed. 2002, 41, 1168-1170.  (18)  Fontecha, J. B.; Goetz, S.; McKee, V. Angew. Chem., Int. Ed. 2002, 41, 45534556.  (19)  Che, C.-M.; Xia, B.-H.; Huang, J.-S.; Chan, C.-K.; Zhou, Z.-Y.; Cheung, K.-K. Chem.--Eur. J. 2001, 7, 3998-4006.  (20)  Doble, D. M. J.; Benison, C. H.; Blake, A. J.; Fenske, D.; Jackson, M. S.; Kay, R. D.; Li, W.-S.; Schroder, M. Angew. Chem., Int. Ed. 1999, 38, 1915-1918.  (21)  Einsle, O.; Tezcan, F. A.; Andrade, S. L. A.; Schmid, B.; Yoshida, M.; Howard, J. B.; Rees, D. C. Science 2002, 297, 1696-1700.  (22)  Shibasaki, M.; Matsunaga, S.; Kumagai, N. Synlett 2008, 1583-1602.  (23)  Gianneschi, N. C.; Masar, M. S., III; Mirkin, C. A. Acc. Chem. Res. 2005, 38, 825-837.  (24)  Roundhill, D. M.; Dickson, M. K.; Dixit, N. S.; Sudha-Dixit, B. P. J. Am. Chem. Soc. 1980, 102, 5538-42.  (25)  Shopsowitz, K. E.; Edwards, D.; Gallant, A. J.; MacLachlan, M. J. Tetrahedron 2009, 65, 8113-8119.  (26)  MacLachlan, M. J. Pure Appl. Chem. 2006, 78, 873-888.  (27)  Gallant, A. J.; Hui, J. K. H.; Zahariev, F. E.; Wang, Y. A.; MacLachlan, M. J. J. Org. Chem. 2005, 70, 7936-7946.  (28)  Gallant, A. J.; Yun, M.; Sauer, M.; Yeung, C. S.; MacLachlan, M. J. Org. Lett. 2005, 7, 4827-4830.  (29)  Ma, C. T. L.; MacLachlan, M. J. Angew. Chem., Int. Ed. 2005, 44, 4178-4182.  (30)  Gallant, A. J.; MacLachlan, M. J. Angew. Chem., Int. Ed. 2003, 42, 5307-5310.  (31)  Akine, S.; Taniguchi, T.; Nabeshima, T. Tetrahedron Lett. 2001, 42, 8861-8864. 196  (32)  Huck, W. T. S.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Recl. Trav. Chim. Pays-Bas 1995, 114, 273-276.  (33)  Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2006, 45, 52485250.  (34)  Frischmann, P. D.; Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2008, 47, 101-112.  (35)  Nabeshima, T.; Miyazaki, H.; Iwasaki, A.; Akine, S.; Saiki, T.; Ikeda, C.; Sato, S. Chem. Lett. 2006, 35, 1070-1071.  (36)  Ohshima, T.; Iwasaki, T.; Maegawa, Y.; Yoshiyama, A.; Mashima, K. J. Am. Chem. Soc. 2008, 130, 2944-2945.  (37)  Iwasaki, T.; Maegawa, Y.; Hayashi, Y.; Ohshima, T.; Mashima, K. J. Org. Chem. 2008, 73, 5147-5150.  (38)  Ohshima, T.; Iwasaki, T.; Mashima, K. Chem. Commun. 2006, 2711-2713.  (39)  Sanville, E.; Burnin, A.; BelBruno, J. J. J. Phys. Chem. A 2006, 110, 2378-2386.  (40)  DeGroot, M. W.; Taylor, N. J.; Corrigan, J. F. Inorg. Chem. 2005, 44, 5447-5458.  (41)  Behrens, S.; Bettenhausen, M.; Eichhofer, A.; Fenske, D. Angew. Chem., Int. Ed. Engl. 1998, 36, 2797-2799.  (42)  Behrens, S.; Bettenhausen, M.; Deveson, A. C.; Eichhoefer, A.; Fenske, D.; Lohde, A.; Woggon, U. Angew. Chem., Int. Ed. Engl. 1996, 35, 2215-2218.  (43)  Dance, I. G.; Choy, A.; Scudder, M. L. J. Am. Chem. Soc. 1984, 106, 6285-6295.  (44)  Liang, X.-Q.; Zhou, X.-H.; Chen, C.; Xiao, H.-P.; Li, Y.-Z.; Zuo, J.-L.; You, X.Z. Cryst. Growth Des. 2009, 9, 1041-1053.  (45)  Wei, G.-H.; Yang, J.; Ma, J.-F.; Liu, Y.-Y.; Li, S.-L.; Zhang, L.-P. Dalton Trans. 2008, 3080-3092.  (46)  Liu, G.-X.; Huang, R.-Y.; Xu, H.; Kong, X.-J.; Huang, L.-F.; Zhu, K.; Ren, X.-M. Polyhedron 2008, 27, 2327-2336.  197  (47)  Yao, Y.-l.; Che, Y.-X.; Zheng, J.-M. Inorg. Chem. Commun. 2008, 11, 12531255.  (48)  Shi, Q.; Sun, Y.; Sheng, L.; Ma, K.; Hu, M.; Hu, X.; Huang, S. Cryst. Growth Des. 2008, 8, 3401-3407.  (49)  Chen, Z.-F.; Zhang, Z.-L.; Tan, Y.-H.; Tang, Y.-Z.; Fun, H.-K.; Zhou, Z.-Y.; Abrahams, B. F.; Liang, H. CrystEngComm 2008, 10, 217-231.  (50)  Wang, L.; Yang, M.; Li, G.; Shi, Z.; Feng, S. Inorg. Chem. 2006, 45, 2474-2478.  (51)  Wu, J.-Y.; Chang, C.-H.; Tseng, T.-W.; Lu, K.-L. J. Mol. Struct. 2006, 796, 6975.  (52)  Xiong, R.-G.; You, X.-Z.; Abrahams, B. F.; Xue, Z.; Che, C.-M. Angew. Chem., Int. Ed. 2001, 40, 4422-4425.  (53)  Kim, J.; Chen, B.; Reineke, T. M.; Li, H.; Eddaoudi, M.; Moler, D. B.; O'Keeffe, M.; Yaghi, O. M. J. Am. Chem. Soc. 2001, 123, 8239-8247.  (54)  Wang, X.-L.; Qin, C.; Lan, Y.-Q.; Shao, K.-Z.; Su, Z.-M.; Wang, E.-B. Chem. Commun. 2009, 410-412.  (55)  Xiao, D.; Yuan, R.; Chai, Y.; Wang, E. Eur. J. Inorg. Chem. 2008, 2610-2615.  (56)  Fang, Q.-R.; Zhu, G.-S.; Jin, Z.; Xue, M.; Wei, X.; Wang, D.-J.; Qiu, S.-L. Angew. Chem., Int. Ed. 2006, 45, 6126-6130.  (57)  Chun, H.; Kim, D.; Dybtsev, D. N.; Kim, K. Angew. Chem., Int. Ed. 2004, 43, 971-974.  (58)  Kajiwara, T.; Kobashi, T.; Shinagawa, R.; Ito, T.; Takaishi, S.; Yamashita, M.; Iki, N. Eur. J. Inorg. Chem. 2006, 1765-1770.  (59)  Zheng, S.-L.; Yang, J.-H.; Yu, X.-L.; Chen, X.-M.; Wong, W.-T. Inorg. Chem. 2004, 43, 830-838.  (60)  Wang, X.-L.; Qin, C.; Wang, E.-B.; Xu, L.; Su, Z.-M.; Hu, C.-W. Angew. Chem., Int. Ed. 2004, 43, 5036-5040.  (61)  Wang, R.; Hong, M.; Luo, J.; Cao, R.; Weng, J. Chem. Commun. 2003, 10181019.  198  (62)  Anantharaman, G.; Walawalkar, M. G.; Murugavel, R.; Gabor, B.; Herbst-Irmer, R.; Baldus, M.; Angerstein, B.; Roesky, H. W. Angew. Chem., Int. Ed. 2003, 42, 4482-4485.  (63)  Frischmann, P. D.; MacLachlan, M. J. Chem. Commun. 2007, 4480-4482.  (64)  Frischmann, P. D.; MacLachlan, M. J. Comments Inorg. Chem. 2008, 29, 26-45.  (65)  Dalgarno, S. J.; Power, N. P.; Atwood, J. L. Coord. Chem. Rev. 2008, 252, 825841.  (66)  Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7-13.  (67)  Mecozzi, S.; Rebek, J., Jr. Chem.--Eur. J. 1998, 4, 1016-1022.  (68)  Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew. Chem., Int. Ed. 2009, 48, 3418-3438.  (69)  Pinacho Crisostomo, F. R.; Lledo, A.; Shenoy, S. R.; Iwasawa, T.; Rebek, J., Jr. J. Am. Chem. Soc. 2009, 131, 7402-7410.  (70)  Gibb, C. L. D.; Sundaresan, A. K.; Ramamurthy, V.; Gibb, B. C. J. Am. Chem. Soc. 2008, 130, 4069-4080.  (71)  Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Science 2007, 316, 85-88.  (72)  Kong, X.-J.; Wu, Y.; Long, L.-S.; Zheng, L.-S.; Zheng, Z. J. Am. Chem. Soc. 2009, 131, 6918-6919.  (73)  Zhao, J.-W.; Jia, H.-P.; Zhang, J.; Zhengk, S.-T.; Yang, G.-Y. Chem.--Eur. J. 2007, 13, 10030-10045.  (74)  Zuhayra, M.; Kampen, W. U.; Henze, E.; Soti, Z.; Zsolnai, L.; Huttner, G.; Oberdorfer, F. J. Am. Chem. Soc. 2006, 128, 424-425.  (75)  Lee, D.; Sorace, L.; Caneschi, A.; Lippard, S. J. Inorg. Chem. 2001, 40, 67746781.  (76)  Ama, T.; Rashid, M. M.; Yonemura, T.; Kawaguchi, H.; Yasui, T. Coord. Chem. Rev. 2000, 198, 101-116.  199  (77)  Zhu, A.-X.; Zhang, J.-P.; Lin, Y.-Y.; Chen, X.-M. Inorg. Chem. 2008, 47, 73897395.  (78)  Chakraborty, J.; Thakurta, S.; Samanta, B.; Ray, A.; Pilet, G.; Batten, S. R.; Jensen, P.; Mitra, S. Polyhedron 2007, 26, 5139-5149.  (79)  Chen, Z.; Liang, F.; Zhou, S.; Xia, C.; Hu, R. J. Mol. Struct. 2007, 827, 20-24.  (80)  Suen, M.-C.; Wang, J.-C. Inorg. Chem. Commun. 2006, 9, 478-481.  (81)  Lo, W.-K.; Wong, W.-K.; Wong, W.-Y.; Guo, J. Eur. J. Inorg. Chem. 2005, 3950-3954.  (82)  Tong, M.-L.; Zheng, S.-L.; Shi, J.-X.; Tong, Y.-X.; Lee, H. K.; Chen, X.-M. J. Chem. Soc., Dalton Trans. 2002, 1727-1734.  (83)  Vinogradova, E. A.; Vassilyeva, O. Y.; Kokozay, V. N.; Skelton, B. W.; Bjernemose, J. K.; Raithby, P. R. J. Chem. Soc., Dalton Trans. 2002, 4248-4252.  (84)  Tong, M.-L.; Lee, H. K.; Zheng, S.-L.; Chen, X.-M. Chem. Lett. 1999, 10871088.  (85)  Lulinski, S.; Madura, I.; Serwatowski, J.; Zachara, J. Inorg. Chem. 1999, 38, 4937-4941.  (86)  Harms, K.; Merle, J.; Maichle-Moessmer, C.; Massa, W.; Krieger, M. Inorg. Chem. 1998, 37, 1099-1104.  (87)  Buesching, I.; Strasdeit, H. J. Chem. Soc., Chem. Commun. 1994, 2789-2790.  (88)  Boulmaaz, S.; Papiernik, R.; Hubert-Pfalzgraf, L. G.; Vaissermann, J.; Daran, J. C. Polyhedron 1992, 11, 1331-1336.  (89)  Liles, D. C.; McPartlin, M.; Tasker, P. A.; Lip, H. C.; Lindoy, L. F. J. Chem. Soc., Chem. Commun. 1976, 549-551.  (90)  Ouellette, W.; Hudson, B. S.; Zubieta, J. Inorg. Chem. 2007, 46, 4887-4904.  (91)  Poul, L.; Fakhfakh, M.; Taibi, M.; Jouini, N.; Herson, P.; Fievet, F. J. Chem. Crystallogr. 2005, 35, 285-291.  200  (92)  Fornies, J.; Gomez, J.; Lalinde, E.; Moreno, M. T. Inorg. Chem. 2001, 40, 54155419.  (93)  Lin, W.; Wang, Z.; Ma, L. J. Am. Chem. Soc. 1999, 121, 11249-11250.  (94)  Duhme, A. K.; Strasdeit, H. Eur. J. Inorg. Chem. 1998, 657-662.  (95)  Weidenbruch, M.; Herrndorf, M.; Schaefer, A.; Pohl, S.; Saak, W. J. Organomet. Chem. 1989, 361, 139-145.  (96)  Shen, X.-Q.; Yao, H.-C.; Yang, R.; Li, Z.-J.; Zhang, H.-Y.; Wu, B.-L.; Hou, H.W. Polyhedron 2007, 27, 203-209.  (97)  Dai, J.-C.; Wu, X.-T.; Fu, Z.-Y.; Cui, C.-P.; Hu, S.-M.; Du, W.-X.; Wu, L.-M.; Zhang, H.-H.; Sun, R.-Q. Inorg. Chem. 2002, 41, 1391-1396.  (98)  Garon, C. N.; Gorelsky, S. I.; Sigouin, O.; Woo, T. K.; Fontaine, F.-G. Inorg. Chem. 2009, 48, 1699-1710.  (99)  Sigouin, O.; Garon, C. N.; Delaunis, G.; Yin, X.; Woo, T. K.; Decken, A.; Fontaine, F.-G. Angew. Chem., Int. Ed. 2007, 46, 4979-4982.  (100) Purcell, K. F.; Kotz, J. C. Inorganic Chemistry; Holt Saunders: Philadelphia, 1977. (101) Ziegler, T. In Metal-ligand Interactions: From Atoms, to Clusters, to Surfaces Salahub, D. R., Russo, N., Eds.; NATO ASI Series 378 Kluwer Academic Publishers: Amsterdam, 1992. (102) Kang, J.; Rebek, J., Jr. Nature 1996, 382, 239-241. (103) Cram, D. J.; Choi, H. J.; Bryant, J. A.; Knobler, C. B. J. Am. Chem. Soc. 1992, 114, 7748-7765. (104) Horman, I.; Dreux, B. Helv. Chim. Acta 1984, 67, 754-764. (105) SPARTAN 04, Wavefunction Inc.: Irvine, CA USA, 2004. (106) Fairchild, R. M.; Holman, K. T. J. Am. Chem. Soc. 2005, 127, 16364-16365. (107) Rebek, J., Jr. Angew. Chem., Int. Ed. 2005, 44, 2068-2078.  201  (108) Power, N.; Dalgarno, S. J.; Atwood, J. L. Angew. Chem., Int. Ed. 2007, 46, 86018604. (109) Kobayashi, K.; Kitagawa, R.; Yamada, Y.; Yamanaka, M.; Suematsu, T.; Sei, Y.; Yamaguchi, K. J. Org. Chem. 2007, 72, 3242-3246. (110) Haino, T.; Kobayashi, M.; Chikaraishi, M.; Fukazawa, Y. Chem. Commun. 2005, 2321-2323. (111) Pluth, M. D.; Raymond, K. N. Chem. Soc. Rev. 2007, 36, 161-171. (112) Palmer, L. C.; Rebek, J., Jr. Org. Biomol. Chem. 2004, 2, 3051-3059. (113) Hof, F.; Craig, S. L.; Nuckolls, C.; Rebek, J., Jr. Angew. Chem., Int. Ed. 2002, 41, 1488-1508. (114) Pons, M.; Millet, O. Prog. Nucl. Magn. Reson. Spectrosc. 2001, 38, 267-324. (115) Tobe, M. L.; Burgess, J. Inorganic Reaction Mechanisms; Addison Wesley Longman Inc.: New York, 1999. (116) Marat, K. In Spinworks Version 3.0; University of Manitoba: 2008. (117) Chierotti, M. R.; Gobetto, R. Chem. Commun. 2008, 1621-1634. (118) Albunia, A. R.; Gaeta, C.; Neri, P.; Grassi, A.; Milano, G. J. Phys. Chem. B 2006, 110, 19207-19214. (119) Wittebort, R. J.; Olejniczak, E. T.; Griffin, R. G. J. Chem. Phys. 1987, 86, 541120. (120) Greenfield, M. S.; Ronemus, A. D.; Vold, R. L.; Vold, R. R.; Ellis, P. D.; Raidy, T. E. J. Magn. Res. (1969) 1987, 72, 89-107. (121) Balakrishnan, N. S. J. Magn. Reson. 1989, 83, 233-245. (122) Wann, M. H.; Harbison, G. S. J. Chem. Phys. 1994, 101, 231-237. (123) Sheldrick, G. M.; SHELXL-97, University of Göttingen: Göttingen, Germany, 1997. (124) Spek, A. L.; PLATON, Utrecht University: Utrecht, The Netherlands, 2002. 202  (125) Davis, J. H.; Jeffrey, K. R.; Bloom, M.; Valic, M. I.; Higgs, T. P. Chem. Phys. Lett. 1976, 42, 390-394. (126) Macho, V.; Brombacher, L.; Spiess, H. W. Appl. Magn. Reson. 2001, 20, 405432. (127) Te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca Guerra, C.; Van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001, 22, 931967. (128) Fonseca Guerra, C.; Snijders, J. G.; Te Velde, G.; Baerends, E. J. Theor. Chem. Acc. 1998, 99, 391-403. (129) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.; Gaussian 03, Revision D.02 ed.; Gaussian Inc.: Wallingford CT, USA, 2004. (130) van Lenthe, E.; Ehlers, A.; Baerends, E.-J. J. Chem. Phys. 1999, 110, 8943-8953. (131) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys. 1993, 99, 45974610. (132) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys. 1994, 101, 97839792. (133) Van Lenthe, E.; Van Leeuwen, R.; Baerends, E. J.; Snijders, J. G. Int. J. Quantum Chem. 1996, 57, 281-293. (134) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865-3868.  203  (135) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200-1211. (136) Schlegel, H. B. J. Comput. Chem. 1982, 3, 214-218. (137) Schlegel, H. B. Modern Electronic Structure Theory; World Scientific Publishing: Singapore, 1995; Vol. 2. (138) Schmider, H. L.; Becke, A. D. J. Chem. Phys. 1998, 108, 9624-9631. (139) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110, 6158-6170. (140) Binkley, J. S.; Pople, J. A.; Hehre, W. J. J. Am. Chem. Soc. 1980, 102, 939-947. (141) Gordon, M. S.; Binkley, J. S.; Pople, J. A.; Pietro, W. J.; Hehre, W. J. J. Am. Chem. Soc. 1982, 104, 2797-2803. (142) Pietro, W. J.; Francl, M. M.; Hehre, W. J.; DeFrees, D. J.; Pople, J. A.; Binkley, J. S. J. Am. Chem. Soc. 1982, 104, 5039-5048. (143) Dobbs, K. D.; Hehre, W. J. J. Comput. Chem. 1986, 7, 359-378. (144) Dobbs, K. D.; Hehre, W. J. J. Comput. Chem. 1987, 8, 880-893. (145) Dunning Jr., T. H.; Hay, P. J. Modern Theoretical Chemistry; Plenum: New York, 1976; Vol. 3. (146) Nicklass, A.; Dolg, M.; Stoll, H.; Preuss, H. J. Chem. Phys. 1995, 102, 89428952. (147) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56, 2257-2261. (148) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54, 724-728. (149) Hariharan, P. C.; Pople, J. A. Mol. Phys. 1974, 27, 209-214. (150) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213-222. (151) Gordon, M. S. Chem. Phys. Lett. 1980, 76, 163-168. (152) Boys, S. F.; Bernardi, F. Mol. Phys. 1970, 19, 553-566.  204  (153) Simon, S.; Duran, M.; Dannenberg, J. J. J. Chem. Phys. 1996, 105, 11024-11031.  205  Chapter 5 Design and Synthesis of New Schiff Base Macrocycles† 5.1  Introduction  5.1.1 Abstract The primary focus of this chapter is on methodology developed for the formation of a new family of metal-free [2+2] Schiff base macrocycles, 72-74, utilizing the differential exchange rates of aldimines and ketimines. The more robust ketimine bond is kinetically inert under the milder conditions used for aldimine bond formation. In particular, this route enables access to the first conjugated macrocycles with four unsymmetrical N2O2 salphen-like pockets. Also the synthesis of a new triptycene substituted [3+3] Schiff base macrocycle, 78, is described later in the chapter. Macrocycle 51c self-assembles into one dimensional nanofibers induced by alkali metal and ammonium salt complexation in the central crown ether-like cavity. The steric bulk of triptycene substituted macrocycle 78 prevents supramolecular fiber formation and is used as a counter example to help prove columnar self-assembly of 51c.  † Versions of this chapter have been published: a) Reproduced in part with permission from Frischmann, P. D.; Jiang, J.; Hui, J. K.-H.; Grzybowski, J. J.; MacLachlan, M. J. “Reversible-Irreversible Approach to Schiff Base Macrocycles: Access to Isomeric Macrocycyles with Multiple Salphen Pockets” Org. Lett. 2008, 10, 1255-1258. Copyright 2008 American Chemical Society. b) This is the pre-peer reviewed version of the following article: Hui, J. K.-H.; Frischmann, P. D.; Tso, C.-H.; Michal, C. A.; MacLachlan, M. J. “Spontaneous Hierarchical Assembly of Crown Ether-like Macrocycles into Nanofibers and Microfibers Induced by Alkali-Metal and Ammonium Salts” Chem. Eur. J. 2010, 16, 2453-2460.  206  207  5.1.2 Background Over the past 30 years many Schiff base macrocycles have been synthesized, often with the ultimate goal of metal complexation.1-9 The polydentate nature of Schiff base macrocycles makes them excellent scaffolds for the coordination of multiple metal ions. These multimetallic complexes frequently exhibit intriguing magnetic,10,11 catalytic,12,13 or supramolecular14-16 behavior. Our group has been especially interested in conjugated macrocycles with multiple salphen-type N2O2 pockets. These macrocycles have proven to be useful precursors to molecular metal clusters14,17-19 and nanotubular assemblies.20,21 The Schiff base condensation of geometrically programmed dialdehydes and diamines is a very convenient route to these macrocycles, as the reversibility of the reaction enables the preparation of macrocycles in high yield and purity without formation of oligomer and polymer. Although the reversibility of Schiff base condensation is convenient for preparing macrocycles, it leads to one significant drawback – this route is usually limited to highly symmetrical macrocycles. For example, our group has reported facile routes to conjugated [3+3] Schiff base macrocycles with average D3h symmetry,22-25 but templatefree routes to larger cycles are virtually unknown.26 Schiff base condensation is not generally useful for obtaining fully conjugated macrocycles with more than one type of imine. Attempts to make macrocycles with chemically distinct imines often leads to a mixture of products unless imines are reduced or coordinated to metals to prevent exchange. Salphens and salens have been synthesized with two different aldimines or one ketimine and one aldimine by stepwise condensation.27-29 This low symmetry in a single N2O2 pocket is useful for developing chirality in metal complexes30 which might in turn influence the products of catalysis. Despite the appeal of unsymmetric N2O2 pockets, methodology to make Schiff base macrocycles possessing them is lacking. This chapter describes a new route to Schiff base macrocycles with multiple salphen pockets, each having two distinct imines. The synthesis is achieved by taking advantage of the differential exchange rates of aldimines and ketimines with primary  208  amines. This route enables preparation of a new family of [2+2] Schiff base macrocycles, each having four unsymmetrical N2O2 pockets.  5.2  Discussion  5.2.1 Model Compounds and Macrocycle Precursors The synthesis of ketimine macrocycles from the condensation of diketones and diamines results in a mixture of products, each in low yield, due to the kinetic stability of the ketimine bond.31 Based on steric considerations, aldimines and ketimines should exhibit very different rates of exchange when reacted with another amine. To test this hypothesis, model substrates, aldimine 62 and ketimine 63, were synthesized by condensing p-anisidine with salicylaldehyde and 2-benzoyl-4-methylphenol, respectively (Fries rearrangement of phenylbenzoate ester heavily favors acylation of the para position so a methyl group is introduced to block para acylation). Solutions of 62 and 63 were prepared in CD3CN and combined with two equivalents of 3,5-dimethylaniline as shown in Scheme 5.1. At equilibrium a mixture of both aldimines or both ketimines should exist due to the similar electron donating properties of p-anisidine and 3,5dimethylaniline.32  Scheme 5.1. Aldimine and ketimine exchange equilibrium.  209  At 20 ºC, no exchange is observed by 1H NMR spectroscopy for either 62 or 63. When a solution of 62 in CD3CN is heated to 57 ºC, resonances assigned to free panisidine appear indicating formation of 64 through imine exchange (measured rate of exchange: 1 x 10-6 mol L-1 s-1). Conversely, when ketimine 63 is subjected to the same conditions for 36 h no exchange is observed. To confirm that this was a kinetic effect and not a thermodynamic equilibrium, a solution of ketimine 65 and two equivalents of panisidine in CH3CN was heated at 57 ºC for 1.5 h. No exchange is observed under these conditions. These results show that in hot acetonitrile, aldimine bonds are labile but ketimines are kinetically inert. This knowledge was applied to the synthesis of Schiff base macrocycles through a cascade  of  ketimine  condensation  followed  by  aldimine  condensation.  Dihydroxydibenzoylbenzenes 6733,34 and 6835 were prepared as shown in Scheme 5.2. Condensation of benzoin with catechol at 260 ºC afforded tetraphenyl-o-benzodifuran 66 in 25% yield. Oxidation with CrO3 followed by hydrolysis of the benzoate ester yields 67. Compound 68 was obtained from the Friedel-Crafts acylation/demethylation of 1,3dimethoxybenzene.  Scheme 5.2. Synthesis of dihydroxydibenzoyl compounds 67 and 68.  210  Scheme 5.3. Synthesis and thermal ellipsoid plot of model compound 69. Ellipsoids are at 50 % probability (C = grey, N = blue, O = red, H = yellow). To examine the reactivity of 67, it was combined with two equivalents of panisidine in refluxing MeOH as outlined in Scheme 5.3. Compound 69 was obtained in 82% yield and characterized by common spectroscopic techniques. Notably, a 3.6 ppm downfield shift is observed in the 1H NMR spectrum for the phenoxy resonance after condensing 67 to 69, indicative of strong intramolecular hydrogen-bonding between the imine and the phenoxy proton. A solid-state structure of 69 obtained from SCXRD analysis further supports the structural assignment and confirms that both ketimines are in the E configuration. It is known that the condensation of a diphenylketone with phenylenediamine stops after only one of the amines has reacted unless harsh conditions are utilized.36 Thus, condensation of 67 and 68 with dialkoxyphenylenediamines, 55b or 55c, afforded diketimine macrocycle precursors 70 and 71a-b, respectively (Scheme 5.4). These are air stable, bright red, crystalline compounds that are easily handled.  211  Scheme 5.4. Synthesis of 70 and 71a-b.  212  5.2.2 Synthesis and Characterization of [2+2] Macrocycles To demonstrate the reversible-irreversible Schiff base condensation approach to macrocycles with unsymmetrical N2O2 pockets, Scheme 5.5 outlines the reaction of diketimine 70 with one equivalent of 4,6-diformyl-1,3-dihydroxybenzene in refluxing CH3CN:CHCl3. It seemed likely that this reaction would yield the rectangular [2+2] Schiff base macrocycle 72, and this was indeed obtained. This macrocycle contains four equivalent N2O2 pockets, each incorporating one aldimine and one ketimine. Two downfield phenoxy resonances and one imine resonance are observed in the 1H NMR spectrum of macrocycle 72. The molecular ion, found at m/z 1946.4, dominates the MALDI-TOF mass spectrum providing further support for the structural assignment. Macrocycle 72 was isolated in 41% yield via the formation of four aldimine bonds with no evidence for scrambling of the ketimines.  Scheme 5.5. Synthesis of macrocycle 72. To further illustrate this approach, macrocycle 73 was synthesized by condensing 71a with 3,6-diformyl-1,2-dihydroxybenzene in CH3CN:CHCl3 as shown in Scheme 5.6. The 1H NMR spectrum of 73 is very similar to that for 72 and mass spectrometry, elemental analysis, and spectroscopy all support the reported structure. Macrocycles 72 and 73 are isomers, with the ketimine phenyl groups and imine protons interchanged. 213  Besides enantiomers and diastereomers, it is not usually possible to make isomeric forms of Schiff base macrocycles since the reversible condensation most often yields the highsymmetry, thermodynamic product.  Scheme 5.6. Synthesis of macrocycle 73. Based on a previous study from our group of [6+6] Schiff base macrocycles made from the reaction of 4,6-diformyl-1,3-dihydroxybenzene with 1,2-dialkoxy-4,5diaminobenzenes, the condensation of 4,6-diformyl-1,3-dihydroxybenzene with 71b was anticipated to yield a hexagon-shaped macrocycle, with the two reagents reacting in a 3:3 ratio.26 Surprisingly, however, 4,6-diformyl-1,3-dihydroxybenzene and 71b react to give exclusively a [2+2] Schiff base macrocycle, 74, in 26% yield (Scheme 5.7). MALDI-TOF mass spectrometry clearly indicates the [2+2] macrocycle is selectively formed, with no evidence of the anticipated [3+3] macrocycle. A MALDI-TOF MS analysis of the filtrate confirmed that no larger cyclization products were present. Macrocycle 74 appears highly strained and must adopt a non-planar geometry. Semi-empirical calculations indicate that it has a bowl shape, but to date, no single crystals suitable for SCXRD analysis have been obtained.  214  Scheme 5.7. Synthesis of Macrocycles 74.  5.2.3 Synthesis of Triptycene-Based [3+3] Macrocycle In an effort to expand the steric bulk at the periphery of macrocycles 51a-e, 1,4diformyl-2,3-dihydroxytriptycene, 77, was prepared as shown in Scheme 5.8. First, the known compound 2,3-dimethoxytriptycene, 75, was synthesized by an alternative procedure where 2,3-dibromotriptycene undergoes nucleophilic aromatic substitution with NaOMe catalyzed by CuBr in MeOH. Compound 75 was dilithiated with four equivalents of nBuLi in dry ether and then quenched with DMF to give 1,4-diformyl-2,3dimethoxytriptycene, 76, after work-up with water and acid. Removal of the methyl groups with BBr3 followed by acidification yielded the target compound 77 in 33% overall yield from 2,3-dibromotriptycene.  215  Scheme 5.8. Synthesis of 1,4-diformyl-2,3-dihydroxytriptycene 77. Schiff base condensation of 77 with diamine 55c in nitrogen sparged CH3CN heated at reflux for 14 h formed the anticipated [3+3] macrocycle, 78, in 51% yield (Scheme 5.9). This macrocycle exhibits much greater solubility in non-polar solvents than the 3,6-diformyl-1,2-dihydroxybenzene analogues 51a-e. ESI-MS confirmed the [3+3] product was isolated with the only peaks found at m/z = 1844.6 (78)+, 1867.6 (78+Na)+, 1883.4 (78+K)+. 1H and  13  C NMR spectroscopy also show macrocycle 78  adopts a D3h symmetric conformation in solution.  Scheme 5.9. Synthesis of triptycene substituted macrocycle 78.  216  5.2.4 Fiber Formation Inhibited by Triptycene Substituents Joseph Hui, a coworker in our research group, has been investigating the alkalimetal induced aggregation of macrocycle 51c. When a solution of macrocycle 51c in CHCl3 is treated with excess NaBF4 (itself nearly insoluble in chloroform), a color change from orange to deep red is observed. After filtration of the excess salts, the solution becomes noticeably viscous after a few minutes (a gel does not form, but a precipitate does after standing for several minutes). Samples of the viscous solution were dried on transmission electron microscopy (TEM) grids. Figures 5.1a,b show TEM images of [Na·51c]BF4. Surprisingly, the sample is organized into a fibrous morphology, where the diameters of the fibers are ca. 170 nm, considerably larger than the diameter of macrocycle 51c (ca. 2–3 nm). Scanning electron microscopy (SEM) also revealed the 3D structure of the fibers in the sample, Figures 5.1c,d. The bundles appear cylindrical in shape. Atomic force microscopy (AFM) in tapping mode showed that the samples are relatively smooth and approximately the same size as observed by TEM, Figures 5.1e,f.  217  Figure 5.1. a,b) TEM, c,d) SEM, and e,f) AFM micorgraphs of [Na·51c]BF4. All samples were prepared by drop-casting a chloroform solution of [Na·51c]BF4 onto formvar carbon-coated grids (TEM and AFM) or aluminum stubs (SEM) and dried at ambient condition.  As further proof for the hierarchical assembly mechanism involved in fiber formation of 51c, I prepared macrocycle 78 with triptycene substituents and Joseph Hui tested the ability of 78 to form fibers with alkali metals. Under identical experimental conditions employed for the assembly of macrocycle 51c, macrocycle 78 displayed no evidence for binding to NaBF4, supporting the assertion that the macrocycle stacking is important in the assembly of the polyelectrolyte. In the case of macrocycle 78, the bulky triptycenyl groups prevent columnar assembly and, therefore, ion binding and nanofiber formation.  218  5.3  Conclusions A new strategy has been developed for making Schiff base macrocycles that have  two inequivalent C=N bonds using a reversible-irreversible aldimine-ketimine condensation approach. This method allowed access to the first conjugated Schiff base macrocycles containing eight imines and multiple unsymmetrical N2O2 salphen-like pockets. It has since been demonstrated that macrocycles 72 and 73 exhibit flexible 1,3alternate geometries in solution, similar to calix[4]arene. In this conformation both 72 and 73 are capable of hosting organic cations such as pyridinium, tetraalkylammonium, or methyl viologen ions in the central cavity with Kassoc ranging from 102 to 105 mol-1 L.37 The reversible-irreversible imine condensation strategy presented here facilitated the synthesis of unsymmetrical isoscoles triangle shaped Schiff base macrocycles that also host organic cations in their interior.38 Triptycene substituted macrocycle 78 was isolated and attempts to self-assemble supramolecular fibers in CHCl3 upon addition of Na+ failed. This proved to be a valuable piece of evidence for understanding the columnar one-dimensional self-assembly of 51c with Na+.  5.4  Experimental  5.4.1 General The synthesis of N-salicylidene-p-anisidine, 62,39 has been previously reported but in this chapter it has been synthesized by a different route and extensively characterized. Full characterization of the previously synthesized but poorly characterized compounds 66, and 6733,34 has been included in this chapter. Synthesis of compounds 68,35 55b-c,22 2benzoyl-4-methylphenol,40 41  dihydroxybenzene  3,6-diformyl-1,2-dihydroxybenzene,24  2,3-dibromotriptycene,  42  4,6-diformyl-1,3-  and 4,5-diamino-1,2-dihexyloxybenzene43  were carried out according to literature procedures. Previously characterized 2,3dimethoxytriptycene,44 75, was prepared by an alternative route reported here. All 219  reactions were carried out under air unless otherwise noted. 1H and  13  C NMR spectra  were recorded on either a Bruker AV-300 or AV-400 spectrometer.  13  C NMR spectra  were recorded using a proton decoupled pulse sequence. 1H and  13  C NMR spectra were  calibrated to the residual protonated solvent at δ 7.27 and δ 77.23, respectively, in CDCl 3 or at δ 5.32 and δ 54.00, respectively, in CD2Cl2. UV-Vis spectra were obtained in CH2Cl2 or MeOH (ca. 1 x 10-6 mol L-1) on a Varian Cary 5000 UV-Vis-near-IR spectrophotometer using a 1 cm quartz cuvette. FT-IR spectra were collected neat in the solid state on a Thermo Nicolet 6700 FT-IR spectrometer or in a KBr pellet on a Nicolet 4700 FT-IR spectrometer. All MS were conducted at the UBC Microanalytical Services Laboratory. EI-MS was conducted on a Kratos MS-50 spectrometer. MALDI-TOF mass spectrometry was performed on a Bruker Biflex IV instrument with a dithranol matrix. Electrospray ionization (ESI) mass spectra were obtained on a Bruker Esquire LC instrument or Waters ZQ LC-MS in MeOH (ca. 1 x 10-4 mol L-1). Single-crystal XRD was performed on Bruker X8 APEX CCD with Mo radiation. Elemental analyses (C,H,N) were performed at the UBC Microanalytical Services Laboratory. Melting points were obtained on a Fisher-John’s melting point apparatus.  220  5.4.2 Procedures and Data Synthesis of N-salicylidene-p-anisidine (62): p-Anisidine (1.0 g, 8.1 mmol) was added to salicylaldehyde (850 μL, 8.1 mmol) while stirring, resulting in immediate precipitation of a yellow solid. The product was filtered and recrystallized from EtOH yielding 1.2 g of 62 as light green/yellow flakes (5.3 mmol, 65%). Data for 62: 13C NMR (100.6 MHz, CDCl3) δ 161.2, 160.7, 159.1, 141.6, 132.9, 132.2, 122.5, 119.6, 117.4, 114.8, 100.2, 55.8. 1H NMR (400 MHz, CDCl3) δ 13.42 (bs, 1H, OH), 8.62 (s, 1H, imine), 7.38 (m, 2H, Ar), 7.29 (m, 2H, Ar), 7.03 (d, 3JHH = 7.6 Hz, 1H, Ar), 6.96 (m, 3H, Ar), 3.85 (s, 3H, OCH3). UV-vis (MeOH) λmax (ε) = 441 (2.5 x 102), 348 (1.6 x 105), 270 (9.5 x 104), 229 (1.7 x 105) nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 1602, 1569, 1490, 1456, 1409, 1279, 1245, 1186, 1150, 1028, 836, 738 cm-1. ESI-MS (MeOH) m/z = 228.2 (62+H)+. Mp = 81 °C. Anal. Calc’d for C14H13NO2: C, 73.99; H 5.77; N, 6.16. Found: C, 73.90; H, 5.86; N, 6.12. Synthesis of p-anisidine ketimine (63): A Schlenk tube, charged with 2-benzoyl-4methylphenol (100 mg, 0.47 mmol) and p-anisidine (116 mg, 0.94 mmol), was heated to 110 °C. After 1 h, the dark brown/yellow solution was removed from heat and diluted with MeOH. The solution was placed in the freezer for 12 h yielding 145 mg of 63 as yellow crystals (0.46 mmol, 98%). Data for 63: 13C NMR (100.6 MHz, CDCl3) δ 172.5, 160.7, 156.9, 140.0, 134.7, 134.1, 132.0, 129.0, 128.5, 127.0, 124.1, 119.9, 117.9, 113.9, 100.2, 55.5, 20.8. 1H NMR (400 MHz, CDCl3) δ 14.68 (bs, 1H, OH), 7.37 (m, 3H, Ar), 7.17 (m, 3H, Ar), 6.98 (d, 3JHH = 8.0 Hz, 1H, Ar), 6.84 (s, 1H, Ar), 6.69 (m, 4H, Ar), 3.73 (s, 3H, OCH3), 2.16 (s, 3H, CH3). UV-vis (MeOH) λmax (ε) = 349 (7.2 x 104), 231 (1.9 x 105) nm (cm-1 mol-1 L). FTIR (neat) ῡ = 2949, 1740, 1605, 1567, 1489, 1458, 1439, 1234, 1033, 889, 830, 790, 540 cm-1. ESI-MS (MeOH) m/z = 318.2 (63+H)+. Mp = 162 °C. Anal. Calc’d for C21H19NO2: C, 79.47; H, 6.03; N, 4.41. Found: C, 79.40; H, 5.90; N, 4.45.  221  Synthesis of N-salicylidene-(3,5-dimethylaniline) (64): Salicylaldehyde (500 μL, 4.8 mmol) and 3,5-dimethylaniline (595 μL, 4.8 mmol) were mixed in a test tube and heated to 150 °C for 1 h. Upon cooling to room temperature, the crude mixture was filtered through a small pad of silica gel (10:1, hexanes:ether). The solvent was removed under reduced pressure followed by heating under reduced pressure to remove excess salicylaldehyde. The remaining 535 mg of yellow, viscous oil was compound 64 (2.4 mmol, 50%). Data for 64: 13C NMR (100.6 MHz, CDCl3) δ 162.4, 161.4, 148.6, 139.3, 133.1, 132.3, 128.8, 119.5, 119.1, 117.4, 100.1, 21.5. 1H NMR (400 MHz, CDCl3) δ 13.36 (bs, 1H, OH), 8.63 (s, 1H, imine), 7.39 (m, 2H, Ar), 7.04 (d, 3JHH 8.0 Hz, 1H, Ar), 6.95 (m, 4H, Ar), 2.38 (s, 6H, CH3). UV-vis (CH2Cl2) λmax (ε) = 341 (1.8 x 104), 270 (1.6 x 104) nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 3010, 2981, 2916, 2863, 1619, 1605, 1590, 1572, 1495, 1458, 1276, 1206, 1147, 843, 751, 685, 583 cm-1. ESI-MS (MeOH) m/z = 226.3 (3+H)+. Anal. Calc’d for C15H15NO: C, 79.97; H, 6.71; N, 6.22. Found: C, 80.02; H, 6.73; N, 6.24. Synthesis of 3,5-dimethylaniline ketimine (65): In a Schlenk tube, 2-benzoyl-4methylphenol (100 mg, 0.47 mmol) and 3,5-dimethylaniline (88 μL, 0.71 mmol) were mixed and heated to 140 °C under reduced pressure. After 30 min, the temperature was reduced to 120 °C where it was kept for 21 h. The crude yellow oil was purified by column chromatography in 10:1 hexanes:ether and then recrystallized from a chilled MeOH solution with a small amount of hexanes. Filtration of the yellow crystals gave 45 mg of compound 65 (0.14 mmol, 30%).  222  Data for 65: 13C NMR (100.6 MHz, CDCl3) δ 173.0, 160.7, 146.9, 138.3, 134.6, 134.2, 132.2, 129.0, 128.3, 127.1, 126.4, 120.5, 119.7, 118.0, 100.2, 21.4, 20.7. 1H NMR (400 MHz, CDCl3) δ 14.43 (bs, 1H, OH), 7.35 (m, 3H, Ar), 7.17 (m, 3H, Ar), 6.99 (d, 3JHH 8.0 Hz, 1H, Ar) 6.85 (s, 1H, Ar), 6.62 (s, 1H, Ar), 6.38 (s, 2H, Ar), 2.17 (s, 3H, CH3), 2.16 (s, 6H, CH3). UV-vis (CH2Cl2) λmax (ε) = 342 (8.4 x 103), 267 (1.5 x 104) nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 3025, 2917, 1734, 1604, 1568, 1488, 1442, 1332, 1285, 1234, 1132, 858, 843, 724, 681 cm-1. ESI-MS (MeOH) m/z = 316.2 (65+H)+. Mp = 111 °C. Anal. Calc’d for C22H21NO: C, 83.78; H, 6.71; N, 4.44. Found: C, 83.60; H, 6.73; N, 4.44. Data for tetraphenyl-o-benzodifuran (66):  13  C NMR (100.6 MHz, CDCl3) δ 150.3,  138.7, 132.9, 130.6, 129.9, 129.8, 129.0, 128.4, 128.2, 127.8, 126.9, 118.7, 115.2. 1H NMR (400 MHz, CDCl3) δ 7.76 (bd, 3JHH 6.9 Hz, 4H, Ar), 7.56 (bd, 3JHH 7.8 Hz, 4H, Ar), 7.51 (bt, JHH 7.3 Hz, 4H, Ar), 7.45 (m, 2H, Ar), 7.36 (m, 6H, Ar), 7.32 (s, 2H, Ar). UVvis (CH2Cl2) λmax (ε) = 340 (3.8 x 104), 263 (3.8 x 104) nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 1601, 1483, 1497, 1451, 1440, 1391, 1370, 1277, 1247, 1096, 1070, 944, 804, 767, 753, 694, 681, 646 cm-1. EI-MS m/z = 462 (66)+. Mp = 240 °C. Anal. Calc’d for C34H22O2: C, 88.29; H, 4.79. Found: C, 88.04; H, 4.82. Data for 3,6-dibenzoylcatechol (67):  13  C NMR (100.6 MHz, CDCl3) δ 201.0, 152.6,  137.4, 132.8, 129.5, 128.6, 122.3, 121.3. 1H NMR (400 MHz, CDCl3) δ 11.68 (s, 2H, OH), 7.75 (d, 3JHH 6.8 Hz, 4H, Ar), 7.63 (t, 3JHH 7.6 Hz, 2H, Ar), 7.53 (t, 3JHH 7.6 Hz, 4H, Ar), 7.12 (s, 2H, Ar). UV-vis (CH2Cl2) λmax (ε) = 405 (2.5 x 103), 301 (2.1 x 104) nm (cm1  mol-1 L). FT-IR (neat) ῡ = 1611, 1599, 1417, 1331, 1317, 1299, 1233, 1184, 988, 939,  861, 839, 793, 766, 625, 690 cm-1. EI-MS m/z = 318 (67)+, 241 (67-Ph)+, 212 (67COPh)+. Mp = 160 °C. Anal. Calc’d for C20H14O4: C, 75.46; H, 4.43. Found: C, 75.05; H, 4.48.  223  Synthesis of model diketimine (69): p-Anisidine (40 mg, 0.32 mmol) was added to a suspension of 3,6-dibenzoylcatechol, 68, (50 mg, 0.15 mmol) in 15 mL of MeOH and the reaction was heated to 70 °C. After 12 h, the reaction was cooled to room temperature. Upon standing, yellow crystals formed on the walls. The crystals were isolated by filtration giving 65 mg of compound 69 (0.12 mmol, 82%). The crystals obtained by this method were subjected to a single-crystal X-ray diffraction study. Data for 69: 13C NMR (100.6 MHz, CDCl3) δ 171.9, 157.3, 153.3, 139.5, 134.7, 129.1, 128.9, 128.6, 124.5, 121.1, 119.4, 114.0, 55.5. 1H NMR (400 MHz, CDCl3) δ 15.30 (s, 2H, OH), 7.34 (m, 6H, Ar) 7.16 (m, 4H, Ar), 6.74 (t, 3JHH 9.5 Hz, 4H, Ar) 6.71 (t, 3JHH 9.5 Hz, 4H, Ar), 6.39 (s, 2H, Ar), 3.74 (s, 6H, OCH3). UV-vis (CH2Cl2) λmax (ε) = 374 (1.6 x 104), 300 (1.8 x 104), 240 (2.2 x 104) nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 2837, 1604, 1491, 1429, 1328, 1240, 1175, 1028, 887, 832, 699, 565 cm-1. ESI-MS (MeOH) m/z = 529.3 (69+H)+. Mp = 233 °C. Anal. Calc’d for C34H28N2O4: C, 77.25; H, 5.34; N, 5.30. Found: C, 76.99; H, 5.39; N, 5.30. Synthesis of 2:1 dipentyloxydiamine:3,6-dibenzoylcatechol (70): In a Schlenk flask, 4,5-diamino-1,2-dipentyloxybenzene, 55b, (500 mg, 1.8 mmol) was dissolved in a mixture of nitrogen sparged toluene (20 mL) and piperidine (200 μL, 2.0 mmol). Addition of 3,6-dibenzoylcatechol, 67, (282 mg, 0.89 mmol) to the Schlenk flask resulted in the immediate evolution of a deep red color. The solution was heated to reflux for 12 h under N2. Upon cooling to room temperature, hexanes (80 mL) was added to the reaction flask resulting in the formation of a red ppt that was isolated by filtration to give 250 mg of 70 (0.30 mmol, 33%).  224  Data for 70: 13C NMR (100.6 MHz, CDCl3) δ 171.3, 152.5, 148.9, 141.1, 135.6, 135.0, 129.2, 128.8, 124.8, 121.6, 119.6, 110.8, 102.0, 70.3, 69.3, 29.1, 29.0, 28.4, 28.3, 22.7, 22.6, 14.3, 14.2. 1H NMR (400 MHz, CDCl3) δ 15.20 (s, 2H, OH), 7.35 (m, 6H, Ar), 7.21 (m, 4H, Ar), 6.44 (s, 2H, Ar), 6.30 (s, 2H, Ar), 5.88 (s, 2H, Ar), 3.89 (t, 3JHH 6.9 Hz, 4H, OCH2), 3.83 (bs, 4H, NH2), 3.37 (t, 3JHH 6.9 Hz, 4H, OCH2), 1.77 (m, 4H, OCH2CH2) 1.51 (m, 4H, OCH2CH2), 1.39 (m, 8H, alkyl), 1.28 (m, 8H, alkyl), 0.91 (m, 12H, CH3). UV-vis (CH2Cl2) λmax (ε) = 461 (1.7 x 104), 309 (4.0 x 104) nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 3319, 2954, 2930, 2858, 1608, 1584, 1506, 1465, 1422, 1328, 1261, 1213, 1124, 735, 697 cm-1. EI-MS m/z = 842 (70)+, 824 (70-H2O)+, 766 (70-Ph)+. Mp = 173 °C. Anal. Calc’d for C52H66N4O6: C, 74.08; H, 7.89; N, 6.65. Found: C, 74.41; H, 7.82; N, 6.76. Synthesis of 2:1 dipentyloxydiamine:4,6-diformyl-1,3-dihydroxybenzene (71a): In a Schlenk tube, under nitrogen, 4,6-diformyl-1,3-dihydroxybenzene, 68 (200 mg, 0.63 mmol), and 4,5-diamino-1,2-dipentyloxybenzene, 55b, (380 mg, 1.4 mmol) were mixed. With a heat gun, the mixture was heated to approximately 210 °C for 15 min until gas evolution was no longer observed from the dark red liquid. After cooling to room temperature, the dark red solid was dissolved in CH2Cl2 and concentrated under vacuum. A yellow precipitate formed upon addition of MeOH. The precipitate was isolated by filtration yielding 117 mg of 71a (0.14 mmol, 22%). Data for 71a: 13C NMR (100.6 MHz, CDCl3) δ 172.9, 169.0, 149.2, 142.2, 139.3, 135.7, 135.3, 129.8, 129.2, 129.1, 125.7, 114.3, 112.3, 106.0, 103.1, 71.1, 70.1, 29.9, 29.7, 29.1, 29.0, 23.4, 23.3, 15.0. 1H NMR (400 MHz, CDCl3): δ 15.74 (s, 2H, OH), 7.04 (m, 10H, Ar), 6.80 (s, 1H, Ar), 6.57 (s, 1H, Ar), 6.24 (s, 2H, Ar), 5.81 (s, 2H, Ar), 3.85 (t, 4H, OCH2), 3.66 (s, 4H, NH2), 3.34 (t, 4H, OCH2), 1.80-0.80 (m, 36H, alkyl). UV-vis (CH2Cl2) λmax (ε) = 394 (8.9 x 104), 306 (1.1 x 105), 243 (8.5 x 104) nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 3465, 3373, 2954, 2930, 2869, 1619, 1568, 1507, 1469, 1331, 1250, 1198, 1113, 988, 918, 846, 698 cm-1. MALDI-TOF MS (dithranol matrix) m/z = 842.2 (71a)+. Mp = 208 °C. Anal. Calc’d for C52H66N4O6: C, 74.08; H, 7.89; N, 6.65. Found: C, 74.06; H, 7.78; N, 6.58.  225  Synthesis of 2:1 dihexyloxydiamine:4,6-diformyl-1,3-dihydroxybenzene (71b): Under nitrogen, a Schlenk tube was charged with 4,6-diformyl-1,3-dihydroxybenzene, 68, (100 mg, 0.32 mmol) and 4,5-diamino-1,2-dihexyloxybenzene, 55c, (210 mg, 0.68 mmol) and heated with a heat gun to approximately 210 °C for 15 mins until gas evolution was no longer observed. After cooling to room temperature, the dark red solid was transferred with CH2Cl2 and concentrated under vacuum. Suspension of the remaining solid in MeOH followed by filtration gave 92 mg of 71b as a yellow powder (0.10 mmol, 32%). Data for 71b: 13C NMR (100.6 MHz, CDCl3) δ 172.7, 168.9, 149.1, 142.9, 139.0, 135.5, 135.1, 129.7, 129.2, 129.1, 125.7, 114.3, 114.1, 112.1, 105.0, 103.1, 78.2, 71.1, 70.1, 32.6, 32.5, 30.2, 30.0, 26.6, 26.5, 23.6, 15.0. 1H NMR (400 MHz, CDCl3) δ 15.76 (s, 2H, OH), 7.06 (m, 10H, Ar), 6.83 (s, 1H, Ar), 6.59 (s, 1H, Ar), 6.26 (s, 2H, Ar), 5.84 (s, 2H, Ar), 3.87 (t, 4H, OCH2), 3.68 (s, 4H, NH2), 3.37 (t, 4H, OCH2), 1.80-0.80 (m, 44H, alkyl). UV-vis (CH2Cl2) λmax (ε) = 422 (2.0 x 104), 283 (4.1 x 104) nm (cm-1 mol-1 L). FTIR (neat) ῡ = 3466, 3374, 2954, 2927, 2858, 1618, 1569, 1506, 1469, 1331, 1249, 1198, 916, 847, 698 cm-1. MALDI-TOF MS (dithranol matrix) m/z = 900.1 (71b+H+). Mp = 182 °C. Anal. Calc’d for C56H74N4O6: C, 74.80; H, 8.29; N, 6.23. Found: C, 73.98; H, 8.15; N, 6.06. Synthesis of [2+2] macrocycle (72): In a Schlenk flask, compound 70 (56.4 mg, 0.07 mmol), 4,6-diformyl-1,3-dihydroxybenzene (11.1 mg, 0.07 mmol), and piperidine (40 μL), were dissolved in 60 mL of dry CH3CN:CHCl3 (1:2). The flask was fit with a condenser and the reaction mixture was refluxed for 16 h under N2. After cooling to room temperature, the solvent was removed under reduced pressure. The remaining solids were suspended in MeOH and filtered yielding a red powder. The crude product was recrystallized from CH2Cl2/MeOH giving 56 mg of macrocycle 72 (0.03 mmol 41%).  226  Data for 72: 13C NMR (100.6 MHz, CDCl3) δ 175.8, 167.1, 158.4, 153.9, 149.4, 147.8, 137.5, 137.2, 135.9, 132.4, 129.7, 128.7, 121.7, 120.8, 114.4, 109.7, 106.1, 104.1, 78.2, 71.2, 70.8, 30.0, 29.8, 29.2, 29.1, 23.5, 23.4, 15.0. 1H NMR (400 MHz, CD2Cl2) δ 14.53 (s, 4H, OH), 14.31 (s, 4H, OH), 8.57 (s, 4H, imine), 7.49 (s, 2H, Ar), 7.20 (m, 20H, Ar), 6.80 (s, 4H, Ar), 6.54 (s, 2H, Ar), 6.45 (s, 4H, Ar), 6.19 (s, 4H, Ar), 3.95 (t, 8H, OCH 2), 3.68 (t, 8H, OCH2), 1.90-0.90 (m, 72H, alkyl). UV-vis (CH2Cl2) λmax (ε) = 420 (1.68 x 105), 284 (2.8 x 105) nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 2952, 2928, 2859, 1624, 1604, 1588, 1497, 1330, 1257, 1176, 1155, 849, 700 cm-1. MALDI-TOF MS (dithranol matrix) m/z = 1946.4 (72)+, 1968.5 (72+Na)+. Mp = dec. at 270 °C. HRMS (ESI) calc’d for C11913CH136N8O16: 1946.0108; found: m/z = 1946.0184 [72+].  12  Synthesis of [2+2] macrocycle (73): Prepared by the same method as macrocycle 72 using 71a (24 mg, 0.03 mmol), 3,6-diformyl-1,2-dihydroxybenzene (4.7 mg, 0.03 mmol), and piperidine (40 μL). The crude product was recrystallized from CH2Cl2/MeOH giving 20 mg of macrocycle 73 (0.01 mmol, 35%). Data for 73: 13C NMR (100.6 MHz, CDCl3) δ 173.5, 169.0, 161.5, 150.9, 149.2, 147.8, 139.9, 135.3, 134.3, 129.3, 129.2, 128.8, 122.2, 121.8, 113.7, 110.2, 106.4, 106.3, 77.7, 70.6, 70.0, 29.9, 29.6, 29.1, 29.0, 23.5, 23.4, 15.0. 1H NMR (400 MHz, CDCl3) δ 15.20 (s, 4H, OH), 12.89 (s, 4H, OH), 8.35 (s, 4H, imine), 6.95 (m, 26H, Ar), 6.61 (s, 4H, Ar), 6.52 (s, 2H, Ar), 6.14 (s, 4H, Ar), 3.91 (t, 8H, OCH2), 3.57 ( t, 8H, OCH2), 1.80-0.80 (m, 72H, alkyl). UV-vis (CH2Cl2) λmax (ε) = 439 (1.54 x 105), 388 (1.65 x 105), 338 (2.0 x 105), 285 (2.9 x 105) nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 2952, 2928, 2858, 1615, 1558, 1305, 1259, 1183, 992, 849, 740, 698 cm-1. MALDI-TOF MS (dithranol matrix) m/z = 1946.4 (73)+. Mp = 245 °C. Anal. Calc’d for C120H136N8O16·2MeOH: C, 72.88; H, 7.22; N, 5.57. Found: C, 72.32; H, 6.88; N, 5.63.  227  Synthesis of [2+2] macrocycle (74): Compound 71b (40 mg, 0.045 mmol) and 4,6diformyl-1,3-dihydroxybenzene (7.4 mg, 0.045 mmol) were dissolved in a mixture of chloroform (7 mL) and acetonitrile (9 mL) under N2. The orange solution was stirred at reflux (85 oC) for 24 h. Orange precipitate was observed after stirring overnight. After cooling to room temperature, the precipitate was isolated by filtration yielding 12 mg of macrocycle 74 (0.012 mmol, 26%). Data for 74: 13C NMR (100.6 MHz, CD2Cl2) δ 174.4, 168.2, 166.7, 157.8, 148.9, 147.5, 139.5, 137.1, 136.2, 134.6, 132.5, 129.0, 128.6, 128.2, 113.9, 113.4, 109.2, 105.3, 105.1, 103.0, 70.2, 69.8, 32.2, 32.1, 29.9, 29.5, 26.3, 26.1, 23.2, 14.4. 1H NMR (300 MHz, CD2Cl2) δ 14.97 (s, 4H, OH), 14.27 (s, 4H, OH), 8.43 (s, 4H, imine), 7.36 (s, 2H, Ar), 6.99 (m, 12H, Ar), 6.86 (s, 4H, Ar), 6.83 (s, 4H, Ar), 6.73 (s, 4H, Ar), 6.68 (d, 4H, Ar), 6.48 (s, 2H, Ar), 6.14 (s, 4H, Ar), 3.92 (t, 8H, OCH2), 3.67 (t, 8H, OCH2), 1.75 (m, 8H, CH2), 1.60 (m, 8H, CH2), 1.42 (m, 8H, CH2), 1.31 (m, 40H, CH2), 0.90 (m, 24H, CH3). UV-vis (CH2Cl2) λmax (ε) = 283 (1.19 x 105), 388 (8.13 x 104) nm (cm-1 mol-1 L). FT-IR (KBr) ῡ = 3459, 3060, 2954, 2930, 2869, 2859, 1625, 1604, 1585, 1504, 1469, 1444, 1388, 1378, 1365, 1328, 1260, 1210, 1179, 1151, 1135, 1104, 1074, 1044, 1013, 962, 916, 890, 850, 774, 741, 725, 699, 647, 610, 543 cm-1. MALDI-TOF MS (dithranol matrix) m/z = 2059.6 (74+H)+. Mp > 270 oC. Anal. Calc’d for C128H152N8O16: C, 74.68; H, 7.44; N, 5.44. Found: C, 74.44; H, 7.39; N, 5.32. Synthesis of 2,3-dimethoxytriptycene (75): A mixture of 2,3-dibromotriptycene (1.000 g, 2.44 mmol), copper(I) bromide (0.079 g, 0.55 mmol), 25 weight% sodium methoxide in methanol (5 mL, 22 mmol), ethyl acetate (0.5 mL), and toluene (10 mL) was refluxed under nitrogen overnight. The solution was cooled and then quenched with the addition of water. After extracting the aqueous layer with dichloromethane, the combined organic layers were dried over anhydrous magnesium sulfate and filtered. The solvent was removed by rotary evaporation to give the desired product as a white solid (0.652 g, 85%).  228  Synthesis of 1,4-diformyl-2,3-dimethoxytriptycene (76): Under a nitrogen atmosphere, a Schlenk flask was charged with 20 mL of dry ether, 2,3-dimethoxytriptycene 75 (4.589 g, 14.7 mmol), and dry TMEDA (5.3 mL, 35.3 mmol). The cloudy solution was chilled to 0° C, nBuLi (1.6 mol L-1 in hexanes, 40 mL, 58.8 mmol) was added dropwise over 30 min, and the brown suspension was then stirred for 16 hours at room temperature. Dry DMF (5.1 mL, 66.0 mmol) was added and the suspension was left to stir for 30 min followed by acidification with HCl/H2O and extraction into ether. The combined organic layers were dried with Na2SO4, filtered, and the solvent was removed under vacuum. The crude mixture was then chromatographed with a 2:1 hexanes:DCM solution. The first yellow fraction was 1,4-diformyl-2,3-dimethoxytriptycene, 76, isolated as a pale yellow solid upon solvent removal under vacuum (2.285 g, 42%). 76 was recrystallized from petroleum ether and diethyl ether for elemental analysis. Data for 76: 13C NMR (100.6 MHz, CDCl3) δ 191.9, 152.8, 144.6, 144.1, 129.1, 125.6, 124.3, 62.2, 47.9. 1H NMR (400 MHz, CDCl3) δ 10.57 (s, 2H, formyl), 7.46 (m, 4H, Ar), 7.04 (m, 4H, Ar), 6.86 (s, 2H, bridgehead), 3.91 (s, 6H, OMe). UV-vis (CH2Cl2) λmax (ε) = 273 (2.2 x 104), 278 (2.2 x 104), 373 (7.8 x 103), nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 3004, 2972, 2937, 2858, 1687, 1562, 1444, 1379, 1300, 1260, 1132, 1156, 1069, 1013, 986, 946, 746, 578 cm-1. APCI-MS (MeOH) m/z = 371 (76)+. Mp = 190 °C. Anal. Calc’d for C24H18O4: C, 77.82; H, 4.90. Found: C, 77.61; H, 4.85. Synthesis of 1,4-diformyl-2,3-dihydroxytriptycene (77): A Schlenk flask was charged with 1,4-diformyl-2,3-dimethoxytriptycene, 76, followed by dissolution in 200 mL of dry DCM. The yellow solution was chilled to 0 °C and BBr3 (2.6 mL, 28 mmol) was added yielding a fuming pink solution that was left stirring at room temperature under nitrogen for 16 h. The solution was then poured over ice, extracted into DCM, the combined organic fractions were dried with MgSO4, filtered, and the solvent was removed under vacuum yielding 1,4-diformyl-2,3-dihydroxytriptycene, 77, as an orange solid (1.962 g, 93%). 77 was recrystallized from cold ether and hexanes for elemental analysis.  229  Data for 77: 13C NMR (100.6 MHz, CDCl3) δ 193.5, 147.0, 144.1, 139.7, 125.9, 123.8, 119.4, 47.3. 1H NMR (400 MHz, CDCl3) δ 10.80 (s, 2H, OH/formyl), 10.69 (s, 2H, OH/formyl), 7.44 (m, 4H, Ar), 7.07 (m, 4H, Ar), 6.23 (s, 2H, bridgehead). UV-vis (CH2Cl2) λmax (ε) = 253 (6.5 x 103), 294 (1.9 x 104), 453 (4.9 x 103), nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 3041, 2996, 2922, 2852, 1644, 1556, 1458, 1436, 2895, 1273, 1198, 962, 921, 761, 675, 625, 600, 568 cm-1. ESI-MS (MeOH) m/z = 341.3 (77)-. Mp = dec. at 220 °C. Anal. Calc’d for C22H14O4: C, 77.18; H, 4.12. Found: C, 76.83; H, 4.19. Synthesis of [3+3] hexyloxy triptycene macrocycle (78): Under a nitrogen atmosphere, a Schlenk flask was charged with 4,5-diamino-1,2-dihexyloxybenzene, 55c (31.5 mg, 0.1 mmol), and 1,4-diformyl-2,3-dihydroxytriptycene, 77, (35 mg, 0.1 mmol) followed by the addition of 6 mL of nitrogen sparged acetonitrile. The solution immediately turned deep red and was refluxed under nitrogen for 14 h, cooled to room temperature, and then chilled at -10 °C for 48 h. The red suspension was filtered and dried under vacuum to yield 78 as an orange powder (32 mg, 51%). Data for 78: 13C NMR (100.6 MHz, CDCl3) δ 159.3, 149.4, 147.5, 145.0, 136.4, 136.0, 125.2, 123.2, 116.6, 107.0, 70.4, 48.5, 31.7, 29.4, 25.8, 22.7, 14.1. 1H NMR (400 MHz, CDCl3) δ 13.04 (s, 6H, OH), 9.27 (s, 6H, imine), 7.18 (m, 12H, Ar), 7.00 (s, 6H, Ar), 6.80 (m, 12H, Ar), 5.83 (s, 6H, bridgehead), 4.23 (t, 3JHH 6.6 Hz, 12H, OCH2), 1.97 (m, 12H, CH2), 1.62 (m, 12H, CH2), 1.45 (m, 24H, CH2), 0.98 (t, 3JHH 6.9 Hz, 18H, CH3). UV-vis (CH2Cl2) λmax (ε) = 284 (5.8 x 104), 330 (7.1 x 104), 417 (1.2 x 105) nm (cm-1 mol-1 L). FT-IR (neat) ῡ = 3726, 3708, 3627, 2925, 2854, 1606, 1506, 1462, 1426, 1375, 1304, 1257, 1173, 1115, 990, 759, 741, 624 cm-1. ESI-MS (MeOH) m/z = 1844.6 (78)+, 1867.6 (78+Na)+, 1883.4 (78+K)+. Mp > 260 °C. Anal. Calc’d for C120H126O12N6·H2O: C, 77.39; H, 6.93; N, 4.51. Found: C, 77.63; H, 7.28; N, 4.52.  230  5.4.3 Kinetic Studies An NMR tube was charged with 500 μL of a 10 mmol L-1 solution of N-salicylidene-panisidine, 62, in CD3CN (2.3 mg in 1.0 mL of CD3CN). Two molar equivalents of 3,5dimethylaniline was introduced to the NMR tube via the addition of 60 μL of a 160 mmol L-1 solution (20 μL of 3,5-dimethylaniline in 1 mL of CD3CN). No change was observed at 20 °C over 20 min by 1H NMR spectroscopy . The NMR spectrometer was preheated to 57 °C followed by introduction of the same sample. Resonances assigned to free p-anisidine were observed after 3 min indicating imine exchange was occurring. The initial 172 s allowed the sample to reach the desired temperature and were subtracted from the total time. Ten spectra were collected over the course of 1.3 h. The concentration of p-anisidine was determined by integration of the 1H NMR resonances and the experimental rate of exchange was found to be 0th order with respect to p-anisidine until 20 mins had elapsed (Figure 5.2). After 20 mins the reverse reaction began competing, leading to a deviation from the initially observed 0th order rate. At 57 °C the same experiment was attempted on p-anisidine ketimine 63. No exchange was observed over 36 h. To ensure this was a kinetic effect and not a thermodynamic one, 3,5-dimethylaniline ketimine 65 was exposed to two equivalents of p-anisidine in CD3CN at 57 °C. No exchange was observed over 1.5 h for this reaction either.  231  Figure 5.2. Concentration of free p-anisidine vs time upon introduction of two equivalents of 3,5-dimethylaniline to N-salicylidene-p-anisidine, 62, at 57 °C.  5.4.4 Crystallography Crystals of model diketimine 69 suitable for X-ray diffraction were obtained from hot MeOH. X-ray crystal data for 69: C34H28N2O4, Mw = 528.58 g mol-1 , orange prism (0.30 x 0.30 x 0.20 mm3), orthorhombic, space group Pbca (#61), a = 16.1351(9), b = 15.7795(12), c = 21.2316(15) Å, D = E = J = 90°, V = 5405.6(6) Å3, Z = 8, ρcalcd = 1.299 g cm-3, F000 = 2224, MoKα radiation, λ = 0.71073 Å, T = 173(2) K, 2θmax = 44.94°, 29621 reflections collected, 5536 were unique (Rint = 0.0609). Final GoF = 0.931, R1 = 0.0437, wR2 = 0.1013, R indices based on 3287 reflections with I>2σ(I). The structure was solved by direct methods45 and the refinement was performed using SHELXL-9746.All nonhydrogen atoms were refined anisotropically. All hydrogen atoms were located through difference mapping and refined isotropically.  232  5.5  References  (1)  Borisova, N. E.; Reshetova, M. D.; Ustynyuk, Y. A. Chem. Rev. 2007, 107, 4679.  (2)  Vigato, P. A.; Tamburini, S.; Bertolo, L. Coord. Chem. Rev. 2007, 251, 13111492.  (3)  Tomat, E.; Cuesta, L.; Lynch, V. M.; Sessler, J. L. Inorg. Chem. 2007, 46, 62246226.  (4)  Croucher, P. D.; Klingele, M. H.; Noble, A.; Brooker, S. Dalton Trans. 2007, 4000-4007.  (5)  Givaja, G.; Volpe, M.; Leeland, J. W.; Edwards, M. A.; Young, T. K.; Darby, S. B.; Reid, S. D.; Blake, A. J.; Wilson, C.; Wolowska, J.; McInnes, E. J. L.; Schroeder, M.; Love, J. B. Chem.--Eur. J. 2007, 13, 3707-3723.  (6)  Kwit, M.; Plutecka, A.; Rychlewska, U.; Gawronski, J.; Khlebnikov, A. F.; Kozhushkov, S. I.; Rauch, K.; de Meijere, A. Chem.--Eur. J. 2007, 13, 86888695.  (7)  Sessler, J. L.; Tomat, E.; Lynch, V. M. Chem. Commun. 2006, 4486-4488.  (8)  Kuhnert, N.; Burzlaff, N.; Patel, C.; Lopez-Periago, A. Org. Biomol. Chem. 2005, 3, 1911-1921.  (9)  Shimakoshi, H.; Takemoto, H.; Aritome, I.; Hisaeda, Y. Tetrahedron Lett. 2002, 43, 4809-4812.  (10)  Tandon, S. S.; Bunge, S. D.; Thompson, L. K. Chem. Commun. 2007, 798-800.  (11)  Tandon, S. S.; Thompson, L. K.; Bridson, J. N.; Benelli, C. Inorg. Chem. 1995, 34, 5507-5515.  (12)  Na, S. J.; Joe, D. J.; Sujith, S.; Han, W.-S.; Kang, S. O.; Lee, B. Y. J. Organomet. Chem. 2006, 691, 611-620.  (13)  Thirumavalavan, M.; Akilan, P.; Kandaswamy, M.; Chinnakali, K.; Kumar, G. S.; Fun, H. K. Inorg. Chem. 2003, 42, 3308-3317.  233  (14)  Frischmann, P. D.; MacLachlan, M. J. Chem. Commun. 2007, 4480-4482.  (15)  MacLachlan, M. J. Pure Appl. Chem. 2006, 78, 873-888.  (16)  Huang, W.; Gou, S.; Hu, D.; Chantrapromma, S.; Fun, H.-K.; Meng, Q. Inorg. Chem. 2001, 40, 1712-1715.  (17)  Frischmann, P. D.; Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2008, 47, 101-112.  (18)  Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2006, 45, 1442-1444.  (19)  Nabeshima, T.; Miyazaki, H.; Iwasaki, A.; Akine, S.; Saiki, T.; Ikeda, C.; Sato, S. Chem. Lett. 2006, 35, 1070-1071.  (20)  Gallant, A. J.; MacLachlan, M. J. Angew. Chem., Int. Ed. 2003, 42, 5307-5310.  (21)  Ma, C. T. L.; MacLachlan, M. J. Angew. Chem., Int. Ed. 2005, 44, 4178-4182.  (22)  Gallant, A. J.; Hui, J. K. H.; Zahariev, F. E.; Wang, Y. A.; MacLachlan, M. J. J. Org. Chem. 2005, 70, 7936-7946.  (23)  Gallant, A. J.; Yun, M.; Sauer, M.; Yeung, C. S.; MacLachlan, M. J. Org. Lett. 2005, 7, 4827-4830.  (24)  Akine, S.; Taniguchi, T.; Nabeshima, T. Tetrahedron Lett. 2001, 42, 8861-8864.  (25)  Huck, W. T. S.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Recl. Trav. Chim. Pays-Bas 1995, 114, 273-276.  (26)  Hui, J. K. H.; MacLachlan, M. J. Chem. Commun. 2006, 2480-2482.  (27)  Curreli, S.; Escudero-Adan, E. C.; Benet-Buchholz, J.; Kleij, A. W. J. Org. Chem. 2007, 72, 7018-7021.  (28)  Holbach, M.; Zheng, X.; Burd, C.; Jones, C. W.; Weck, M. J. Org. Chem. 2006, 71, 2903-2906.  (29)  Dalla Cort, A.; Gasparrini, F.; Lunazzi, L.; Mandolini, L.; Mazzanti, A.; Pasquini, C.; Pierini, M.; Rompietti, R.; Schiaffino, L. J. Org. Chem. 2005, 70, 8877-8883.  234  (30)  Dalla Cort, A.; Mandolini, L.; Palmieri, G.; Pasquini, C.; Schiaffino, L. Chem. Commun. 2003, 2178-2179.  (31)  Higuchi, M.; Kanazawa, H.; Yamamoto, K. Org. Lett. 2003, 5, 345-347.  (32)  Schultz, D.; Nitschke, J. R. J. Am. Chem. Soc. 2006, 128, 9887-9892.  (33)  Dischendorfer, O.; Limontschew, W. Monatsh. Chem. 1949, 80, 741-748.  (34)  Limontschew, W.; Wiesenberger, E. Monatsh. Chem. 1952, 83, 137-143.  (35)  Abdul-Aziz, M.; Auping, J. V.; Meador, M. A. J. Org. Chem. 1995, 60, 13031308.  (36)  Atkins, R.; Brewer, G.; Kokot, E.; Mockler, G. M.; Sinn, E. Inorg. Chem. 1985, 24, 127-134.  (37)  Jiang, J.; MacLachlan, M. J. Chem. Commun. 2009, 5695-5697.  (38)  Jiang, J.; MacLachlan, M. J. Org. Lett. 2010, 12, 1020-1023.  (39)  Ibrahim, M. N.; Hamad, K. J.; Al-Joroshi, S. H. Asian J. Chem. 2006, 18, 24042406.  (40)  Wei, J.-f.; Chen, G.; Shi, X.-y.; Lei, N. Shenyang Huagong Xueyuan Xuebao 2004, 18, 259-262.  (41)  Worden, L. R.; Kaufman, K. D.; Smith, P. J.; Widiger, G. N. J. Chem. Soc. C 1970, 227-230.  (42)  Hart, H.; Bashir-Hashemi, A.; Luo, J.; Meador, M. A. Tetrahedron 1986, 42, 1641-1654.  (43)  Kim, D.-H.; Choi, M. J.; Chang, S.-K. Bull. Korean Chem. Soc. 2000, 21, 145147.  (44)  Gong, K.; Zhu, X.; Zhao, R.; Xiong, S.; Mao, L.; Chen, C. Anal. Chem. 2005, 77, 8158-8165.  (45)  Burla, M. C.; Camalli, M.; Carrozzini, B.; Cascarano, G. L.; Giacovazzo, C.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 2003, 36, 1103.  235  (46)  Sheldrick, G. M.; SHELXL-97, University of Göttingen: Göttingen, Germany, 1997.  236  Chapter 6 Columnar Organization of Head-to-Tail SelfAssembled Pt4 Rings† 6.1  Introduction  6.1.1 Abstract Coordination of Pt2+ to a family of tunable Schiff base pro-ligands directs the twelve-component self-assembly of disc-shaped Pt4 rings, 90a-d, in a head-to-tail fashion. Aggregation of these S4 symmetric Pt4 macrocycles into columnar architectures was investigated by dynamic and static light scattering, NMR spectroscopy, powder Xray diffraction, and transmission electron microscopy. Data from these experiments supports the formation of columnar architectures for all of the structures studied, except when bulky tris(4-tbutylphenyl)methyl substituents were present. In this case, aggregation was limited to dimers in CHCl3 (Kdim = 3200 ± 200 L mol-1 at 25 °C) and a thermodynamic analysis revealed dimerization is an entropy driven process. Columnar architectures of Pt4 rings with branched 2-hexyldecyl substituents organize into lyotropic mesophases in non-polar organic solvents. These new self-assembled supramolecules are promising candidates to access nanotubes with multiple linear arrays of Pt2+ ions.  †A version of this chapter has been published: Reproduced in part with permission from Frischmann, P. D.; Guieu, S.; Tabeshi, R.; MacLachlan, M. J. “Columnar Organization of Head-to-Tail Self-Assembled Pt4 Rings” J. Am. Chem. Soc. 2010, 132, 7668-7675. Copyright 2010 American Chemical Society.  237  6.1.2 Background Self-assembly has emerged as a powerful technique to organize small molecules into functional materials with minimal effort.1-4 The morphology of self-assembled materials is dictated by the geometry and chemical functionality of the monomeric constituents.5,6 Disc-shaped molecules composed primarily of aromatic units often code for the assembly of one-dimensional fibers or non-covalent nanotubes and are being pursued as materials for photovoltaic active layers, sensors, and molecular wires.7-15 Solvophobic and weak intermolecular forces such as hydrogen-bonding or π-π interactions dominate the self-assembly process of these systems. Planar coordination complexes also assemble into fibers where anisotropic metalmetal interactions aid in assembly and impart unique properties, often inaccessible with purely  organic  analogues.  Phosphorescent,  electrophosphorescent,  luminescent,  electroluminescent, vapoluminescent, and semi-conducting wires have been constructed via self-assembly of neutral or cationic Pt2+ complexes coordinated by chelating aromatic ligands.16-22 Similar complexes form luminescent metallogels, chromonic liquid crystals,  238  nanosheets, and wheel-shaped superstructures through a subtle balance of Coulombic, platinum(II)-platinum(II), and/or π-π interactions.23-30 The exciting properties exhibited by these systems hold promise for development of novel materials. Platinum-pyridyl chemistry has been extensively investigated in the synthesis of complex molecules. Typically, dipyridyl-based ligands are combined with cis or trans protected Pt2+ complexes, and highly charged metallocycles are formed.31-34 This approach has become a paradigm for development of functional polyhedra, catenanes, and other unique architectures.35-40 Despite the variety of interesting structures that have emerged from this method, significant limitations remain. For example, the high charge of the metallocycle has prevented the observation of stacking in solution, a property that could be used to assemble nanotubes that include metal-metal bonding.41 Also, only high symmetry objects are usually accessible through this method. Neutral Schiff base platinum(II) complexes with trans-N2O2 donors, such as the monomer depicted in Figure 6.1, have been reported42 and Bosnich demonstrated that similar complexes are sterically unencumbered for one-dimensional assembly.43-45 A solid-state structure revealed close axial Pt···Pt contacts (3.26 Å) between a platinum(II) Schiff base complex and a metallated bis(tertpyridyl) pincer host, facilitated by π-π and metallophilic interactions.46 With the goal of assembling cycles that may ultimately show metal-metal bonding in a columnar orientation, I envisioned a ligand system that codes for the head-to-tail self-assembly of disc-shaped platinum-containing metallocycles as illustrated in Figure 6.1.  239  Figure 6.1. Reported N2O2 Pt2+ Schiff base monomer and conceptual evolution of the monomer into a head-to-tail self-assembling metallocycle. When a single molecule is outfitted with a donor-acceptor pair appropriately distributed to prohibit intramolecular recognition, one end of the molecule recognizes the other end in an intermolecular fashion. The spatial arrangement of this self-recognition determines whether polymers or macrocycles are isolated. Head-to-tail self-assembly facilitated in this fashion has been used to prepare a variety of supramolecules.47-59 Through hydrogen bonds and the aid of an alkali metal, G-quartets are a natural example of head-to-tail assembling supramolecules, and synthetic analogues are known to exhibit columnar aggregation.60,61 In this chapter, a new route to Pt-pyridyl type metallocycles using a head-to-tail synthetic strategy is discussed. Self-recognition is achieved by incorporating a coordinating pyridyl ligand and an open (or solvent-occupied) Pt coordination site into the same molecule. Specifically, tetrameric metallocycles are constructed in a selective twelve component, one-pot self-assembly directed by tetradentate Schiff base N-ONO donor pro-ligands formed either in situ or prior to metallation. Supramolecular aggregation of individual Pt4 metallocycles into columnar arrays is observed and bestows liquid crystalline properties upon the materials. These parallel columnar arrays are exciting materials with potential for anisotropic Pt-Pt interactions.  240  6.2  Discussion  6.2.1 Model Complex Synthesis, Solid-State Structure, and Pyridine Binding I set out to construct Pt-containing metallocycles like the one shown on the righthand side of Figure 6.1. To test the feasibility of this approach and to determine the expected ring size, model complex 80a was prepared, where the pyridyl group, essential for self-recognition, was substituted by a phenyl group (Scheme 6.1). The results of a single-crystal X-ray diffraction (SCXRD) analysis on complex 80a, depicted in Figure 6.2, shows that Pt2+ is complexed by the ONO Schiff base pocket, and the fourth coordination site is occupied by an S-bound DMSO molecule. Besides the expected rotation of the peripheral phenyl group to relieve steric repulsion between protons (normal for a biphenyl-type system), the molecule is nearly planar. Moreover, the angle of interest for self-assembly, shown in Figure 6.2a, is approximately 90°, indicating that head-to-tail self-assembly should direct formation of cyclic tetramers if the phenyl group is replaced by a 3-pyridyl unit.  Scheme 6.1. Synthesis of model complex 80a.  241  Figure 6.2. Solid-state structure of complex 80a. a) Top-down view highlighting the 90° angle that directs tetrameric self-assembly when the peripheral phenyl group is replaced by a 3-pyridyl group. b) Side-on view. Hydrogen atoms have been omitted (C = green, N = blue, O = red, Pt = yellow, S = purple). When a solution of model complex 80a in DMSO-d6 was titrated with pyridine and monitored by 1H NMR spectroscopy, resonances assigned to both free and coordinated pyridine were evident (Figure 6.3). From integration of these resonances, the equilibrium constant (Kpyr) was determined to be 35 ± 14 mol-1 L at 25 °C. This favorable Pt2+-pyridine interaction, even in a competing coordinating solvent, combined with the entropy gain expected upon forming the metallocycle (from displaced solvent) was anticipated to drive the self-assembly of 90a-d.  242  Figure 6.3. a) Equilibrium between DMSO-bound model complex 80a and pyridinebound model complex 80b. b) 1H NMR spectra of model complex 80a in DMSO-d6 when titrated with pyridine (25 mmol L-1, 400 MHz). Equivalents of pyridine are given on the left and the resonances integrated for thermodynamic analysis are color coded on top.  243  6.2.2 Synthesis of Head-to-Tail Directing N-ONO Pro-Ligands Imine  condensation  of  substituted  o-aminophenols  with  5-(3-pyridyl)-  salicylaldehyde in EtOH afforded the desired head-to-tail directing N-ONO proligands 85a-d in 67-99% yield as shown in Scheme 6.2. One advantage of this simple twocomponent assembly is the R group of the o-aminophenol is easily replaced to tune the molecule for desired solubility/function. Unsubstituted, 85a, branched alkyl, 85b, and sterically bulky, 85c-d, imines have been isolated for this study.  Scheme 6.2. Synthesis of N-ONO imines 85a-d. Imine 85a was isolated from commercially available o-aminophenol, 81, as a bright orange solid in 82% yield. The synthesis of 2-hexyldecyl substituted oaminophenol, 82, is outlined in Scheme 6.3. Iron mediated cross-coupling between the Grignard complex of p-bromoanisole and 1-bromo-2-hexyldecane followed by removal of the methyl group with BBr3 yielded 4-(2-hexyldecyl)phenol, 86. Nitration of 86 with “claycop” and subsequent reduction of the nitro group by Pd/C with hydrazine yielded the target o-aminophenol, 82. Bulky tris(phenyl)(3-amino-4-hydroxyphenyl)methane, 83 was nitrated and reduced via a similar route starting from the known compound, 88,62,63 as shown in Scheme 6.4.  244  Scheme 6.3. Synthesis of 2-amino-4-(2-hexydecyl)phenol, 82.  Scheme 6.4. Synthesis of tris(phenyl)(3-amino-4-hydroxyphenyl)methane, 83. Initially  I  attempted  to  synthesize  tris(4-tbutylphenyl)(3-amino-4-  hydroxyphenyl)methane, 84, by the same route as 83. Surprisingly, tris(4t  butylphenyl)(4-hydroxyphenyl)methane could not be nitrated in appreciable yield by a  variety of nitration strategies. Compound 83 has also been synthesized with an electrophilic aromatic substitution strategy.62,63 This approach involves acid catalyzed in situ dehydration of tris(phenyl)methanol yielding a triphenylcarbenium cation that undergoes electrophilic aromatic substitution of o-aminophenol at the position para to the hydroxyl  group.  Attempts  to  isolate  t  butyl  substituted  84  from  tris(4245  t  butylphenyl)methanol with the same reaction conditions (refluxing HCl/AcOH) gave a  1:1 mixture of 84 and methyl amide protected 84 in < 5% yield. The mixture was difficult to separate and the forcing conditions necessary for amide hydrolysis generally led to decomposition of the product. Overall this route was not viable for isolating reasonable quantities of 84. Successful literature procedures for installing tris(4-tbutylphenyl)methyl groups on aromatic rings use the desired activated aromatic ring (e.g., phenol or aniline) as the solvent during triarylcarbenium cation generation (conducted at 100 °C).64 In these cases the hydroxyl or amine functional group directs tritylation at the para position. Unfortunately, attempts to isolate 84 via the same route with o-aminophenol as solvent (conducted at 180 °C) were unsuccessful and yielded unidentifiable black tar-like residues. The observed decomposition was likely due to the elevated reaction temperature required to reach the melting point of o-aminophenol (172 °C) compared to phenol (41 °C) or aniline (-6 °C). To reduce the reaction temperature, o-aminophenol was replaced by N-(2-hydroxyphenyl)carbamate (Mp = 142 °C)  and tris(4-tbutylphenyl)methyl  chloride replaced tris(4-tbutylphenyl)methanol to avoid the otherwise necessary addition of an acid catalyst. These two reagents were mixed in the solid state under N2 and heated to 165 °C as shown in Scheme 6.5. Over a period of 15 minutes the solids melted, the solution turned deep red, gas evolved, and finally the reaction solidified and was then cooled to room temperature. Column chromatography of the crude pink solid gave 84 in 83% yield.  Scheme 6.5. Synthesis of tris(4-tbutylphenyl)(3-amino-4-hydroxyphenyl)methane, 84.  246  6.2.3 Self-assembly of Pt4 Rings In a one-pot experiment where pro-ligand 85a was generated in situ, K2PtCl4, K2CO3, 5-(3-pyridyl)salicylaldehyde, and o-aminophenol were reacted in nitrogen sparged DMSO at 150 °C for 2 h. Upon cooling, an air-stable yellow solid precipitated that was isolated by centrifugation and washed with water and MeOH. Solvent free matrix assisted laser desorption ionization time-of-flight mass spectrometry (MALDITOF MS) showed the selective formation of the cyclic tetramer, 90a, (m/z = 1934) with no other oligomers or ring sizes observed. Most interestingly, the only higher mass peaks observed correspond to sequential aggregates from dimer, [90a2]+, up to hexamer, [90a6]+, due to columnar aggregation, as shown in Figure 6.4.  Scheme 6.6. Head-to-tail self-assembly of Pt4 rings 90a-d. i) N-ONO salicylaldimine pro-ligand 85a-d (or corresponding amine and aldehyde), K2CO3, and K2PtCl4, are heated in DMSO at 150 °C for 2-4 h.  247  Figure 6.4. Aggregates of Pt4 ring 90a observed with MALDI-TOF MS. As Pt4 ring 90a is insoluble in all common solvents, 2-hexyldecyl substituted Pt4 ring 90b was pursued as a soluble alternative in order to perform solution NMR spectroscopic studies. Although 90a-d may all be prepared in situ from the imine precursors (i.e., the aldehyde and aminophenol), starting from pro-ligands 85a-d gives better results. In this manner, cyclic tetramer 90b was selectively synthesized (Scheme 6.6) and its identity confirmed by MALDI-TOF MS. Aggregation of Pt4 rings was again evident, with a series of peaks observed up to m/z = 14168, corresponding to [90b5]+. 1H NMR spectroscopy of 90b in CDCl3, C6D6, and toluene-d8 produced only broad unidentifiable resonances even at elevated temperatures. This broadening is attributed to a substantial decrease in the T2 relaxation time,65 providing further evidence for strong interactions between Pt4 rings resulting in polymeric aggregation. Clearly bulkier substituents are necessary to inhibit stacking for an NMR spectroscopic study. Pt4 ring 90c with bulky trityl substituents was synthesized (Scheme 6.6). Unfortunately, 90c is also poorly soluble. MALDI-TOF MS analysis, however, revealed formation of not only cyclic tetramer 90c but also aggregates up to [90c6]+. Pro-ligand 85d with even bulkier tris(4-tbutylphenyl)methyl substituents was prepared, and used to  248  synthesize Pt4 ring 90d. Despite observing aggregates up to [90d5]+ in the MALDI-TOF MS, 90d is very soluble in organic solvents. Metallocycles 90b-d are all brightly colored yellow solids which show absorption bands around 300-310 nm, 425-430 nm, and 450455 nm. The 1H NMR spectrum of 90d in CDCl3 (see Figure 6.5a) exhibits sharp resonances that were assigned by COSY and ROESY 2D NMR spectroscopy. The low intensity resonances in the spectrum that appear to be impurities exhibit strong cross peaks with the identified resonances in the ROESY spectrum, indicative of chemical exchange (Figure 6.6). A variable concentration 1H NMR spectroscopic experiment revealed that the identified resonances are in fact due to an aggregate and the low intensity resonances are assigned to monomers. Equilibrium constants calculated by integration of 1H NMR spectra collected between -45 and 25 °C fit best to a monomerdimer rather than infinite aggregate model,66 as might be expected for such sharp resonances (Figure 6.5b). At 25 °C, Kdim = 3200 ± 200 L mol-1 and a van’t Hoff analysis (Figure 6.5c) gave ΔH° = 13 ± 1 kJ mol-1 and ΔS° 110 ± 30 J mol-1 K-1 (r2 = 0.990). This result indicates that dimerization is enthalpy opposed, but entropy favored. The entropy increase in this process is attributed to the release of solvating CDCl3 molecules upon dimerization.67-70 Extended aggregation of 90d in the solid state, as observed by MALDITOF MS, is inhibited in solution by rapid rotation of the tris(4-tBuPh) substituents. As very bulky substituents are necessary to limit the aggregation even to dimers, as in the case of 90d, macrocycle 90b with much smaller substituents must aggregate extensively in solution, explaining its broad 1H NMR spectrum.  249  Figure 6.5. a) 1H NMR spectrum of 90d in CDCl3 (400 MHz) and an inset of the chemical structure. Resonances of the peripheral, R = tris(4-tBuPh), aromatic rings overlap with the residual CHCl3 resonance calibrated to 7.27 ppm. b) Variable temperature 1H NMR spectra of 90d in CDCl3 (2 mmol L-1) from -45 to 25 °C. Monomer-dimer equilibrium is inset and resonances are color coded (monomer = red, dimer = blue). c) Van’t Hoff plot for dimerization of 90d in CDCl3. 250  Figure 6.6. 2D ROESY 1H NMR spectrum of 90d in CDCl3 with some of the most intense exchange cross peaks circled in red (400 MHz). I was unable to obtain single crystals of 90a-d, so an ab initio DFT study of complex 90a was performed to learn more about its structure and to provide a rough model for aggregation. Figure 6.7a shows the puckered conformation of 90a determined by DFT optimization (optimization with B3LYP level of theory – LanL2dZ basis sets for Pt, 6-31G* for other atoms).71 This conformation exhibits a 60° deviation from planarity, a maximum outer diameter of 2.4 nm, and a 0.7 nm inner pore. The intermetallic Pt-Pt distances in the macrocycle are 1.5 nm (diagonal) and 1.1 nm (edge). Belonging to the rare point group S4, this geometry is imposed by the 38° dihedral angle between the 3pyridyl and phenyl rings of the salicylidene unit.72,73  251  Figure 6.7. a) DFT optimized geometry of Pt4 ring 90a. b) Computer model of two possible orientations for columnar aggregation. Each ring is stacked directly on the other in the syn orientation whereas the anti orientation exhibits alternating AB type stacking.  252  Modeling studies reveal that the puckered Pt4 rings may organize into columns either with a single orientation (syn) or with alternating orientations (anti), depending on the steric bulk of the peripheral R group (Figure 6.7b). In the syn case, the repeat unit is a simple translation along the columnar axis whereas the anti orientation requires the same translation, a 90° rotation about the columnar axis, and a 180° rotation perpendicular to the columnar axis. With small substituents, the cycles may adopt either stacking motif, but with sterically demanding substituents, only the anti alternating, AB pattern, is possible since this reduces intermolecular interactions between the substituents. Potential for intermolecular Pt-Pt interaction exists in the syn case.  6.2.4 Evidence for Columnar Organization of Pt4 Rings Solution aggregation of Pt4 rings 90b and 90d in CHCl3 was further investigated by dynamic and multi-angle laser light scattering (DLS and MALLS, respectively). The 1  H NMR studies of 90d described above indicated that aggregation in solution was  limited to dimers. DLS of 90d (0.6 mg mL-1) showed that only aggregates < 10 nm in diameter are present, confirming dimerization and not infinite aggregation is occurring. On the other hand, DLS of 90b (0.6 mg mL-1) resulted in a broad peak corresponding to a hydrodynamic radius (RH) of 60 nm (assuming spherical particles).74 To obtain information about the shape of the aggregates of 90b in solution, a MALLS study was undertaken. Columnar aggregation of 90b was confirmed by the Kratky plot, shown in Figure 6.8a, that exhibits linear angular dependence over the light scattering intensity of the aggregates.75-80 A Zimm plot was also constructed from MALLS data over a 0.4 to 0.8 mg mL-1 concentration range and is shown in Figure 6.8b with extrapolations to zero concentration and angle. From the Zimm analysis an average molecular weight, Mw = 1.2 ± 0.4 x 107 g mol-1, radius of gyration, Rg = 150 ± 17, and a second virial coefficient, A2 = 5.4 ± 0.4 x 10-4 mol mL g-2 were calculated. Although the aggregates in this system are dynamic in nature (owing to non-covalent interactions), making these values non-quantitative, they are still reasonably reliable for making predictions about the system.81 In particular, spherical or random-coil aggregates exhibit Rg/RH ratios below 1.5 whereas for rigid rod-shaped aggregates the ratio is greater.82,83  253  The Rg/RH ratio for Pt4 ring 90b is 2.5, indicating these are highly anisotropic rigid column/rod-shaped aggregates. Attempts to elucidate the thermodynamics of columnar aggregation in CHCl3 by variable concentration UV-vis experiments were inhibited because there is no change in the absorption spectra from 3.4 x 10-4 to 2.6 x 10-6 mol L-1, suggesting that even at very low concentrations 90b is extensively aggregated.  Figure 6.8. MALLS analysis of 90b in CHCl3 a) Kratky plot (0.6 mg mL-1). b) Zimm plot (0.4, 0.6, and 0.8 mg mL-1). Columnar aggregates of Pt4 ring 90b organize into lyotropic mesophases upon concentration in non-polar organic solvents. Liquid crystalline (LC) behavior was observed in CHCl3, C6H6, PhCl, trichloroethylene, CS2, and pyridine; however, the highest quality mesophases, as judged by the homogeneity of the texture, are obtained in chlorobenzene. Birefringence of these LCs was observed with a polarized optical microscope (POM) upon slow evaporation of dropcast solutions on a microscope slide, 254  and typical POM images of 90b are depicted in Figure 6.9. The critical concentration for birefringence is difficult to estimate, as it is observed only upon evaporation. In each example shown in Figure 6.9 the concentration of 90b was > 1 mg mL-1. The steric bulk of Pt4 rings 90c and 90d inhibits extended columnar aggregation and accordingly little to no birefringence was observed for these metallocycles in the same solvents.  Figure 6.9. POM images observed under crossed polarizers of growing LC textures for 90b from a) CHCl3, and b,c) PhCl. Black indicates an isotropic phase. Most often, thread-like or Schlieren textures were observed for 90b suggesting the adoption of a columnar nematic LC phase.84 In Figure 6.9c, a fan-shaped texture was observed near the isotropic phase, potentially due to adoption of a disordered hexagonal columnar mesophase.85 The S4 symmetry of 90b may inhibit organization of a high fidelity hexagonal columnar LC phase leading to a more disordered phase. In all cases, the anisotropy of rod-shaped aggregates self-assembled from 90b leads to long range parallel orientation of columns in concentrated solutions and adoption of nematic and/or disordered hexagonal columnar LC phases as outlined in Scheme 6.7.86,87 This is the first report of lyotropic liquid crystallinity for Pt---pyridyl metallocycles, and a rare example of a multimetallic LC.88-96  255  Scheme 6.7. Dynamic assembly of Pt4 ring 90b into randomly oriented oligomers, elongated and oriented columns, and finally columnar nematic or disordered hexagonal columnar mesophases upon concentration. Organization of the Pt4 rings was also studied in the solid state by powder X-ray diffraction (PXRD). Figure 8a-d shows the PXRD patterns of microcrystalline 90a-d immediately after isolation. Unsubstituted Pt4 ring 90a displays the most crystallinity, with several sharp peaks present in the powder pattern. Unfortunately, there are too few peaks to obtain a space group or even a definitive unit cell, but the best fits to the experimental data were for tetragonal unit cells (D= E= J= 90º) with a = b = 2.043 or 2.890 nm and c = 0.5-0.7 nm.97 The distinct peaks observed at low angle in the diffraction pattern fit best to a tetragonal unit cell. The lack of well-defined peaks beyond ca. 19º 2θ, which give information about the 3rd parameter (c), made it very difficult to obtain any reliable measure of c. For unit cells found with a = b = 2.043 nm, the best value of c was either 0.44 or 0.49 nm. For the unit cells found with a = b = 2.890 nm, the value of c was in the 0.57 – 0.68 nm range. This pattern certainly does not prove a columnar stacking of the macrocycles, but the data is consistent with such an arrangement. If the cycles of 90a 256  were stacked one on top of another as shown in Fig. 6.7b into parallel columns, the intercolumn center-to-center distance would be about 2.1 nm and the intermacrocycle separation in the columns would be around 0.5 nm. Low-angle peaks are observed at 3.1, 2.9, and 3.2 nm for 90b, 90c, and 90d, respectively. These roughly correspond to the expected inter-columnar distance for parallel aligned aggregates, allowing for interdigitation of the peripheral substituents. Additional low-intensity peaks at higher angle support the presence of some order within each column. It was not possible to directly identify the phases of the lyotropic LCs from PXRD patterns collected on solutions of 90b in capillary tubes. Instead, PXRD data was collected from samples of 90b in CHCl3 and PhCl that were dropcast onto amorphous silicon plates and left to dry. Diffractograms of the dried LC phases are shown in Figure 6.10e-f. The low-angle diffraction observed for the sample obtained from CHCl3 (with a peak at 3.1 nm) is nearly identical to that observed in the PXRD pattern of as-isolated 90b, suggesting that columnar alignment of the nematic phase is maintained upon drying. Disorder is evident from broad peaks centered about 1.6 and 1.1 nm, preventing any further structural elucidation. PXRD of 90b dropcast from PhCl shows significantly more order and a sharp peak is present at 3.5 nm. Close inspection of the low-angle region reveals a broad peak centered about 3.1 nm (10) followed by relatively sharp peaks at 1.8 (11), 1.5 (20), and 1.2 nm (21), diagnostic spacing for hexagonal columnar ordering. These observations are in agreement with POM images that show 90b adopts both columnar nematic and hexagonal columnar mesophases in PhCl. Inter-columnar spacing is greater for the less ordered columnar nematic phase (3.5 vs 3.1 nm). Overall, the PXRD studies of the films dried from the lyotropic LC phases support the retention of columnar organization in the metallocycles.  257  Figure 6.10. Normalized wide-angle PXRD patterns of as-prepared Pt4 rings: a) 90a, b) 90b, c) 90c, d) 90d, e) dried mesophase of 90b drop-cast from CHCl3 onto amorphous silicon, and f) dried mesophase of 90b dropcast from PhCl onto amorphous silicon with assigned indices for hexagonal ordering. All data is depicted from 2º to 30º 2θ. Solid-state organization of Pt4 ring 90b was also investigated by transmission electron microscopy (TEM); representative micrographs are shown in Figure 6.11a-e. At low magnification, oblong “pill-shaped” aggregates assembled from cyclohexanone were observed. High magnification of the same sample reveals each “pill” is composed of individual stacks of Pt4 rings organized into parallel columnar arrays, as shown in Figure 6.11d-e. The parallel columns are spaced by roughly 4 nm and extend hundreds of nanometers. When 90b was deposited on the TEM grid from CHCl3, bundles of randomly  258  oriented rigid rods with dimensions from tens to hundreds of nanometers were observed. This is in agreement with the MALLS data that confirmed the existence of rigid rodshaped aggregates of 90b in CHCl3. Micrographs of 90b deposited from C6H6 depict large flexible fibers that are hundreds of nanometers in diameter and span microns. The breadth of the rods and fibers observed for 90b cast from CHCl3 and C6H6 suggests that they are also composed of individual columnar arrays, similar to those observed from cyclohexanone; however, it was not possible to achieve a similar image resolution in these cases. Organization of individual Pt4 rings to micron length fibers represents an impressive hierarchical self-assembly that spans several orders of magnitude in length, all in one-pot. Only diffuse, globular aggregates were present in TEM images of sterically encumbered Pt4 ring 90d.  259  Figure 6.11. Low magnification TEM images of Pt4 ring 90b dried from various solvents. a) “Pill-shaped” oblate aggregates from cyclohexanone. b) Rigid rod-shaped bundles from CHCl3. c) Micron length flexible fibers from C6H6. High magnification TEM images of Pt4 ring 90b from cyclohexanone. d) Individual columnar arrays are visible with periodic spacing of roughly 4 nm. e) Model of columnar aggregates inset.  260  6.3  Conclusions A new class of highly tunable pro-ligands that selectively self-assemble into novel  S4 symmetric disc-shaped metallocycles upon platinum(II) coordination has been developed. This head-to-tail approach offers a very flexible, simple method to access Ptcontaining macrocycles en route to nanotubes. Data from light scattering, X-ray diffraction, and transmission electron microscopy supports the formation of 1-D columnar aggregates from the rings. Although some molecular squares formed with Pt--pyridyl bonding exhibit columnar organization in the solid state, to the best of my knowledge no reports of solution aggregation exist, a fact I attribute to Coulombic repulsion. Neutral Pt4 rings presented here stack strongly both in the solid state and in solution, where they form lyotropic LCs when concentrated.  6.4  Experimental  6.4.1 General New synthetic routes for previously reported tris(phenyl)(4-hydroxy-3-nitro-phenyl) methane, 89,98 and tris(phenyl)(3-amino-4-hydroxyphenyl) methane, 83,62,63 have been described and each compound was fully characterized (only Mp included in references). Known  compounds  phenylsalicylaldehyde, t  4-tritylphenol 99  (88),62,63  5-(3-pyridyl)salicylaldehyde,99,100  tert-butyl-N-(2-hydroxyphenyl)carbamate,  101  5-  tris(4-  butylphenyl)methyl chloride,64 [Fe(TMEDA)3]Cl3,102 and “claycop”103 were synthesized  according to literature procedures. Compound 81, o-aminophenol, was purchased from Aldrich. All reactions were carried out under air unless otherwise noted. 1H and 13C NMR spectra were recorded on a Bruker AV-300, AV-400, or AV-600 spectrometer. 13C NMR spectra were recorded using a proton decoupled pulse sequence. 1H and 13C NMR spectra were calibrated to the residual protonated solvent at δ 7.27 and δ 77.0, respectively, in CDCl3, at δ 3.30 and δ 49.0, respectively, in MeOD, or at δ 2.50 and δ 39.51, respectively, for DMSO-d6. Figures 6.12 and 6.13 depict the 2D COSY NMR spectrum 261  and 1D 1H NMR spectrum of 90d with proton assignments, respectively. UV-vis spectra were obtained in CH2Cl2 or MeOH (ca. 1 x 10-6 M) on a Varian Cary 5000 UV-vis-nearIR spectrophotometer using a 1 cm quartz cuvette. FT-IR spectra were obtained neat with a Thermo Nicolet 6700 FT-IR spectrometer. MALDI-TOF mass spectra were obtained in a trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]-malonitrile (DCTB) matrix (1:10 analyte:matrix, solvent free) at the UBC Microanalytical Services Laboratory on a Bruker Biflex IV instrument. The best mass accuracy was achieved at low laser power, and some accuracy was sacrificed when collecting with the high laser power required to observe high mass aggregates. Electrospray ionization (ESI) mass spectra were obtained on a Bruker Esquire LC instrument. Elemental analyses (C,H,N) were performed at the UBC Microanalytical Services Laboratory. Single-crystal XRD was performed on Bruker X8 APEX CCD with Mo radiation. A Bruker D8 Advance with Cu radiation and a NaI scintillation detector was used for powder X-ray diffraction analysis. Birefringence of liquid crystal textures was captured with an Olympus BX41 polarized optical microscope. Transmission electron microscopy was performed at UBC BioImaging Facility on a Hitachi H7600 electron microscope. Multi-angle laser light scattering was performed on a DAWN EOS spectrometer in batch-mode with a λ = 690 nm laser (Wyatt Technology Corp). Determination of dn/dc was performed on an OPTILAB DSP Wyatt Interferometric Refractometer and found to be 0.080 mL g-1. A Wyatt Technology Corp. Dyna Pro Titan with a Temperature-Controlled Microsampler was used for dynamic light scattering measurements.  262  6.4.2 Procedures and Data Synthesis of ONO imine pro-ligand (79): 5-Phenylsalicylaldehyde (161 mg, 0.81 mmol) was added to a solution of 2-aminophenol (89 mg, 0.81 mmol) in 10 mL of EtOH and the reaction was heated to reflux. After 45 min, the reaction was cooled to room temperature and the solvent was removed under vacuum giving crude product. Recrystallization from MeOH gave 79 as an orange solid (202 mg, 0.70 mmol, 86%). Data for 79: 13C NMR (100.6 MHz, MeOD) δ 163.8, 162.9, 152.3, 141.7, 136.1, 133.2, 133.1, 131.8, 130.0, 129.4, 127.9, 127.6, 121.4, 121.0, 120.8, 119.2, 117.7. 1H NMR (400 MHz, MeOD) δ 8.97 (s, 1H, imine), 7.77 (d, 4JHH = 2.0 Hz, 1H, Ar), 7.65 (dd, 3JHH = 8.4, 4  JHH = 2.0 Hz, 1H, Ar), 7.61 (d, 3JHH = 7.2 Hz, 2H, Ar), 7.41 (t, 3JHH = 7.2 Hz, 2H, Ar),  7.35 (dd, 3JHH = 8.0, 4JHH = 1.6 Hz, 1H, Ar), 7.29 (t, 3JHH = 7.2 Hz, 1H, Ar), 7.13 (t, 3JHH = 7.6 Hz, 1H, Ar), 7.00 (d, 3JHH = 8.4 Hz, 1H, Ar), 6.93 (m, 2H, Ar). UV-vis (MeOH) λmax (ε) = 432 (9.0 x 102), 370 (2.0 x 103), 260 (8.0 x 103) nm (cm-1 mol-1 L). FT-IR (neat) ν = 3375, 3304, 3033, 1621, 1487, 1306, 1284, 1266, 1237, 1147, 1128, 923, 835, 764, 740, 725, 696, 530, 515, 483 cm-1. ESI-MS (MeOH) m/z = 290.1 [79+H]+. Mp = 158 °C. Anal. Calc’d for 79: C19H15NO2: C, 78.87; H, 5.23; N, 4.84. Found: C, 78.43; H, 5.23; N, 4.78. Synthesis Pt-DMSO (S bound) ONO model complex (80a): A solution of K2PtCl4 (99 mg, 0.24 mmol) in 5 mL of DMSO at 100 °C. was charged with 79 (69 mg, 0.24 mmol) and Na2CO3 (60 mg, 0.56 mmol) and the reaction was heated to 140 °C for 15 min, cooled to 100 °C and stirred for 15 more min before it was cooled to room temperature. Addition of H2O (6 mL) produced a precipitate that was isolated by filtration, redissolved in 15 mL of EtOAc, and filtered through celite. Removal of solvent under reduced pressure afforded 80a as a yellow solid (56 mg, 42%).  263  Data for 80a:  13  C NMR (100.6 MHz, DMSO-d6) δ 166.5, 160.8, 148.0, 139.0, 138.1,  132.5, 132.4, 128.9, 128.8, 128.4, 126.6, 125.7, 121.6, 121.2, 117.8, 116.3, 116.0, 40.9, 40.3. 1H NMR (300 MHz, DMSO-d6) δ 9.44 (s, 1H, imine), 8.32 (d, 4JHH = 2.4 Hz, 1H, Ar), 8.16 (d, 3JHH = 7.2 Hz, 1H, Ar), 7.89 (dd, 3JHH = 9.0, 4JHH = 2.4 Hz, 1H, Ar), 7.71 (d, 3  JHH =7.2 Hz, 2H, Ar), 7.48 (t, 3JHH = 7.5 Hz, 2H, Ar), 7.32 (m, 2H, Ar), 7.11 (t, 3JHH =  8.1 Hz, 1H, Ar), 7.01 (d, 3JHH = 7.2 Hz, 1H, Ar), 6.76 (t, 3JHH = 8.4 Hz, 1H, Ar), 3.53 (bs, 6H, DMSO). UV-vis (MeOH) λmax (ε) = 440 (1.7 x 104), 410 (1.8 x 104), 330 (8.3 x 103), 280 (4.1 x 104), 220 (2.8 x 104) nm (cm-1 mol-1 L). FT-IR (neat) ν = 2991, 2908, 1597, 1583, 1501, 1476, 1311, 1262, 1165, 1126, 844, 826, 806, 758, 748, 732, 690, 671, 577, 547 cm-1. ESI-MS (MeOH) m/z = 583.0 (80a+Na)+. Mp = dec. at 229 °C. Anal. Calc’d for 80a: C21H19NO3PtS: C, 45.00; H, 3.42; N, 2.50. Found: C, 44.97; H, 3.50; N, 2.46. Synthesis of 4-(2-hexyldecyl)-2-aminophenol (82): A Schlenk flask was charged with 4-(2-hexyldecyl)-2-nitrophenol 87 (200 mg, 0.55 mmol) and Pd/C (roughly 30 mg) then put through two vacuum/N2 purge cycles. Nitrogen sparged THF (15 mL) was added to the Schlenk flask followed by hydrazine monohydrate (0.05 mL, 1.0 mmol) and the reaction was left to stir at room temperature. After 40 h the suspension was filtered through a frit and the solvent of the filtrate was removed under reduced pressure. The crude residue was purified via column chromatography in 10:1 DCM:EtOAc giving 82 as a colorless oil (142 mg, 77%).  Data for 82:  13  C NMR (100 MHz, CDCl3) G 142.0, 135.2, 134.0, 120.2, 118.1, 115.0,  39.9, 39.7, 33.1, 31.9, 30.0, 29.7, 29.6, 29.3, 26.6, 26.5, 22.7, 14.1. 1H NMR (400 MHz, CDCl3) G 6.64 (d, 3JHH = 8.0 Hz, 1H, Ar CH), 6.56 (d, 4JHH = 2.0 Hz, 1H, Ar CH), 6.46 (dd, 3JHH = 8.0, 4JHH 2.0 Hz, 1H, Ar CH), 4.10 (bs, 3H, NH2, OH), 2.38 (d, 3JHH = 6.8 Hz, 2H, CH2), 1.59-1.48 (m, 1H, CH), 1.30-1.20 (m, 24H, CH2), 0.91-0.87 (m 6H, CH3). UVvis (CH2Cl2) λmax (ε) = 240 (1.1 x 105), 296 (6.0 x 103) nm (L mol-1 cm-1). ESI-MS (MeOH) m/z = 334.5 [82+H]+. FT-IR (neat): ῡ = 3380, 3315, 2921, 2852, 1609, 1517, 1456, 1377, 1283, 1215, 1141, 875, 804 cm-1. HRMS (ESI) Calc’d for C22H40NO: 334.3110 ; Found: 334.3116 (-1.8 ppm).  264  Synthesis of tris(phenyl)(3-amino-4-hydroxyphenyl)methane (83): A Schlenk flask was charged with tris(phenyl)(4-hydroxy-3-nitrophenyl)methane (89) (1.015 g, 2.6 mmol) and Pd/C (roughly 200 mg) and put through two vacuum/N2 purge cycles. Nitrogen sparged THF (40 mL) was added followed by hydrazine (0.300 mL, 6.1 mmol) and the reaction was stirred at room temperature for 48 h. The suspension was filtered and the filtrate was concentrated under vacuum. Recrystallization of the crude residue from hot EtOH gave 83 as a white solid (680 mg, 74%). Data for 83: 13C NMR (100.6 MHz, CDCl3) G 147.1, 142.1, 137.3, 135.5, 130.5, 127.3, 125.6, 118.6, 117.5, 113.1, 63.9. 1H NMR (300 MHz, DMSO-d6) G 8.94 (s, 1H, OH), 7.29-7.13 (m, 15H, Ar CH), 6.52 (d, 3JHH = 8.2 Hz, 1H, Ar CH), 6.38 (d, 4JHH = 2.3 Hz, 1H, Ar CH), 6.15 (dd, 3JHH = 8.2, 4JHH = 2.3 Hz, 1H, Ar CH), 4.41 (bs, 2H, NH2). UV-vis (CH2Cl2) λmax (ε) = 302 (4.2 x 103) nm (L mol-1 cm-1). ESI-MS (MeOH) m/z = 350.4 [83H]-, 352.4 [83+H]+. FT-IR (neat) ῡ = 3413, 3329, 3058, 3020, 1629, 1588, 1533, 1517, 1489, 1440, 1320, 1272, 1245, 1184, 1148, 1080, 1033, 833, 823 cm-1. Mp = 226 ºC. Anal. Calc’d for 83: C25H21NO· 1/3 H2O: C, 84.00; H, 6.11; N, 3.92. Found: C, 83.79; H, 6.55; N, 4.16.  265  Synthesis of tris(4-tbutylphenyl)(3-amino-4-hydroxyphenyl)methane (84): Intimate mixing of the white powders tris(4-tbutylphenyl)methyl chloride (300 mg, 0.67 mmol) and tert-butyl-N-(2-hydroxyphenyl)carbamate (480 mg, 2.3 mmol) in a Schlenk tube was followed by two cycles of vacuum/N2 purging. The Schlenk tube was then submerged in an oil bath preheated to 150 °C while stirring under inert atmosphere. Over 15 min, the reaction  was  heated  to  165  °C  and  during  that  time  tert-butyl-N-(2-  hydroxyphenyl)carbamate melted, the solution turned deep red, gas evolved, and then the mixture rapidly solidified. When no more bubbling/boiling was observed, the heat was turned off and upon cooling to room temperature a pink solid was isolated. The crude mixture was dissolved in DCM with a little MeOH, dry loaded onto silica gel, and subjected to column chromatography in 40:1 DCM:CH3CN. (product eluted with an rf of ~0.5). Desired fractions were combined and dried under reduced pressure to give 84 as a pink solid (289 mg, 83%). Recrystallization from toluene/hexanes gave 84 a white solid for elemental analysis. Data for 84: 13C NMR (100.6 MHz, CDCl3) δ 148.3, 144.2, 142.7, 132.0, 130.8, 124.4, 124.0, 118.5, 115.7, 63.1, 34.3, 31.4.  1  H NMR (400 MHz, DMSO-d6) δ 8.76 (bs, 1H,  OH), 7.27 (d, 3JHH = 8.5 Hz, 6H, Ar), 7.09 (d, 3JHH = 8.5 Hz, 6H, Ar), 6.50 (d, 4JHH = 1.7 Hz, 1H, Ar), 6.44 (d, 3JHH = 7.9 Hz, 1H, Ar), 6.33 (dd, 3JHH = 7.9, 4JHH = 1.7 Hz, 1H, Ar), 4.40 (bs, 2H, RNH2), 1.26 (s, 27H, tBu). UV-vis (CH2Cl2) λmax (ε) = 303 (3.8 x 103) nm (L mol-1 cm-1). ESI-MS (MeOH) m/z = 520.6 [84+H]+, 411.5 [84-C6H6NO]+. FT-IR (neat) v = 3085, 2057, 2958, 2901, 2866, 1506, 1475, 1441, 1285, 1269, 1202, 1018, 823, 705, 578 cm-1. Mp = dec. at 300 °C. Anal. Calc’d for 84: C37H45NO: C, 85.50; H, 8.73; N, 2.69. Found: C, 85.22; H, 8.73; N, 2.68. Synthesis of N-ONO imine pro-ligand R=H (85a): A round bottom flask was charged with 5-(pyridyl)salicylaldehyde (100 mg, 0.50 mmol), 2-aminophenol (55 mg, 0.50 mmol), and 5 mL of EtOH then heated to reflux for 4 h.  Upon cooling to room  temperature 3 mL of hexanes was added and the flask was placed in a freezer at -10 °C for 2 d. Imine 85a was isolated by filtration as a bright orange powder (120 mg, 82%).  266  Data for 85a: 13C NMR (100.6 MHz, MeOD) δ 165.0, 162.4, 152.3, 148.3, 147.9, 136.0, 135.6, 133.1, 132.1, 129.6, 128.8, 121.4, 120.8, 119.9, 117.7, 117.2, 115.9, 31.1. 1H NMR (400 MHz, MeOD) δ 9.02 (s, 1H, imine), 8.84 (bs, 1H, Ar), 8.50 (bs, 1H, Ar), 8.12 (d, 3JHH = 8.0 Hz, 1H, Ar), 7.87 (d, 4JHH = 4.0 Hz, 1H, Ar), 7.72 (dd, 3JHH = 8.0, 4JHH = 4.0 Hz, 1H, Ar), 7.52 (t, 3JHH = 8.0 Hz, 1H, Ar), 7.38 (d, 3JHH = 8.0 Hz, 1H, Ar), 7.15 (t, JHH = 38.0 Hz, 1H, Ar), 7.06 (d, = 3JHH 8.0 Hz, 1H, Ar), 6.94 (m, 2H, Ar). UV-vis (MeOH) λmax (ε) = 440 (3.5 x 103), 360 (1.1 x 104), 270 (2.6 x 104) nm (cm-1 mol-1 L). ESI-MS (MeOH) m/z = 291.1 [85a+H]+. HRMS (ESI) calc’d for C18H15N2O2: 291.1134; found: m/z = 291.1140 [85a+H]+. FT-IR (neat) ν = 3055, 1606, 1591, 1504, 1472, 1463, 1272, 822, 806, 752, 741, 712, 611 cm-1. Mp = 137 °C. Anal. Calc’d for 85a: C18H14O2N2·1/3(H2O):  C, 72.96; H, 4.99; N, 9.45. Found: C, 73.19; H, 5.03; N, 9.32.  Synthesis of N-ONO imine pro-ligand R=2-hexyldecyl (85b): A round bottom flask was charged with 5-(pyridyl)salicylaldehyde (81 mg, 0.41 mmol), 2-amino-4-(2hexyldecyl)phenol (82, 137 mg, 0.41 mmol), and 20 mL of EtOH then heated to reflux for 18 h. Upon cooling to room temperature the solvent was removed under vacuum leaving imine 85b as a tacky red solid (217 mg, 99%). Data for 85b: 13C NMR (100.6 MHz, CDCl3) δ 162.3, 161.3, 148.3, 147.5, 147.3, 135.8, 134.9, 134.4, 134.0, 131.7, 130.6, 129.5, 128.5, 123.8, 119.9, 119.7, 118.2, 116.2, 39.8, 33.1,. 1H NMR (400 MHz, DMSO-d6) G 14.03 (bs, 1H, OH), 9.59 (s, 1H, OH), 9.07 (s, 1H, imine), 8.94 (bs, 1H, Ar), 8.56 (bs, 1H, Ar), 8.12 (m, 1H, Ar), 8.06 (d, 4JHH = 2.8 Hz, 1H, Ar), 7.78 (dd, 3JHH = 8.4 Hz, 4JHH = 2.8 Hz, 1H, Ar), 7.52 (m, 1H, Ar), 7.17 (d, 4JHH = 2.0 Hz, 1H, Ar), 7.06 (d, 3JHH = 8.4 Hz, 1H, Ar), 6.92 (dd, 3JHH = 8.0 Hz, 4JHH = 2.0 Hz, 1H, Ar), 6.88 (d, 3JHH = 8.0 Hz, 1H, Ar), 2.44 (d, 3JHH = 6.8 Hz, 2H, CH2), 1.59 (m, 1H, CH), 1.21 (bs, 24H, CH2), 0.81 (m, 6H, CH3). UV-vis (CH2Cl2) λmax (ε) = 264 (3.5 x 104), 373 (1.7 x 104) nm (L mol-1 cm-1). ESIMS (MeOH) m/z = 537.6 [85b+Na]+, 513.6 [85b-H]-. FT-IR (neat) ῡ = 3055, 2953, 2921, 2851, 1621, 1573, 1501, 1467, 1377, 1355, 1274, 1177, 1120, 802, 707 cm-1. Mp = 25 °C. Anal. Calc’d for 85b: C34H46O2N2: C, 79.33; H, 9.01; N, 5.44. Found: C, 79.30; H, 9.00; N, 5.63.  267  Synthesis of N-ONO imine pro-ligand R=CPh3 (85c): A round bottom flask was charged with 5-(pyridyl)salicylaldehyde (113 mg, 0.57 mmol), tris(phenyl)(3-amino-4hydroxyphenyl)methane (83, 200 mg, 0.57 mmol), and 20 mL of EtOH then heated to reflux for 17 h. The reaction was cooled to room temperature and imine 85c was isolated by filtration as a yellow solid (257 mg, 85%). Data for 85c: 13C NMR (100.6 MHz, DMSO-d6) G 161.4, 160.7, 149.0, 147.8, 147.1, 146.6, 137.7, 134.7, 134.5, 133.7, 133.4, 131.2, 130.4, 127.7, 125.9, 123.7, 122.2, 119.8, 117.5, 115.7, 63.9. 1H NMR (400 MHz, DMSO-d6) G 13.66 (bs, 1H, OH), 9.92 (s, 1H, OH), 8.89 (d, 4JHH = 2.0 Hz, 1H, Ar), 8.84 (s, 1H, imine), 8.51 (m, 1H, Ar), 8.06 (m, 1H, Ar), 8.02 (d, 4JHH = 2.8 Hz, 1H, Ar), 7.76 (dd, 3JHH = 8.4 Hz, 4JHH = 2.8 Hz, 1H, Ar), 7.45 (m, 1H, Ar), 7.31 (m, 6H, Ar), 7.21 (m, 9H, Ar), 7.06 (d, 4JHH = 2.0 Hz, 1H, Ar), 7.03 (d, 3  JHH = 8.4, 1H, Ar), 6.91 (bs, 2H, Ar). UV-vis (CH2Cl2) λmax (ε) = 264 (3.2 x 104), 377  (2.0 x 104) nm (L mol-1 cm-1). ESI-MS (MeOH) m/z = 533.3 [85c+H]+. FT-IR (neat) ῡ = 3031, 1621, 1577, 1496, 1414, 1289, 1274, 1174, 1121, 1030, 803, 761, 705, cm-1. Mp = 280 °C. Anal. Calc’d for 85c: C37H28O2N2: C, 83.43; H, 5.30; N 5.26. Found: C, 83.06; H, 5.56; N, 5.42. Synthesis of N-ONO imine pro-ligand R=C(4-tBuPh)3 (85d): A round bottom flask was charged with 5-(pyridyl)salicylaldehyde (63 mg, 0.32 mmol), tris(4-tbutylphenyl)(3amino-4-hydroxyphenyl)methane (84, 165 mg, 0.32 mmol), and 20 mL of EtOH then heated to reflux for 15 h. The reaction was cooled to room temperature and imine 85d was isolated by filtration as an orange solid (151 mg, 67%).  268  Data for 85d: 13C NMR (100.6 MHz, CDCl3) G 162.6, 160.8, 149.2, 148.9, 148.2, 147.8, 143.5, 133.8, 132.7, 132.0, 130.8, 130.7, 130.6, 129.3, 124.6, 124.2, 124.0, 119.8, 119.1, 118.2, 117.2, 63.7, 34.3, 31.4.+NMR (400 MHz, DMSO-d6) G 13.96 (s, 1H, OH), 9.70 (s, 1H, OH), 9.04 (s, 1H, imine), 8.90 (bs, 1H, Ar), 8.53 (bs, 1H, Ar), 8.06 (m, 1H, Ar), 7.99 (d, 4JHH = 2.8 Hz, 1H, Ar), 7.76 (dd, 3JHH = 8.8 Hz, 4JHH = 2.8 Hz, 1H, Ar), 7.47 (m, 1H, Ar), 7.33 (d, 3JHH = 8.8 Hz, 6H, Ar), 7.28 (d, 3JHH = 8.8 Hz, 1H, Ar), 7.15 (d, 3JHH = 8.8 Hz, 6H, Ar), 7.05 (d, 3JHH = 8.4 Hz, 1H, Ar), 6.92 (d, 4JHH = 2.0 Hz, 1H, Ar), 6.69 (dd, 3JHH = 8.4 Hz, 4JHH = 2.0 Hz, 1H, Ar), 1.27 (s, 27H, CH3). UV-vis (CH2Cl2) λmax (ε) = 263 (5.7 x 104), 378 (2.9 x 104) nm (L mol-1 cm-1). ESI-MS (MeOH) m/z = 701.7 [85d+H]+, 699.7 [85d-H]-. FT-IR (neat) ῡ = 3084, 3057, 2959, 2902, 2866, 1623, 1587, 1532, 1505, 1476, 1418, 1269, 1244, 1171, 1109, 1017, 833, 822, 801, 750, 705, 680, 635, 586 cm-1. Mp = dec. at 300 °C. Anal. Calc’d for 85d: C49H52O2N2· ½ H2O : C, 82.90; H, 7.52; N, 3.95; Found: C, 82.69; H, 7.38; N, 3.91. Synthesis of 4-(2-hexyldecyl)phenol (86): 4-Bromoanisole (5.0 mL, 7.47 g, 40.0 mmol, 1 eq.) and magnesium turning (916 mg, 40.0 mmol, 1 eq.) were heated to reflux for 1 h in dry THF (40 mL) under nitrogen to form the Grignard reagent. This Grignard reagent was added dropwise to a solution of 1-bromo-2-hexyldecane (12.21 g, 40 mmol, 1 eq.) and FeCl3(TMEDA)3 (460 mg, 0.9 mmol, 0.03 eq.) in dry THF (20 mL) at 0 ºC under nitrogen. After stirring 18 h at RT, diluted HCl (0.1 mol L-1, 30 mL) was added until everything dissolved, then THF was evaporated and the aqueous phase extracted with dichloromethane (3 x 30 mL). The combined organic phases were dried over MgSO4 and concentrated under reduced pressure. The crude product was dissolved in dry DCM (60 mL) under nitrogen, and treated with BBr3 (5 mL, 60 mmol, 1.5 eq.) at room temperature. After stirring 18 h at room temperature, water (40 mL) was added and the aqueous phase was extracted with DCM (3 x 30 mL). The combined organic phases were dried over MgSO4 and concentrated under reduced pressure. Silica gel flash chromatography of the residue (DCM) gave 86 as a light brown oil (748 mg, 5.9% over 2 steps).  269  Data for 86:  13  C NMR (100.6 MHz, CDCl3) G 153.2, 134.0, 130.2, 115.3, 39.8, 39.6,  33.1, 31.9, 30.0, 29.7, 29.6, 29.4, 26.6, 26.6, 22.7, 14.1. 1H NMR (400 MHz, CDCl3) G 7.01 (d, 3JHH = 8.4 Hz, 2H, Ar), 6.75 (d, 3JHH = 8.4 Hz, 2H, Ar), 4.55 (s, 1H, OH), 2.46 (d, 3  JHH = 6.8 Hz, 2H, CH2), 1.57-1.52 (m, 1H, CH), 1.35-1.19 (m, 24H, CH2), 0.91-0.87 (m  6H, CH3). UV-vis (CH2Cl2) λmax (ε) = 279 (2.3 x 103), 285 (1.9 x 103) nm (L mol-1 cm-1). ESI-MS (MeOH) m/z = 317.4 [86-H]-. FT-IR (neat): ῡ = 3335, 2955, 2921, 2852, 1613, 1598, 1512, 1456, 1377, 1224, 1171, 816 cm-1. Anal. Calc’d for 86: C22H38O: C, 82.95; H, 12.02. Found: C, 83.19; H, 12.05.  Synthesis of 4-(2-hexyldecyl)-2-nitrophenol (87): A suspension of claycop (188 mg, 300 mg/mmol of reagent to be nitrated) in 11 mL of 10:1 ether:acetic anhydride was added to a solution of 4-(2-hexyldecyl)phenol (86, 217 mg, 0.6 mmol) in 11 mL of the same 10:1 mixture and left to stir at room temperature. After 34 h, 20 mL of water was added to the reaction and the aqueous layer was extracted with ether (3 x 30 mL), dried over Na2SO4, filtered, and concentrated under reduced pressure. Column chromatography in 2:1 petroleum ether:DCM afforded 87 as a yellow oil (200 mg, 92%).  Data for 87:  13  C NMR (100 MHz, CDCl3) G 153.3, 138.7, 134.3, 133.3, 124.5, 119.5,  39.5, 39.3, 33.0, 32.9, 31.9, 31.8, 29.9, 29.6, 29.6, 29.3, 26.5, 26.4, 22.7, 22.6, 14.1, 14.0. 1  H NMR (400 MHz, CDCl3) G 10.47 (s, 1H, OH), 7.86 (d, 4JHH = 2.4 Hz, 1H, Ar), 7.38  (dd, 3JHH = 8.8, 4JHH = 2.4 Hz, 1H, Ar), 7.08 (d, 3JHH = 8.8 Hz, 1H, Ar), 2.52 (d, 3JHH = 7.4 Hz, 2H, CH2), 1.59-1.57 (m, 1H, CH), 1.35-1.15 (m, 24H, CH2), 0.91-0.87 (m 6H, CH3). UV-vis (CH2Cl2) λmax (ε) = 285 (5.9 x 103), 377 (6.4 x 103) nm (L mol-1 cm-1). ESIMS (MeOH) m/z = 362.5 [87-H]-. FT-IR (neat): ῡ = 3241, 2922, 2853, 1629, 1584, 1537, 1487, 1465, 1426, 1320, 1245, 1180, 1131, 1079, 823 cm-1. HRMS (ESI) Calc’d for C22H36NO3: 362.2695; Found: 362.2700 (-1.3 ppm).  270  Synthesis of tris(phenyl)(4-hydroxy-3-nitrophenyl)methane (89): 4-Tritylphenol, 88, (2.0 g, 5.94 mmol) was added to a suspension of claycop (3.0g, 500 mg/mmol of reagent to nitrate) in dry ether (100 mL) and acetic anhydride (9 mL) under nitrogen at room temperature. After 4 h stirring, the suspension was quenched with diluted HCl (0.1 mol L1  , 30 mL) and the aqueous phase was extracted with DCM (3 x 30 mL). The combined  organic phases were dried over MgSO4 and concentrated under reduced pressure. Silica gel flash chromatography of the residue (DCM) followed by recrystallization from EtOH gave 89 as a light yellow solid (1.238 g, 3.25 mmol, 55%).  Data for 89:  13  C NMR (75 MHz, CDCl3) G 153.5, 146.5, 145.6, 141.3, 131.1, 130.8,  127.9, 127.5, 126.4, 126.0. 1H NMR (300 MHz, CDCl3) G 10.61 (s, 1H, OH), 8.07 (d, 4  JHH = 2.7 Hz, 1H, Ar CH), 7.41 (dd, 3JHH = 9.0, 4JHH = 2.7 Hz, 1H, Ar CH), 7.33-7.16  (m, 15H, Ar CH), 7.05 (d, 3JHH = 9.0 Hz, 1H, Ar CH). UV-vis (CH2Cl2) λmax (ε) = 292 (7.2 x 103), 377 (3.9 x 103) nm (L mol-1 cm-1). ESI-MS (MeOH) m/z = 380.5 [89-H]-. FTIR (neat) ῡ = 3029, 1739, 1628, 1586, 1537, 1489, 1442, 1418, 1324, 1246, 1201, 1080, 1035, 834 cm-1. Mp = 142 ºC. Anal. Calc’d for C25H19NO3: C, 78.72; H, 5.02; N, 3.67. Found: C, 78.92; H, 5.37; N, 3.68. General synthetic procedure for head-to-tail assembled Pt4 rings (90a-d): A suspension of K2PtCl4 in DMSO was sparged with N2 and then heated to 100 °C until all of the salt dissolved. A separate Schlenk flask was charged with the corresponding NONO imine and K2CO3 and two cycles of evacuation/N2 purging were conducted. The K2PtCl4 solution was transferred via syringe to the flask containing imine and K2CO3 and the mixture was heated to 150-160 °C for 2-4 h. The yellow suspension was cooled to room temperature and 30-60 mL of water was added followed by centrifugation. After decanting the supernatant solution, four cycles of washing, centrifuging, and decanting were performed, once with water and three times with MeOH. Upon drying the precipitate, tetramers 90a-d were isolated as yellow powders.  271  Synthesis of head-to-tail Pt4 ring R=H (90a): Performed with K2PtCl4 (114 mg, 0.28 mmol), K2CO3 (58 mg, 0.42 mmol), N-ONO imine 85a (80 mg, 0.28 mmol), and 10 mL of DMSO. Heated to 160 °C for 4 h and isolated as a pale yellow solid (117 mg, 86%). Data for 90a: MALDI-TOF MS low laser power m/z = 1934 [90a]+, 3868 [90a2]+, 5802 [90a3]+, 7731 [90a4]+, 9662 [90a5]+; high laser power m/z = 1948 [90a]+, 3896 [90a2]+, 5842 [90a3]+, 7785 [90a4]+, 9732 [90a5]+, 11678 [90a6]+, 13620 [90a7]+. FT-IR (neat) ῡ = 3079, 3022, 2916, 1599, 1584, 1479, 1456, 1454, 1378, 1317, 1294, 1269, 1165, 1139, 801, 730, 692, 548 cm-1. Mp = dec. at 380 °C. Anal. Calc’d for 90a: C72H48O8N8Pt4· 2(H2O): C, 43.91; H, 2.66; N, 5.69. Found: C, 43.77; H, 2.60; N, 5.34. Synthesis of head-to-tail Pt4 ring R=2-hexyldecyl (90b): Performed with K2PtCl4 (80.6 mg, 0.19 mmol), K2CO3 (39 mg, 0.29 mmol), N-ONO imine 85b (100 mg, 0.19 mmol), and 15 mL of DMSO. Heated to 155 °C for 3 h and isolated as a bright yellow solid (94 mg, 70%). Data for 90b: MALDI-TOF MS m/z = 1415 [(85bPt)2-4H]+, 2123 [(85bPt)3-6H]+, 2620 [90b-C15H30]+, 2831 [90b]+, 5666 [90b2]+, 8501 [90b3]+, 11331 [90b4]+, 14168 [90b5]+. UV-vis (CH2Cl2) λmax (ε) = 242 (7.0 x 104), 306 (7.3 x 104), 429 (8.4 x 104), 454 (6.8 x 104) nm (L mol-1 cm-1). FT-IR (neat) ῡ = 3078, 3020, 2953, 2919, 2850, 1607, 1589, 1575, 1496, 1488, 1462, 1456, 1443, 1377, 1316, 1292, 1271, 1179, 815, 795, 687, 672, 553 cm-1. Mp = dec. at 380 °C. Anal. Calc’d for 90b: C136H176O8N8Pt4: C, 57.69; H, 6.27; N, 3.96. Found: C, 57.65; H, 6.28; N, 3.93. Synthesis of head-to-tail Pt4 ring R=CPh3 (90c): Performed with K2PtCl4 (117 mg, 0.28 mmol), K2CO3 (58 mg, 0.42 mmol), N-ONO imine 85c (150 mg, 0.28 mmol), and 15 mL of DMSO. Heated to 160 °C for 2 h and isolated as a bright yellow solid (153 mg, 75%).  272  Data for 90c: MALDI-TOF MS m/z = 725 [85cPt -2H]+. 1451 [(85cPt)2-4H]+. 2178 [(85cPt)3-5H]+. 2902 [90c]+, 5800 [90c2]+, 8700 [90c3]+, 11587 [90c4]+, 14489 [90c5]+, 17390 [90c6]+. UV-vis (CH2Cl2) λmax (ε) = 302 (1.4 x 105), 429 (1.0 x 105), 454 (9.4 x 104) nm (L mol-1 cm-1). FT-IR (neat) ῡ = 3054, 3028, 2920, 2851, 1603, 1575, 1487, 1464, 1443, 1296, 1175, 797, 744, 700, 554 cm-1. Mp > 300 °C. Anal. Calc’d for 90c: C148H104O8N8Pt4: C, 61.24; H, 3.61; N, 3.86. Found: C, 61.51; H, 3.44; N, 3.72. Synthesis of head-to-tail Pt4 ring R=C(4-tBuPh)3 (90d): Performed with K2PtCl4 (83.5 mg, 0.20 mmol), K2CO3 (42 mg, 0.30 mmol), N-ONO imine 85d (141 mg, 0.20 mmol), and 15 mL of DMSO. Heated to 155 °C for 3 h and isolated as a bright yellow solid (117 mg, 65%). Data for 90d: Dimerization and chemical shift equivalence prohibited full assignment of the  13  C NMR data.  13  C NMR (151 MHz, CDCl3) δ 166.9, 162.3, 149.2, 148.6, 148.5,  148.4, 148.2, 147.7, 144.3, 144.3, 144.1, 143.2, 137.6, 137.3, 135.6, 134.0, 131.1, 131.0, 129.6, 124.9, 124.5, 124.4, 124.2, 123.6, 123.1, 122.8, 122.3, 120.5, 120.5, 114.3, 114.1, 64.1, 34.6, 34.5, 31.7. Dimer 1H NMR (400 MHz, CDCl3) δ 8.89 (d, 3JHH 6.0 Hz, 1H, Ar), 8.52 (s, 1H, Ar), 8.22 (s, 1H, imine), 7.80 (s, 1H, Ar), 7.63 (m, 1H, Ar), 7.58 (m, 1H, Ar), 7.38 (d, 3JHH = 9.2 Hz, 1H, Ar), 7.30-7.25 (m, 13H, Ar), 7.21 (d, 3JHH = 8.4 Hz, 1H, Ar), 6.63 (m, 1H, Ar), 6.54 (m, 1H, Ar), 1.32 (s, 27H, CH3) ppm. MALDI-TOF MS m/z = 849 [85dPt-2H]+, 1789 [(85dPt)2-3H]+, 3576 [90d]+, 7152 [90d2]+, 10725 [90d3]+, 14297 [90d4]+, 17869 [90d5]+. UV-vis (CH2Cl2) λmax (ε) = 312 (2.1 x 105), 426 (9.8 x 104), 452 (9.0 x 104) nm (L mol-1 cm-1). FT-IR (neat) ῡ = 3084, 3029, 2958, 2903, 2866, 1604, 1575, 1506, 1482, 1463, 1362, 1317, 1269, 1174, 1018, 812, 800, 690, 576, 550 cm-1. Mp > 300 °C. Anal. Calc’d for 90d: C196H200O8N8Pt4: C, 65.83; H, 5.64; N, 3.13. Found: C, 66.16; H, 5.98; N, 2.99.  273  Figure 6.12. 2D COSY NMR spectrum of 90d in CDCl3 (400 MHz).  Figure 6.13. 1H NMR spectroscopic assignment from 2D COSY and ROESY of 90d2 in CDCl3 (400 MHz, 7.27 ppm). 274  6.4.3 Crystallography Crystals of model complex 80a suitable for single-crystal x-ray diffraction were grown from DMSO. X-ray crystal data for 80a: C21H18.5NO3PtS, Mw = 560.02 g mol-1, yellow needle (1.20 x 0.1 x 0.1 mm3), orthorhombic, space group P12121 (#19), a = 5.5909(5), b = 17.6944(17), c = 19.0577(18) Å, D = E = J = 90.000(5)°, V = 1885.3(3) Å3, Z = 4 ρcalcd = 1.973 g cm-3, F000 = 1078, MoKα radiation, λ = 0.71073 Å, T = 173(2) K, 2θmax = 58.81°, 25739 reflections collected, 5149 were unique (Rint = 0.0761). Final GoF = 1.037, R1 = 0.0441, wR2 = 0.1146, R indices based on 4434 reflections with I>2σ(I). The structure was solved by direct methods and the refinement was performed using SHELXL-97.104 All non-hydrogen atoms were refined anisotropically. Hydrogen atoms were included at fixed positions. The phenyl ring including carbons 1-6 was disordered about 2 positions. Each position was modeled with half occupancy. The thermal ellipsoid plot of 80a is shown in Figure 6.14.  Figure 6.14. Thermal ellipsoid plot of 80a top and side view (C = green, N = blue, O = red, Pt = yellow, S = beige, H = white).  275  6.4.4 Computational Details The optimized geometry of 90a was calculated by Angela Crane using the B3LYP level of theory and a split LanL2DZ/6-31G* basis set as implemented by the Gaussian 03 program.105 The LanL2DZ basis set and relativistic core potential were applied to Pt, and the 6-31G* basis set was used for all other atoms.  276  6.5  References  (1)  Drain, C. M.; Varotto, A.; Radivojevic, I. Chem. Rev. 2009, 109, 1630-1658.  (2)  Lehn, J.-M. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 4763-4768.  (3)  Whitesides George, M.; Boncheva, M. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 4769-4774.  (4)  Reinhoudt, D. N.; Crego-Calama, M. Science 2002, 295, 2403-2407.  (5)  Hosseini, M. W. Acc. Chem. Res. 2005, 38, 313-323.  (6)  Lehn, J.-M. Science 2002, 295, 2400-2403.  (7)  Wong, W. W. H.; Ma, C.-Q.; Pisula, W.; Yan, C.; Feng, X.; Jones, D. J.; Mullen, K.; Janssen, R. A. J.; Bauerle, P.; Holmes, A. B. Chem. Mater. 2010, 22, 457-466.  (8)  Klosterman, J. K.; Yamauchi, Y.; Fujita, M. Chem. Soc. Rev. 2009, 38, 17141725.  (9)  Yin, M.; Shen, J.; Pisula, W.; Liang, M.; Zhi, L.; Mullen, K. J. Am. Chem. Soc. 2009, 131, 14618-14619.  (10)  Reiriz, C.; Brea, R. J.; Arranz, R.; Carrascosa, J. L.; Garibotti, A.; Manning, B.; Valpuesta, J. M.; Eritja, R.; Castedo, L.; Granja, J. R. J. Am. Chem. Soc. 2009, 131, 11335-11337.  (11)  Schmaltz, B.; Weil, T.; Mullen, K. Adv. Mater. 2009, 21, 1067-1078.  (12)  van Hameren, R.; van Buul, A. M.; Castriciano, M. A.; Villari, V.; Micali, N.; Schoen, P.; Speller, S.; Scolaro, L. M.; Rowan, A. E.; Elemans, J. A. A. W.; Nolte, R. J. M. Nano Lett. 2008, 8, 253-259.  (13)  Zhao, D.; Moore, J. S. Chem. Commun. 2003, 807-818.  (14)  Zang, L.; Che, Y.; Moore, J. S. Acc. Chem. Res. 2008, 41, 1596-1608.  (15)  Bong, D. T.; Clark, T. D.; Granja, J. R.; Ghadiri, M. R. Angew. Chem., Int. Ed. 2001, 40, 988-1011.  277  (16)  Kozhevnikov, V. N.; Donnio, B.; Bruce, D. W. Angew. Chem., Int. Ed. 2008, 47, 6286-6289.  (17)  Yuen, M.-Y.; Roy, V. A. L.; Lu, W.; Kui, S. C. F.; Tong, G. S. M.; So, M.-H.; Chui, S. S.-Y.; Muccini, M.; Ning, J. Q.; Xu, S. J.; Che, C.-M. Angew. Chem., Int. Ed. 2008, 47, 9895-9899.  (18)  Sun, Y.; Ye, K.; Zhang, H.; Zhang, J.; Zhao, L.; Li, B.; Yang, G.; Yang, B.; Wang, Y.; Lai, S.-W.; Che, C.-M. Angew. Chem., Int. Ed. 2006, 45, 5610-5613.  (19)  Kui, S. C. F.; Chui, S. S.-Y.; Che, C.-M.; Zhu, N. J. Am. Chem. Soc. 2006, 128, 8297-8309.  (20)  Kwok, C.-C.; Ngai, H. M. Y.; Chan, S.-C.; Sham, I. H. T.; Che, C.-M.; Zhu, N. Inorg. Chem. 2005, 44, 4442-4444.  (21)  Lu, W.; Mi, B.-X.; Chan, M. C. W.; Hui, Z.; Che, C.-M.; Zhu, N.; Lee, S.-T. J. Am. Chem. Soc. 2004, 126, 4958-4971.  (22)  Lin, Y.-Y.; Chan, S.-C.; Chan, M. C. W.; Hou, Y.-J.; Zhu, N.; Che, C.-M.; Liu, Y.; Wang, Y. Chem.--Eur. J. 2003, 9, 1263-1272.  (23)  Tam, A. Y.-Y.; Wong, K. M.-C.; Yam, V. W.-W. J. Am. Chem. Soc. 2009, 131, 6253-6260.  (24)  Tam, A. Y.-Y.; Wong, K. M.-C.; Zhu, N.; Wang, G.; Yam, V. W.-W. Langmuir 2009, 25, 8685-8695.  (25)  Chen, Y.; Li, K.; Lu, W.; Chui, S. S.-Y.; Ma, C.-W.; Che, C.-M. Angew. Chem., Int. Ed. 2009, 48, 9909-9913.  (26)  Lu, W.; Chen, Y.; Roy, V. A. L.; Chui, S. S.-Y.; Che, C.-M. Angew. Chem., Int. Ed. 2009, 48, 7621-7625.  (27)  Cardolaccia, T.; Li, Y.; Schanze, K. S. J. Am. Chem. Soc. 2008, 130, 2535-2545.  (28)  Lu, W.; Chui, S. S.-Y.; Ng, K.-M.; Che, C.-M. Angew. Chem., Int. Ed. 2008, 47, 4568-4572.  (29)  Camerel, F.; Ziessel, R.; Donnio, B.; Bourgogne, C.; Guillon, D.; Schmutz, M.; Iacovita, C.; Bucher, J.-P. Angew. Chem., Int. Ed. 2007, 46, 2659-2662.  278  (30)  Tam, A. Y.-Y.; Wong, K. M.-C.; Wang, G.; Yam, V. W.-W. Chem. Commun. 2007, 2028-2030.  (31)  Lee, J.; Ghosh, K.; Stang, P. J. J. Am. Chem. Soc. 2009, 131, 12028-12029.  (32)  Northrop, B. H.; Zheng, Y.-R.; Chi, K.-W.; Stang, P. J. Acc. Chem. Res. 2009, 42, 1554-1563.  (33)  Northrop, B. H.; Yang, H.-B.; Stang, P. J. Chem. Commun. 2008, 5896-5908.  (34)  Stang, P. J.; Cao, D. H. J. Am. Chem. Soc. 1994, 116, 4981-4982.  (35)  Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew. Chem., Int. Ed. 2009, 48, 3418-3438.  (36)  Ghosh, K.; Hu, J.; White, H. S.; Stang, P. J. J. Am. Chem. Soc. 2009, 131, 66956697.  (37)  Yamashita, K.-i.; Kawano, M.; Fujita, M. J. Am. Chem. Soc. 2007, 129, 18501851.  (38)  Yoshizawa, M.; Ono, K.; Kumazawa, K.; Kato, T.; Fujita, M. J. Am. Chem. Soc. 2005, 127, 10800-10801.  (39)  Sun, S.-S.; Stern, C. L.; Nguyen, S.-B. T.; Hupp, J. T. J. Am. Chem. Soc. 2004, 126, 6314-6326.  (40)  Leininger, S.; Olenyuk, B.; Stang, P. J. Chem. Rev. 2000, 100, 853-907.  (41)  Stang, P. J.; Cao, D. H.; Saito, S.; Arif, A. M. J. Am. Chem. Soc. 1995, 117, 62736283.  (42)  Motschi, H.; Nussbaumer, C.; Pregosin, P. S.; Bachechi, F.; Mura, P.; Zambonelli, L. Helv. Chim. Acta 1980, 63, 2071-2186.  (43)  Crowley, J. D.; Steele, I. M.; Bosnich, B. Inorg. Chem. 2005, 44, 2989-2991.  (44)  Crowley, J. D.; Steele, I. M.; Bosnich, B. Eur. J. Inorg. Chem. 2005, 3907-3917.  (45)  Crowley, J. D.; Goshe, A. J.; Bosnich, B. Chem. Commun. 2003, 392-393.  (46)  Goshe, A. J.; Steele, I. M.; Bosnich, B. J. Am. Chem. Soc. 2003, 125, 444-451. 279  (47)  Berggren, G.; Thapper, A.; Huang, P.; Kurz, P.; Eriksson, L.; Styring, S.; Anderlund, M. F. Dalton Trans. 2009, 10044-10054.  (48)  Fernandez, G.; Perez, E. M.; Sanchez, L.; Martin, N. Angew. Chem., Int. Ed. 2008, 47, 1094-1097.  (49)  Asadi, A.; Patrick, B. O.; Perrin, D. M. J. Am. Chem. Soc. 2008, 130, 1286012861.  (50)  Jensen, R. A.; Kelley, R. F.; Lee, S. J.; Wasielewski, M. R.; Hupp, J. T.; Tiede, D. M. Chem. Commun. 2008, 1886-1888.  (51)  Willison, S. A.; Krause, J. A.; Connick, W. B. Inorg. Chem. 2008, 47, 1258-1260.  (52)  Fernandez, J. J.; Fernandez, A.; Vazquez-Garcia, D.; Lopez-Torres, M.; Suarez, A.; Gomez-Blanco, N.; Vila, J. M. Eur. J. Inorg. Chem. 2007, 5408-5418.  (53)  Asadi, A.; Patrick, B. O.; Perrin, D. M. J. Org. Chem. 2007, 72, 466-475.  (54)  Brasey, T.; Scopelliti, R.; Severin, K. Inorg. Chem. 2005, 44, 160-162.  (55)  Fenniri, H.; Mathivanan, P.; Vidale, K. L.; Sherman, D. M.; Hallenga, K.; Wood, K. V.; Stowell, J. G. J. Am. Chem. Soc. 2001, 123, 3854-3855.  (56)  Carina, R. F.; Williams, A. F.; Bernardinelli, G. Inorg. Chem. 2001, 40, 18261832.  (57)  Matsumoto, N.; Motoda, Y.; Matsuo, T.; Nakashima, T.; Re, N.; Dahan, F.; Tuchagues, J.-P. Inorg. Chem. 1999, 38, 1165-1173.  (58)  Stang, P. J.; Whiteford, J. A. Res. Chem. Intermed. 1996, 22, 659-665.  (59)  Kajiwara, T.; Ito, T. J. Chem. Soc., Chem. Commun. 1994, 1773-1774.  (60)  Davis, J. T. Angew. Chem., Int. Ed. 2004, 43, 668-698.  (61)  Mezzina, E.; Mariani, P.; Itri, R.; Masiero, S.; Pieraccini, S.; Spada, G. P.; Spinozzi, F.; Davis, J. T.; Gottarelli, G. Chem.--Eur. J. 2001, 7, 388-395.  (62)  Chuchani, G. J. Chem. Soc. 1959, 1753-1756.  (63)  Chuchani, G. J. Chem. Soc. 1960, 325-326. 280  (64)  Gibson, H. W.; Lee, S. H.; Engen, P. T.; Lecavalier, P.; Sze, J.; Shen, Y. X.; Bheda, M. J. Org. Chem. 1993, 58, 3748-3756.  (65)  Evertsson, H.; Nilsson, S.; Welch, C. J.; Sundeloef, L.-O. Langmuir 1998, 14, 6403-6408.  (66)  Martin, R. B. Chem. Rev. 1996, 96, 3043-3064.  (67)  Frischmann, P. D.; MacLachlan, M. J. Chem. Commun. 2007, 4480-4482.  (68)  Frischmann, P. D.; Facey, G. A.; Ghi, P. Y.; Gallant, A. J.; Bryce, D. L.; Lelj, F.; MacLachlan, M. J. J. Am. Chem. Soc. 2010, 132, 3893-3908.  (69)  Kang, J.; Rebek, J., Jr. Nature 1996, 382, 239-241.  (70)  Cram, D. J.; Choi, H. J.; Bryant, J. A.; Knobler, C. B. J. Am. Chem. Soc. 1992, 114, 7748-7765.  (71)  Yamashita, K.-i.; Sato, K.-i.; Kawano, M.; Fujita, M. New J. Chem. 2009, 33, 264-270.  (72)  Su, C.-Y.; Smith, M. D.; zur Loye, H.-C. Angew. Chem., Int. Ed. 2003, 42, 40854089.  (73)  Beissel, T.; Powers, R. E.; Raymond, K. N. Angew. Chem., Int. Ed. Engl. 1996, 35, 1084-1086.  (74)  Goldin, A. A.; DYNALS v 2.0 Software for particle size distribution analysis in photon correlation spectroscopy.  (75)  Ryu, J.-H.; Tang, L.; Lee, E.; Kim, H.-J.; Lee, M. Chem.--Eur. J. 2008, 14, 871881.  (76)  Ryu, J.-H.; Lee, E.; Lim, Y.-b.; Lee, M. J. Am. Chem. Soc. 2007, 129, 4808-4814.  (77)  Yang, W.-Y.; Lee, E.; Lee, M. J. Am. Chem. Soc. 2006, 128, 3484-3485.  (78)  Kim, B.-S.; Hong, D.-J.; Bae, J.; Lee, M. J. Am. Chem. Soc. 2005, 127, 1633316337.  281  (79)  Rosselli, S.; Ramminger, A.-D.; Wagner, T.; Silier, B.; Wiegand, S.; Haussler, W.; Lieser, G.; Scheumann, V.; Hoger, S. Angew. Chem., Int. Ed. 2001, 40, 31383141.  (80)  Bockstaller, M.; Koehler, W.; Wegner, G.; Vlassopoulos, D.; Fytas, G. Macromolecules 2000, 33, 3951-3953.  (81)  Lortie, F.; Boileau, S.; Bouteiller, L.; Chassenieux, C.; Laupretre, F. Macromolecules 2005, 38, 5283-5287.  (82)  Runge, M. B.; Dutta, S.; Bowden, N. B. Macromolecules 2006, 39, 498-508.  (83)  Schärtl, W. Light Scattering from Polymer Solutions and Nanoparticle Dispersions; Springer-Verlag: Berline, Heidelberg, 2007.  (84)  Dierking, I. Textures of Liquid Crystals; Wiley-VCH Verlag: Weinheim, 2003.  (85)  Saez, I. M.; Goodby, J. W.; Richardson, R. M. Chem.--Eur. J. 2001, 7, 27582764.  (86)  Kato, T.; Yasuda, T.; Kamikawa, Y.; Yoshio, M. Chem. Commun. 2009, 729-739.  (87)  Laschat, S.; Baro, A.; Steinke, N.; Giesselmann, F.; Haegele, C.; Scalia, G.; Judele, R.; Kapatsina, E.; Sauer, S.; Schreivogel, A.; Tosoni, M. Angew. Chem., Int. Ed. 2007, 46, 4832-4887.  (88)  Binnemans, K.; Lodewyckx, K.; Donnio, B.; Guillon, D. Chem.--Eur. J. 2002, 8, 1101-1105.  (89)  Coco, S.; Cordovilla, C.; Donnio, B.; Espinet, P.; Garcia-Casas, M. J.; Guillon, D. Chem.--Eur. J. 2008, 14, 3544-3552.  (90)  Binnemans, K.; Lodewyckx, K.; Cardinaels, T.; Parac-Vogt, T. N.; Bourgogne, C.; Guillon, D.; Donnio, B. Eur. J. Inorg. Chem. 2006, 150-157.  (91)  Cordovilla, C.; Coco, S.; Espinet, P.; Donnio, B. J. Am. Chem. Soc. 2010, 132, 1424-1431.  (92)  Domracheva, N.; Mirea, A.; Schwoerer, M.; Torre-Lorente, L.; Lattermann, G. ChemPhysChem 2006, 7, 2567-2577.  282  (93)  Bilgin-Eran, B.; Tschierske, C.; Diele, S.; Baumeister, U. J. Mater. Chem. 2006, 16, 1136-1144.  (94)  Nesrullajev, A.; Bilgin-Eran, B.; Kazanci, N. Mater. Chem. Phys. 2002, 76, 7-14.  (95)  Nesrullajev, A.; Bilgin-Eran, B.; Singer, D.; Kazanci, N.; Praefcke, K. Mater. Res. Bull. 2002, 37, 2467-2482.  (96)  Donnio, B. Curr. Opin. Colloid Interface Sci. 2002, 7, 371-394.  (97)  Shirley, R.; Crysfire "An Interactive Powder Indexing Support System": 2004.  (98)  Zincke, T.; Wugk, E. Justus Liebigs Ann. Chem. 1909, 363, 284-301.  (99)  Morris, G. A.; Nguyen, S. T. Tetrahedron Lett. 2001, 42, 2093-2096.  (100) Johnson, R. A.; Nidy, E. G.; Aiken, J. W.; Crittenden, N. J.; Gorman, R. R. J. Med. Chem. 1986, 29, 1461-1468. (101) Varala, R.; Nuvula, S.; Adapa, S. R. J. Org. Chem. 2006, 71, 8283-8286. (102) Cahiez, G.; Habiak, V.; Duplais, C.; Moyeux, A. Angew. Chem., Int. Ed. 2007, 46, 4364-4366. (103) Cornelis, A.; Laszlo, P. Synthesis 1985, 909-918. (104) Sheldrick, G. M.; SHELXL-97, University of Göttingen: Göttingen, Germany, 1997. (105) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; 283  Wong, M. W.; Gonzalez, C.; Pople, J. A.; Gaussian 03, Revision D.02 ed.; Gaussian Inc.: Wallingford CT, USA, 2004.  284  Chapter 7 Conclusions and Future Directions 7.1  Conclusions Self-assembly is the synthetic strategy of choice for constructing complex  architectures and is featured throughout this thesis. Chapter one covered the historical and contemporary development of supramolecular chemistry with an emphasis on coordination chemistry and host-guest systems. Coordination cages and cavitands were discussed, and the previously diffuse field of metallocavitands was introduced. The synthesis of metallocavitands was compartmentalized into two general methods, aggregation and macrocycle templation. These host molecules are defined as multimetallic complexes where metal coordination is necessary for cavity formation and are currently broadening the horizon of traditional host-guest chemistry. The endless variety of tunable cavity geometries, high potential for free metal coordination sites accessible to encapsulated guest molecules, and one-pot self-assembly approach are all distinct advantages metallocavitands may possess over traditional organic cavitands. Macrocycle-templated routes to metallocavitands have been limited and previous solidstate studies identified Schiff base macrocycles as prospective metallocavitand templates. A large portion of my research efforts has been focused on the multimetallic coordination chemistry of [3+3] Schiff base macrocycle 51 and the supramolecular chemistry of the isolated heptametallic complexes. Building on my predecessor’s report of a heptazinc complex coordinated to the macrocycle, I first proved macrocycle 51 templates formation of the heptazinc cluster. This was done by isolating tetrazinc intermediates and demonstrating they were reactive toward additional metal salts yielding the heptazinc metallocavitand or mixed heptametallic Zn/Co clusters, identified by mass spectrometry. A solid-state structure of heptazinc metallocavitand, 52d, showed this metallocavitand organizes into capsules with a single DMF molecule located in the cavity. Capsule formation was also observed in solution, and the thermodynamics of  285  dimerization were quantified by non-linear curve-fitting of imine chemical shift data from variable-temperature variable-concentration  1  H NMR spectroscopic experiments.  Interestingly, dimerization of octyloxy substituted metallocavitand, 52e, in aromatic solvents is an entropy-driven event that I attribute to the expulsion of solvent from the monomeric cavity upon dimerization. My interest in developing the host-guest chemistry of heptazinc metallocavitands 52a-e was hindered by the small cavity size so I sought a simple method to increase the cavity volume. Reacting macrocycle 51 with Cd(OAc)2 instead of Zn(OAc)2 yielded heptacadmium metallocavitands, 61a-e, that have roughly 50% larger cavities due to the increased puckering of the macrocycle template that is necessary to coordinate the larger Cd2+ cation. A cursory comparison of the solid-state structures of heptazinc and heptacadmium metallocavitands would suggest they are nearly identical; however, the subtle differences dramatically alter the supramolecular chemistry of 61a-e. In particular a bridging μ4-O vs μ3-OH ligand is found in the heptazinc vs heptacadmium metallocavitands, respectively. It was challenging to prove the existence of the μ3-OH ligand, and I initially reported it as a μ3-O ligand. Convincing evidence for the μ3-OH ligand was provided by a combination of low-temperature 1H NMR spectroscopy, NMR simulation, SCXRD, and DFT calculations. In the solid state, 61 also organizes into dimers with two DMF molecules encapsulated. Within each capsule of 612 are eight Lewis-acidic sites, six Cd2+ ions and two μ3-OH ligands, accessible to Lewis-basic guest molecules. Crystallography revealed encapsulated DMF molecules simultaneously participate in both a hydrogen bond with the μ3-OH ligand and a weak metal-ligand interaction with a Cd2+ ion in a macrocyclic N2O2  pocket.  This  host-guest  hydrogen-bonding/metal-coordinating synergy is  responsible for the exceptionally high packing coefficient (0.80) observed for DMF within capsules of 61. Solution dimerization in DMF-d7 and aromatic solvents was observed for heptacadmium metallocavitands and the thermodynamics of self-assembly were quantified. Dimerization of 61 is entropy-driven and, notably, the rate of dimerization is significantly slower in DMF-d7. Using an unequally populated two-site exchange model, the rates of dimerization and activation parameters were calculated from coalescence  286  temperatures observed by variable-temperature  1  H NMR spectroscopy. Besides  dimerization, acetate exchange between two distinct acetate environments on the cadmium cluster was also observed by NMR spectroscopy. Only in the presence of a Lewis-basic guest molecule (DMF) was an acetate exchange pathway active. Linewidth simulation of the acetate resonances observed over a series of variable-temperature 1H NMR spectra afforded exchange rates and activation parameters. These two events, dimerization and acetate exchange, share the same sign and magnitude of ΔH‡ and ΔS‡. I proposed a common rate-determining step that involves intricate host-guest DMF-cavity interactions. Specifically, scission of the DMF- μ3-OH hydrogen-bond and formation of a DMF-Cd2+ coordination bond was proposed as the rate-determining step. Although this process is likely more complex, my hypothesis was supported by DFT calculations and the gathered kinetic data. Three-fold rotation of a single encapsulated DMF molecule was also observed in the solid-state by magic angle spinning 2D NMR spectroscopy. In an effort to create new macrocyclic metallocavitand templates, I developed methodology to access large diameter [2+2] Schiff base macrocycles 72-74. By relying on the differential exchange rates of aldimines and ketimines, the N2O2 salphen pockets of these [2+2] Schiff base macrocycles are composed of one aldimine and one ketimine, rendering each pocket unsymmetrical. This was a first for Schiff base macrocycles and holds promise for synthesizing metallocavitands with chiral N2O2 coordination environments. A [3+3] macrocycle, 78, with triptycene units built into the backbone was also described, and the three-dimensional nature of this macrocycle prevented alkalimetal induced columnar aggregation. Coordination of alkali metals in the central crown ether-like cavity of sterically unhindered macrocycle 51c yielded supramolecular fibers, and macrocycle 78 proved to be a valuable counter example for proving the mechanism of assembly. While studying heptazinc and heptacadmium metallocavitands I was (and still am) fascinated by the architectures constructed with platinum(II)/palladium(II)-pyridyl bonding reported primarily by Peter Stang and Makoto Fujita. This curiosity, and an interest in creating molecular wires, led me to pursue head-to-tail self-assembled Pt4 rings. A variety of N-ONO imine pro-ligands were conceived and I demonstrated that forming them in situ with K2PtCl4 directs the twelve component self-assembly of  287  individual Pt4 rings. These metallocycles further organize into non-covalent nanotubes that span microns in length both in solution and the solid state as observed by light scattering and transmission electron microscopy, respectively. Columnar assembly of Pt4 rings also drives the adoption of lyotropic columnar nematic and/or hexagonal columnar mesophases in a variety of non-polar solvents.  7.2  Reflection I feel that the goals of this thesis have been achieved. Despite occasionally feeling  the acute frustrations inherent in chemical research, the joy of addressing problems, devising logical strategies for collecting relevant facts, carrying out and simultaneously revising the strategies, and ultimately proving/disproving hypotheses has been exhilarating. I hope that the results presented here represent a respectable contribution of chemical knowledge. Host-guest chemistry has increasingly gained attention and realworld applications in sensing, catalysis, and reactive molecule sequestering now exist thanks to the development of this field. My research efforts have uncovered a reliable macrocycle templating method for developing container molecules with metal sites accessible to encapsulated guest molecules. To popularize this general concept and make the field more approachable to future investigators, I proposed uniting similar examples under the broad subject metallocavitands. Given the interesting applications of synthetic organic cavitands (that have been explored for at least 30 years) I believe the continued pursuit of metallocavitand systems will ultimately yield breakthroughs that benefit humanity. My broad interest in supramolecular chemistry led me to initiate the research presented in the previous chapter on Pt4 rings. Although a variety of molecular squares have been previously reported, the Schiff base platinum-chelating head-to-tail approach is fundamentally different than these examples and allows for the preparation of neutral metallocycles that exhibit columnar self-assembly in solution. I envision a variety of future investigations related to molecular wires, sensing, and light-harvesting functions, particularly if intracolumnar platinum(II)-platinum(II) interactions exist.  288  If I had to express any regrets related to my research it would be the lack of current applications extracted from my efforts. I accept this shortcoming for two reasons: fundamental knowledge is valuable in its own right, and the future holds unanticipated challenges with currently inconceivable solutions where my efforts may yet contribute. Overall, the challenges, failures, and successes I have experienced, and documented in this thesis, have expanded my perspective of the vital role chemistry still must play in solving the health, environmental, and energy problems currently facing society and galvanized my commitment to act as a concerned scientist and citizen of the world. For this I am grateful.  7.3  Future Direction  7.3.1 Metallocavitands Throughout this thesis I have illustrated the advantages metallocavitands may offer over traditional organic cavitands and coordination complexes of organic cavitands. To summarize, metallocavitands are often isolated in one-pot and high yield via selfassembly strategies, frequently have free metal coordination sites accessible to encapsulated guest molecules (enhancing potential for innovative host-guest catalysis and molecular sensing applications), and a near limitless variety of cavity geometries may be accessed, as opposed to the more limited geometries imposed by organic molecules. One interesting avenue of metallocavitand chemistry I think my system may be steered down is to simultaneously act as both a molecular flask and a multimetallic catalyst. To achieve this goal, Schiff base macrocycle templated metallocavitands with catalytic function need to be realized and the cavity volume should be expanded to accommodate a wider variety of substrates. In 2006 Ohshima and coworkers reported the catalytic conversion of esters and βamino alcohols into oxazolines using basic zinc acetate, Zn4(OAc)4O, and the trifluoroacetate analogue, Zn4(OCOCF3)4O, as a catalyst (Scheme 7.1a).1 Optimal yields were achieved with the trifluoroacetate version due to the enhanced Lewis acidity of the  289  cluster. Since that initial report, Zn4(OCOCF3)4O has also been used as a functional group tolerant catalyst for the 99% chemoselective acylation of alcohols in the presence of amines (Scheme 7.1b) and the solvent free transesterification of methyl esters.2,3  Scheme 7.1. Reported reactions catalyzed by Zn4(OCOCF3)6O. a) Oxazoline formation from methylbenzoate and (S)-valinol. b) Chemoselective acylation of an alcohol in the presence of a primary amine. In Chapter two I drew parallels between basic zinc acetate and the tetrazinc acetate cluster stabilized in the center of metallocavitand 52. With this similarity in mind,  290  I propose pursuing heptazinc-hexatrifluoroacetate metallocavitand, 91, shown in Scheme 7.2, and assessing its catalytic usefulness toward the same reactions reportedly catalyzed by Zn4(OCOCF3)6O. If 91 is indeed comparable to Zn4(OCOCF3)6O, I see distant potential for a host-guest responsive catalytic system where in the presence of specific guest molecules, bound inside the cavity, the catalytic function of the peripheral cluster may be tuned, or more dramatically, switched on/off. This system could then act as a spectroscopic sensor for specific guest molecules if non-emissive substrates that yield emissive products are chosen. Many oxazolines are emissive whereas most β-amino alcohols and many esters are not. Simply demonstrating this enzyme-like allosteric response in a synthetic system would be conceptually stimulating.  Scheme 7.2. Proposed synthesis of heptazinc-hexatrifluoroacetate metallocavitand 91. Enhancing the cavity volume of heptazinc and heptacadmium metallocavitands 52 and 61, respectively, would significantly increase the potential for interesting host-guest chemistry by allowing encapsulation of larger guest molecules. The cavity walls of metallocavitands 52 and 61 are composed of the diformyldihydroxybenzene units of the macrocycle  scaffold.  I  propose  expanding  the  aromatic  character  of  the  diformyldihydroxybenzene macrocycle precursor from benzene to triphenylene as outlined in Scheme 7.3. This strategy begins with dilithiating the known compound 2,3dimethoxytriphenylene, 92,4-6 followed by quenching with DMF to give 1,4-diformyl2,3-dimethoxytriphenylene, 93. If direct o-lithiation of 92 is unachievable, bromination at the one and four positions of 2,3-dialkoxytriphenylene derivatives is known to proceed in  291  good yield and I propose synthesizing molecule 94 using similar conditions.7,8 Brominelithium exchange between nBuLi and 94 followed by DMF quenching should also yield 93. Deprotection of the methyl groups with BBr3 will yield 1,4-diformyl-2,3dihydroxytriphenylene.  Scheme 7.3. Proposed synthesis of 1,4-diformyl-2,3-dihydroxytriphenylene, 95. As the functional group arrangement of 95 is identical to the benzene analogue, [3+3] Schiff base macrocycle formation should proceed in a similar fashion when 95 is reacted with 1,2-dialkoxy-4,5-diaminobenzene, 55, yielding macrocycle 96 (Scheme 7.4). Some twisting of 96 may be observed to reduce steric repulsion between the imine and triphenylene protons. Coordination of seven equivalents of cadmium(II) to 96 would yield metallocavitand 97 with a roughly 300% larger cavity (cavity volume estimated to be 300 Å3 for 97 compared to 100 Å3 for 61). A computer model of the envisioned heptacadmium metallocavitand, 97, templated by macrocycle 96 is shown in Figure 7.1. Upon isolation of these metallocavitands, the dimerization potential and scope of guest binding should be evaluated.  292  Scheme 7.4. Proposed synthesis of triphenylene containing macrocycle 96.  Figure 7.1. Computer model of a heptacadmium metallocavitand, 97, templated by macrocycle 96 (R = CH2CH3). a) Side-on view. b) Looking into the cavity (C = green, N = blue, O = red, H = white, Cd = yellow).  293  7.3.2 Pt4 Rings The first report of these supramolecules was focused on the synthesis and general self-assembly properties associated with the system. Now these properties may be harnessed to access new materials with potential applications. Determining whether or not intracolumnar platinum(II)-platinum(II) interactions exist is a top priority for this system. A crystallographic study of any Pt4 ring would provide strong evidence for or against this type of interaction, at least in the solid-state. Shorter alkyl chain derivatives may be synthesized in hopes of promoting crystallization. Variable-temperature variableconcentration (VTVC) UV-vis spectroscopy may also provide information about metalmetal interactions in solution. Many literature reports show that Beer’s Law is not obeyed by a variety of platinum(II)-terpy complexes, providing evidence for columnar assembly.9-13 More importantly, at high concentration the absorption spectra of these same complexes show the emergence of a low-energy metal-metal-to-ligand chargetransfer band that proves intracolumnar platinum(II)-platinum(II) interactions exist.14,15 To conduct a similar experiment with Pt4 rings, a solvent needs to be identified that reduces columnar assembly, allowing for observation of concentration based spectral changes. If this proves impossible, high temperature VCVT UV-vis spectroscopy may sufficiently disaggregate the assemblies, enabling observation of concentration based spectral changes. If Pt4 rings show intracolumnar platinum(II)-platinum(II) interactions, their potential anisotropic conductivity should be measured as observed conductivity is important for developing electronic device applications. Monometallic bipyridyl-phenyl NNC chelated platinum(II) complexes, known to assemble into wire structures, exhibit conductivity as high as 10-2 cm2 V-1 s-1.13 Similar complexes have comprised the active layer of field-effect transistors.16,17 Independent of metal-metal interactions, the central cavity of non-covalent nanotubes assembled from Pt4 rings might be used as a host vessel. If the individual Pt4 rings could be derivatized with ionic functional groups they may be dissolved in water or polar-organic solvents. In such media, hydrophobic guest molecules will be sequestered in the non-polar nanotubes, driven by solvophobic forces. Using commercially available  294  3-amino-4-hydroxybenzenesulfonic acid as the o-aminophenol, Scheme 7.5 depicts the synthesis of tetraanionic metallocycle, 98, accessed via the standard synthetic route for Pt4 rings. Although Coulombic repulsion between metallocycles may decrease the degree of columnar assembly, intermolecular Coulombic attraction to cations may compensate this effect. As the calculated inner diameter of Pt4 rings is 0.7 nm, nanotubes composed of 98 might host luminescent aromatic molecules such as naphthalene or anthracene, potentially altering or quenching the encapsulated guest molecule’s emission spectra. Disaggregation of the nanotubes, induced by an external stimulus, would free the trapped guest molecules that may then be observed spectroscopically. This concept of a stimulus responsive nanotube is the basis for a supramolecular sensor and is depicted in Figure 7.2.  Scheme 7.5. Proposed synthesis of water soluble Pt4 ring 98.  295  Figure 7.2. Proposed supramolecular sensor composed of anthracene “loaded” noncovalent nanotubes. A stimulus that triggers disaggregation of the nanotubes would free encapsulated anthracene, turning on or altering the luminescence. DFT geometry optimization of Pt4 ring 90 reported in Chapter six, showed 90 exists in an S4 symmetric conformation. This deviation from planarity was attributed to steric repulsion between protons of the 5-(3-pyridyl)salicylidene biphenyl unit. To alleviate this repulsion and potentially access planar Pt4 rings, ethynyl spaced pyridyl salicylaldehyde, 99, may be synthesized by Sonogashira coupling of commercially available 5-bromosalicylaldehyde and 3-ethynylpyridine. Under standard Pt4 ring forming conditions, ethynyl spaced Pt4 ring 100 may be isolated as show in Scheme 7.6. If 100 is planar, columnar assemblies will surely exhibit axial platinum(II)platinum(II) interactions and the conductivity of this cycle should be investigated with the goal of producing functional materials. Another experiment that may produce interesting results is the coordination of silver(I) to the ethynyl spacers via an  296  intermolecular π-tweezer effect.18,19 These intermolecular silver(I)-ethynyl bonds would link individual rings within each non-covalent nanotube and may give rise to new properties. In non-polar solvents, addition of excess silver(I) salts to nanotubes of 100 with hydrophobic alkyl chains may yield high concentrations of silver(I) cations inside the nanotubes. Chemical reduction of nanotube sequestered silver(I) with sodium borohydride or sodium citrate may yield sub-nanometer silver(0) wires templated inside the nanotube.20,21  Scheme 7.6. Proposed Synthesis of ethynyl extended Pt4 ring 100.  297  7.4  References  (1)  Ohshima, T.; Iwasaki, T.; Mashima, K. Chem. Commun. 2006, 2711-2713.  (2)  Ohshima, T.; Iwasaki, T.; Maegawa, Y.; Yoshiyama, A.; Mashima, K. J. Am. Chem. Soc. 2008, 130, 2944-2945.  (3)  Iwasaki, T.; Maegawa, Y.; Hayashi, Y.; Ohshima, T.; Mashima, K. J. Org. Chem. 2008, 73, 5147-5150.  (4)  Liu, Z.; Larock, R. C. J. Org. Chem. 2007, 72, 223-232.  (5)  Liu, Z.; Zhang, X.; Larock, R. C. J. Am. Chem. Soc. 2005, 127, 15716-15717.  (6)  Bushby, R. J.; Hardy, C. J. Chem. Soc., Perkin Trans. 1 1986, 721-723.  (7)  Li, Z.; Zhi, L.; Lucas, N. T.; Wang, Z. Tetrahedron 2009, 65, 3417-3424.  (8)  Li, Z.; Lucas, N. T.; Wang, Z.; Zhu, D. J. Org. Chem. 2007, 72, 3917-3920.  (9)  Yam Vivian, W.-W.; Chan Kenneth, H.-Y.; Wong Keith, M.-C.; Zhu, N. Chem.-Eur. J. 2005, 11, 4535-4543.  (10)  Tam, A. Y.-Y.; Wong, K. M.-C.; Wang, G.; Yam, V. W.-W. Chem. Commun. 2007, 2028-2030.  (11)  Tam, A. Y.-Y.; Lam, W. H.; Wong, K. M.-C.; Zhu, N.; Yam, V. W.-W. Chem.-Eur. J. 2008, 14, 4562-4576.  (12)  Sun, Y.; Ye, K.; Zhang, H.; Zhang, J.; Zhao, L.; Li, B.; Yang, G.; Yang, B.; Wang, Y.; Lai, S.-W.; Che, C.-M. Angew. Chem., Int. Ed. 2006, 45, 5610-5613.  (13)  Lu, W.; Roy, V. A. L.; Che, C.-M. Chem. Commun. 2006, 3972-3974.  (14)  Yam, V. W.-W.; Wong, K. M.-C.; Zhu, N. J. Am. Chem. Soc. 2002, 124, 65066507.  298  (15)  Yu, C.; Chan, K. H.-Y.; Wong, K. M.-C.; Yam, V. W.-W. Chem.--Eur. J. 2008, 14, 4577-4584.  (16)  Yuen, M.-Y.; Roy, V. A. L.; Lu, W.; Kui, S. C. F.; Tong, G. S. M.; So, M.-H.; Chui, S. S.-Y.; Muccini, M.; Ning, J. Q.; Xu, S. J.; Che, C.-M. Angew. Chem., Int. Ed. 2008, 47, 9895-9899.  (17)  Lu, W.; Chen, Y.; Roy, V. A. L.; Chui, S. S.-Y.; Che, C.-M. Angew. Chem., Int. Ed. 2009, 48, 7621-7625.  (18)  Mueller, C.; Whiteford, J. A.; Stang, P. J. J. Am. Chem. Soc. 1998, 120, 98279837.  (19)  Whiteford, J. A.; Stang, P. J.; Huang, S. D. Inorg. Chem. 1998, 37, 5595-5601.  (20)  Bois, L.; Chassagneux, F.; Desroches, C.; Battie, Y.; Destouches, N.; Gilon, N.; Parola, S.; Stephan, O. Langmuir 2010, 26, 8729-8736.  (21)  Eisele, D. M.; Berlepsch, H.; Bottcher, C.; Stevenson, K. J.; Vanden Bout, D. A.; Kirstein, S.; Rabe, J. P. J. Am. Chem. Soc. 2010, 132, 2104-2105.  299  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)